{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "196c74c9-07fc-4458-a371-aef9c2cb3f3c",
   "metadata": {},
   "source": [
    "### CNN"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b7613d6c-a8c0-4d73-b3c1-0c3a0683251b",
   "metadata": {},
   "source": [
    "<img src=\"https://media.istockphoto.com/id/1313334189/de/vektor/cartoon-vektor-pilze-clipart-doodle-symbol-set-schwarze-silhouetten-natur-illustration.jpg?s=612x612&w=0&k=20&c=8mNNIIO9V3hMXQdblh4rap5mS1-7ZkHo2b9AWLbzkOY=\" width=750>"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5e2a8764-8cdd-4315-b79e-e4ba97a918eb",
   "metadata": {},
   "source": [
    "### Introduction\n",
    "In this project, a deep learning model was built to classify mushroom images into different classes.\n",
    "The dataset contains images of mushrooms, and the goal is to predict the correct mushroom category for each image.\n",
    "\n",
    "First, a custom Convolutional Neural Network (CNN) was tested. After that, a more powerful Transfer Learning approach was used with ResNet50, a pretrained model trained on ImageNet.\n",
    "Data augmentation techniques such as rotation, zoom, and flipping were applied to improve generalization.\n",
    "The model was trained using sparse categorical cross-entropy and evaluated with accuracy and Top-3 accuracy."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "64941a01-4ba0-401e-97f1-cb1fb110f582",
   "metadata": {},
   "source": [
    "### Import Libraries"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "id": "0209fe55-73f2-4ece-9cc7-c3352ad1b1e5",
   "metadata": {},
   "outputs": [],
   "source": [
    "from tensorflow import keras\n",
    "from tensorflow.keras.preprocessing.image import ImageDataGenerator\n",
    "from tensorflow.keras.models import Sequential\n",
    "from tensorflow.keras.layers import Input, Conv2D, MaxPooling2D, Rescaling, RandomFlip, RandomRotation, RandomZoom, Flatten, Dense, Dropout, BatchNormalization\n",
    "from tensorflow.keras.optimizers import Adam\n",
    "from tensorflow.keras import regularizers\n",
    "from tensorflow.keras.callbacks import EarlyStopping, ModelCheckpoint, ReduceLROnPlateau\n",
    "from tensorflow.keras.layers import GlobalAveragePooling2D\n",
    "from sklearn.model_selection import train_test_split\n",
    "from PIL import Image\n",
    "from pathlib import Path\n",
    "from collections import Counter\n",
    "from tensorflow.keras.models import load_model\n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn.metrics import confusion_matrix, classification_report\n",
    "from sklearn.metrics import ConfusionMatrixDisplay\n",
    "import seaborn as sns\n",
    "import numpy as np\n",
    "import tensorflow as tf\n",
    "import pickle\n",
    "import warnings\n",
    "warnings.filterwarnings(\"ignore\")  "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e641d1c0-d88d-4a00-a0c2-993a30c7df0c",
   "metadata": {},
   "source": [
    "### Load Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "56035867-bb4b-48fb-98a8-f89f703e8390",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "BASE_DIR: /workspace/MushroomClassificationCNN\n",
      "train exists: True\n",
      "test exists : True\n",
      "sub exists  : True\n",
      "img_dir exists: True\n"
     ]
    }
   ],
   "source": [
    "import os\n",
    "import pandas as pd\n",
    "from pathlib import Path\n",
    "\n",
    "BASE_DIR = Path.cwd()  # aktueller Notebook-Ordner\n",
    "TRAIN_CSV = BASE_DIR / \"train.csv\"\n",
    "TEST_CSV  = BASE_DIR / \"test.csv\"\n",
    "SUB_CSV   = BASE_DIR / \"sample_submission.csv\"\n",
    "IMG_DIR   = BASE_DIR / \"dataset\"\n",
    "\n",
    "print(\"BASE_DIR:\", BASE_DIR)\n",
    "print(\"train exists:\", TRAIN_CSV.exists())\n",
    "print(\"test exists :\", TEST_CSV.exists())\n",
    "print(\"sub exists  :\", SUB_CSV.exists())\n",
    "print(\"img_dir exists:\", IMG_DIR.exists())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "c45067f3-9dc7-4b18-ac47-63c31c71b457",
   "metadata": {},
   "outputs": [],
   "source": [
    "train_df=pd.read_csv(TRAIN_CSV)\n",
    "test_df=pd.read_csv(TEST_CSV)\n",
    "sub_df=pd.read_csv(SUB_CSV)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "6daf3ed4-7f7d-431f-9e72-6cd92cb06d5b",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train shape: (2365, 2)\n",
      "Test shape : (598, 1)\n",
      "Sub shape  : (598, 2)\n"
     ]
    }
   ],
   "source": [
    "print('Train shape:', train_df.shape)\n",
    "print('Test shape :', test_df.shape)\n",
    "print('Sub shape  :', sub_df.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "11229145-b0ef-4e22-9fbd-e6e65cc9ce82",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(Index(['Image', 'Mushroom'], dtype='object'),\n",
       " Index(['Image'], dtype='object'),\n",
       " Index(['Id', 'Predicted'], dtype='object'))"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_df.columns, test_df.columns, sub_df.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "750a1bc1-a86b-4ce9-8bea-fb4cdb84936f",
   "metadata": {},
   "outputs": [],
   "source": [
    "train_df['filepath'] = train_df['Image'].astype(int).astype(str).str.zfill(5).radd('dataset/') + '.jpg'\n",
    "test_df['filepath']  = test_df['Image'].astype(int).astype(str).str.zfill(5).radd('dataset/') + '.jpg'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "b0e13c00-1191-4f4a-b1bf-5fb928a8bfaf",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Missing train: 0\n",
      "Missing test : 0\n"
     ]
    }
   ],
   "source": [
    "print('Missing train:', (~train_df['filepath'].map(os.path.exists)).sum())\n",
    "print('Missing test :', (~test_df['filepath'].map(os.path.exists)).sum())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "900a34f8-ff1d-4e71-aa7c-565a2debdd4d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(0    0\n",
       " 1    0\n",
       " 2    0\n",
       " 3    0\n",
       " 4    0\n",
       " Name: Mushroom, dtype: int64,\n",
       " dtype('int64'))"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_df['Mushroom'].head(), train_df['Mushroom'].dtype"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "dfb3497a-2779-41c9-8003-fd579d3797f4",
   "metadata": {},
   "source": [
    "### EDA"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "921fbf9e-0cfd-42d6-9aee-ef916a5b0b05",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAjsAAAHCCAYAAAAJowgXAAAAOnRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjEwLjcsIGh0dHBzOi8vbWF0cGxvdGxpYi5vcmcvTLEjVAAAAAlwSFlzAAAPYQAAD2EBqD+naQAAMspJREFUeJzt3XtYVVX+x/HPARGQqyjXSVHMC5ipeQsxL0mimUXRlI2Z9pQ2BlY6aWNjoY6lY/eUtJrSmrLrlE1WmvfGQkdxyrujaGIhaDpAmqDC+v3R4/l1EhURPIc179fz7Odh77X23t+FKB/3XnsfhzHGCAAAwFJe7i4AAACgNhF2AACA1Qg7AADAaoQdAABgNcIOAACwGmEHAABYjbADAACsRtgBAABWI+wAAACrEXaA/1HNmjXT8OHD3V1GlUyaNEkOh8Nl28Wq/9tvv5XD4dC8efOc24YPH67AwMBaP/cpDodDkyZNumjnA2xD2AEsk5ubq3vuuUdxcXHy8/NTcHCwkpKS9Nxzz+nYsWPuLs+tPv30U48NDZ5cG1DX1XN3AQBqzieffKLf/va38vX11R133KHLLrtMx48f1+rVqzVu3Dht2bJFL730krvLrBE7duyQl9f5/X/t008/VVZW1nmFitjYWB07dkw+Pj7nWeH5OVttx44dU716/HMNVBd/ewBL7NmzR4MHD1ZsbKyWL1+u6OhoZ1t6erp27dqlTz75xI0V1ixfX99aPf7JkydVUVGh+vXry8/Pr1bPdS7uPj9Q13EbC7DEjBkzdOTIEb3yyisuQeeUSy+9VPfff/8Z9z98+LAefPBBtWvXToGBgQoODtaAAQP0zTffnNZ35syZatu2rRo0aKCGDRuqc+fOmj9/vrP9xx9/1AMPPKBmzZrJ19dXERERuuaaa7Rhw4ZzjmP16tXq0qWL/Pz81KJFC7344ouV9vv1nJ0TJ05o8uTJatmypfz8/NSoUSP16NFDS5YskfTzPJusrCxJP8+BObVI/z8v58knn9Szzz6rFi1ayNfXV1u3bq10zs4pu3fvVkpKigICAhQTE6MpU6bIGONsX7lypRwOh1auXOmy36+PebbaTm379RWff//73xowYICCg4MVGBiovn37as2aNS595s2bJ4fDoS+//FJjx45VeHi4AgICdOONN+rgwYOV/wEAFuLKDmCJjz/+WHFxcerevXu19t+9e7cWLFig3/72t2revLkKCwv14osvqlevXtq6datiYmIkSS+//LLuu+8+3Xzzzbr//vtVWlqqjRs3au3atfrd734nSfr973+v999/XxkZGUpISNChQ4e0evVqbdu2TVdcccUZa9i0aZP69eun8PBwTZo0SSdPnlRmZqYiIyPPWf+kSZM0bdo03X333eratatKSkq0fv16bdiwQddcc43uuece5efna8mSJfrb3/5W6THmzp2r0tJSjRw5Ur6+vgoLC1NFRUWlfcvLy9W/f39deeWVmjFjhhYtWqTMzEydPHlSU6ZMOWe9v1SV2n5py5YtuuqqqxQcHKzx48fLx8dHL774onr37q1Vq1apW7duLv1Hjx6thg0bKjMzU99++62effZZZWRk6J133jmvOoE6ywCo84qLi40kc8MNN1R5n9jYWDNs2DDnemlpqSkvL3fps2fPHuPr62umTJni3HbDDTeYtm3bnvXYISEhJj09vcq1nJKammr8/PzM3r17ndu2bt1qvL29za//ufp1/e3btzcDBw486/HT09NPO44xP49TkgkODjYHDhyotG3u3LnObcOGDTOSzOjRo53bKioqzMCBA039+vXNwYMHjTHGrFixwkgyK1asOOcxz1SbMcZIMpmZmc711NRUU79+fZObm+vclp+fb4KCgkzPnj2d2+bOnWskmeTkZFNRUeHcPmbMGOPt7W2KiooqPR9gG25jARYoKSmRJAUFBVX7GL6+vs4Jv+Xl5Tp06JACAwPVunVrl9tPoaGh+u6777Ru3bozHis0NFRr165Vfn5+lc9fXl6uxYsXKzU1VU2bNnVuj4+PV0pKyjn3Dw0N1ZYtW7Rz584qn/PX0tLSFB4eXuX+GRkZzq8dDocyMjJ0/PhxLV26tNo1nEt5ebk+//xzpaamKi4uzrk9Ojpav/vd77R69Wrnz8MpI0eOdLktdtVVV6m8vFx79+6ttToBT0LYASwQHBws6ee5MtVVUVGhZ555Ri1btpSvr68aN26s8PBwbdy4UcXFxc5+Dz30kAIDA9W1a1e1bNlS6enp+vLLL12ONWPGDG3evFlNmjRR165dNWnSJO3evfus5z948KCOHTumli1bntbWunXrc9Y/ZcoUFRUVqVWrVmrXrp3GjRunjRs3VnH0P2vevHmV+3p5ebmEDUlq1aqVpJ/n5NSWgwcP6qeffqr0exIfH6+Kigrt27fPZfsvw6MkNWzYUJL03//+t9bqBDwJYQewQHBwsGJiYrR58+ZqH+Pxxx/X2LFj1bNnT73xxhtavHixlixZorZt27rMW4mPj9eOHTv09ttvq0ePHvr73/+uHj16KDMz09nnlltu0e7duzVz5kzFxMToiSeeUNu2bfXZZ59d0DjPpmfPnsrNzdWrr76qyy67TH/96191xRVX6K9//WuVj+Hv71+jNf36RYinlJeX1+h5zsXb27vS7eYXk6kBmxF2AEtcd911ys3NVXZ2drX2f//999WnTx+98sorGjx4sPr166fk5GQVFRWd1jcgIEC33nqr5s6dq7y8PA0cOFCPPfaYSktLnX2io6N17733asGCBdqzZ48aNWqkxx577IznDw8Pl7+/f6W3oXbs2FGlMYSFhenOO+/UW2+9pX379unyyy93eYrpTOGjOioqKk67WvWf//xH0s9Pikn/fwXl19/Dym4fVbW28PBwNWjQoNLvyfbt2+Xl5aUmTZpU6VjA/wrCDmCJ8ePHKyAgQHfffbcKCwtPa8/NzdVzzz13xv29vb1P+5/+e++9p++//95l26FDh1zW69evr4SEBBljdOLECZWXl7vc9pKkiIgIxcTEqKys7KznT0lJ0YIFC5SXl+fcvm3bNi1evPiM+52prsDAQF166aUu5wwICJB0eviorlmzZjm/NsZo1qxZ8vHxUd++fSX9/EJCb29vffHFFy77vfDCC6cdq6q1eXt7q1+/fvroo49cbpcVFhZq/vz56tGjh/O2JoCf8eg5YIkWLVpo/vz5uvXWWxUfH+/yBuWvvvpK77333lk/S+q6667TlClTdOedd6p79+7atGmT3nzzzdPmpfTr109RUVFKSkpSZGSktm3bplmzZmngwIEKCgpSUVGRLrnkEt18881q3769AgMDtXTpUq1bt05PPfXUWccwefJkLVq0SFdddZXuvfdenTx50vlOn3PNv0lISFDv3r3VqVMnhYWFaf369c7H30/p1KmTJOm+++5TSkqKvL29NXjw4HN8Zyvn5+enRYsWadiwYerWrZs+++wzffLJJ3r44Yedk5xDQkL029/+VjNnzpTD4VCLFi20cOFCHThw4LTjnU9tU6dO1ZIlS9SjRw/de++9qlevnl588UWVlZVpxowZ1RoPYDX3PgwGoKb95z//MSNGjDDNmjUz9evXN0FBQSYpKcnMnDnTlJaWOvtV9uj5H/7wBxMdHW38/f1NUlKSyc7ONr169TK9evVy9nvxxRdNz549TaNGjYyvr69p0aKFGTdunCkuLjbGGFNWVmbGjRtn2rdvb4KCgkxAQIBp3769eeGFF6pU/6pVq0ynTp1M/fr1TVxcnJkzZ47JzMw856PnU6dONV27djWhoaHG39/ftGnTxjz22GPm+PHjzj4nT540o0ePNuHh4cbhcDiPeepR8CeeeOK0es706HlAQIDJzc01/fr1Mw0aNDCRkZEmMzPztMf3Dx48aNLS0kyDBg1Mw4YNzT333GM2b9582jHPVJsxpz96bowxGzZsMCkpKSYwMNA0aNDA9OnTx3z11VcufU49er5u3TqX7Wd6JB6wlcMYZqgBAAB7MWcHAABYjbADAACsRtgBAABWI+wAAACrEXYAAIDVCDsAAMBqbn2p4LRp0/TBBx9o+/bt8vf3V/fu3fWXv/zF5QPuevfurVWrVrnsd88992jOnDnO9by8PI0aNUorVqxQYGCghg0bpmnTpqlevaoNr6KiQvn5+QoKCqrR18kDAIDaY4zRjz/+qJiYGHl5nfn6jVvDzqpVq5Senq4uXbro5MmTevjhh9WvXz9t3brV+ep0SRoxYoSmTJniXG/QoIHz6/Lycg0cOFBRUVH66quvtH//ft1xxx3y8fHR448/XqU68vPz+SwZAADqqH379umSSy45Y7tHvVTw4MGDioiI0KpVq9SzZ09JP1/Z6dChg5599tlK9/nss8903XXXKT8/X5GRkZKkOXPm6KGHHtLBgwdVv379c563uLhYoaGh2rdvH58pAwBAHVFSUqImTZqoqKhIISEhZ+znUZ+NderDA8PCwly2v/nmm3rjjTcUFRWlQYMG6ZFHHnFe3cnOzla7du2cQUeSUlJSNGrUKG3ZskUdO3Y87TxlZWUuHw74448/SpKCg4MJOwAA1DHnmoLiMWGnoqJCDzzwgJKSknTZZZc5t//ud79TbGysYmJitHHjRj300EPasWOHPvjgA0lSQUGBS9CR5FwvKCio9FzTpk3T5MmTa2kkAADAk3hM2ElPT9fmzZu1evVql+0jR450ft2uXTtFR0erb9++ys3NVYsWLap1rgkTJmjs2LHO9VOXwQAAgH084tHzjIwMLVy4UCtWrDjrBCNJ6tatmyRp165dkqSoqCgVFha69Dm1HhUVVekxfH19nbesuHUFAIDd3Bp2jDHKyMjQhx9+qOXLl6t58+bn3Ofrr7+WJEVHR0uSEhMTtWnTJh04cMDZZ8mSJQoODlZCQkKt1A0AAOoOt97GSk9P1/z58/XRRx8pKCjIOccmJCRE/v7+ys3N1fz583XttdeqUaNG2rhxo8aMGaOePXvq8ssvlyT169dPCQkJGjp0qGbMmKGCggJNnDhR6enp8vX1defwAACAB3Dro+dnmj09d+5cDR8+XPv27dPtt9+uzZs36+jRo2rSpIluvPFGTZw40eXW0969ezVq1CitXLlSAQEBGjZsmKZPn17llwqWlJQoJCRExcXF3NICAKCOqOrvb496z467EHYAAKh7qvr72yMmKAMAANQWwg4AALAaYQcAAFiNsAMAAKxG2AEAAFYj7AAAAKsRdgAAgNUIOwAAwGoe86nncI9mf/ykVo//7fSBtXp8qfbHIF2cccBz8DPlOWz5s7BhHHV5DFzZAQAAViPsAAAAq3Ebq5rq8uU8AP8bbLhNDdQEruwAAACrEXYAAIDVCDsAAMBqhB0AAGA1wg4AALAaYQcAAFiNsAMAAKxG2AEAAFYj7AAAAKsRdgAAgNUIOwAAwGqEHQAAYDXCDgAAsBphBwAAWI2wAwAArFbP3QUAsEezP35S6+f4dvrAWj8HALtwZQcAAFiNsAMAAKxG2AEAAFYj7AAAAKsRdgAAgNUIOwAAwGo8eg54CB7bBoDawZUdAABgNcIOAACwGmEHAABYjbADAACsRtgBAABWI+wAAACrEXYAAIDVCDsAAMBqhB0AAGA1wg4AALAaYQcAAFiNsAMAAKxG2AEAAFYj7AAAAKsRdgAAgNUIOwAAwGqEHQAAYDXCDgAAsBphBwAAWI2wAwAArEbYAQAAViPsAAAAqxF2AACA1Qg7AADAaoQdAABgNcIOAACwGmEHAABYza1hZ9q0aerSpYuCgoIUERGh1NRU7dixw6VPaWmp0tPT1ahRIwUGBiotLU2FhYUuffLy8jRw4EA1aNBAERERGjdunE6ePHkxhwIAADyUW8POqlWrlJ6erjVr1mjJkiU6ceKE+vXrp6NHjzr7jBkzRh9//LHee+89rVq1Svn5+brpppuc7eXl5Ro4cKCOHz+ur776Sq+99prmzZunRx991B1DAgAAHqaeO0++aNEil/V58+YpIiJCOTk56tmzp4qLi/XKK69o/vz5uvrqqyVJc+fOVXx8vNasWaMrr7xSn3/+ubZu3aqlS5cqMjJSHTp00J///Gc99NBDmjRpkurXr++OoQEAAA/hUXN2iouLJUlhYWGSpJycHJ04cULJycnOPm3atFHTpk2VnZ0tScrOzla7du0UGRnp7JOSkqKSkhJt2bLlIlYPAAA8kVuv7PxSRUWFHnjgASUlJemyyy6TJBUUFKh+/foKDQ116RsZGamCggJnn18GnVPtp9oqU1ZWprKyMud6SUlJTQ0DAAB4GI+5spOenq7Nmzfr7bffrvVzTZs2TSEhIc6lSZMmtX5OAADgHh4RdjIyMrRw4UKtWLFCl1xyiXN7VFSUjh8/rqKiIpf+hYWFioqKcvb59dNZp9ZP9fm1CRMmqLi42Lns27evBkcDAAA8iVvDjjFGGRkZ+vDDD7V8+XI1b97cpb1Tp07y8fHRsmXLnNt27NihvLw8JSYmSpISExO1adMmHThwwNlnyZIlCg4OVkJCQqXn9fX1VXBwsMsCAADs5NY5O+np6Zo/f74++ugjBQUFOefYhISEyN/fXyEhIbrrrrs0duxYhYWFKTg4WKNHj1ZiYqKuvPJKSVK/fv2UkJCgoUOHasaMGSooKNDEiROVnp4uX19fdw4PAAB4ALeGndmzZ0uSevfu7bJ97ty5Gj58uCTpmWeekZeXl9LS0lRWVqaUlBS98MILzr7e3t5auHChRo0apcTERAUEBGjYsGGaMmXKxRoGAADwYG4NO8aYc/bx8/NTVlaWsrKyztgnNjZWn376aU2WBgAALOERE5QBAABqC2EHAABYjbADAACsRtgBAABWI+wAAACrEXYAAIDVCDsAAMBqhB0AAGA1wg4AALAaYQcAAFiNsAMAAKxG2AEAAFYj7AAAAKsRdgAAgNUIOwAAwGqEHQAAYDXCDgAAsBphBwAAWI2wAwAArEbYAQAAViPsAAAAqxF2AACA1Qg7AADAaoQdAABgNcIOAACwGmEHAABYjbADAACsRtgBAABWI+wAAACrEXYAAIDVCDsAAMBqhB0AAGA1wg4AALAaYQcAAFiNsAMAAKxG2AEAAFYj7AAAAKsRdgAAgNUIOwAAwGqEHQAAYDXCDgAAsBphBwAAWI2wAwAArEbYAQAAViPsAAAAqxF2AACA1Qg7AADAaoQdAABgNcIOAACwGmEHAABYjbADAACsRtgBAABWI+wAAACrEXYAAIDVCDsAAMBqhB0AAGA1wg4AALAaYQcAAFiNsAMAAKxG2AEAAFYj7AAAAKsRdgAAgNXcGna++OILDRo0SDExMXI4HFqwYIFL+/Dhw+VwOFyW/v37u/Q5fPiwhgwZouDgYIWGhuquu+7SkSNHLuIoAACAJ3Nr2Dl69Kjat2+vrKysM/bp37+/9u/f71zeeustl/YhQ4Zoy5YtWrJkiRYuXKgvvvhCI0eOrO3SAQBAHVHPnScfMGCABgwYcNY+vr6+ioqKqrRt27ZtWrRokdatW6fOnTtLkmbOnKlrr71WTz75pGJiYmq8ZgAAULd4/JydlStXKiIiQq1bt9aoUaN06NAhZ1t2drZCQ0OdQUeSkpOT5eXlpbVr17qjXAAA4GHcemXnXPr376+bbrpJzZs3V25urh5++GENGDBA2dnZ8vb2VkFBgSIiIlz2qVevnsLCwlRQUHDG45aVlamsrMy5XlJSUmtjAAAA7uXRYWfw4MHOr9u1a6fLL79cLVq00MqVK9W3b99qH3fatGmaPHlyTZQIAAA8nMffxvqluLg4NW7cWLt27ZIkRUVF6cCBAy59Tp48qcOHD59xno8kTZgwQcXFxc5l3759tVo3AABwnzoVdr777jsdOnRI0dHRkqTExEQVFRUpJyfH2Wf58uWqqKhQt27dzngcX19fBQcHuywAAMBObr2NdeTIEedVGknas2ePvv76a4WFhSksLEyTJ09WWlqaoqKilJubq/Hjx+vSSy9VSkqKJCk+Pl79+/fXiBEjNGfOHJ04cUIZGRkaPHgwT2IBAABJbr6ys379enXs2FEdO3aUJI0dO1YdO3bUo48+Km9vb23cuFHXX3+9WrVqpbvuukudOnXSP//5T/n6+jqP8eabb6pNmzbq27evrr32WvXo0UMvvfSSu4YEAAA8jFuv7PTu3VvGmDO2L168+JzHCAsL0/z582uyLAAAYJE6NWcHAADgfBF2AACA1Qg7AADAaoQdAABgNcIOAACwGmEHAABYjbADAACsRtgBAABWI+wAAACrEXYAAIDVCDsAAMBq1Qo7cXFxOnTo0Gnbi4qKFBcXd8FFAQAA1JRqhZ1vv/1W5eXlp20vKyvT999/f8FFAQAA1JTz+tTzf/zjH86vFy9erJCQEOd6eXm5li1bpmbNmtVYcQAAABfqvMJOamqqJMnhcGjYsGEubT4+PmrWrJmeeuqpGisOAADgQp1X2KmoqJAkNW/eXOvWrVPjxo1rpSgAAICacl5h55Q9e/bUdB0AAAC1olphR5KWLVumZcuW6cCBA84rPqe8+uqrF1wYAABATahW2Jk8ebKmTJmizp07Kzo6Wg6Ho6brAgAAqBHVCjtz5szRvHnzNHTo0JquBwAAoEZV6z07x48fV/fu3Wu6FgAAgBpXrbBz9913a/78+TVdCwAAQI2r1m2s0tJSvfTSS1q6dKkuv/xy+fj4uLQ//fTTNVIcAADAhapW2Nm4caM6dOggSdq8ebNLG5OVAQCAJ6lW2FmxYkVN1wEAAFArqjVnBwAAoK6o1pWdPn36nPV21fLly6tdEAAAQE2qVtg5NV/nlBMnTujrr7/W5s2bT/uAUAAAAHeqVth55plnKt0+adIkHTly5IIKAgAAqEk1Omfn9ttv53OxAACAR6nRsJOdnS0/P7+aPCQAAMAFqdZtrJtuusll3Rij/fv3a/369XrkkUdqpDAAAICaUK2wExIS4rLu5eWl1q1ba8qUKerXr1+NFAYAAFATqhV25s6dW9N1AAAA1IpqhZ1TcnJytG3bNklS27Zt1bFjxxopCgAAoKZUK+wcOHBAgwcP1sqVKxUaGipJKioqUp8+ffT2228rPDy8JmsEAACotmo9jTV69Gj9+OOP2rJliw4fPqzDhw9r8+bNKikp0X333VfTNQIAAFRbta7sLFq0SEuXLlV8fLxzW0JCgrKyspigDAAAPEq1ruxUVFTIx8fntO0+Pj6qqKi44KIAAABqSrXCztVXX637779f+fn5zm3ff/+9xowZo759+9ZYcQAAABeqWmFn1qxZKikpUbNmzdSiRQu1aNFCzZs3V0lJiWbOnFnTNQIAAFRbtebsNGnSRBs2bNDSpUu1fft2SVJ8fLySk5NrtDgAAIALdV5XdpYvX66EhASVlJTI4XDommuu0ejRozV69Gh16dJFbdu21T//+c/aqhUAAOC8nVfYefbZZzVixAgFBwef1hYSEqJ77rlHTz/9dI0VBwAAcKHOK+x888036t+//xnb+/Xrp5ycnAsuCgAAoKacV9gpLCys9JHzU+rVq6eDBw9ecFEAAAA15bzCzm9+8xtt3rz5jO0bN25UdHT0BRcFAABQU84r7Fx77bV65JFHVFpaelrbsWPHlJmZqeuuu67GigMAALhQ5/Xo+cSJE/XBBx+oVatWysjIUOvWrSVJ27dvV1ZWlsrLy/WnP/2pVgoFAACojvMKO5GRkfrqq680atQoTZgwQcYYSZLD4VBKSoqysrIUGRlZK4UCAABUx3m/VDA2Nlaffvqp/vvf/2rXrl0yxqhly5Zq2LBhbdQHAABwQar1BmVJatiwobp06VKTtQAAANS4an02FgAAQF1B2AEAAFYj7AAAAKsRdgAAgNUIOwAAwGqEHQAAYDXCDgAAsBphBwAAWI2wAwAArEbYAQAAViPsAAAAq7k17HzxxRcaNGiQYmJi5HA4tGDBApd2Y4weffRRRUdHy9/fX8nJydq5c6dLn8OHD2vIkCEKDg5WaGio7rrrLh05cuQijgIAAHgyt4ado0ePqn379srKyqq0fcaMGXr++ec1Z84crV27VgEBAUpJSVFpaamzz5AhQ7RlyxYtWbJECxcu1BdffKGRI0derCEAAAAPV+1PPa8JAwYM0IABAyptM8bo2Wef1cSJE3XDDTdIkl5//XVFRkZqwYIFGjx4sLZt26ZFixZp3bp16ty5syRp5syZuvbaa/Xkk08qJibmoo0FAAB4Jo+ds7Nnzx4VFBQoOTnZuS0kJETdunVTdna2JCk7O1uhoaHOoCNJycnJ8vLy0tq1a8947LKyMpWUlLgsAADATh4bdgoKCiRJkZGRLtsjIyOdbQUFBYqIiHBpr1evnsLCwpx9KjNt2jSFhIQ4lyZNmtRw9QAAwFN4bNipTRMmTFBxcbFz2bdvn7tLAgAAtcRjw05UVJQkqbCw0GV7YWGhsy0qKkoHDhxwaT958qQOHz7s7FMZX19fBQcHuywAAMBOHht2mjdvrqioKC1btsy5raSkRGvXrlViYqIkKTExUUVFRcrJyXH2Wb58uSoqKtStW7eLXjMAAPA8bn0a68iRI9q1a5dzfc+ePfr6668VFhampk2b6oEHHtDUqVPVsmVLNW/eXI888ohiYmKUmpoqSYqPj1f//v01YsQIzZkzRydOnFBGRoYGDx7Mk1gAAECSm8PO+vXr1adPH+f62LFjJUnDhg3TvHnzNH78eB09elQjR45UUVGRevTooUWLFsnPz8+5z5tvvqmMjAz17dtXXl5eSktL0/PPP3/RxwIAADyTW8NO7969ZYw5Y7vD4dCUKVM0ZcqUM/YJCwvT/Pnza6M8AABgAY+dswMAAFATCDsAAMBqhB0AAGA1wg4AALAaYQcAAFiNsAMAAKxG2AEAAFYj7AAAAKsRdgAAgNUIOwAAwGqEHQAAYDXCDgAAsBphBwAAWI2wAwAArEbYAQAAViPsAAAAqxF2AACA1Qg7AADAaoQdAABgNcIOAACwGmEHAABYjbADAACsRtgBAABWI+wAAACrEXYAAIDVCDsAAMBqhB0AAGA1wg4AALAaYQcAAFiNsAMAAKxG2AEAAFYj7AAAAKsRdgAAgNUIOwAAwGqEHQAAYDXCDgAAsBphBwAAWI2wAwAArEbYAQAAViPsAAAAqxF2AACA1Qg7AADAaoQdAABgNcIOAACwGmEHAABYjbADAACsRtgBAABWI+wAAACrEXYAAIDVCDsAAMBqhB0AAGA1wg4AALAaYQcAAFiNsAMAAKxG2AEAAFYj7AAAAKsRdgAAgNUIOwAAwGqEHQAAYDXCDgAAsBphBwAAWI2wAwAArObRYWfSpElyOBwuS5s2bZztpaWlSk9PV6NGjRQYGKi0tDQVFha6sWIAAOBpPDrsSFLbtm21f/9+57J69Wpn25gxY/Txxx/rvffe06pVq5Sfn6+bbrrJjdUCAABPU8/dBZxLvXr1FBUVddr24uJivfLKK5o/f76uvvpqSdLcuXMVHx+vNWvW6Morr7zYpQIAAA/k8Vd2du7cqZiYGMXFxWnIkCHKy8uTJOXk5OjEiRNKTk529m3Tpo2aNm2q7Oxsd5ULAAA8jEdf2enWrZvmzZun1q1ba//+/Zo8ebKuuuoqbd68WQUFBapfv75CQ0Nd9omMjFRBQcFZj1tWVqaysjLneklJSW2UDwAAPIBHh50BAwY4v7788svVrVs3xcbG6t1335W/v3+1jztt2jRNnjy5JkoEAAAezuNvY/1SaGioWrVqpV27dikqKkrHjx9XUVGRS5/CwsJK5/j80oQJE1RcXOxc9u3bV4tVAwAAd6pTYefIkSPKzc1VdHS0OnXqJB8fHy1btszZvmPHDuXl5SkxMfGsx/H19VVwcLDLAgAA7OTRt7EefPBBDRo0SLGxscrPz1dmZqa8vb112223KSQkRHfddZfGjh2rsLAwBQcHa/To0UpMTORJLAAA4OTRYee7777TbbfdpkOHDik8PFw9evTQmjVrFB4eLkl65pln5OXlpbS0NJWVlSklJUUvvPCCm6sGAACexKPDzttvv33Wdj8/P2VlZSkrK+siVQQAAOqaOjVnBwAA4HwRdgAAgNUIOwAAwGqEHQAAYDXCDgAAsBphBwAAWI2wAwAArEbYAQAAViPsAAAAqxF2AACA1Qg7AADAaoQdAABgNcIOAACwGmEHAABYjbADAACsRtgBAABWI+wAAACrEXYAAIDVCDsAAMBqhB0AAGA1wg4AALAaYQcAAFiNsAMAAKxG2AEAAFYj7AAAAKsRdgAAgNUIOwAAwGqEHQAAYDXCDgAAsBphBwAAWI2wAwAArEbYAQAAViPsAAAAqxF2AACA1Qg7AADAaoQdAABgNcIOAACwGmEHAABYjbADAACsRtgBAABWI+wAAACrEXYAAIDVCDsAAMBqhB0AAGA1wg4AALAaYQcAAFiNsAMAAKxG2AEAAFYj7AAAAKsRdgAAgNUIOwAAwGqEHQAAYDXCDgAAsBphBwAAWI2wAwAArEbYAQAAViPsAAAAqxF2AACA1Qg7AADAaoQdAABgNcIOAACwGmEHAABYzZqwk5WVpWbNmsnPz0/dunXTv/71L3eXBAAAPIAVYeedd97R2LFjlZmZqQ0bNqh9+/ZKSUnRgQMH3F0aAABwMyvCztNPP60RI0bozjvvVEJCgubMmaMGDRro1VdfdXdpAADAzep82Dl+/LhycnKUnJzs3Obl5aXk5GRlZ2e7sTIAAOAJ6rm7gAv1ww8/qLy8XJGRkS7bIyMjtX379kr3KSsrU1lZmXO9uLhYklRSUlLl81aU/VSNas/P+dRTXbU9DhvGIDGOqrJhDBLjqCobxiAxjqryxDGc6m+MOXtHU8d9//33RpL56quvXLaPGzfOdO3atdJ9MjMzjSQWFhYWFhYWC5Z9+/adNSvU+Ss7jRs3lre3twoLC122FxYWKioqqtJ9JkyYoLFjxzrXKyoqdPjwYTVq1EgOh6PGaywpKVGTJk20b98+BQcH1/jxLxbG4TlsGINkxzhsGIPEODyJDWOQLs44jDH68ccfFRMTc9Z+dT7s1K9fX506ddKyZcuUmpoq6efwsmzZMmVkZFS6j6+vr3x9fV22hYaG1nKlUnBwcJ3+wT2FcXgOG8Yg2TEOG8YgMQ5PYsMYpNofR0hIyDn71PmwI0ljx47VsGHD1LlzZ3Xt2lXPPvusjh49qjvvvNPdpQEAADezIuzceuutOnjwoB599FEVFBSoQ4cOWrRo0WmTlgEAwP8eK8KOJGVkZJzxtpW7+fr6KjMz87RbZ3UN4/AcNoxBsmMcNoxBYhyexIYxSJ41Docx53peCwAAoO6q8y8VBAAAOBvCDgAAsBphBwAAWI2wA8CtmDYIoLZZ8zSWJ/nhhx/06quvKjs7WwUFBZKkqKgode/eXcOHD1d4eLibKwQ8h6+vr7755hvFx8e7uxQAluJprBq2bt06paSkqEGDBkpOTna+66ewsFDLli3TTz/9pMWLF6tz585urvR/x7Fjx5STk6OwsDAlJCS4tJWWlurdd9/VHXfc4abqqmbbtm1as2aNEhMT1aZNG23fvl3PPfecysrKdPvtt+vqq692d4nn9MuPaPml5557TrfffrsaNWokSXr66acvZlkX7OjRo3r33Xe1a9cuRUdH67bbbnOOBbVr9OjRuuWWW3TVVVe5uxRI2r9/v2bPnq3Vq1dr//798vLyUlxcnFJTUzV8+HB5e3u7r7ga+CxO/EK3bt3MyJEjTUVFxWltFRUVZuTIkebKK690Q2U1Ly8vz9x5553uLuOsduzYYWJjY43D4TBeXl6mZ8+eJj8/39leUFBgvLy83FjhuX322Wemfv36JiwszPj5+ZnPPvvMhIeHm+TkZHP11Vcbb29vs2zZMneXeU4Oh8N06NDB9O7d22VxOBymS5cupnfv3qZPnz7uLvOc4uPjzaFDh4wxP/8daNasmQkJCTFdunQxYWFhJiIiwuzevdvNVZ5dTk6OS42vv/666d69u7nkkktMUlKSeeutt9xYXdWd+nvdsmVLM336dLN//353l1QtM2fONEOHDnV+319//XUTHx9vWrdubSZMmGBOnDjh5grPbd26dSYkJMR06tTJ9OjRw3h7e5uhQ4eaW2+91YSGhpru3bubkpISt9VH2Klhfn5+Ztu2bWds37Ztm/Hz87uIFdWer7/+2uODQmpqqhk4cKA5ePCg2blzpxk4cKBp3ry52bt3rzGmboSdxMRE86c//ckYY8xbb71lGjZsaB5++GFn+x//+EdzzTXXuKu8Kps2bZpp3rz5acGsXr16ZsuWLW6q6vw5HA5TWFhojDFmyJAhpnv37qaoqMgYY8yPP/5okpOTzW233ebOEs/p8ssvN0uWLDHGGPPyyy8bf39/c99995nZs2ebBx54wAQGBppXXnnFzVWem8PhMEuXLjX333+/ady4sfHx8THXX3+9+fjjj015ebm7y6uSP//5zyYoKMikpaWZqKgoM336dNOoUSMzdepU8/jjj5vw8HDz6KOPurvMc0pKSjKTJk1yrv/tb38z3bp1M8YYc/jwYdOhQwdz3333uas8wk5Na9asmXnttdfO2P7aa6+Z2NjYi1fQBfjoo4/OujzzzDMeHxQiIiLMxo0bnesVFRXm97//vWnatKnJzc2tE2EnODjY7Ny50xhjTHl5ualXr57ZsGGDs33Tpk0mMjLSXeWdl3/961+mVatW5g9/+IM5fvy4MaZuh524uDjz+eefu7R/+eWXpkmTJu4orcr8/f3Nt99+a4wxpmPHjuall15yaX/zzTdNQkKCO0o7L7/8szh+/Lh55513TEpKivH29jYxMTHm4Ycfdv7d8VQtWrQwf//7340xP/8H0tvb27zxxhvO9g8++MBceuml7iqvyvz9/U1ubq5zvby83Pj4+JiCggJjjDGff/65iYmJcVd5hgnKNezBBx/UyJEjlZOTo759+542Z+fll1/Wk08+6eYqqyY1NVUOh+OsT8s4HI6LWNH5O3bsmOrV+/8fc4fDodmzZysjI0O9evXS/Pnz3Vhd1Z36Pnt5ecnPz8/lU36DgoJUXFzsrtLOS5cuXZSTk6P09HR17txZb775psf/DFXmVM2lpaWKjo52afvNb36jgwcPuqOsKmvQoIF++OEHxcbG6vvvv1fXrl1d2rt166Y9e/a4qbrq8fHx0S233KJbbrlFeXl5evXVVzVv3jxNnz5d5eXl7i7vjPLz851zONu3by8vLy916NDB2X7FFVcoPz/fTdVVXUREhPbv36+4uDhJP//OO3nypPPTzlu2bKnDhw+7rT4ePa9h6enpeu2117R27VqlpaUpMTFRiYmJSktL09q1azVv3jzde++97i6zSqKjo/XBBx+ooqKi0mXDhg3uLvGc2rRpo/Xr15+2fdasWbrhhht0/fXXu6Gq89OsWTPt3LnTuZ6dna2mTZs61/Py8k77hevJAgMD9dprr2nChAlKTk726F9EZ9K3b19dccUVKikp0Y4dO1za9u7d6/ETlAcMGKDZs2dLknr16qX333/fpf3dd9/VpZde6o7SakTTpk01adIk7dmzR4sWLXJ3OWcVFRWlrVu3SpJ27typ8vJy57okbdmyRREREe4qr8pSU1P1+9//XosWLdKKFSs0ZMgQ9erVS/7+/pKkHTt26De/+Y3b6uPKTi249dZbdeutt+rEiRP64YcfJEmNGzeWj4+Pmys7P506dVJOTo5uuOGGStvPddXHE9x444166623NHTo0NPaZs2apYqKCs2ZM8cNlVXdqFGjXALBZZdd5tL+2Wef1YmnsX5t8ODB6tGjh3JychQbG+vucqosMzPTZT0wMNBl/eOPP/b4p4P+8pe/KCkpSb169VLnzp311FNPaeXKlYqPj9eOHTu0Zs0affjhh+4u85xiY2PP+oSPw+HQNddccxErOn9DhgzRHXfcoRtuuEHLli3T+PHj9eCDD+rQoUNyOBx67LHHdPPNN7u7zHOaOnWq9u/fr0GDBqm8vFyJiYl64403nO0Oh0PTpk1zW308eo4z+uc//6mjR4+qf//+lbYfPXpU69evV69evS5yZQAuVFFRkaZPn66PP/5Yu3fvVkVFhaKjo5WUlKQxY8bweoyLpKKiQtOnT1d2dra6d++uP/7xj3rnnXc0fvx4/fTTTxo0aJBmzZqlgIAAd5daJaWlpTp58uRp/wlwN8IOAACwGnN2AACA1Qg7AADAaoQdAABgNcIOgDrP4XBowYIF7i4DgIci7ADweAUFBRo9erTi4uLk6+urJk2aaNCgQVq2bJm7SwNQB/CeHQAe7dtvv1VSUpJCQ0P1xBNPqF27djpx4oQWL16s9PR0bd++3d0lAvBwXNkB4NHuvfdeORwO/etf/1JaWppatWqltm3bauzYsVqzZk2l+zz00ENq1aqVGjRooLi4OD3yyCM6ceKEs/2bb75Rnz59FBQUpODgYHXq1Mn5pu29e/dq0KBBatiwoQICAtS2bVt9+umnF2WsAGoHV3YAeKzDhw9r0aJFeuyxxyp9qVpoaGil+wUFBWnevHmKiYnRpk2bNGLECAUFBWn8+PGSfn5rbceOHTV79mx5e3vr66+/dr7hPD09XcePH9cXX3yhgIAAbd261eNekAbg/BB2AHisXbt2yRijNm3anNd+EydOdH7drFkzPfjgg3r77bedYScvL0/jxo1zHrdly5bO/nl5eUpLS1O7du0kyfnBhgDqLm5jAfBY1X3B+zvvvKOkpCRFRUUpMDBQEydOVF5enrN97Nixuvvuu5WcnKzp06crNzfX2Xbfffdp6tSpSkpKUmZmpjZu3HjB4wDgXoQdAB6rZcuWcjgc5zUJOTs7W0OGDNG1116rhQsX6t///rf+9Kc/6fjx484+kyZN0pYtWzRw4EAtX75cCQkJzg++vPvuu7V7924NHTpUmzZtUufOnTVz5swaHxuAi4fPxgLg0QYMGKBNmzZpx44dp83bKSoqUmhoqBwOhz788EOlpqbqqaee0gsvvOBytebuu+/W+++/r6KiokrPcdttt+no0aP6xz/+cVrbhAkT9Mknn3CFB6jDuLIDwKNlZWWpvLxcXbt21d///nft3LlT27Zt0/PPP6/ExMTT+rds2VJ5eXl6++23lZubq+eff9551UaSjh07poyMDK1cuVJ79+7Vl19+qXXr1ik+Pl6S9MADD2jx4sXas2ePNmzYoBUrVjjbANRNTFAG4NHi4uK0YcMGPfbYY/rDH/6g/fv3Kzw8XJ06ddLs2bNP63/99ddrzJgxysjIUFlZmQYOHKhHHnlEkyZNkiR5e3vr0KFDuuOOO1RYWKjGjRvrpptu0uTJkyVJ5eXlSk9P13fffafg4GD1799fzzzzzMUcMoAaxm0sAABgNW5jAQAAqxF2AACA1Qg7AADAaoQdAABgNcIOAACwGmEHAABYjbADAACsRtgBAABWI+wAAACrEXYAAIDVCDsAAMBqhB0AAGC1/wPv+4N19Q5lSwAAAABJRU5ErkJggg==",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "train_df['Mushroom'].value_counts().sort_index().plot(kind='bar')\n",
    "plt.title('Class distribution')\n",
    "plt.xlabel('Class')\n",
    "plt.ylabel('Count')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "949dc595-572b-45b9-a8a2-fc6969ee06cc",
   "metadata": {},
   "source": [
    "The class distribution is nearly balanced across all 10 mushroom classes, which reduces the risk of class imbalance during training."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "f3059c2b-21e9-48aa-88f3-8c3e7ef20212",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1500x600 with 10 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, axes = plt.subplots(2, 5, figsize=(15,6))\n",
    "\n",
    "for cls, ax in zip(sorted(train_df['Mushroom'].unique()), axes.flatten()):\n",
    "    img_path = train_df[train_df['Mushroom']==cls].iloc[0]['filepath']\n",
    "    ax.imshow(Image.open(img_path))\n",
    "    ax.set_title(f'Class {cls}')\n",
    "    ax.axis('off')\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4567c866-26b3-4206-b595-cd96757214a6",
   "metadata": {},
   "source": [
    "Sample images show clear visual differences between mushroom classes in shape, color, and texture, indicating that the classes are visually distinguishable."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "c0ea38f4-b985-45f3-aba3-e256813c902c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sizes = train_df['filepath'].sample(100).apply(lambda p: Image.open(p).size)\n",
    "sizes = list(sizes)\n",
    "\n",
    "plt.hist([s[0] for s in sizes], bins=20)\n",
    "plt.title('Image width distribution')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4e6ee3b3-8196-461f-97a6-9f68aed52f36",
   "metadata": {},
   "source": [
    "The image width distribution shows moderate variation, justifying the need for resizing all images to a fixed input size."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "406208ed-ba90-4b58-a25c-2c82b0756cf4",
   "metadata": {},
   "source": [
    "### Modeling"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "003ad7d0-499f-4a0f-a678-5a3833069d53",
   "metadata": {},
   "outputs": [],
   "source": [
    "IMG_SIZE=(224,224)\n",
    "BATCH=16\n",
    "SEED=42"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "552e9698-a767-4619-b48d-dc5ee43271e9",
   "metadata": {},
   "outputs": [],
   "source": [
    "tr_df,va_df=train_test_split(train_df,test_size=0.2,random_state=SEED,stratify=train_df['Mushroom'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "7acc8455-bed4-4597-a0b7-6e8839e50fdd",
   "metadata": {},
   "outputs": [],
   "source": [
    "train_gen=ImageDataGenerator(rescale=1./255,rotation_range=30,zoom_range=0.3,brightness_range=[0.8,1.2],width_shift_range=0.1,height_shift_range=0.1,horizontal_flip=True)\n",
    "val_gen=ImageDataGenerator(rescale=1./255)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "3cb92beb-9adc-4cd4-b830-81ae8c50f95b",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Found 1892 validated image filenames.\n",
      "Found 473 validated image filenames.\n"
     ]
    }
   ],
   "source": [
    "x_train=train_gen.flow_from_dataframe(tr_df,x_col='filepath',y_col='Mushroom',target_size=IMG_SIZE,class_mode='raw',batch_size=BATCH,shuffle=True,seed=SEED)\n",
    "x_val=val_gen.flow_from_dataframe(va_df,x_col='filepath',y_col='Mushroom',target_size=IMG_SIZE,class_mode='raw',batch_size=BATCH,shuffle=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "6a6ab460-53d6-4132-9d8e-e5b2b9955ff1",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING: All log messages before absl::InitializeLog() is called are written to STDERR\n",
      "I0000 00:00:1766594263.018160   19261 cuda_executor.cc:1001] could not open file to read NUMA node: /sys/bus/pci/devices/0000:02:00.0/numa_node\n",
      "Your kernel may have been built without NUMA support.\n",
      "I0000 00:00:1766594263.117632   19261 cuda_executor.cc:1001] could not open file to read NUMA node: /sys/bus/pci/devices/0000:02:00.0/numa_node\n",
      "Your kernel may have been built without NUMA support.\n",
      "I0000 00:00:1766594263.117669   19261 cuda_executor.cc:1001] could not open file to read NUMA node: /sys/bus/pci/devices/0000:02:00.0/numa_node\n",
      "Your kernel may have been built without NUMA support.\n",
      "I0000 00:00:1766594263.119691   19261 cuda_executor.cc:1001] could not open file to read NUMA node: /sys/bus/pci/devices/0000:02:00.0/numa_node\n",
      "Your kernel may have been built without NUMA support.\n",
      "I0000 00:00:1766594263.119719   19261 cuda_executor.cc:1001] could not open file to read NUMA node: /sys/bus/pci/devices/0000:02:00.0/numa_node\n",
      "Your kernel may have been built without NUMA support.\n",
      "I0000 00:00:1766594263.119727   19261 cuda_executor.cc:1001] could not open file to read NUMA node: /sys/bus/pci/devices/0000:02:00.0/numa_node\n",
      "Your kernel may have been built without NUMA support.\n",
      "I0000 00:00:1766594263.696717   19261 cuda_executor.cc:1001] could not open file to read NUMA node: /sys/bus/pci/devices/0000:02:00.0/numa_node\n",
      "Your kernel may have been built without NUMA support.\n",
      "I0000 00:00:1766594263.696764   19261 cuda_executor.cc:1001] could not open file to read NUMA node: /sys/bus/pci/devices/0000:02:00.0/numa_node\n",
      "Your kernel may have been built without NUMA support.\n",
      "2025-12-24 16:37:43.696773: I tensorflow/core/common_runtime/gpu/gpu_device.cc:2112] Could not identify NUMA node of platform GPU id 0, defaulting to 0.  Your kernel may not have been built with NUMA support.\n",
      "I0000 00:00:1766594263.696808   19261 cuda_executor.cc:1001] could not open file to read NUMA node: /sys/bus/pci/devices/0000:02:00.0/numa_node\n",
      "Your kernel may have been built without NUMA support.\n",
      "2025-12-24 16:37:43.696829: I tensorflow/core/common_runtime/gpu/gpu_device.cc:2021] Created device /job:localhost/replica:0/task:0/device:GPU:0 with 10861 MB memory:  -> device: 0, name: NVIDIA GeForce RTX 5080 Laptop GPU, pci bus id: 0000:02:00.0, compute capability: 12.0\n"
     ]
    }
   ],
   "source": [
    "model=Sequential()\n",
    "model.add(Input(shape=(IMG_SIZE[0], IMG_SIZE[1], 3)))\n",
    "model.add(Conv2D(64, (3,3), padding='same', activation='relu'))\n",
    "model.add(BatchNormalization())\n",
    "model.add(MaxPooling2D((2,2)))\n",
    "model.add(Conv2D(128, (3,3), padding='same', activation='relu'))\n",
    "model.add(BatchNormalization())\n",
    "model.add(MaxPooling2D((2,2)))\n",
    "model.add(Conv2D(256, (3,3), padding='same', activation='relu'))\n",
    "model.add(BatchNormalization())\n",
    "model.add(MaxPooling2D((2,2)))\n",
    "model.add(Conv2D(512, (3,3), padding='same', activation='relu'))\n",
    "model.add(BatchNormalization())\n",
    "model.add(MaxPooling2D((2,2)))\n",
    "model.add(GlobalAveragePooling2D())\n",
    "model.add(Dense(256, activation='relu', kernel_regularizer=regularizers.l2(1e-4)))\n",
    "model.add(Dropout(0.3))\n",
    "model.add(Dense(10, activation='softmax'))\n",
    "model.compile(optimizer=tf.keras.optimizers.Adam(learning_rate=1e-4),loss='sparse_categorical_crossentropy',metrics=['accuracy',tf.keras.metrics.TopKCategoricalAccuracy(k=3)])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "a968ced6-2cbd-4521-b1f6-7ba0a7079531",
   "metadata": {},
   "outputs": [],
   "source": [
    "callbacks = [\n",
    "    tf.keras.callbacks.ModelCheckpoint(\n",
    "        'best_model.keras',\n",
    "        monitor='val_accuracy',\n",
    "        save_best_only=True,\n",
    "        mode='max'\n",
    "    ),\n",
    "    tf.keras.callbacks.EarlyStopping(\n",
    "        patience=6,\n",
    "        restore_best_weights=True,\n",
    "        monitor='val_loss'\n",
    "    ),\n",
    "    tf.keras.callbacks.ReduceLROnPlateau(\n",
    "        monitor='val_loss',\n",
    "        factor=0.5,\n",
    "        patience=3,\n",
    "        verbose=1\n",
    "    )\n",
    "]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "3311ef4b-d26e-4e39-b733-5ebd0321da93",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/50\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2025-12-24 16:37:45.483491: I external/local_xla/xla/stream_executor/cuda/cuda_dnn.cc:531] Loaded cuDNN version 90701\n",
      "W0000 00:00:1766594265.706193   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594265.731074   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594265.731797   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594265.739431   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594265.740527   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594265.741538   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594265.764379   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594265.766690   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594265.769069   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594265.771730   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594265.787362   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594265.795529   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594265.894327   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594265.895843   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594265.897867   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594265.900398   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594265.902226   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594265.904137   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594265.906834   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594265.909132   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594265.911684   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594265.914254   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594265.916940   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594265.920537   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594265.932908   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594265.937454   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594265.950823   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594265.954348   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594265.955852   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594265.957830   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594265.959458   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594265.961984   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594265.964800   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594265.968043   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594265.971685   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594265.975385   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594265.978779   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594265.982149   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594265.984284   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594265.988540   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594265.997495   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594266.002131   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594266.003254   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594266.004900   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594266.006456   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594266.007825   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594266.009707   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594266.011742   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594266.014352   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594266.016959   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594266.020445   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594266.023945   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594266.027154   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594266.030413   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594266.034499   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594266.037196   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "WARNING: All log messages before absl::InitializeLog() is called are written to STDERR\n",
      "I0000 00:00:1766594266.614987   19374 service.cc:146] XLA service 0x716db51f36c0 initialized for platform CUDA (this does not guarantee that XLA will be used). Devices:\n",
      "I0000 00:00:1766594266.615011   19374 service.cc:154]   StreamExecutor device (0): NVIDIA GeForce RTX 5080 Laptop GPU, Compute Capability 12.0\n",
      "2025-12-24 16:37:46.624856: I tensorflow/compiler/mlir/tensorflow/utils/dump_mlir_util.cc:268] disabling MLIR crash reproducer, set env var `MLIR_CRASH_REPRODUCER_DIRECTORY` to enable.\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "I0000 00:00:1766594266.684383   19374 device_compiler.h:188] Compiled cluster using XLA!  This line is logged at most once for the lifetime of the process.\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "W0000 00:00:1766594266.914581   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594266.918653   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594266.921450   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594266.922804   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594266.924007   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594266.926475   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594266.927929   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594266.930672   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594266.935696   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594266.937545   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594266.940194   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594266.942521   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594266.945021   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594266.947945   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594266.955657   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594266.958968   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594266.962518   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594266.978524   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594266.979380   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594266.980430   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594266.984071   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594266.985401   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594266.986772   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594266.988273   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594266.989821   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594266.991642   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.006783   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.009184   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.016398   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.021263   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "W0000 00:00:1766594267.147318   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.148395   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.149454   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.150587   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.151672   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.152790   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.153790   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.154941   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.156661   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.158972   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.160963   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.163584   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.165783   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.170486   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.171679   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.173504   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.175337   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.176619   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.177707   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.179457   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.181310   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.183929   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.187175   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.189746   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.193542   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.196765   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.201065   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.204477   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.208368   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "W0000 00:00:1766594267.354953   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.356318   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.357577   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.358886   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.360898   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.362338   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.363767   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.365512   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.367208   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.369422   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.371936   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.376714   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.378679   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.380582   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.382497   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.384021   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.386078   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.388817   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.394029   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.398465   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.403002   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.408190   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.413063   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.414966   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.419614   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.423967   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.431526   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.437845   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  1/119 [..............................] - ETA: 6:52 - loss: 2.9121 - accuracy: 0.1250 - top_k_categorical_accuracy: 0.0000e+00"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "W0000 00:00:1766594267.599402   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.600276   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.601050   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.602929   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.604005   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.604836   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.605855   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.607154   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.609050   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.610948   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.612505   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.616434   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.618135   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.626595   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "W0000 00:00:1766594267.706863   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.707594   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.708202   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.709052   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.709935   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.710745   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.712184   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.713223   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.714725   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.716021   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.719629   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.721335   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.727298   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.728389   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.729993   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.732024   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.733417   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.735340   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.737696   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.740421   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.743799   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.747150   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.750660   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.754588   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.756756   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.761398   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.772416   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.775600   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.777216   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.779385   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.781212   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.783901   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.787019   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.790685   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.794874   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.799114   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.802917   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.807004   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.809671   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.814730   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.820658   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.824680   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.826084   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.828044   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.829899   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.831553   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.834246   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.836751   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.839925   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.843055   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.846943   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.850931   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.854459   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.858039   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.862489   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.865158   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.873807   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.874933   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.876601   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.878388   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.880120   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.881707   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.884185   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.886735   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.889312   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.892013   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.895499   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.898958   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.902605   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.915655   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.919750   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.931218   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.935267   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.949497   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.950641   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.952199   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.962101   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.963748   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.965227   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.966914   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.968582   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.970520   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.973261   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.975713   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.978592   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.981036   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.984672   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.986415   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.987852   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.989444   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.991010   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.992509   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.993603   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.995140   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594267.997167   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.000335   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.002604   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.004358   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.007076   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.013504   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.015130   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.017703   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.020263   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.022145   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.023503   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.025997   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.028807   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.031783   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.035494   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.039508   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.043898   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.047670   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.053060   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.057272   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.061892   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.070266   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.072526   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.074536   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.076574   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.080075   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.082282   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.084435   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.087095   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.089166   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.092518   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.096130   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.102279   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.105217   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.108107   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.110947   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.113189   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.116376   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.120283   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.126720   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  2/119 [..............................] - ETA: 1:07 - loss: 2.5994 - accuracy: 0.1875 - top_k_categorical_accuracy: 0.0000e+00"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "W0000 00:00:1766594268.132182   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.137782   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.144158   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.150211   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.152392   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.158247   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.165244   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.178035   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.186319   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.228802   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.229880   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.231331   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.233321   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.235269   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.237319   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.239287   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.242791   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.247870   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.252775   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.255315   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.260839   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.262779   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.270230   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.291781   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.292877   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.294444   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.295962   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.297429   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.298791   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.300954   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.303150   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.305399   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.307769   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.310743   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.313728   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.316837   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.319797   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.323146   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.326870   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.330253   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.335945   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.342375   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.344660   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.346765   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.350422   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.352765   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.355130   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.358054   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.360304   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.363976   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.367990   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.373617   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.379291   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.385491   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.392134   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.396769   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.403452   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.412524   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.425153   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.436009   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.447324   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.459982   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.471834   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.474803   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.484780   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.493668   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.508176   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.517106   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  3/119 [..............................] - ETA: 53s - loss: 2.7610 - accuracy: 0.1667 - top_k_categorical_accuracy: 0.0000e+00 "
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "W0000 00:00:1766594268.555164   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.562917   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.564509   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.566245   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.567832   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.569124   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.570720   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.572887   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.576240   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.579511   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.581451   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.589584   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.593697   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.610490   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.634819   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.636014   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.637928   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.639765   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.641526   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.643176   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.645583   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.647925   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.650289   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.652823   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.656093   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.659334   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.662714   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.665892   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.669481   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.673431   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.677015   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.683404   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.690116   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.691959   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.693838   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.697289   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.699495   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.701785   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.704498   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.706625   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.710058   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.713762   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.719940   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.722864   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.725841   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.728751   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.731078   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.734371   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.738459   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.744286   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.748985   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.753778   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  6/119 [>.............................] - ETA: 27s - loss: 2.7966 - accuracy: 0.1562 - top_k_categorical_accuracy: 0.0521    "
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "W0000 00:00:1766594268.758856   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.764014   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.765928   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.770848   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.775524   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.784003   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.790778   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.817458   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.819605   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.820570   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.822089   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.823672   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.824963   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.826550   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.828723   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.832101   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.835372   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.837305   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.843375   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.846042   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594268.858423   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " 45/119 [==========>...................] - ETA: 6s - loss: 2.5965 - accuracy: 0.1921 - top_k_categorical_accuracy: 0.2444 "
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "W0000 00:00:1766594271.653548   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.653837   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.654059   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.654355   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.654647   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.654925   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.655338   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.655700   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.656120   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.656632   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.657895   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.659598   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.665379   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.665865   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.666428   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.666936   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.667553   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.668531   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.669268   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.670016   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.670836   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.671665   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.672490   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.673329   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.674381   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.675808   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.677439   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.685853   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.686546   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.687081   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.687537   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.688134   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.688765   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.689572   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.690661   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.691381   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.692217   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.693124   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.694093   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.694977   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.696093   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.697736   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.704375   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.704949   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.705479   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.705934   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.706504   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.707142   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.707880   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.708915   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.710004   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.710955   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.711907   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.712934   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.713923   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.715007   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.717322   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.724461   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.724863   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.725416   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.725926   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.726389   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.726884   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.727496   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.728135   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.728788   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.729484   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.730362   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.731230   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.732097   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.732953   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.734047   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.735163   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.737747   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.738235   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.738697   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.739152   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.739604   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.740082   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.740594   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.741217   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.741961   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.742623   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.743561   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.744564   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.745563   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.749106   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.749524   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.749971   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.750412   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.750853   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.751291   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.751766   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.752312   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.752955   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.753639   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.754666   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.755573   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.757782   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.760700   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.761376   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.761989   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.762583   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.763025   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.763323   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.763962   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.764619   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.765317   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.766212   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.767143   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.768123   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.768929   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.770090   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.771156   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.772286   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.775156   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.775636   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.776160   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.776629   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.777202   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.777891   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.778716   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.779516   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.780358   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.781105   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.781987   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.783554   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.786575   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.787273   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.787944   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.788627   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.789185   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.789916   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.790835   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.792059   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.793135   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.794194   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.795267   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.796364   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.796819   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.798133   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.799264   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.800964   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.808756   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.809037   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.809338   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.809718   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.810019   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.810341   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.810722   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.811231   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.811845   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.812542   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.813159   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.813865   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594271.815706   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "119/119 [==============================] - ETA: 0s - loss: 2.3787 - accuracy: 0.2125 - top_k_categorical_accuracy: 0.2822"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "W0000 00:00:1766594277.532216   19372 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594277.532722   19372 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594277.533165   19372 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594277.533746   19372 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594277.534337   19372 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594277.534950   19372 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594277.535959   19372 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594277.536729   19372 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594277.537770   19372 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594277.538810   19372 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594277.541727   19372 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594277.543651   19372 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594277.549249   19372 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594277.550114   19372 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594277.551278   19372 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594277.552708   19372 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594277.554226   19372 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594277.555603   19372 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594277.557328   19372 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594277.558926   19372 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594277.560622   19372 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594277.562316   19372 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594277.564038   19372 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594277.566243   19372 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594277.568111   19372 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594277.570795   19372 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594277.580009   19372 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594277.581063   19372 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594277.582367   19372 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594277.583374   19372 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594277.584269   19372 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594277.585535   19372 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594277.586996   19372 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594277.588714   19372 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594277.590765   19372 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594277.592859   19372 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594277.594776   19372 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594277.596944   19372 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594277.598318   19372 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594277.600918   19372 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594277.606289   19372 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594277.609930   19372 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594277.611093   19372 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594277.612249   19372 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594277.613246   19372 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594277.614127   19372 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594277.615343   19372 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594277.616665   19372 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594277.618280   19372 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594277.620128   19372 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594277.622173   19372 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594277.624268   19372 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594277.626259   19372 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594277.628422   19372 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594277.630880   19372 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594277.633395   19372 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "119/119 [==============================] - 14s 85ms/step - loss: 2.3787 - accuracy: 0.2125 - top_k_categorical_accuracy: 0.2822 - val_loss: 2.4838 - val_accuracy: 0.1015 - val_top_k_categorical_accuracy: 0.8795 - lr: 1.0000e-04\n",
      "Epoch 2/50\n",
      " 45/119 [==========>...................] - ETA: 5s - loss: 2.0606 - accuracy: 0.2910 - top_k_categorical_accuracy: 0.2980 "
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "W0000 00:00:1766594280.674088   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.674597   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.674874   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.675237   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.675577   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.675886   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.676414   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.676827   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.677359   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.677906   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.679317   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.681306   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.686410   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.687020   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.687583   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.688118   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.688753   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.689677   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.690325   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.691044   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.691816   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.692616   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.693378   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.694160   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.695170   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.696841   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.698347   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.707142   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.707620   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.708054   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.708564   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.709085   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.709678   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.710405   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.711558   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.712288   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.713115   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.713961   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.714837   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.715628   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.716634   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.718276   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.724792   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.725409   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.725840   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.726245   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.726762   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.727346   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.728017   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.729046   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.730041   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.730882   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.731743   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.732681   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.733557   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.734528   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.736889   19375 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.743089   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.743630   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.744166   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.744645   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.745053   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.745506   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.746073   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.746653   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.747263   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.747882   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.748665   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.749425   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.750190   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.750946   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.751803   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.752727   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.755080   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.755707   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.756113   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.756533   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.756942   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.757365   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.757781   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.758325   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.758999   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.759720   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.760654   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.761574   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.762468   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.766107   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.766665   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.767067   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.767476   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.767929   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.768351   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.768821   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.769329   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.769927   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.770551   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.771532   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.772345   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.774370   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.777151   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.777801   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.778376   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.778932   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.779343   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.779648   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.780270   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.780890   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.781537   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.782338   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.783216   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.784140   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.784899   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.785978   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.786935   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.787984   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.790994   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.791639   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.792146   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.792605   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.793152   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.793774   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.794554   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.795274   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.796087   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.796809   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.797633   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.799080   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.801780   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.802599   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.803250   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.803879   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.804417   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.805115   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.806017   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.807208   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.808201   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.809176   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.810243   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.811298   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.811778   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.812986   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.813997   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.815609   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.823326   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.823648   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.823955   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.824351   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.824686   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.825068   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.825460   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.825955   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.826565   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.827250   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.827871   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.828549   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594280.830278   19377 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "119/119 [==============================] - 8s 70ms/step - loss: 2.0309 - accuracy: 0.2944 - top_k_categorical_accuracy: 0.2844 - val_loss: 2.7441 - val_accuracy: 0.1099 - val_top_k_categorical_accuracy: 0.0042 - lr: 1.0000e-04\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "W0000 00:00:1766594285.877285   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594285.877812   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594285.878280   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594285.878905   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594285.879534   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594285.880149   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594285.881145   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594285.881917   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594285.882954   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594285.884006   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594285.886950   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594285.888851   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594285.894239   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594285.895319   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594285.896473   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594285.897915   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594285.899430   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594285.900743   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594285.902463   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594285.904075   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594285.905786   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594285.907473   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594285.909237   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594285.911422   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594285.913236   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594285.915807   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594285.924756   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594285.925778   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594285.926997   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594285.927947   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594285.928799   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594285.930075   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594285.931627   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594285.933310   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594285.935285   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594285.937298   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594285.939149   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594285.941161   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594285.942475   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594285.944953   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594285.950096   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594285.953370   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594285.954536   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594285.955502   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594285.956510   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594285.957411   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594285.958609   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594285.959911   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594285.961516   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594285.963218   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594285.965226   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594285.967245   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594285.969134   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594285.971159   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594285.973426   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594285.975822   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 3/50\n",
      " 50/119 [===========>..................] - ETA: 4s - loss: 1.9927 - accuracy: 0.3109 - top_k_categorical_accuracy: 0.2754 "
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "W0000 00:00:1766594289.265110   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594289.265419   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594289.265643   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594289.265925   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594289.266288   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594289.266600   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594289.267138   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594289.267525   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594289.268050   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594289.268600   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594289.269973   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594289.271943   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594289.277827   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594289.278565   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594289.279012   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594289.279420   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594289.279898   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594289.280473   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594289.281173   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594289.282237   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594289.282897   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594289.283652   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594289.284426   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594289.285217   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594289.285952   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594289.286865   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594289.288433   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594289.296851   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594289.297391   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594289.297771   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594289.298162   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594289.298551   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594289.298948   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594289.299347   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594289.299858   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594289.300492   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594289.301083   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594289.301980   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594289.302909   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594289.303817   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594289.308227   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594289.310331   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594289.310590   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594289.310991   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594289.311316   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594289.311695   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594289.312077   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594289.312569   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594289.313186   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594289.313879   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594289.314498   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594289.315184   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594289.316961   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "119/119 [==============================] - 9s 79ms/step - loss: 1.9247 - accuracy: 0.3467 - top_k_categorical_accuracy: 0.2711 - val_loss: 2.4091 - val_accuracy: 0.1776 - val_top_k_categorical_accuracy: 0.1057 - lr: 1.0000e-04\n",
      "Epoch 4/50\n",
      "119/119 [==============================] - ETA: 0s - loss: 1.8321 - accuracy: 0.3853 - top_k_categorical_accuracy: 0.2680 "
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "W0000 00:00:1766594303.298294   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594303.302082   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594303.302344   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594303.302674   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594303.302997   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594303.303297   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594303.303864   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594303.304242   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594303.304790   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594303.305452   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594303.306759   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594303.308658   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594303.313871   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594303.314481   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594303.314901   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594303.315318   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594303.315807   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594303.316371   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594303.317062   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594303.318170   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594303.318866   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594303.319636   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594303.320422   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594303.321254   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594303.322006   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594303.322956   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594303.324525   19373 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594303.333416   19376 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594303.334023   19376 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594303.334397   19376 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594303.334775   19376 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594303.335142   19376 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594303.335532   19376 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594303.335921   19376 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594303.336421   19376 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594303.337059   19376 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594303.337659   19376 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594303.338492   19376 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594303.339341   19376 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594303.340159   19376 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594303.346027   19376 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594303.346472   19376 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594303.346728   19376 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594303.347107   19376 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594303.347424   19376 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594303.348011   19376 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594303.348407   19376 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594303.348885   19376 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594303.349483   19376 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594303.350167   19376 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594303.350765   19376 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594303.351417   19376 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594303.353099   19376 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "119/119 [==============================] - 8s 70ms/step - loss: 1.8321 - accuracy: 0.3853 - top_k_categorical_accuracy: 0.2680 - val_loss: 2.0983 - val_accuracy: 0.2791 - val_top_k_categorical_accuracy: 0.2600 - lr: 1.0000e-04\n",
      "Epoch 5/50\n",
      " 76/119 [==================>...........] - ETA: 2s - loss: 1.7888 - accuracy: 0.3870 - top_k_categorical_accuracy: 0.3065 "
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "W0000 00:00:1766594308.755690   19376 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594308.756257   19376 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594308.756725   19376 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594308.757138   19376 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594308.757517   19376 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594308.757917   19376 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594308.758317   19376 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594308.758834   19376 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594308.759470   19376 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594308.760075   19376 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594308.760965   19376 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594308.761883   19376 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594308.762781   19376 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "119/119 [==============================] - 8s 70ms/step - loss: 1.7875 - accuracy: 0.3906 - top_k_categorical_accuracy: 0.2849 - val_loss: 1.8314 - val_accuracy: 0.3594 - val_top_k_categorical_accuracy: 0.0951 - lr: 1.0000e-04\n",
      "Epoch 6/50\n",
      " 10/119 [=>............................] - ETA: 6s - loss: 1.7113 - accuracy: 0.4122 - top_k_categorical_accuracy: 0.2162 "
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "W0000 00:00:1766594312.740043   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594312.740517   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594312.740932   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594312.741337   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594312.741733   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594312.742141   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594312.742548   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594312.743069   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594312.743721   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594312.744327   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594312.745196   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594312.746088   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594312.746953   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "119/119 [==============================] - 8s 70ms/step - loss: 1.7249 - accuracy: 0.4080 - top_k_categorical_accuracy: 0.2817 - val_loss: 1.7688 - val_accuracy: 0.3932 - val_top_k_categorical_accuracy: 0.2072 - lr: 1.0000e-04\n",
      "Epoch 7/50\n",
      " 11/119 [=>............................] - ETA: 6s - loss: 1.8717 - accuracy: 0.3598 - top_k_categorical_accuracy: 0.2866 "
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "W0000 00:00:1766594321.187256   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594321.187732   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594321.188115   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594321.188502   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594321.188898   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594321.189302   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594321.189734   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594321.190255   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594321.190917   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594321.191520   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594321.192398   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594321.193286   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594321.194177   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "119/119 [==============================] - 9s 79ms/step - loss: 1.7406 - accuracy: 0.4112 - top_k_categorical_accuracy: 0.2891 - val_loss: 1.7564 - val_accuracy: 0.4144 - val_top_k_categorical_accuracy: 0.3298 - lr: 1.0000e-04\n",
      "Epoch 8/50\n",
      "116/119 [============================>.] - ETA: 0s - loss: 1.6721 - accuracy: 0.4225 - top_k_categorical_accuracy: 0.2874 "
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "W0000 00:00:1766594337.531232   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594337.531926   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594337.532408   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594337.532900   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594337.533394   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594337.533907   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594337.534410   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594337.535082   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594337.535913   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594337.536677   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594337.537736   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594337.538869   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766594337.539983   19374 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "119/119 [==============================] - 8s 69ms/step - loss: 1.6740 - accuracy: 0.4234 - top_k_categorical_accuracy: 0.2881 - val_loss: 1.9701 - val_accuracy: 0.3531 - val_top_k_categorical_accuracy: 0.2685 - lr: 1.0000e-04\n",
      "Epoch 9/50\n",
      "119/119 [==============================] - 9s 73ms/step - loss: 1.6321 - accuracy: 0.4434 - top_k_categorical_accuracy: 0.2865 - val_loss: 1.7480 - val_accuracy: 0.4186 - val_top_k_categorical_accuracy: 0.2537 - lr: 1.0000e-04\n",
      "Epoch 10/50\n",
      "119/119 [==============================] - 9s 71ms/step - loss: 1.6114 - accuracy: 0.4567 - top_k_categorical_accuracy: 0.2680 - val_loss: 1.6742 - val_accuracy: 0.4334 - val_top_k_categorical_accuracy: 0.2622 - lr: 1.0000e-04\n",
      "Epoch 11/50\n",
      "119/119 [==============================] - 10s 80ms/step - loss: 1.5791 - accuracy: 0.4688 - top_k_categorical_accuracy: 0.2870 - val_loss: 1.5262 - val_accuracy: 0.5032 - val_top_k_categorical_accuracy: 0.2685 - lr: 1.0000e-04\n",
      "Epoch 12/50\n",
      "119/119 [==============================] - 8s 69ms/step - loss: 1.5842 - accuracy: 0.4693 - top_k_categorical_accuracy: 0.2563 - val_loss: 1.6413 - val_accuracy: 0.4609 - val_top_k_categorical_accuracy: 0.1903 - lr: 1.0000e-04\n",
      "Epoch 13/50\n",
      "119/119 [==============================] - 8s 71ms/step - loss: 1.4924 - accuracy: 0.5005 - top_k_categorical_accuracy: 0.2812 - val_loss: 1.6066 - val_accuracy: 0.4884 - val_top_k_categorical_accuracy: 0.3066 - lr: 1.0000e-04\n",
      "Epoch 14/50\n",
      "119/119 [==============================] - ETA: 0s - loss: 1.4800 - accuracy: 0.5042 - top_k_categorical_accuracy: 0.2632 \n",
      "Epoch 14: ReduceLROnPlateau reducing learning rate to 4.999999873689376e-05.\n",
      "119/119 [==============================] - 10s 81ms/step - loss: 1.4800 - accuracy: 0.5042 - top_k_categorical_accuracy: 0.2632 - val_loss: 1.5883 - val_accuracy: 0.4968 - val_top_k_categorical_accuracy: 0.2114 - lr: 1.0000e-04\n",
      "Epoch 15/50\n",
      "119/119 [==============================] - 9s 72ms/step - loss: 1.4149 - accuracy: 0.5317 - top_k_categorical_accuracy: 0.2923 - val_loss: 1.5099 - val_accuracy: 0.4968 - val_top_k_categorical_accuracy: 0.2431 - lr: 5.0000e-05\n",
      "Epoch 16/50\n",
      "119/119 [==============================] - 8s 70ms/step - loss: 1.3782 - accuracy: 0.5444 - top_k_categorical_accuracy: 0.2537 - val_loss: 1.4837 - val_accuracy: 0.5095 - val_top_k_categorical_accuracy: 0.2812 - lr: 5.0000e-05\n",
      "Epoch 17/50\n",
      "119/119 [==============================] - 9s 71ms/step - loss: 1.3632 - accuracy: 0.5407 - top_k_categorical_accuracy: 0.2553 - val_loss: 1.5128 - val_accuracy: 0.4947 - val_top_k_categorical_accuracy: 0.2537 - lr: 5.0000e-05\n",
      "Epoch 18/50\n",
      "119/119 [==============================] - 10s 80ms/step - loss: 1.3488 - accuracy: 0.5534 - top_k_categorical_accuracy: 0.2690 - val_loss: 1.4904 - val_accuracy: 0.5011 - val_top_k_categorical_accuracy: 0.2368 - lr: 5.0000e-05\n",
      "Epoch 19/50\n",
      "119/119 [==============================] - 9s 73ms/step - loss: 1.3306 - accuracy: 0.5539 - top_k_categorical_accuracy: 0.2822 - val_loss: 1.4682 - val_accuracy: 0.5370 - val_top_k_categorical_accuracy: 0.2199 - lr: 5.0000e-05\n",
      "Epoch 20/50\n",
      "119/119 [==============================] - 8s 71ms/step - loss: 1.3147 - accuracy: 0.5698 - top_k_categorical_accuracy: 0.2791 - val_loss: 1.4869 - val_accuracy: 0.4968 - val_top_k_categorical_accuracy: 0.2706 - lr: 5.0000e-05\n",
      "Epoch 21/50\n",
      "119/119 [==============================] - 8s 71ms/step - loss: 1.3108 - accuracy: 0.5745 - top_k_categorical_accuracy: 0.2822 - val_loss: 1.5401 - val_accuracy: 0.5053 - val_top_k_categorical_accuracy: 0.2389 - lr: 5.0000e-05\n",
      "Epoch 22/50\n",
      "119/119 [==============================] - 10s 81ms/step - loss: 1.2811 - accuracy: 0.5862 - top_k_categorical_accuracy: 0.2881 - val_loss: 1.4580 - val_accuracy: 0.5370 - val_top_k_categorical_accuracy: 0.2495 - lr: 5.0000e-05\n",
      "Epoch 23/50\n",
      "119/119 [==============================] - 9s 71ms/step - loss: 1.2687 - accuracy: 0.5862 - top_k_categorical_accuracy: 0.2648 - val_loss: 1.4672 - val_accuracy: 0.5370 - val_top_k_categorical_accuracy: 0.2685 - lr: 5.0000e-05\n",
      "Epoch 24/50\n",
      "119/119 [==============================] - 8s 71ms/step - loss: 1.2413 - accuracy: 0.5925 - top_k_categorical_accuracy: 0.2706 - val_loss: 1.3962 - val_accuracy: 0.5433 - val_top_k_categorical_accuracy: 0.2727 - lr: 5.0000e-05\n",
      "Epoch 25/50\n",
      "119/119 [==============================] - 9s 72ms/step - loss: 1.2692 - accuracy: 0.5851 - top_k_categorical_accuracy: 0.2902 - val_loss: 1.4322 - val_accuracy: 0.5645 - val_top_k_categorical_accuracy: 0.1797 - lr: 5.0000e-05\n",
      "Epoch 26/50\n",
      "119/119 [==============================] - 10s 83ms/step - loss: 1.2351 - accuracy: 0.5946 - top_k_categorical_accuracy: 0.2896 - val_loss: 1.3684 - val_accuracy: 0.5729 - val_top_k_categorical_accuracy: 0.2114 - lr: 5.0000e-05\n",
      "Epoch 27/50\n",
      "119/119 [==============================] - 9s 72ms/step - loss: 1.1944 - accuracy: 0.6158 - top_k_categorical_accuracy: 0.2865 - val_loss: 1.3723 - val_accuracy: 0.5835 - val_top_k_categorical_accuracy: 0.2579 - lr: 5.0000e-05\n",
      "Epoch 28/50\n",
      "119/119 [==============================] - 8s 70ms/step - loss: 1.1899 - accuracy: 0.6205 - top_k_categorical_accuracy: 0.2870 - val_loss: 1.5276 - val_accuracy: 0.4947 - val_top_k_categorical_accuracy: 0.2008 - lr: 5.0000e-05\n",
      "Epoch 29/50\n",
      "119/119 [==============================] - ETA: 0s - loss: 1.1938 - accuracy: 0.6094 - top_k_categorical_accuracy: 0.2791 \n",
      "Epoch 29: ReduceLROnPlateau reducing learning rate to 2.499999936844688e-05.\n",
      "119/119 [==============================] - 9s 72ms/step - loss: 1.1938 - accuracy: 0.6094 - top_k_categorical_accuracy: 0.2791 - val_loss: 1.4557 - val_accuracy: 0.5539 - val_top_k_categorical_accuracy: 0.3340 - lr: 5.0000e-05\n",
      "Epoch 30/50\n",
      "119/119 [==============================] - 10s 84ms/step - loss: 1.1428 - accuracy: 0.6353 - top_k_categorical_accuracy: 0.2801 - val_loss: 1.4430 - val_accuracy: 0.5370 - val_top_k_categorical_accuracy: 0.1987 - lr: 2.5000e-05\n",
      "Epoch 31/50\n",
      "119/119 [==============================] - 9s 71ms/step - loss: 1.1155 - accuracy: 0.6533 - top_k_categorical_accuracy: 0.2674 - val_loss: 1.3925 - val_accuracy: 0.5772 - val_top_k_categorical_accuracy: 0.1903 - lr: 2.5000e-05\n",
      "Epoch 32/50\n",
      "119/119 [==============================] - 8s 70ms/step - loss: 1.0792 - accuracy: 0.6686 - top_k_categorical_accuracy: 0.2812 - val_loss: 1.3683 - val_accuracy: 0.5603 - val_top_k_categorical_accuracy: 0.2051 - lr: 2.5000e-05\n",
      "Epoch 33/50\n",
      "119/119 [==============================] - 10s 82ms/step - loss: 1.0655 - accuracy: 0.6559 - top_k_categorical_accuracy: 0.2854 - val_loss: 1.3923 - val_accuracy: 0.5539 - val_top_k_categorical_accuracy: 0.1987 - lr: 2.5000e-05\n",
      "Epoch 34/50\n",
      "119/119 [==============================] - 9s 74ms/step - loss: 1.0643 - accuracy: 0.6601 - top_k_categorical_accuracy: 0.2854 - val_loss: 1.3401 - val_accuracy: 0.5603 - val_top_k_categorical_accuracy: 0.2262 - lr: 2.5000e-05\n",
      "Epoch 35/50\n",
      "119/119 [==============================] - 8s 71ms/step - loss: 1.0642 - accuracy: 0.6675 - top_k_categorical_accuracy: 0.2907 - val_loss: 1.3293 - val_accuracy: 0.5751 - val_top_k_categorical_accuracy: 0.3171 - lr: 2.5000e-05\n",
      "Epoch 36/50\n",
      "119/119 [==============================] - 9s 72ms/step - loss: 1.0478 - accuracy: 0.6675 - top_k_categorical_accuracy: 0.2981 - val_loss: 1.3622 - val_accuracy: 0.5539 - val_top_k_categorical_accuracy: 0.2431 - lr: 2.5000e-05\n",
      "Epoch 37/50\n",
      "119/119 [==============================] - 10s 82ms/step - loss: 1.0438 - accuracy: 0.6665 - top_k_categorical_accuracy: 0.2738 - val_loss: 1.3310 - val_accuracy: 0.5814 - val_top_k_categorical_accuracy: 0.1924 - lr: 2.5000e-05\n",
      "Epoch 38/50\n",
      "119/119 [==============================] - 9s 71ms/step - loss: 1.0293 - accuracy: 0.6686 - top_k_categorical_accuracy: 0.3018 - val_loss: 1.3290 - val_accuracy: 0.5729 - val_top_k_categorical_accuracy: 0.2283 - lr: 2.5000e-05\n",
      "Epoch 39/50\n",
      "119/119 [==============================] - 8s 71ms/step - loss: 1.0267 - accuracy: 0.6623 - top_k_categorical_accuracy: 0.3002 - val_loss: 1.3630 - val_accuracy: 0.5645 - val_top_k_categorical_accuracy: 0.2283 - lr: 2.5000e-05\n",
      "Epoch 40/50\n",
      "119/119 [==============================] - 9s 72ms/step - loss: 1.0107 - accuracy: 0.6760 - top_k_categorical_accuracy: 0.2854 - val_loss: 1.4598 - val_accuracy: 0.5518 - val_top_k_categorical_accuracy: 0.1734 - lr: 2.5000e-05\n",
      "Epoch 41/50\n",
      "119/119 [==============================] - ETA: 0s - loss: 0.9819 - accuracy: 0.6755 - top_k_categorical_accuracy: 0.3023 \n",
      "Epoch 41: ReduceLROnPlateau reducing learning rate to 1.249999968422344e-05.\n",
      "119/119 [==============================] - 10s 84ms/step - loss: 0.9819 - accuracy: 0.6755 - top_k_categorical_accuracy: 0.3023 - val_loss: 1.3712 - val_accuracy: 0.5624 - val_top_k_categorical_accuracy: 0.2178 - lr: 2.5000e-05\n",
      "Epoch 42/50\n",
      "119/119 [==============================] - 9s 72ms/step - loss: 0.9782 - accuracy: 0.6982 - top_k_categorical_accuracy: 0.2822 - val_loss: 1.3275 - val_accuracy: 0.5835 - val_top_k_categorical_accuracy: 0.2558 - lr: 1.2500e-05\n",
      "Epoch 43/50\n",
      "119/119 [==============================] - 8s 70ms/step - loss: 0.9640 - accuracy: 0.6871 - top_k_categorical_accuracy: 0.2939 - val_loss: 1.3391 - val_accuracy: 0.5666 - val_top_k_categorical_accuracy: 0.2643 - lr: 1.2500e-05\n",
      "Epoch 44/50\n",
      "119/119 [==============================] - 9s 73ms/step - loss: 0.9781 - accuracy: 0.6908 - top_k_categorical_accuracy: 0.2902 - val_loss: 1.3349 - val_accuracy: 0.5877 - val_top_k_categorical_accuracy: 0.2262 - lr: 1.2500e-05\n",
      "Epoch 45/50\n",
      "119/119 [==============================] - ETA: 0s - loss: 0.9261 - accuracy: 0.7082 - top_k_categorical_accuracy: 0.2844 \n",
      "Epoch 45: ReduceLROnPlateau reducing learning rate to 6.24999984211172e-06.\n",
      "119/119 [==============================] - 10s 84ms/step - loss: 0.9261 - accuracy: 0.7082 - top_k_categorical_accuracy: 0.2844 - val_loss: 1.3317 - val_accuracy: 0.5581 - val_top_k_categorical_accuracy: 0.3023 - lr: 1.2500e-05\n",
      "Epoch 46/50\n",
      "119/119 [==============================] - 8s 71ms/step - loss: 0.9500 - accuracy: 0.7003 - top_k_categorical_accuracy: 0.2844 - val_loss: 1.3324 - val_accuracy: 0.5751 - val_top_k_categorical_accuracy: 0.2178 - lr: 6.2500e-06\n",
      "Epoch 47/50\n",
      "119/119 [==============================] - 9s 72ms/step - loss: 0.9063 - accuracy: 0.7236 - top_k_categorical_accuracy: 0.2690 - val_loss: 1.2985 - val_accuracy: 0.5983 - val_top_k_categorical_accuracy: 0.2178 - lr: 6.2500e-06\n",
      "Epoch 48/50\n",
      "119/119 [==============================] - 10s 83ms/step - loss: 0.9065 - accuracy: 0.7156 - top_k_categorical_accuracy: 0.3118 - val_loss: 1.3026 - val_accuracy: 0.5899 - val_top_k_categorical_accuracy: 0.2178 - lr: 6.2500e-06\n",
      "Epoch 49/50\n",
      "119/119 [==============================] - 9s 72ms/step - loss: 0.9199 - accuracy: 0.7172 - top_k_categorical_accuracy: 0.2859 - val_loss: 1.3097 - val_accuracy: 0.5920 - val_top_k_categorical_accuracy: 0.2304 - lr: 6.2500e-06\n",
      "Epoch 50/50\n",
      "119/119 [==============================] - ETA: 0s - loss: 0.9246 - accuracy: 0.7056 - top_k_categorical_accuracy: 0.2907 \n",
      "Epoch 50: ReduceLROnPlateau reducing learning rate to 3.12499992105586e-06.\n",
      "119/119 [==============================] - 8s 71ms/step - loss: 0.9246 - accuracy: 0.7056 - top_k_categorical_accuracy: 0.2907 - val_loss: 1.2989 - val_accuracy: 0.5877 - val_top_k_categorical_accuracy: 0.2410 - lr: 6.2500e-06\n"
     ]
    }
   ],
   "source": [
    "history=model.fit(x_train,validation_data=x_val,epochs=50,callbacks=callbacks)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "a044cce9-ff03-484f-bdc6-9e07dc17a64a",
   "metadata": {},
   "outputs": [],
   "source": [
    "model=load_model('best_model.keras')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "3afc6745-b7e0-4da6-94bb-c2703b241044",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "30/30 [==============================] - 1s 13ms/step - loss: 1.2985 - accuracy: 0.5983 - top_k_categorical_accuracy: 0.2178\n",
      "\n",
      "Test-Resultate:\n",
      "loss: 1.2985\n",
      "accuracy: 0.5983\n",
      "top_k_categorical_accuracy: 0.2178\n"
     ]
    }
   ],
   "source": [
    "results=model.evaluate(x_val)\n",
    "\n",
    "print(\"\\nTest-Resultate:\")\n",
    "for name, value in zip(model.metrics_names, results):\n",
    "    print(f\"{name}: {value:.4f}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "bb37e257-c661-46d7-a290-fc2bb4b06900",
   "metadata": {},
   "outputs": [],
   "source": [
    "with open(\"mushroom_cnn_history.pkl\", \"wb\") as f:\n",
    "    pickle.dump(history.history, f)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "3d106eeb-e01f-452b-a2b4-36afc7a707b2",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x500 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "epochs = range(len(history.history[\"accuracy\"]))\n",
    "\n",
    "plt.figure(figsize=(12, 5))\n",
    "\n",
    "# Accuracy\n",
    "plt.subplot(1, 2, 1)\n",
    "plt.plot(epochs, history.history[\"accuracy\"], label=\"Train Accuracy\")\n",
    "plt.plot(epochs, history.history[\"val_accuracy\"], label=\"Val Accuracy\")\n",
    "plt.title(\"Training vs Validation Accuracy\")\n",
    "plt.xlabel(\"Epoch\")\n",
    "plt.ylabel(\"Accuracy\")\n",
    "plt.legend()\n",
    "\n",
    "# Loss \n",
    "plt.subplot(1, 2, 2)\n",
    "plt.plot(epochs, history.history[\"loss\"], label=\"Train Loss\")\n",
    "plt.plot(epochs, history.history[\"val_loss\"], label=\"Val Loss\")\n",
    "plt.title(\"Training vs Validation Loss\")\n",
    "plt.xlabel(\"Epoch\")\n",
    "plt.ylabel(\"Loss\")\n",
    "plt.legend()\n",
    "\n",
    "plt.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "7e58bd83-cd7b-4c31-8035-12c35ab830c5",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "30/30 [==============================] - 0s 11ms/step\n"
     ]
    }
   ],
   "source": [
    "y_true=x_val.labels\n",
    "\n",
    "y_pred_probs = model.predict(x_val)\n",
    "y_pred=np.argmax(y_pred_probs, axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "5231f39a-345b-4a62-8420-e6b4110a5450",
   "metadata": {},
   "outputs": [],
   "source": [
    "y_true=va_df[\"Mushroom\"].astype(int).values\n",
    "\n",
    "y_pred_probs=model.predict(x_val, verbose=0)\n",
    "y_pred=np.argmax(y_pred_probs, axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "8166e7fd-66b7-4888-8289-454cb59edf4e",
   "metadata": {},
   "outputs": [],
   "source": [
    "class_names=[str(i) for i in sorted(train_df[\"Mushroom\"].unique())]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "0657e7c9-cd3f-4cf1-8caa-02fdebdf43ff",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x800 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "cm = confusion_matrix(y_true, y_pred)\n",
    "\n",
    "plt.figure(figsize=(10,8))\n",
    "sns.heatmap(cm, annot=True, fmt=\"d\",\n",
    "            xticklabels=class_names, yticklabels=class_names)\n",
    "plt.xlabel(\"Predicted\")\n",
    "plt.ylabel(\"True\")\n",
    "plt.title(\"Confusion Matrix\")\n",
    "plt.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "7d2833c9-a316-41cf-aec2-bd34a0f2ba5f",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.79      0.56      0.66        48\n",
      "           1       0.35      0.31      0.33        48\n",
      "           2       0.71      0.62      0.67        48\n",
      "           3       0.56      0.52      0.54        44\n",
      "           4       0.49      0.67      0.57        45\n",
      "           5       0.65      0.62      0.64        48\n",
      "           6       0.66      0.44      0.53        48\n",
      "           7       0.66      0.79      0.72        48\n",
      "           8       0.44      0.60      0.51        48\n",
      "           9       0.80      0.83      0.82        48\n",
      "\n",
      "    accuracy                           0.60       473\n",
      "   macro avg       0.61      0.60      0.60       473\n",
      "weighted avg       0.61      0.60      0.60       473\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(classification_report(y_true, y_pred, target_names=class_names))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cf375f6b-f7f6-4b4a-b2ed-29b36b3694be",
   "metadata": {},
   "source": [
    "### ResNet50 Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "id": "f73d3225-6462-41d1-b4c1-65b6b6d7587b",
   "metadata": {},
   "outputs": [],
   "source": [
    "from tensorflow.keras.applications import ResNet50\n",
    "from tensorflow.keras.applications.resnet50 import preprocess_input"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "id": "a8d5a5ec-4d57-447f-bb49-0e6dfe24a9ca",
   "metadata": {},
   "outputs": [],
   "source": [
    "IMG_SIZE=(224, 224)\n",
    "BATCH=16\n",
    "SEED=42"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "id": "c37bd51e-d97e-4792-b6f0-655ef99dc31b",
   "metadata": {},
   "outputs": [],
   "source": [
    "tr_df, va_df=train_test_split(train_df,test_size=0.2,random_state=SEED,stratify=train_df[\"Mushroom\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "id": "dc8cae42-aa0c-47ea-a7f5-6fe8d4e66640",
   "metadata": {},
   "outputs": [],
   "source": [
    "tr_df[\"Mushroom\"]=tr_df[\"Mushroom\"].astype(str)\n",
    "va_df[\"Mushroom\"]=va_df[\"Mushroom\"].astype(str)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "id": "144e5c40-7ff4-47fc-bdad-239f7565dc81",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Found 1892 validated image filenames belonging to 10 classes.\n",
      "Found 473 validated image filenames belonging to 10 classes.\n"
     ]
    }
   ],
   "source": [
    "train_gen=ImageDataGenerator(\n",
    "    preprocessing_function=preprocess_input,\n",
    "    rotation_range=25,\n",
    "    zoom_range=0.2,\n",
    "    width_shift_range=0.1,\n",
    "    height_shift_range=0.1,\n",
    "    horizontal_flip=True,\n",
    "    brightness_range=(0.8, 1.2),\n",
    "    shear_range=0.1\n",
    ")\n",
    "\n",
    "val_gen=ImageDataGenerator(preprocessing_function=preprocess_input)\n",
    "\n",
    "x_train=train_gen.flow_from_dataframe(\n",
    "    tr_df,\n",
    "    x_col=\"filepath\",\n",
    "    y_col=\"Mushroom\",\n",
    "    target_size=IMG_SIZE,\n",
    "    class_mode=\"sparse\",\n",
    "    batch_size=BATCH,\n",
    "    shuffle=True,\n",
    "    seed=SEED\n",
    ")\n",
    "\n",
    "x_val=val_gen.flow_from_dataframe(\n",
    "    va_df,\n",
    "    x_col=\"filepath\",\n",
    "    y_col=\"Mushroom\",\n",
    "    target_size=IMG_SIZE,\n",
    "    class_mode=\"sparse\",\n",
    "    batch_size=BATCH,\n",
    "    shuffle=False\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "id": "e7ebe457-5d5f-4885-80b3-a8f5ae9062ae",
   "metadata": {},
   "outputs": [],
   "source": [
    "base_model = ResNet50(\n",
    "    weights='imagenet',\n",
    "    include_top=False,\n",
    "    input_shape=(224, 224, 3)\n",
    ")\n",
    "\n",
    "base_model.trainable = False\n",
    "\n",
    "model = Sequential()\n",
    "model.add(base_model)\n",
    "model.add(GlobalAveragePooling2D())\n",
    "BatchNormalization()\n",
    "model.add(Dense(256, activation='relu'))\n",
    "model.add(Dropout(0.5))\n",
    "model.add(Dense(10, activation='softmax'))\n",
    "\n",
    "model.compile(\n",
    "    optimizer=Adam(learning_rate=1e-3),\n",
    "    loss='sparse_categorical_crossentropy',\n",
    "    metrics=['accuracy', tf.keras.metrics.SparseTopKCategoricalAccuracy(k=3, name='top3_acc')]\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "id": "dc688c72-6836-4739-9f6c-06bb9c348d6a",
   "metadata": {},
   "outputs": [],
   "source": [
    "callbacks = [\n",
    "    ModelCheckpoint(\"best_resnet50.keras\", save_best_only=True, monitor=\"val_accuracy\", mode=\"max\"),\n",
    "    EarlyStopping(patience=8, restore_best_weights=True, monitor=\"val_accuracy\", mode=\"max\"),\n",
    "    ReduceLROnPlateau(patience=3, factor=0.5, monitor=\"val_loss\", verbose=1)\n",
    "]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "id": "ab57d431-30b0-4e4d-8cf2-b240a8ddeb64",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/50\n",
      "  4/119 [>.............................] - ETA: 3s - loss: 3.6599 - accuracy: 0.1719 - top3_acc: 0.3438  "
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "119/119 [==============================] - 12s 79ms/step - loss: 1.7810 - accuracy: 0.4160 - top3_acc: 0.6987 - val_loss: 1.1216 - val_accuracy: 0.6173 - val_top3_acc: 0.9091 - lr: 0.0010\n",
      "Epoch 2/50\n",
      "119/119 [==============================] - 9s 74ms/step - loss: 1.1703 - accuracy: 0.5936 - top3_acc: 0.8631 - val_loss: 0.8756 - val_accuracy: 0.7104 - val_top3_acc: 0.9175 - lr: 0.0010\n",
      "Epoch 3/50\n",
      "119/119 [==============================] - 9s 77ms/step - loss: 0.9743 - accuracy: 0.6570 - top3_acc: 0.9080 - val_loss: 0.7526 - val_accuracy: 0.7569 - val_top3_acc: 0.9408 - lr: 0.0010\n",
      "Epoch 4/50\n",
      "119/119 [==============================] - 10s 81ms/step - loss: 0.8386 - accuracy: 0.7252 - top3_acc: 0.9244 - val_loss: 0.7821 - val_accuracy: 0.7188 - val_top3_acc: 0.9218 - lr: 0.0010\n",
      "Epoch 5/50\n",
      "119/119 [==============================] - 9s 73ms/step - loss: 0.7982 - accuracy: 0.7172 - top3_acc: 0.9313 - val_loss: 0.7299 - val_accuracy: 0.7590 - val_top3_acc: 0.9429 - lr: 0.0010\n",
      "Epoch 6/50\n",
      "119/119 [==============================] - 9s 74ms/step - loss: 0.6862 - accuracy: 0.7648 - top3_acc: 0.9471 - val_loss: 0.6672 - val_accuracy: 0.7886 - val_top3_acc: 0.9408 - lr: 0.0010\n",
      "Epoch 7/50\n",
      "119/119 [==============================] - 10s 85ms/step - loss: 0.6570 - accuracy: 0.7701 - top3_acc: 0.9572 - val_loss: 0.6762 - val_accuracy: 0.7907 - val_top3_acc: 0.9471 - lr: 0.0010\n",
      "Epoch 8/50\n",
      "119/119 [==============================] - 8s 71ms/step - loss: 0.6496 - accuracy: 0.7748 - top3_acc: 0.9466 - val_loss: 0.6747 - val_accuracy: 0.7738 - val_top3_acc: 0.9471 - lr: 0.0010\n",
      "Epoch 9/50\n",
      "119/119 [==============================] - ETA: 0s - loss: 0.5629 - accuracy: 0.8002 - top3_acc: 0.9625 \n",
      "Epoch 9: ReduceLROnPlateau reducing learning rate to 0.0005000000237487257.\n",
      "119/119 [==============================] - 8s 70ms/step - loss: 0.5629 - accuracy: 0.8002 - top3_acc: 0.9625 - val_loss: 0.6816 - val_accuracy: 0.7738 - val_top3_acc: 0.9471 - lr: 0.0010\n",
      "Epoch 10/50\n",
      "119/119 [==============================] - 9s 76ms/step - loss: 0.5627 - accuracy: 0.8050 - top3_acc: 0.9715 - val_loss: 0.6320 - val_accuracy: 0.7992 - val_top3_acc: 0.9535 - lr: 5.0000e-04\n",
      "Epoch 11/50\n",
      "119/119 [==============================] - 10s 83ms/step - loss: 0.4621 - accuracy: 0.8340 - top3_acc: 0.9757 - val_loss: 0.6005 - val_accuracy: 0.7928 - val_top3_acc: 0.9556 - lr: 5.0000e-04\n",
      "Epoch 12/50\n",
      "119/119 [==============================] - 9s 76ms/step - loss: 0.4713 - accuracy: 0.8356 - top3_acc: 0.9767 - val_loss: 0.6162 - val_accuracy: 0.8097 - val_top3_acc: 0.9577 - lr: 5.0000e-04\n",
      "Epoch 13/50\n",
      "119/119 [==============================] - 8s 70ms/step - loss: 0.4330 - accuracy: 0.8473 - top3_acc: 0.9799 - val_loss: 0.6319 - val_accuracy: 0.8013 - val_top3_acc: 0.9556 - lr: 5.0000e-04\n",
      "Epoch 14/50\n",
      "119/119 [==============================] - ETA: 0s - loss: 0.4081 - accuracy: 0.8515 - top3_acc: 0.9847 \n",
      "Epoch 14: ReduceLROnPlateau reducing learning rate to 0.0002500000118743628.\n",
      "119/119 [==============================] - 9s 75ms/step - loss: 0.4081 - accuracy: 0.8515 - top3_acc: 0.9847 - val_loss: 0.6269 - val_accuracy: 0.8118 - val_top3_acc: 0.9535 - lr: 5.0000e-04\n",
      "Epoch 15/50\n",
      "119/119 [==============================] - 10s 80ms/step - loss: 0.3987 - accuracy: 0.8578 - top3_acc: 0.9773 - val_loss: 0.6177 - val_accuracy: 0.7865 - val_top3_acc: 0.9641 - lr: 2.5000e-04\n",
      "Epoch 16/50\n",
      "119/119 [==============================] - 8s 71ms/step - loss: 0.3791 - accuracy: 0.8726 - top3_acc: 0.9810 - val_loss: 0.6125 - val_accuracy: 0.8076 - val_top3_acc: 0.9598 - lr: 2.5000e-04\n",
      "Epoch 17/50\n",
      "119/119 [==============================] - 9s 75ms/step - loss: 0.3697 - accuracy: 0.8647 - top3_acc: 0.9826 - val_loss: 0.5907 - val_accuracy: 0.8140 - val_top3_acc: 0.9577 - lr: 2.5000e-04\n",
      "Epoch 18/50\n",
      "119/119 [==============================] - 10s 81ms/step - loss: 0.3556 - accuracy: 0.8805 - top3_acc: 0.9836 - val_loss: 0.5979 - val_accuracy: 0.8118 - val_top3_acc: 0.9535 - lr: 2.5000e-04\n",
      "Epoch 19/50\n",
      "119/119 [==============================] - 9s 72ms/step - loss: 0.3739 - accuracy: 0.8800 - top3_acc: 0.9841 - val_loss: 0.5964 - val_accuracy: 0.8118 - val_top3_acc: 0.9577 - lr: 2.5000e-04\n",
      "Epoch 20/50\n",
      "119/119 [==============================] - ETA: 0s - loss: 0.3501 - accuracy: 0.8774 - top3_acc: 0.9841 \n",
      "Epoch 20: ReduceLROnPlateau reducing learning rate to 0.0001250000059371814.\n",
      "119/119 [==============================] - 8s 71ms/step - loss: 0.3501 - accuracy: 0.8774 - top3_acc: 0.9841 - val_loss: 0.5999 - val_accuracy: 0.8118 - val_top3_acc: 0.9619 - lr: 2.5000e-04\n",
      "Epoch 21/50\n",
      "119/119 [==============================] - 8s 71ms/step - loss: 0.3217 - accuracy: 0.8927 - top3_acc: 0.9873 - val_loss: 0.5937 - val_accuracy: 0.8140 - val_top3_acc: 0.9619 - lr: 1.2500e-04\n",
      "Epoch 22/50\n",
      "119/119 [==============================] - 10s 82ms/step - loss: 0.3143 - accuracy: 0.8948 - top3_acc: 0.9863 - val_loss: 0.6006 - val_accuracy: 0.8076 - val_top3_acc: 0.9577 - lr: 1.2500e-04\n",
      "Epoch 23/50\n",
      "119/119 [==============================] - ETA: 0s - loss: 0.3074 - accuracy: 0.8885 - top3_acc: 0.9889 \n",
      "Epoch 23: ReduceLROnPlateau reducing learning rate to 6.25000029685907e-05.\n",
      "119/119 [==============================] - 9s 71ms/step - loss: 0.3074 - accuracy: 0.8885 - top3_acc: 0.9889 - val_loss: 0.5921 - val_accuracy: 0.8097 - val_top3_acc: 0.9577 - lr: 1.2500e-04\n",
      "Epoch 24/50\n",
      "119/119 [==============================] - 8s 71ms/step - loss: 0.2972 - accuracy: 0.8938 - top3_acc: 0.9937 - val_loss: 0.5879 - val_accuracy: 0.8097 - val_top3_acc: 0.9619 - lr: 6.2500e-05\n",
      "Epoch 25/50\n",
      "119/119 [==============================] - 9s 72ms/step - loss: 0.2946 - accuracy: 0.8938 - top3_acc: 0.9900 - val_loss: 0.5915 - val_accuracy: 0.8076 - val_top3_acc: 0.9556 - lr: 6.2500e-05\n"
     ]
    }
   ],
   "source": [
    "history2=model.fit(x_train,validation_data=x_val,epochs=50,callbacks=callbacks)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "id": "cb966619-6e89-40d6-b72e-563be35ad910",
   "metadata": {},
   "outputs": [],
   "source": [
    "base_model.trainable = True\n",
    "\n",
    "for layer in base_model.layers[:-30]:\n",
    "    layer.trainable = False\n",
    "\n",
    "for layer in base_model.layers:\n",
    "    if isinstance(layer, tf.keras.layers.BatchNormalization):\n",
    "        layer.trainable = False"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "id": "b9bdcf9e-5be5-478d-b853-7b3dd9e0f85d",
   "metadata": {},
   "outputs": [],
   "source": [
    "model.compile(optimizer=tf.keras.optimizers.Adam(1e-5), loss=\"sparse_categorical_crossentropy\",metrics=[\"accuracy\", tf.keras.metrics.SparseTopKCategoricalAccuracy(k=3, name=\"top3_acc\")])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "id": "0e5c095a-500b-4a38-ba22-8a3f21d58c47",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/40\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  1/119 [..............................] - ETA: 7:29 - loss: 0.3300 - accuracy: 0.8750 - top3_acc: 1.0000"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n",
      "'+ptx85' is not a recognized feature for this target (ignoring feature)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "119/119 [==============================] - 14s 84ms/step - loss: 0.3455 - accuracy: 0.8879 - top3_acc: 0.9836 - val_loss: 0.6332 - val_accuracy: 0.8140 - val_top3_acc: 0.9514 - lr: 1.0000e-05\n",
      "Epoch 2/40\n",
      "119/119 [==============================] - 9s 72ms/step - loss: 0.3214 - accuracy: 0.8922 - top3_acc: 0.9868 - val_loss: 0.5918 - val_accuracy: 0.8097 - val_top3_acc: 0.9619 - lr: 1.0000e-05\n",
      "Epoch 3/40\n",
      "119/119 [==============================] - 11s 88ms/step - loss: 0.2636 - accuracy: 0.9091 - top3_acc: 0.9947 - val_loss: 0.5799 - val_accuracy: 0.8203 - val_top3_acc: 0.9577 - lr: 1.0000e-05\n",
      "Epoch 4/40\n",
      "119/119 [==============================] - 9s 72ms/step - loss: 0.2564 - accuracy: 0.9107 - top3_acc: 0.9926 - val_loss: 0.5964 - val_accuracy: 0.7992 - val_top3_acc: 0.9556 - lr: 1.0000e-05\n",
      "Epoch 5/40\n",
      "119/119 [==============================] - 9s 73ms/step - loss: 0.2202 - accuracy: 0.9260 - top3_acc: 0.9952 - val_loss: 0.6165 - val_accuracy: 0.8076 - val_top3_acc: 0.9683 - lr: 1.0000e-05\n",
      "Epoch 6/40\n",
      "119/119 [==============================] - 10s 80ms/step - loss: 0.2101 - accuracy: 0.9271 - top3_acc: 0.9963 - val_loss: 0.5741 - val_accuracy: 0.8224 - val_top3_acc: 0.9662 - lr: 1.0000e-05\n",
      "Epoch 7/40\n",
      "119/119 [==============================] - 11s 88ms/step - loss: 0.1904 - accuracy: 0.9313 - top3_acc: 0.9952 - val_loss: 0.5808 - val_accuracy: 0.8309 - val_top3_acc: 0.9704 - lr: 1.0000e-05\n",
      "Epoch 8/40\n",
      "119/119 [==============================] - 9s 74ms/step - loss: 0.1620 - accuracy: 0.9419 - top3_acc: 0.9974 - val_loss: 0.6014 - val_accuracy: 0.8245 - val_top3_acc: 0.9704 - lr: 1.0000e-05\n",
      "Epoch 9/40\n",
      "119/119 [==============================] - ETA: 0s - loss: 0.1688 - accuracy: 0.9397 - top3_acc: 0.9963 \n",
      "Epoch 9: ReduceLROnPlateau reducing learning rate to 4.999999873689376e-06.\n",
      "119/119 [==============================] - 9s 73ms/step - loss: 0.1688 - accuracy: 0.9397 - top3_acc: 0.9963 - val_loss: 0.6036 - val_accuracy: 0.8266 - val_top3_acc: 0.9704 - lr: 1.0000e-05\n",
      "Epoch 10/40\n",
      "119/119 [==============================] - 11s 90ms/step - loss: 0.1517 - accuracy: 0.9498 - top3_acc: 0.9974 - val_loss: 0.5914 - val_accuracy: 0.8372 - val_top3_acc: 0.9641 - lr: 5.0000e-06\n",
      "Epoch 11/40\n",
      "119/119 [==============================] - 10s 81ms/step - loss: 0.1282 - accuracy: 0.9582 - top3_acc: 0.9984 - val_loss: 0.5695 - val_accuracy: 0.8478 - val_top3_acc: 0.9619 - lr: 5.0000e-06\n",
      "Epoch 12/40\n",
      "119/119 [==============================] - 9s 75ms/step - loss: 0.1319 - accuracy: 0.9561 - top3_acc: 0.9984 - val_loss: 0.5888 - val_accuracy: 0.8457 - val_top3_acc: 0.9746 - lr: 5.0000e-06\n",
      "Epoch 13/40\n",
      "119/119 [==============================] - 10s 80ms/step - loss: 0.1270 - accuracy: 0.9551 - top3_acc: 0.9979 - val_loss: 0.6097 - val_accuracy: 0.8520 - val_top3_acc: 0.9704 - lr: 5.0000e-06\n",
      "Epoch 14/40\n",
      "119/119 [==============================] - ETA: 0s - loss: 0.1180 - accuracy: 0.9614 - top3_acc: 0.9979 \n",
      "Epoch 14: ReduceLROnPlateau reducing learning rate to 2.499999936844688e-06.\n",
      "119/119 [==============================] - 10s 86ms/step - loss: 0.1180 - accuracy: 0.9614 - top3_acc: 0.9979 - val_loss: 0.5956 - val_accuracy: 0.8436 - val_top3_acc: 0.9683 - lr: 5.0000e-06\n",
      "Epoch 15/40\n",
      "119/119 [==============================] - 9s 75ms/step - loss: 0.0981 - accuracy: 0.9699 - top3_acc: 0.9989 - val_loss: 0.5972 - val_accuracy: 0.8478 - val_top3_acc: 0.9662 - lr: 2.5000e-06\n",
      "Epoch 16/40\n",
      "119/119 [==============================] - 9s 74ms/step - loss: 0.0929 - accuracy: 0.9699 - top3_acc: 0.9995 - val_loss: 0.6012 - val_accuracy: 0.8457 - val_top3_acc: 0.9704 - lr: 2.5000e-06\n",
      "Epoch 17/40\n",
      "119/119 [==============================] - ETA: 0s - loss: 0.0762 - accuracy: 0.9730 - top3_acc: 0.9989 \n",
      "Epoch 17: ReduceLROnPlateau reducing learning rate to 1.249999968422344e-06.\n",
      "119/119 [==============================] - 10s 85ms/step - loss: 0.0762 - accuracy: 0.9730 - top3_acc: 0.9989 - val_loss: 0.6073 - val_accuracy: 0.8520 - val_top3_acc: 0.9662 - lr: 2.5000e-06\n",
      "Epoch 18/40\n",
      "119/119 [==============================] - 9s 75ms/step - loss: 0.0836 - accuracy: 0.9715 - top3_acc: 1.0000 - val_loss: 0.6088 - val_accuracy: 0.8520 - val_top3_acc: 0.9683 - lr: 1.2500e-06\n",
      "Epoch 19/40\n",
      "119/119 [==============================] - 9s 74ms/step - loss: 0.0929 - accuracy: 0.9672 - top3_acc: 0.9989 - val_loss: 0.6178 - val_accuracy: 0.8499 - val_top3_acc: 0.9662 - lr: 1.2500e-06\n",
      "Epoch 20/40\n",
      "119/119 [==============================] - ETA: 0s - loss: 0.0845 - accuracy: 0.9725 - top3_acc: 0.9974 \n",
      "Epoch 20: ReduceLROnPlateau reducing learning rate to 6.24999984211172e-07.\n",
      "119/119 [==============================] - 9s 77ms/step - loss: 0.0845 - accuracy: 0.9725 - top3_acc: 0.9974 - val_loss: 0.6239 - val_accuracy: 0.8520 - val_top3_acc: 0.9704 - lr: 1.2500e-06\n",
      "Epoch 21/40\n",
      "119/119 [==============================] - 10s 88ms/step - loss: 0.0642 - accuracy: 0.9810 - top3_acc: 1.0000 - val_loss: 0.6286 - val_accuracy: 0.8520 - val_top3_acc: 0.9662 - lr: 6.2500e-07\n"
     ]
    }
   ],
   "source": [
    "history2=model.fit(x_train, validation_data=x_val, epochs=40, callbacks=callbacks)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "id": "a574eb89-3528-4336-9a52-2d3cc2f20d41",
   "metadata": {},
   "outputs": [],
   "source": [
    "model.save(\"resnet50_mushroom_best.keras\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "id": "c48fecf8-d041-4a9c-9946-e732d322490a",
   "metadata": {},
   "outputs": [],
   "source": [
    "model=load_model(\"resnet50_mushroom_best.keras\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "id": "732f4c85-05b8-4e39-a285-0d6d22ad9a88",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "30/30 [==============================] - 2s 17ms/step - loss: 0.6097 - accuracy: 0.8520 - top3_acc: 0.9704\n",
      "\n",
      "Validation Results:\n",
      "loss: 0.6097\n",
      "accuracy: 0.8520\n",
      "top3_acc: 0.9704\n"
     ]
    }
   ],
   "source": [
    "results=model.evaluate(x_val, verbose=1)\n",
    "print(\"\\nValidation Results:\")\n",
    "for name, value in zip(model.metrics_names, results):\n",
    "    print(f\"{name}: {value:.4f}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "id": "b12e37a8-c9de-435b-86d9-a9eed768718e",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "30/30 [==============================] - 1s 22ms/step\n"
     ]
    }
   ],
   "source": [
    "y_true = x_val.labels\n",
    "\n",
    "y_pred_probs = model.predict(x_val, verbose=1)\n",
    "y_pred = np.argmax(y_pred_probs, axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "id": "b227164e-0626-4112-a652-bc925128325b",
   "metadata": {},
   "outputs": [],
   "source": [
    "class_names=[str(i) for i in range(10)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "id": "bbcadf14-7cf1-418f-9c5a-ce1b405167f9",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Unique y_true: [0 1 2 3 4 5 6 7 8 9]\n",
      "Num classes predicted: 10\n"
     ]
    }
   ],
   "source": [
    "print(\"Unique y_true:\", np.unique(y_true))\n",
    "print(\"Num classes predicted:\", y_pred_probs.shape[1])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "id": "79647af3-6474-45f2-9fb0-2daee7eb1a8b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x800 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "cm = confusion_matrix(y_true, y_pred, labels=np.arange(len(class_names)))\n",
    "\n",
    "disp = ConfusionMatrixDisplay(confusion_matrix=cm, display_labels=class_names)\n",
    "fig, ax = plt.subplots(figsize=(10, 8))\n",
    "disp.plot(ax=ax, cmap=\"Blues\", colorbar=True, values_format=\"d\")\n",
    "plt.title(\"Confusion Matrix (Validation)\")\n",
    "plt.xticks(rotation=45)\n",
    "plt.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "id": "594f6a46-773b-4f28-9bb4-c7942d24ef17",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0     0.9130    0.8750    0.8936        48\n",
      "           1     0.7800    0.8125    0.7959        48\n",
      "           2     0.9149    0.8958    0.9053        48\n",
      "           3     0.7333    0.7500    0.7416        44\n",
      "           4     0.8372    0.8000    0.8182        45\n",
      "           5     0.8636    0.7917    0.8261        48\n",
      "           6     0.8980    0.9167    0.9072        48\n",
      "           7     0.9200    0.9583    0.9388        48\n",
      "           8     0.7255    0.7708    0.7475        48\n",
      "           9     0.9375    0.9375    0.9375        48\n",
      "\n",
      "    accuracy                         0.8520       473\n",
      "   macro avg     0.8523    0.8508    0.8512       473\n",
      "weighted avg     0.8534    0.8520    0.8523       473\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(classification_report(y_true, y_pred, target_names=class_names, digits=4))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "id": "600558f3-4e94-4cdd-abff-79211ce65277",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x500 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "epochs=range(len(history2.history[\"accuracy\"]))\n",
    "\n",
    "plt.figure(figsize=(12, 5))\n",
    "\n",
    "# Accuracy\n",
    "plt.subplot(1, 2, 1)\n",
    "plt.plot(epochs, history2.history[\"accuracy\"], label=\"Train Accuracy\")\n",
    "plt.plot(epochs, history2.history[\"val_accuracy\"], label=\"Val Accuracy\")\n",
    "plt.title(\"Training vs Validation Accuracy\")\n",
    "plt.xlabel(\"Epoch\")\n",
    "plt.ylabel(\"Accuracy\")\n",
    "plt.legend()\n",
    "\n",
    "# Loss\n",
    "plt.subplot(1, 2, 2)\n",
    "plt.plot(epochs, history2.history[\"loss\"], label=\"Train Loss\")\n",
    "plt.plot(epochs, history2.history[\"val_loss\"], label=\"Val Loss\")\n",
    "plt.title(\"Training vs Validation Loss\")\n",
    "plt.xlabel(\"Epoch\")\n",
    "plt.ylabel(\"Loss\")\n",
    "plt.legend()\n",
    "\n",
    "plt.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "id": "c3e8d2a0-dda2-492a-a1cb-fad6d789c8a5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Image</th>\n",
       "      <th>filepath</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2372</td>\n",
       "      <td>dataset/02372.jpg</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2373</td>\n",
       "      <td>dataset/02373.jpg</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2374</td>\n",
       "      <td>dataset/02374.jpg</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2375</td>\n",
       "      <td>dataset/02375.jpg</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2376</td>\n",
       "      <td>dataset/02376.jpg</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Image           filepath\n",
       "0   2372  dataset/02372.jpg\n",
       "1   2373  dataset/02373.jpg\n",
       "2   2374  dataset/02374.jpg\n",
       "3   2375  dataset/02375.jpg\n",
       "4   2376  dataset/02376.jpg"
      ]
     },
     "execution_count": 89,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "id": "b447c2cd-4b17-4f2e-854f-a0c4ef45c39f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Id</th>\n",
       "      <th>Predicted</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Id  Predicted\n",
       "0   0          0\n",
       "1   1          1\n",
       "2   2          2\n",
       "3   3          3\n",
       "4   4          4"
      ]
     },
     "execution_count": 94,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sub_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 95,
   "id": "cd4761b6-67ba-45db-8038-909563b4ae03",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Image</th>\n",
       "      <th>Mushroom</th>\n",
       "      <th>filepath</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>dataset/00001.jpg</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>dataset/00002.jpg</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>dataset/00003.jpg</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>dataset/00004.jpg</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>0</td>\n",
       "      <td>dataset/00005.jpg</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Image  Mushroom           filepath\n",
       "0      1         0  dataset/00001.jpg\n",
       "1      2         0  dataset/00002.jpg\n",
       "2      3         0  dataset/00003.jpg\n",
       "3      4         0  dataset/00004.jpg\n",
       "4      5         0  dataset/00005.jpg"
      ]
     },
     "execution_count": 95,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "id": "723ac071-90bc-4731-9ebe-95f23584de11",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Found 598 validated image filenames.\n"
     ]
    }
   ],
   "source": [
    "test_gen=ImageDataGenerator(\n",
    "    preprocessing_function=preprocess_input\n",
    ")\n",
    "\n",
    "x_test=test_gen.flow_from_dataframe(\n",
    "    test_df,\n",
    "    x_col=\"filepath\",\n",
    "    y_col=None,\n",
    "    target_size=IMG_SIZE,\n",
    "    class_mode=None,\n",
    "    batch_size=BATCH,\n",
    "    shuffle=False\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "id": "0555cac0-6406-45f8-b5ed-51b1706bf5cb",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "38/38 [==============================] - 2s 25ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "W0000 00:00:1766602629.193990   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.194493   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.195014   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.195374   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.195783   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.196055   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.196464   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.197073   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.197468   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.197897   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.198240   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.199051   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.204456   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.204670   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.204847   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.205075   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.205228   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.205436   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.205657   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.205822   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.206020   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.206716   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.206999   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.207326   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.207569   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.207975   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.208334   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.208702   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.210925   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.217163   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.219795   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.219956   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.220114   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.220261   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.220428   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.220599   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.220777   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.220987   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.221187   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.221401   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.221622   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.221844   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.222103   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.222360   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.222683   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.223885   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.223999   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.224126   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.224257   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.224375   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.224510   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.224598   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.224690   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.224793   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.224901   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.225021   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.225138   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.225276   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.225415   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.225538   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.225650   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.228325   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.228475   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.228637   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.228797   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.228934   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.229114   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.229361   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.229548   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.229690   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.230044   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.230387   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.230734   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.230893   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.231164   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.231481   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.231784   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.232317   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.233129   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.240266   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.240463   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.240651   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.240812   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.240981   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.241157   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.241343   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.241538   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.241725   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.241912   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.242092   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.242306   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.242508   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.242761   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.243067   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.244211   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.244331   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.244429   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.244520   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.244615   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.244710   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.244819   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.244921   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.245038   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.245169   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.245299   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.245412   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.245521   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.245626   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.245734   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.245858   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.248370   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.248527   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.248660   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.248804   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.248968   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.249130   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.249364   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.249615   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.249812   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.250037   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.250367   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.250687   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.251105   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.251423   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.251750   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.252080   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.253160   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.258083   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.260033   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.260306   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.260573   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.260838   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.261006   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.261312   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.261626   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.261896   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.262189   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.262413   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.262642   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.262952   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.263163   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.263529   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.263851   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.264900   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.265020   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.265128   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.265244   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.265364   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.265492   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.265607   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.265736   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.265870   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.266013   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.266158   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.266302   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.266462   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.266618   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.266789   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.266970   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.269732   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.269944   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.270080   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.270279   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.270535   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.270807   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.271094   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.271419   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.271773   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.272135   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.272510   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.272930   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.273565   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.274131   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.274680   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.275305   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.279095   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n",
      "W0000 00:00:1766602629.284378   19378 gpu_timer.cc:114] Skipping the delay kernel, measurement accuracy will be reduced\n"
     ]
    }
   ],
   "source": [
    "y_test_probs=model.predict(x_test)\n",
    "y_test_pred=np.argmax(y_test_probs, axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 96,
   "id": "6bb78e36-1fb5-4547-9039-7bbd7104fddc",
   "metadata": {},
   "outputs": [],
   "source": [
    "submission=pd.DataFrame({'Id': test_df['Image'],'Mushroom': y_test_pred})\n",
    "submission.to_csv('submission.csv', index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 97,
   "id": "27f2eab0-7c0c-40ef-b89f-f7057c6a6c39",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Id</th>\n",
       "      <th>Mushroom</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2372</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2373</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2374</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2375</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2376</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     Id  Mushroom\n",
       "0  2372         8\n",
       "1  2373         2\n",
       "2  2374         2\n",
       "3  2375         8\n",
       "4  2376         4"
      ]
     },
     "execution_count": 97,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "submission.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "011c94f5-177c-4fa6-8e4b-9933c496d3cd",
   "metadata": {},
   "source": [
    "### Conclusion\n",
    "The final model based on ResNet50 with fine-tuning achieved strong results.\n",
    "On the validation dataset, the model reached an accuracy of 85.2% and a Top-3 accuracy of 97.0%.\n",
    "\n",
    "These results show that transfer learning significantly improved performance compared to the basic CNN.\n",
    "The high Top-3 accuracy indicates that the correct class is almost always among the top predictions.\n",
    "Overall, the model performs well and is suitable for real-world mushroom image classification tasks."
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.12.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}