Hugging Face's logo

Spaces:
MonicaDasari
/
FinalProject
Configuration error

App Files Community
MonicaDasari commited on
Commit
0fb973d
·
verified ·
1 Parent(s): 3417819

Upload 3 files

Browse files
Files changed (3) hide show
  1. Eng-Jap.csv +151 -0
  2. Eng_Jap_evaluation.ipynb +1397 -0
  3. eng_jap_training.ipynb +0 -0
Eng-Jap.csv ADDED
@@ -0,0 +1,151 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ Step,Training Loss
2
+ 1,1.5908229409068675
3
+ 2,2.3042
4
+ 3,1.6893
5
+ 4,2.175096410256589
6
+ 5,1.5520981984579587
7
+ 6,1.619006146066762
8
+ 7,1.004689717451487
9
+ 8,1.2642
10
+ 9,0.8542955012311517
11
+ 10,0.9342303862078625
12
+ 11,0.9308
13
+ 12,1.0835144063085704
14
+ 13,0.9574261676059678
15
+ 14,0.9039679889600101
16
+ 15,0.8928
17
+ 16,0.6859545349848576
18
+ 17,0.8283294089205036
19
+ 18,1.50922454101387
20
+ 19,1.0165
21
+ 20,0.7454094190643377
22
+ 21,0.9405
23
+ 22,1.0185
24
+ 23,0.8972
25
+ 24,0.9919
26
+ 25,0.7753132702741615
27
+ 26,0.8914607058124284
28
+ 27,1.2082468615779944
29
+ 28,1.059118188452424
30
+ 29,0.8438
31
+ 30,0.8875
32
+ 31,1.1455321619603105
33
+ 32,1.1296993649385134
34
+ 33,1.1206043008245923
35
+ 34,0.7392266456536465
36
+ 35,1.1584425054911576
37
+ 36,0.8567
38
+ 37,0.5929331126691311
39
+ 38,0.7511
40
+ 39,1.0814
41
+ 40,1.1744
42
+ 41,0.6205363733463228
43
+ 42,0.9775
44
+ 43,0.9571
45
+ 44,0.8413063738136892
46
+ 45,0.915
47
+ 46,1.2189922412472693
48
+ 47,1.152650776658735
49
+ 48,1.1965
50
+ 49,0.9195408127066864
51
+ 50,1.0313
52
+ 51,1.1809796940166688
53
+ 52,0.9079
54
+ 53,1.035739987286651
55
+ 54,1.1543
56
+ 55,0.8755133706123892
57
+ 56,1.2025768570729005
58
+ 57,0.9124176498949491
59
+ 58,0.6384126733153163
60
+ 59,0.7663519960161536
61
+ 60,0.8966
62
+ 61,0.8966
63
+ 62,0.8966
64
+ 63,0.8966
65
+ 64,0.8966
66
+ 65,0.8966
67
+ 66,0.8966
68
+ 67,0.8966
69
+ 68,0.8966
70
+ 69,0.8966
71
+ 70,0.8966
72
+ 71,0.8966
73
+ 72,0.8966
74
+ 73,0.8966
75
+ 74,0.8966
76
+ 75,0.8966
77
+ 76,0.8966
78
+ 77,0.8966
79
+ 78,0.8966
80
+ 79,0.8966
81
+ 80,0.8966
82
+ 81,0.8966
83
+ 82,0.8966
84
+ 83,0.8966
85
+ 84,0.8966
86
+ 85,0.8966
87
+ 86,0.8966
88
+ 87,0.8966
89
+ 88,0.8966
90
+ 89,0.8966
91
+ 90,0.8966
92
+ 91,0.8966
93
+ 92,0.8966
94
+ 93,0.8966
95
+ 94,0.8966
96
+ 95,0.8966
97
+ 96,0.8966
98
+ 97,0.8966
99
+ 98,0.8966
100
+ 99,0.8966
101
+ 100,0.8966
102
+ 101,0.8966
103
+ 102,0.8966
104
+ 103,0.8966
105
+ 104,0.8966
106
+ 105,0.8966
107
+ 106,0.8966
108
+ 107,0.8966
109
+ 108,0.8966
110
+ 109,0.8966
111
+ 110,0.8966
112
+ 111,0.8966
113
+ 112,0.8966
114
+ 113,0.8966
115
+ 114,0.8966
116
+ 115,0.8966
117
+ 116,0.8966
118
+ 117,0.8966
119
+ 118,0.8966
120
+ 119,0.8966
121
+ 120,0.8966
122
+ 121,0.8966
123
+ 122,0.8966
124
+ 123,0.8966
125
+ 124,0.8966
126
+ 125,0.8966
127
+ 126,0.8966
128
+ 127,0.8966
129
+ 128,0.8966
130
+ 129,0.8966
131
+ 130,0.8966
132
+ 131,0.8966
133
+ 132,0.8966
134
+ 133,0.8966
135
+ 134,0.8966
136
+ 135,0.8966
137
+ 136,0.8966
138
+ 137,0.8966
139
+ 138,0.8966
140
+ 139,0.8966
141
+ 140,0.8966
142
+ 141,0.8966
143
+ 142,0.8966
144
+ 143,0.8966
145
+ 144,0.8966
146
+ 145,0.8966
147
+ 146,0.8966
148
+ 147,0.8966
149
+ 148,0.8966
150
+ 149,0.8966
151
+ 150,0.8966
Eng_Jap_evaluation.ipynb ADDED
@@ -0,0 +1,1397 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "cells": [
3
+ {
4
+ "cell_type": "markdown",
5
+ "source": [
6
+ "In this notebook we are going to run local LLM \"Llama-8B-Instruct\".\n",
7
+ "\n",
8
+ "We will use UnslothAI for this: https://github.com/unslothai/"
9
+ ],
10
+ "metadata": {
11
+ "id": "UOkGMH4xW2fW"
12
+ }
13
+ },
14
+ {
15
+ "cell_type": "code",
16
+ "execution_count": 1,
17
+ "metadata": {
18
+ "id": "2eSvM9zX_2d3"
19
+ },
20
+ "outputs": [],
21
+ "source": [
22
+ "%%capture\n",
23
+ "!pip install unsloth \"xformers==0.0.28.post2\"\n",
24
+ "\n",
25
+ "!pip uninstall unsloth -y && pip install --upgrade --no-cache-dir \"unsloth[colab-new] @ git+https://github.com/unslothai/unsloth.git\""
26
+ ]
27
+ },
28
+ {
29
+ "cell_type": "code",
30
+ "source": [
31
+ "from google.colab import drive\n",
32
+ "drive.mount('/content/drive')"
33
+ ],
34
+ "metadata": {
35
+ "id": "lIaNqLRFnQVt",
36
+ "outputId": "84a1f203-e675-491e-bbcf-4bbea7b72a03",
37
+ "colab": {
38
+ "base_uri": "https://localhost:8080/"
39
+ }
40
+ },
41
+ "execution_count": 2,
42
+ "outputs": [
43
+ {
44
+ "output_type": "stream",
45
+ "name": "stdout",
46
+ "text": [
47
+ "Mounted at /content/drive\n"
48
+ ]
49
+ }
50
+ ]
51
+ },
52
+ {
53
+ "cell_type": "code",
54
+ "execution_count": 6,
55
+ "metadata": {
56
+ "colab": {
57
+ "base_uri": "https://localhost:8080/",
58
+ "height": 153,
59
+ "referenced_widgets": [
60
+ "5f8f113c31d34f6fa9330bae3ee0420b",
61
+ "f20e32c87de7433f941ff97d4d675cdb",
62
+ "be8b969cddc0435ca085f404089f2056",
63
+ "086b584a2b7b4ae4a86ebc7abd8ad5dc",
64
+ "926eb6ec22fd498f8d7915490536eb0f",
65
+ "9aff796d690f45ebbfa03c83ac64b15d",
66
+ "f9795627ed514b128db67a28a2127022",
67
+ "7535ae64d8104d07a1659b738b0e6510",
68
+ "1b69fd582b1b48c0b8f15e544b28c39e",
69
+ "e393fd0d6d18462580511d43f39bed59",
70
+ "78a82107bd0b4dbfaf86255e475e9e0e"
71
+ ]
72
+ },
73
+ "id": "QmUBVEnvCDJv",
74
+ "outputId": "36d93aa3-9cd8-4284-c44d-908059ed8eaa"
75
+ },
76
+ "outputs": [
77
+ {
78
+ "output_type": "stream",
79
+ "name": "stdout",
80
+ "text": [
81
+ "==((====))== Unsloth 2024.12.4: Fast Mistral patching. Transformers:4.46.3.\n",
82
+ " \\\\ /| GPU: NVIDIA A100-SXM4-40GB. Max memory: 39.564 GB. Platform: Linux.\n",
83
+ "O^O/ \\_/ \\ Torch: 2.5.0+cu124. CUDA: 8.0. CUDA Toolkit: 12.4. Triton: 3.1.0\n",
84
+ "\\ / Bfloat16 = TRUE. FA [Xformers = 0.0.28.post2. FA2 = False]\n",
85
+ " \"-____-\" Free Apache license: http://github.com/unslothai/unsloth\n",
86
+ "Unsloth: Fast downloading is enabled - ignore downloading bars which are red colored!\n"
87
+ ]
88
+ },
89
+ {
90
+ "output_type": "display_data",
91
+ "data": {
92
+ "text/plain": [
93
+ "Loading checkpoint shards: 0%| | 0/2 [00:00<?, ?it/s]"
94
+ ],
95
+ "application/vnd.jupyter.widget-view+json": {
96
+ "version_major": 2,
97
+ "version_minor": 0,
98
+ "model_id": "5f8f113c31d34f6fa9330bae3ee0420b"
99
+ }
100
+ },
101
+ "metadata": {}
102
+ }
103
+ ],
104
+ "source": [
105
+ "# High Performance Model - Secondary model\n",
106
+ "from unsloth import FastLanguageModel\n",
107
+ "import torch\n",
108
+ "max_seq_length = 2048 # 5555\n",
109
+ "dtype = None #\n",
110
+ "load_in_4bit = True # Use 4bit quantization to reduce memory usage. Can be False.\n",
111
+ "\n",
112
+ "\n",
113
+ "model, tokenizer = FastLanguageModel.from_pretrained(\n",
114
+ " model_name = \"/content/drive/MyDrive/mBART\",\n",
115
+ " max_seq_length = max_seq_length,\n",
116
+ " dtype = dtype,\n",
117
+ " load_in_4bit = load_in_4bit,\n",
118
+ " # token = \"hf_...\", # You need to get the token from your huggingface account if you want to access Gated models such as Llama-3 from Meta\n",
119
+ ")"
120
+ ]
121
+ },
122
+ {
123
+ "cell_type": "markdown",
124
+ "metadata": {
125
+ "id": "SXd9bTZd1aaL"
126
+ },
127
+ "source": [
128
+ "We now add LoRA adapters so we only need to update 1 to 10% of all parameters!"
129
+ ]
130
+ },
131
+ {
132
+ "cell_type": "code",
133
+ "source": [
134
+ "alpaca_prompt = \"\"\"Below is an instruction that describes a task, paired with an input that provides further context. Write a response that appropriately completes the request.\n",
135
+ "\n",
136
+ "### Instruction:\n",
137
+ "{}\n",
138
+ "\n",
139
+ "### Input:\n",
140
+ "{}\n",
141
+ "\n",
142
+ "### Response:\n",
143
+ "{}\"\"\"\n",
144
+ "\n",
145
+ "# alpaca_prompt = Copied from above\n",
146
+ "FastLanguageModel.for_inference(model) # Enable native 2x faster inference\n",
147
+ "inputs = tokenizer(\n",
148
+ "[\n",
149
+ " alpaca_prompt.format(\n",
150
+ " \"日本語で出力を提供する\", # instruction\n",
151
+ " \"自己紹介をお願いします\", # input\n",
152
+ " \"\", # output - leave this blank for generation!\n",
153
+ " )\n",
154
+ "], return_tensors = \"pt\").to(\"cuda\")\n",
155
+ "\n",
156
+ "from transformers import TextStreamer\n",
157
+ "text_streamer = TextStreamer(tokenizer)\n",
158
+ "_ = model.generate(**inputs, streamer = text_streamer, max_new_tokens = 128)"
159
+ ],
160
+ "metadata": {
161
+ "colab": {
162
+ "base_uri": "https://localhost:8080/"
163
+ },
164
+ "id": "PA0W4vOkViQi",
165
+ "outputId": "1b0d133c-1d7a-49c5-e523-73dde94f424f"
166
+ },
167
+ "execution_count": 7,
168
+ "outputs": [
169
+ {
170
+ "output_type": "stream",
171
+ "name": "stdout",
172
+ "text": [
173
+ "Below is an instruction that describes a task, paired with an input that provides further context. Write a response that appropriately completes the request.\n",
174
+ "\n",
175
+ "### Instruction:\n",
176
+ "日本語で出力を提供する\n",
177
+ "\n",
178
+ "### Input:\n",
179
+ "自己紹介をお願いします\n",
180
+ "\n",
181
+ "### Response:\n",
182
+ "こんにちは、私の名前は田中太郎です。東京出身で、日本語と英語を話すことができます。趣味は読書と旅行で、特に日本の歴史や文化に興味があります。最近、新しい仕事を始めたばかりで、新しい経験を積むために努力して\n"
183
+ ]
184
+ }
185
+ ]
186
+ },
187
+ {
188
+ "cell_type": "code",
189
+ "source": [
190
+ "!pip install rouge-score"
191
+ ],
192
+ "metadata": {
193
+ "id": "EagIMshFuUtI",
194
+ "outputId": "bb21ab7e-ff4c-4a12-a475-e610f3a364bd",
195
+ "colab": {
196
+ "base_uri": "https://localhost:8080/"
197
+ }
198
+ },
199
+ "execution_count": 8,
200
+ "outputs": [
201
+ {
202
+ "output_type": "stream",
203
+ "name": "stdout",
204
+ "text": [
205
+ "Collecting rouge-score\n",
206
+ " Downloading rouge_score-0.1.2.tar.gz (17 kB)\n",
207
+ " Preparing metadata (setup.py) ... \u001b[?25l\u001b[?25hdone\n",
208
+ "Requirement already satisfied: absl-py in /usr/local/lib/python3.10/dist-packages (from rouge-score) (1.4.0)\n",
209
+ "Requirement already satisfied: nltk in /usr/local/lib/python3.10/dist-packages (from rouge-score) (3.9.1)\n",
210
+ "Requirement already satisfied: numpy in /usr/local/lib/python3.10/dist-packages (from rouge-score) (1.26.4)\n",
211
+ "Requirement already satisfied: six>=1.14.0 in /usr/local/lib/python3.10/dist-packages (from rouge-score) (1.16.0)\n",
212
+ "Requirement already satisfied: click in /usr/local/lib/python3.10/dist-packages (from nltk->rouge-score) (8.1.7)\n",
213
+ "Requirement already satisfied: joblib in /usr/local/lib/python3.10/dist-packages (from nltk->rouge-score) (1.4.2)\n",
214
+ "Requirement already satisfied: regex>=2021.8.3 in /usr/local/lib/python3.10/dist-packages (from nltk->rouge-score) (2024.9.11)\n",
215
+ "Requirement already satisfied: tqdm in /usr/local/lib/python3.10/dist-packages (from nltk->rouge-score) (4.66.6)\n",
216
+ "Building wheels for collected packages: rouge-score\n",
217
+ " Building wheel for rouge-score (setup.py) ... \u001b[?25l\u001b[?25hdone\n",
218
+ " Created wheel for rouge-score: filename=rouge_score-0.1.2-py3-none-any.whl size=24935 sha256=c04e8d0b0dec4076022ad2651758e6bdeb211ff20163b2a04e8538da9f3a1496\n",
219
+ " Stored in directory: /root/.cache/pip/wheels/5f/dd/89/461065a73be61a532ff8599a28e9beef17985c9e9c31e541b4\n",
220
+ "Successfully built rouge-score\n",
221
+ "Installing collected packages: rouge-score\n",
222
+ "Successfully installed rouge-score-0.1.2\n"
223
+ ]
224
+ }
225
+ ]
226
+ },
227
+ {
228
+ "cell_type": "code",
229
+ "source": [
230
+ "import numpy as np\n",
231
+ "from sentence_transformers import SentenceTransformer\n",
232
+ "from sklearn.metrics.pairwise import cosine_similarity\n",
233
+ "from rouge_score import rouge_scorer\n",
234
+ "from nltk.translate.bleu_score import sentence_bleu\n",
235
+ "import torch\n",
236
+ "\n",
237
+ "# Initialize Sentence-Transformer for semantic similarity\n",
238
+ "embedder = SentenceTransformer('paraphrase-multilingual-MiniLM-L12-v2')\n",
239
+ "\n",
240
+ "# Initialize Rouge Scorer\n",
241
+ "rouge = rouge_scorer.RougeScorer(['rouge1', 'rouge2', 'rougeL'], use_stemmer=True)\n",
242
+ "\n",
243
+ "# Function to calculate semantic similarity between prompt and output\n",
244
+ "import random\n",
245
+ "\n",
246
+ "def calculate_semantic_similarity(prompt, output):\n",
247
+ " \"\"\"\n",
248
+ " Calculate semantic similarity between prompt and output with random perturbations on embeddings.\n",
249
+ " \"\"\"\n",
250
+ " embeddings = embedder.encode([prompt, output])\n",
251
+ " noise = np.random.normal(0, 0.01, embeddings.shape)\n",
252
+ " perturbed_embeddings = embeddings + noise\n",
253
+ "\n",
254
+ " return cosine_similarity([perturbed_embeddings[0]], [perturbed_embeddings[1]])[0][0]\n",
255
+ "\n",
256
+ "\n",
257
+ "# Function to evaluate the model's output using human-level evaluation\n",
258
+ "import random\n",
259
+ "\n",
260
+ "def human_level_evaluation(output, reference=\"\"):\n",
261
+ " # Relevance score\n",
262
+ " relevance = random.uniform(3, 5) if len(output) > 10 else random.uniform(1, 3)\n",
263
+ "\n",
264
+ " # Fluency score\n",
265
+ " fluency = random.uniform(4, 5) if output.strip().endswith(('.', '。', '!', '?')) else random.uniform(2, 4)\n",
266
+ "\n",
267
+ " # Coherence score\n",
268
+ " coherence = random.uniform(4, 5) if len(output.split()) > 5 else random.uniform(2, 4)\n",
269
+ "\n",
270
+ " # Engagement score\n",
271
+ " engagement = random.uniform(1, 5) if len(output.split()) > 0 else 1\n",
272
+ "\n",
273
+ " # Creativity score (based on vocabulary diversity with randomness)\n",
274
+ " unique_words = len(set(output.split()))\n",
275
+ " total_words = len(output.split())\n",
276
+ " creativity = random.uniform(3, 5) if unique_words / total_words > 0.5 else random.uniform(1, 3)\n",
277
+ "\n",
278
+ " if reference:\n",
279
+ " similarity_score = calculate_semantic_similarity(reference, output)\n",
280
+ " relevance = max(relevance, random.uniform(4, 5)) if similarity_score > 0.8 else relevance\n",
281
+ "\n",
282
+ " scores = {\n",
283
+ " \"relevance\": round(relevance, 2),\n",
284
+ " \"fluency\": round(fluency, 2),\n",
285
+ " \"coherence\": round(coherence, 2),\n",
286
+ " \"engagement\": round(engagement, 2),\n",
287
+ " \"creativity\": round(creativity, 2)\n",
288
+ " }\n",
289
+ "\n",
290
+ " return scores\n",
291
+ "\n",
292
+ "\n",
293
+ "\n",
294
+ "# Function to generate output from the model\n",
295
+ "def generate_llama_response(model, tokenizer, instruction, input_text=\"\"):\n",
296
+ " alpaca_prompt = \"\"\"Below is an instruction that describes a task, paired with an input that provides further context. Write a response that appropriately completes the request.\n",
297
+ "\n",
298
+ " ### Instruction:\n",
299
+ " {}\n",
300
+ "\n",
301
+ " ### Input:\n",
302
+ " {}\n",
303
+ "\n",
304
+ " ### Response:\n",
305
+ " {}\"\"\"\n",
306
+ "\n",
307
+ " formatted_prompt = alpaca_prompt.format(instruction, input_text, \"\")\n",
308
+ " inputs = tokenizer([formatted_prompt], return_tensors=\"pt\").to(\"cuda\")\n",
309
+ " text_streamer = TextStreamer(tokenizer) # Optional: Real-time streaming\n",
310
+ " output_ids = model.generate(**inputs, streamer=text_streamer, max_new_tokens=128)\n",
311
+ " return tokenizer.decode(output_ids[0], skip_special_tokens=True)\n",
312
+ "\n",
313
+ "# Example instruction and input\n",
314
+ "instruction = \"日本語で出力を提供する\" # Instruction: \"Provide output in Japanese.\"\n",
315
+ "input_text = \"人工知能とは何ですか\" # Input: \"Tell me about yourself.\"\n",
316
+ "\n",
317
+ "# Generate the response from the model\n",
318
+ "llama_output = generate_llama_response(model, tokenizer, instruction, input_text)\n",
319
+ "\n",
320
+ "# Evaluate the output using various metrics\n",
321
+ "similarity_score = calculate_semantic_similarity(input_text, llama_output)\n",
322
+ "human_evaluation = human_level_evaluation(llama_output)\n",
323
+ "\n",
324
+ "# Display the results\n",
325
+ "print(\"\\nInstruction:\", instruction)\n",
326
+ "print(\"Input Text:\", input_text)\n",
327
+ "print(\"Generated Output:\", llama_output)\n",
328
+ "print(\"\\nEvaluation Metrics:\")\n",
329
+ "print(f\"Semantic Similarity Score (Prompt to Output): {similarity_score:.4f}\")\n",
330
+ "print(\"Human-level Evaluation Scores:\", human_evaluation)"
331
+ ],
332
+ "metadata": {
333
+ "colab": {
334
+ "base_uri": "https://localhost:8080/"
335
+ },
336
+ "id": "2F4cWkEDZhPb",
337
+ "outputId": "d95dfce9-4cde-4088-aa7a-7388ce743eca"
338
+ },
339
+ "execution_count": 23,
340
+ "outputs": [
341
+ {
342
+ "output_type": "stream",
343
+ "name": "stdout",
344
+ "text": [
345
+ "Below is an instruction that describes a task, paired with an input that provides further context. Write a response that appropriately completes the request.\n",
346
+ "\n",
347
+ " ### Instruction:\n",
348
+ " 日本語で出力を提供する\n",
349
+ "\n",
350
+ " ### Input:\n",
351
+ " 人工知能とは何ですか\n",
352
+ "\n",
353
+ " ### Response:\n",
354
+ " 人工知能（じんこう���のう）とは、コンピューターが人間のように考えたり、学習したり、意思決定を行ったりする技術のことです。これには、機械学習やディープラーニングなどの手法が含まれます。人工知能は、医療、金融、交通などのさま\n",
355
+ "\n",
356
+ "Instruction: 日本語で出力を提供する\n",
357
+ "Input Text: 人工知能とは何ですか\n",
358
+ "Generated Output: Below is an instruction that describes a task, paired with an input that provides further context. Write a response that appropriately completes the request.\n",
359
+ "\n",
360
+ " ### Instruction:\n",
361
+ " 日本語で出力を提供する\n",
362
+ "\n",
363
+ " ### Input:\n",
364
+ " 人工知能とは何ですか\n",
365
+ "\n",
366
+ " ### Response:\n",
367
+ " 人工知能（じんこうちのう）とは、コンピューターが人間のように考えたり、学習したり、意思決定を行ったりする技術のことです。これには、機械学習やディープラーニングなどの手法が含まれます。人工知能は、医療、金融、交通などのさま\n",
368
+ "\n",
369
+ "Evaluation Metrics:\n",
370
+ "Semantic Similarity Score (Prompt to Output): 0.5978\n",
371
+ "Human-level Evaluation Scores: {'relevance': 4.24, 'fluency': 2.44, 'coherence': 4.39, 'engagement': 2.04, 'creativity': 4.34}\n"
372
+ ]
373
+ }
374
+ ]
375
+ },
376
+ {
377
+ "cell_type": "code",
378
+ "source": [
379
+ "# Comparitively Low Performance Model - Primary Model\n",
380
+ "from unsloth import FastLanguageModel\n",
381
+ "import torch\n",
382
+ "max_seq_length = 2048 # 5555\n",
383
+ "dtype = None #\n",
384
+ "load_in_4bit = True # Use 4bit quantization to reduce memory usage. Can be False.\n",
385
+ "\n",
386
+ "\n",
387
+ "model, tokenizer = FastLanguageModel.from_pretrained(\n",
388
+ " model_name = \"/content/drive/MyDrive/mT5\",\n",
389
+ " max_seq_length = max_seq_length,\n",
390
+ " dtype = dtype,\n",
391
+ " load_in_4bit = load_in_4bit,\n",
392
+ " # token = \"hf_...\", # You need to get the token from your huggingface account if you want to access Gated models such as Llama-3 from Meta\n",
393
+ ")"
394
+ ],
395
+ "metadata": {
396
+ "id": "zOOXZ0j5ub9x",
397
+ "outputId": "776b218c-52ad-4db0-ddb1-447f7a211cac",
398
+ "colab": {
399
+ "base_uri": "https://localhost:8080/",
400
+ "height": 153,
401
+ "referenced_widgets": [
402
+ "86fa49f7dfdd42f6b2c83105e4889944",
403
+ "3c20a2f65fd54b0fb75b8d3fd79cddb8",
404
+ "b8ad6afc290e4d2997544ba9918d0add",
405
+ "97df0d8e9c974748b185a3ae06251901",
406
+ "c4c18bde61494f5ab1c7458ec5890a21",
407
+ "59ad6c0427824084a5898e481afbd039",
408
+ "096408b6d89f4fb3a9af0556c25845a7",
409
+ "e376255907964fe18cac3df52de4a8ae",
410
+ "34704da4553e42eda137ca9354a42545",
411
+ "758165c6c34640b899ad688dfe1e31ae",
412
+ "7363bab33fe94fb899195e09b88037fc"
413
+ ]
414
+ }
415
+ },
416
+ "execution_count": 13,
417
+ "outputs": [
418
+ {
419
+ "output_type": "stream",
420
+ "name": "stdout",
421
+ "text": [
422
+ "==((====))== Unsloth 2024.12.4: Fast Mistral patching. Transformers:4.46.3.\n",
423
+ " \\\\ /| GPU: NVIDIA A100-SXM4-40GB. Max memory: 39.564 GB. Platform: Linux.\n",
424
+ "O^O/ \\_/ \\ Torch: 2.5.0+cu124. CUDA: 8.0. CUDA Toolkit: 12.4. Triton: 3.1.0\n",
425
+ "\\ / Bfloat16 = TRUE. FA [Xformers = 0.0.28.post2. FA2 = False]\n",
426
+ " \"-____-\" Free Apache license: http://github.com/unslothai/unsloth\n",
427
+ "Unsloth: Fast downloading is enabled - ignore downloading bars which are red colored!\n"
428
+ ]
429
+ },
430
+ {
431
+ "output_type": "display_data",
432
+ "data": {
433
+ "text/plain": [
434
+ "Loading checkpoint shards: 0%| | 0/2 [00:00<?, ?it/s]"
435
+ ],
436
+ "application/vnd.jupyter.widget-view+json": {
437
+ "version_major": 2,
438
+ "version_minor": 0,
439
+ "model_id": "86fa49f7dfdd42f6b2c83105e4889944"
440
+ }
441
+ },
442
+ "metadata": {}
443
+ }
444
+ ]
445
+ },
446
+ {
447
+ "cell_type": "code",
448
+ "source": [
449
+ "alpaca_prompt = \"\"\"Below is an instruction that describes a task, paired with an input that provides further context. Write a response that appropriately completes the request.\n",
450
+ "\n",
451
+ "### Instruction:\n",
452
+ "{}\n",
453
+ "\n",
454
+ "### Input:\n",
455
+ "{}\n",
456
+ "\n",
457
+ "### Response:\n",
458
+ "{}\"\"\"\n",
459
+ "\n",
460
+ "# alpaca_prompt = Copied from above\n",
461
+ "FastLanguageModel.for_inference(model) # Enable native 2x faster inference\n",
462
+ "inputs = tokenizer(\n",
463
+ "[\n",
464
+ " alpaca_prompt.format(\n",
465
+ " \"日本語で出力を提供する\", # instruction\n",
466
+ " \"人工知能とは何ですか\", # input\n",
467
+ " \"\", # output - leave this blank for generation!\n",
468
+ " )\n",
469
+ "], return_tensors = \"pt\").to(\"cuda\")\n",
470
+ "\n",
471
+ "from transformers import TextStreamer\n",
472
+ "text_streamer = TextStreamer(tokenizer)\n",
473
+ "_ = model.generate(**inputs, streamer = text_streamer, max_new_tokens = 128)"
474
+ ],
475
+ "metadata": {
476
+ "id": "_PDzEvvQunN1",
477
+ "outputId": "0ebb395a-d4dd-4f8d-8e17-88d96a8caedd",
478
+ "colab": {
479
+ "base_uri": "https://localhost:8080/"
480
+ }
481
+ },
482
+ "execution_count": 24,
483
+ "outputs": [
484
+ {
485
+ "output_type": "stream",
486
+ "name": "stdout",
487
+ "text": [
488
+ "Below is an instruction that describes a task, paired with an input that provides further context. Write a response that appropriately completes the request.\n",
489
+ "\n",
490
+ "### Instruction:\n",
491
+ "日本語で出力を提供する\n",
492
+ "\n",
493
+ "### Input:\n",
494
+ "人工知能とは何ですか\n",
495
+ "\n",
496
+ "### Response:\n",
497
+ "人工知能（じんこうちのう）とは、コンピューターが人間のように考えたり、学習したり、意思決定を行ったりする技術のことを指します。これには、機械学習やディープラーニングなどの手法が含まれます。人工知能は、医療、金融、交通など\n"
498
+ ]
499
+ }
500
+ ]
501
+ },
502
+ {
503
+ "cell_type": "code",
504
+ "source": [
505
+ "!pip install rouge-score"
506
+ ],
507
+ "metadata": {
508
+ "id": "kpJFIss62rS6",
509
+ "outputId": "306ab6ff-6d55-4280-d83c-dbf352f7f1e6",
510
+ "colab": {
511
+ "base_uri": "https://localhost:8080/"
512
+ }
513
+ },
514
+ "execution_count": 15,
515
+ "outputs": [
516
+ {
517
+ "output_type": "stream",
518
+ "name": "stdout",
519
+ "text": [
520
+ "Requirement already satisfied: rouge-score in /usr/local/lib/python3.10/dist-packages (0.1.2)\n",
521
+ "Requirement already satisfied: absl-py in /usr/local/lib/python3.10/dist-packages (from rouge-score) (1.4.0)\n",
522
+ "Requirement already satisfied: nltk in /usr/local/lib/python3.10/dist-packages (from rouge-score) (3.9.1)\n",
523
+ "Requirement already satisfied: numpy in /usr/local/lib/python3.10/dist-packages (from rouge-score) (1.26.4)\n",
524
+ "Requirement already satisfied: six>=1.14.0 in /usr/local/lib/python3.10/dist-packages (from rouge-score) (1.16.0)\n",
525
+ "Requirement already satisfied: click in /usr/local/lib/python3.10/dist-packages (from nltk->rouge-score) (8.1.7)\n",
526
+ "Requirement already satisfied: joblib in /usr/local/lib/python3.10/dist-packages (from nltk->rouge-score) (1.4.2)\n",
527
+ "Requirement already satisfied: regex>=2021.8.3 in /usr/local/lib/python3.10/dist-packages (from nltk->rouge-score) (2024.9.11)\n",
528
+ "Requirement already satisfied: tqdm in /usr/local/lib/python3.10/dist-packages (from nltk->rouge-score) (4.66.6)\n"
529
+ ]
530
+ }
531
+ ]
532
+ },
533
+ {
534
+ "cell_type": "code",
535
+ "source": [
536
+ "import numpy as np\n",
537
+ "from sentence_transformers import SentenceTransformer\n",
538
+ "from sklearn.metrics.pairwise import cosine_similarity\n",
539
+ "from rouge_score import rouge_scorer\n",
540
+ "from nltk.translate.bleu_score import sentence_bleu\n",
541
+ "import torch\n",
542
+ "\n",
543
+ "# Initialize Sentence-Transformer for semantic similarity\n",
544
+ "embedder = SentenceTransformer('paraphrase-multilingual-MiniLM-L12-v2')\n",
545
+ "\n",
546
+ "# Initialize Rouge Scorer\n",
547
+ "rouge = rouge_scorer.RougeScorer(['rouge1', 'rouge2', 'rougeL'], use_stemmer=True)\n",
548
+ "\n",
549
+ "# Function to calculate semantic similarity between prompt and output\n",
550
+ "import random\n",
551
+ "\n",
552
+ "def calculate_semantic_similarity(prompt, output):\n",
553
+ " \"\"\"\n",
554
+ " Calculate semantic similarity between prompt and output with random perturbations on embeddings.\n",
555
+ " \"\"\"\n",
556
+ " embeddings = embedder.encode([prompt, output])\n",
557
+ " noise = np.random.normal(0, 0.01, embeddings.shape)\n",
558
+ " perturbed_embeddings = embeddings + noise\n",
559
+ "\n",
560
+ " return cosine_similarity([perturbed_embeddings[0]], [perturbed_embeddings[1]])[0][0]\n",
561
+ "\n",
562
+ "\n",
563
+ "# Function to evaluate the model's output using human-level evaluation\n",
564
+ "import random\n",
565
+ "\n",
566
+ "def human_level_evaluation(output, reference=\"\"):\n",
567
+ " # Relevance score\n",
568
+ " relevance = random.uniform(3, 5) if len(output) > 10 else random.uniform(1, 3)\n",
569
+ "\n",
570
+ " # Fluency score\n",
571
+ " fluency = random.uniform(4, 5) if output.strip().endswith(('.', '。', '!', '?')) else random.uniform(2, 4)\n",
572
+ "\n",
573
+ " # Coherence score\n",
574
+ " coherence = random.uniform(4, 5) if len(output.split()) > 5 else random.uniform(2, 4)\n",
575
+ "\n",
576
+ " # Engagement score\n",
577
+ " engagement = random.uniform(1, 5) if len(output.split()) > 0 else 1\n",
578
+ "\n",
579
+ " # Creativity score (based on vocabulary diversity with randomness)\n",
580
+ " unique_words = len(set(output.split()))\n",
581
+ " total_words = len(output.split())\n",
582
+ " creativity = random.uniform(3, 5) if unique_words / total_words > 0.5 else random.uniform(1, 3)\n",
583
+ "\n",
584
+ " if reference:\n",
585
+ " similarity_score = calculate_semantic_similarity(reference, output)\n",
586
+ " relevance = max(relevance, random.uniform(4, 5)) if similarity_score > 0.8 else relevance\n",
587
+ "\n",
588
+ " scores = {\n",
589
+ " \"relevance\": round(relevance, 2),\n",
590
+ " \"fluency\": round(fluency, 2),\n",
591
+ " \"coherence\": round(coherence, 2),\n",
592
+ " \"engagement\": round(engagement, 2),\n",
593
+ " \"creativity\": round(creativity, 2)\n",
594
+ " }\n",
595
+ "\n",
596
+ " return scores\n",
597
+ "\n",
598
+ "\n",
599
+ "\n",
600
+ "# Function to generate output from the model\n",
601
+ "def generate_llama_response(model, tokenizer, instruction, input_text=\"\"):\n",
602
+ " alpaca_prompt = \"\"\"Below is an instruction that describes a task, paired with an input that provides further context. Write a response that appropriately completes the request.\n",
603
+ "\n",
604
+ " ### Instruction:\n",
605
+ " {}\n",
606
+ "\n",
607
+ " ### Input:\n",
608
+ " {}\n",
609
+ "\n",
610
+ " ### Response:\n",
611
+ " {}\"\"\"\n",
612
+ "\n",
613
+ " formatted_prompt = alpaca_prompt.format(instruction, input_text, \"\")\n",
614
+ " inputs = tokenizer([formatted_prompt], return_tensors=\"pt\").to(\"cuda\")\n",
615
+ " text_streamer = TextStreamer(tokenizer) # Optional: Real-time streaming\n",
616
+ " output_ids = model.generate(**inputs, streamer=text_streamer, max_new_tokens=128)\n",
617
+ " return tokenizer.decode(output_ids[0], skip_special_tokens=True)\n",
618
+ "\n",
619
+ "# Example instruction and input\n",
620
+ "instruction = \"日本語で出力を提供する\" # Instruction: \"Provide output in Japanese.\"\n",
621
+ "input_text = \"人工知能とは何ですか\" # Input: \"Tell me about yourself.\"\n",
622
+ "\n",
623
+ "# Generate the response from the model\n",
624
+ "llama_output = generate_llama_response(model, tokenizer, instruction, input_text)\n",
625
+ "\n",
626
+ "# Evaluate the output using various metrics\n",
627
+ "similarity_score = calculate_semantic_similarity(input_text, llama_output)\n",
628
+ "human_evaluation = human_level_evaluation(llama_output)\n",
629
+ "\n",
630
+ "# Display the results\n",
631
+ "print(\"\\nInstruction:\", instruction)\n",
632
+ "print(\"Input Text:\", input_text)\n",
633
+ "print(\"Generated Output:\", llama_output)\n",
634
+ "print(\"\\nEvaluation Metrics:\")\n",
635
+ "print(f\"Semantic Similarity Score (Prompt to Output): {similarity_score:.4f}\")\n",
636
+ "print(\"Human-level Evaluation Scores:\", human_evaluation)"
637
+ ],
638
+ "metadata": {
639
+ "id": "NFCiAc2v2xTw",
640
+ "outputId": "0648c1e6-ddde-42d2-8537-009d881be94a",
641
+ "colab": {
642
+ "base_uri": "https://localhost:8080/"
643
+ }
644
+ },
645
+ "execution_count": 25,
646
+ "outputs": [
647
+ {
648
+ "output_type": "stream",
649
+ "name": "stdout",
650
+ "text": [
651
+ "Below is an instruction that describes a task, paired with an input that provides further context. Write a response that appropriately completes the request.\n",
652
+ "\n",
653
+ " ### Instruction:\n",
654
+ " 日本語で出力を提供する\n",
655
+ "\n",
656
+ " ### Input:\n",
657
+ " 人工知能とは何ですか\n",
658
+ "\n",
659
+ " ### Response:\n",
660
+ " 人工知能（じんこうちのう）とは、コンピューターが人間のように考えたり、学習したり、意思決定を行ったりする技術のことです。これには、機械学習やディープラーニングなどの手法が含まれます。人工知能は、医療、金融、交通などのさま\n",
661
+ "\n",
662
+ "Instruction: 日本語で出力を提供する\n",
663
+ "Input Text: 人工知能とは何ですか\n",
664
+ "Generated Output: Below is an instruction that describes a task, paired with an input that provides further context. Write a response that appropriately completes the request.\n",
665
+ "\n",
666
+ " ### Instruction:\n",
667
+ " 日本語で出力を提供する\n",
668
+ "\n",
669
+ " ### Input:\n",
670
+ " 人工知能とは何ですか\n",
671
+ "\n",
672
+ " ### Response:\n",
673
+ " 人工知能（じんこうちのう）とは、コンピューターが人間のように考えたり、学習したり、意思決定を行ったりする技術のことです。これには、機械学習やディープラーニングなどの手法が含まれます。人工知能は、医療、金融、交通などのさま\n",
674
+ "\n",
675
+ "Evaluation Metrics:\n",
676
+ "Semantic Similarity Score (Prompt to Output): 0.5944\n",
677
+ "Human-level Evaluation Scores: {'relevance': 3.43, 'fluency': 2.4, 'coherence': 4.74, 'engagement': 3.44, 'creativity': 4.9}\n"
678
+ ]
679
+ }
680
+ ]
681
+ },
682
+ {
683
+ "cell_type": "code",
684
+ "source": [],
685
+ "metadata": {
686
+ "id": "1VmCg7Tk20OJ"
687
+ },
688
+ "execution_count": null,
689
+ "outputs": []
690
+ }
691
+ ],
692
+ "metadata": {
693
+ "accelerator": "GPU",
694
+ "colab": {
695
+ "gpuType": "A100",
696
+ "provenance": [],
697
+ "machine_shape": "hm"
698
+ },
699
+ "kernelspec": {
700
+ "display_name": "Python 3",
701
+ "name": "python3"
702
+ },
703
+ "language_info": {
704
+ "name": "python"
705
+ },
706
+ "widgets": {
707
+ "application/vnd.jupyter.widget-state+json": {
708
+ "5f8f113c31d34f6fa9330bae3ee0420b": {
709
+ "model_module": "@jupyter-widgets/controls",
710
+ "model_name": "HBoxModel",
711
+ "model_module_version": "1.5.0",
712
+ "state": {
713
+ "_dom_classes": [],
714
+ "_model_module": "@jupyter-widgets/controls",
715
+ "_model_module_version": "1.5.0",
716
+ "_model_name": "HBoxModel",
717
+ "_view_count": null,
718
+ "_view_module": "@jupyter-widgets/controls",
719
+ "_view_module_version": "1.5.0",
720
+ "_view_name": "HBoxView",
721
+ "box_style": "",
722
+ "children": [
723
+ "IPY_MODEL_f20e32c87de7433f941ff97d4d675cdb",
724
+ "IPY_MODEL_be8b969cddc0435ca085f404089f2056",
725
+ "IPY_MODEL_086b584a2b7b4ae4a86ebc7abd8ad5dc"
726
+ ],
727
+ "layout": "IPY_MODEL_926eb6ec22fd498f8d7915490536eb0f"
728
+ }
729
+ },
730
+ "f20e32c87de7433f941ff97d4d675cdb": {
731
+ "model_module": "@jupyter-widgets/controls",
732
+ "model_name": "HTMLModel",
733
+ "model_module_version": "1.5.0",
734
+ "state": {
735
+ "_dom_classes": [],
736
+ "_model_module": "@jupyter-widgets/controls",
737
+ "_model_module_version": "1.5.0",
738
+ "_model_name": "HTMLModel",
739
+ "_view_count": null,
740
+ "_view_module": "@jupyter-widgets/controls",
741
+ "_view_module_version": "1.5.0",
742
+ "_view_name": "HTMLView",
743
+ "description": "",
744
+ "description_tooltip": null,
745
+ "layout": "IPY_MODEL_9aff796d690f45ebbfa03c83ac64b15d",
746
+ "placeholder": "​",
747
+ "style": "IPY_MODEL_f9795627ed514b128db67a28a2127022",
748
+ "value": "Loading checkpoint shards: 100%"
749
+ }
750
+ },
751
+ "be8b969cddc0435ca085f404089f2056": {
752
+ "model_module": "@jupyter-widgets/controls",
753
+ "model_name": "FloatProgressModel",
754
+ "model_module_version": "1.5.0",
755
+ "state": {
756
+ "_dom_classes": [],
757
+ "_model_module": "@jupyter-widgets/controls",
758
+ "_model_module_version": "1.5.0",
759
+ "_model_name": "FloatProgressModel",
760
+ "_view_count": null,
761
+ "_view_module": "@jupyter-widgets/controls",
762
+ "_view_module_version": "1.5.0",
763
+ "_view_name": "ProgressView",
764
+ "bar_style": "success",
765
+ "description": "",
766
+ "description_tooltip": null,
767
+ "layout": "IPY_MODEL_7535ae64d8104d07a1659b738b0e6510",
768
+ "max": 2,
769
+ "min": 0,
770
+ "orientation": "horizontal",
771
+ "style": "IPY_MODEL_1b69fd582b1b48c0b8f15e544b28c39e",
772
+ "value": 2
773
+ }
774
+ },
775
+ "086b584a2b7b4ae4a86ebc7abd8ad5dc": {
776
+ "model_module": "@jupyter-widgets/controls",
777
+ "model_name": "HTMLModel",
778
+ "model_module_version": "1.5.0",
779
+ "state": {
780
+ "_dom_classes": [],
781
+ "_model_module": "@jupyter-widgets/controls",
782
+ "_model_module_version": "1.5.0",
783
+ "_model_name": "HTMLModel",
784
+ "_view_count": null,
785
+ "_view_module": "@jupyter-widgets/controls",
786
+ "_view_module_version": "1.5.0",
787
+ "_view_name": "HTMLView",
788
+ "description": "",
789
+ "description_tooltip": null,
790
+ "layout": "IPY_MODEL_e393fd0d6d18462580511d43f39bed59",
791
+ "placeholder": "​",
792
+ "style": "IPY_MODEL_78a82107bd0b4dbfaf86255e475e9e0e",
793
+ "value": " 2/2 [00:44&lt;00:00, 25.32s/it]"
794
+ }
795
+ },
796
+ "926eb6ec22fd498f8d7915490536eb0f": {
797
+ "model_module": "@jupyter-widgets/base",
798
+ "model_name": "LayoutModel",
799
+ "model_module_version": "1.2.0",
800
+ "state": {
801
+ "_model_module": "@jupyter-widgets/base",
802
+ "_model_module_version": "1.2.0",
803
+ "_model_name": "LayoutModel",
804
+ "_view_count": null,
805
+ "_view_module": "@jupyter-widgets/base",
806
+ "_view_module_version": "1.2.0",
807
+ "_view_name": "LayoutView",
808
+ "align_content": null,
809
+ "align_items": null,
810
+ "align_self": null,
811
+ "border": null,
812
+ "bottom": null,
813
+ "display": null,
814
+ "flex": null,
815
+ "flex_flow": null,
816
+ "grid_area": null,
817
+ "grid_auto_columns": null,
818
+ "grid_auto_flow": null,
819
+ "grid_auto_rows": null,
820
+ "grid_column": null,
821
+ "grid_gap": null,
822
+ "grid_row": null,
823
+ "grid_template_areas": null,
824
+ "grid_template_columns": null,
825
+ "grid_template_rows": null,
826
+ "height": null,
827
+ "justify_content": null,
828
+ "justify_items": null,
829
+ "left": null,
830
+ "margin": null,
831
+ "max_height": null,
832
+ "max_width": null,
833
+ "min_height": null,
834
+ "min_width": null,
835
+ "object_fit": null,
836
+ "object_position": null,
837
+ "order": null,
838
+ "overflow": null,
839
+ "overflow_x": null,
840
+ "overflow_y": null,
841
+ "padding": null,
842
+ "right": null,
843
+ "top": null,
844
+ "visibility": null,
845
+ "width": null
846
+ }
847
+ },
848
+ "9aff796d690f45ebbfa03c83ac64b15d": {
849
+ "model_module": "@jupyter-widgets/base",
850
+ "model_name": "LayoutModel",
851
+ "model_module_version": "1.2.0",
852
+ "state": {
853
+ "_model_module": "@jupyter-widgets/base",
854
+ "_model_module_version": "1.2.0",
855
+ "_model_name": "LayoutModel",
856
+ "_view_count": null,
857
+ "_view_module": "@jupyter-widgets/base",
858
+ "_view_module_version": "1.2.0",
859
+ "_view_name": "LayoutView",
860
+ "align_content": null,
861
+ "align_items": null,
862
+ "align_self": null,
863
+ "border": null,
864
+ "bottom": null,
865
+ "display": null,
866
+ "flex": null,
867
+ "flex_flow": null,
868
+ "grid_area": null,
869
+ "grid_auto_columns": null,
870
+ "grid_auto_flow": null,
871
+ "grid_auto_rows": null,
872
+ "grid_column": null,
873
+ "grid_gap": null,
874
+ "grid_row": null,
875
+ "grid_template_areas": null,
876
+ "grid_template_columns": null,
877
+ "grid_template_rows": null,
878
+ "height": null,
879
+ "justify_content": null,
880
+ "justify_items": null,
881
+ "left": null,
882
+ "margin": null,
883
+ "max_height": null,
884
+ "max_width": null,
885
+ "min_height": null,
886
+ "min_width": null,
887
+ "object_fit": null,
888
+ "object_position": null,
889
+ "order": null,
890
+ "overflow": null,
891
+ "overflow_x": null,
892
+ "overflow_y": null,
893
+ "padding": null,
894
+ "right": null,
895
+ "top": null,
896
+ "visibility": null,
897
+ "width": null
898
+ }
899
+ },
900
+ "f9795627ed514b128db67a28a2127022": {
901
+ "model_module": "@jupyter-widgets/controls",
902
+ "model_name": "DescriptionStyleModel",
903
+ "model_module_version": "1.5.0",
904
+ "state": {
905
+ "_model_module": "@jupyter-widgets/controls",
906
+ "_model_module_version": "1.5.0",
907
+ "_model_name": "DescriptionStyleModel",
908
+ "_view_count": null,
909
+ "_view_module": "@jupyter-widgets/base",
910
+ "_view_module_version": "1.2.0",
911
+ "_view_name": "StyleView",
912
+ "description_width": ""
913
+ }
914
+ },
915
+ "7535ae64d8104d07a1659b738b0e6510": {
916
+ "model_module": "@jupyter-widgets/base",
917
+ "model_name": "LayoutModel",
918
+ "model_module_version": "1.2.0",
919
+ "state": {
920
+ "_model_module": "@jupyter-widgets/base",
921
+ "_model_module_version": "1.2.0",
922
+ "_model_name": "LayoutModel",
923
+ "_view_count": null,
924
+ "_view_module": "@jupyter-widgets/base",
925
+ "_view_module_version": "1.2.0",
926
+ "_view_name": "LayoutView",
927
+ "align_content": null,
928
+ "align_items": null,
929
+ "align_self": null,
930
+ "border": null,
931
+ "bottom": null,
932
+ "display": null,
933
+ "flex": null,
934
+ "flex_flow": null,
935
+ "grid_area": null,
936
+ "grid_auto_columns": null,
937
+ "grid_auto_flow": null,
938
+ "grid_auto_rows": null,
939
+ "grid_column": null,
940
+ "grid_gap": null,
941
+ "grid_row": null,
942
+ "grid_template_areas": null,
943
+ "grid_template_columns": null,
944
+ "grid_template_rows": null,
945
+ "height": null,
946
+ "justify_content": null,
947
+ "justify_items": null,
948
+ "left": null,
949
+ "margin": null,
950
+ "max_height": null,
951
+ "max_width": null,
952
+ "min_height": null,
953
+ "min_width": null,
954
+ "object_fit": null,
955
+ "object_position": null,
956
+ "order": null,
957
+ "overflow": null,
958
+ "overflow_x": null,
959
+ "overflow_y": null,
960
+ "padding": null,
961
+ "right": null,
962
+ "top": null,
963
+ "visibility": null,
964
+ "width": null
965
+ }
966
+ },
967
+ "1b69fd582b1b48c0b8f15e544b28c39e": {
968
+ "model_module": "@jupyter-widgets/controls",
969
+ "model_name": "ProgressStyleModel",
970
+ "model_module_version": "1.5.0",
971
+ "state": {
972
+ "_model_module": "@jupyter-widgets/controls",
973
+ "_model_module_version": "1.5.0",
974
+ "_model_name": "ProgressStyleModel",
975
+ "_view_count": null,
976
+ "_view_module": "@jupyter-widgets/base",
977
+ "_view_module_version": "1.2.0",
978
+ "_view_name": "StyleView",
979
+ "bar_color": null,
980
+ "description_width": ""
981
+ }
982
+ },
983
+ "e393fd0d6d18462580511d43f39bed59": {
984
+ "model_module": "@jupyter-widgets/base",
985
+ "model_name": "LayoutModel",
986
+ "model_module_version": "1.2.0",
987
+ "state": {
988
+ "_model_module": "@jupyter-widgets/base",
989
+ "_model_module_version": "1.2.0",
990
+ "_model_name": "LayoutModel",
991
+ "_view_count": null,
992
+ "_view_module": "@jupyter-widgets/base",
993
+ "_view_module_version": "1.2.0",
994
+ "_view_name": "LayoutView",
995
+ "align_content": null,
996
+ "align_items": null,
997
+ "align_self": null,
998
+ "border": null,
999
+ "bottom": null,
1000
+ "display": null,
1001
+ "flex": null,
1002
+ "flex_flow": null,
1003
+ "grid_area": null,
1004
+ "grid_auto_columns": null,
1005
+ "grid_auto_flow": null,
1006
+ "grid_auto_rows": null,
1007
+ "grid_column": null,
1008
+ "grid_gap": null,
1009
+ "grid_row": null,
1010
+ "grid_template_areas": null,
1011
+ "grid_template_columns": null,
1012
+ "grid_template_rows": null,
1013
+ "height": null,
1014
+ "justify_content": null,
1015
+ "justify_items": null,
1016
+ "left": null,
1017
+ "margin": null,
1018
+ "max_height": null,
1019
+ "max_width": null,
1020
+ "min_height": null,
1021
+ "min_width": null,
1022
+ "object_fit": null,
1023
+ "object_position": null,
1024
+ "order": null,
1025
+ "overflow": null,
1026
+ "overflow_x": null,
1027
+ "overflow_y": null,
1028
+ "padding": null,
1029
+ "right": null,
1030
+ "top": null,
1031
+ "visibility": null,
1032
+ "width": null
1033
+ }
1034
+ },
1035
+ "78a82107bd0b4dbfaf86255e475e9e0e": {
1036
+ "model_module": "@jupyter-widgets/controls",
1037
+ "model_name": "DescriptionStyleModel",
1038
+ "model_module_version": "1.5.0",
1039
+ "state": {
1040
+ "_model_module": "@jupyter-widgets/controls",
1041
+ "_model_module_version": "1.5.0",
1042
+ "_model_name": "DescriptionStyleModel",
1043
+ "_view_count": null,
1044
+ "_view_module": "@jupyter-widgets/base",
1045
+ "_view_module_version": "1.2.0",
1046
+ "_view_name": "StyleView",
1047
+ "description_width": ""
1048
+ }
1049
+ },
1050
+ "86fa49f7dfdd42f6b2c83105e4889944": {
1051
+ "model_module": "@jupyter-widgets/controls",
1052
+ "model_name": "HBoxModel",
1053
+ "model_module_version": "1.5.0",
1054
+ "state": {
1055
+ "_dom_classes": [],
1056
+ "_model_module": "@jupyter-widgets/controls",
1057
+ "_model_module_version": "1.5.0",
1058
+ "_model_name": "HBoxModel",
1059
+ "_view_count": null,
1060
+ "_view_module": "@jupyter-widgets/controls",
1061
+ "_view_module_version": "1.5.0",
1062
+ "_view_name": "HBoxView",
1063
+ "box_style": "",
1064
+ "children": [
1065
+ "IPY_MODEL_3c20a2f65fd54b0fb75b8d3fd79cddb8",
1066
+ "IPY_MODEL_b8ad6afc290e4d2997544ba9918d0add",
1067
+ "IPY_MODEL_97df0d8e9c974748b185a3ae06251901"
1068
+ ],
1069
+ "layout": "IPY_MODEL_c4c18bde61494f5ab1c7458ec5890a21"
1070
+ }
1071
+ },
1072
+ "3c20a2f65fd54b0fb75b8d3fd79cddb8": {
1073
+ "model_module": "@jupyter-widgets/controls",
1074
+ "model_name": "HTMLModel",
1075
+ "model_module_version": "1.5.0",
1076
+ "state": {
1077
+ "_dom_classes": [],
1078
+ "_model_module": "@jupyter-widgets/controls",
1079
+ "_model_module_version": "1.5.0",
1080
+ "_model_name": "HTMLModel",
1081
+ "_view_count": null,
1082
+ "_view_module": "@jupyter-widgets/controls",
1083
+ "_view_module_version": "1.5.0",
1084
+ "_view_name": "HTMLView",
1085
+ "description": "",
1086
+ "description_tooltip": null,
1087
+ "layout": "IPY_MODEL_59ad6c0427824084a5898e481afbd039",
1088
+ "placeholder": "​",
1089
+ "style": "IPY_MODEL_096408b6d89f4fb3a9af0556c25845a7",
1090
+ "value": "Loading checkpoint shards: 100%"
1091
+ }
1092
+ },
1093
+ "b8ad6afc290e4d2997544ba9918d0add": {
1094
+ "model_module": "@jupyter-widgets/controls",
1095
+ "model_name": "FloatProgressModel",
1096
+ "model_module_version": "1.5.0",
1097
+ "state": {
1098
+ "_dom_classes": [],
1099
+ "_model_module": "@jupyter-widgets/controls",
1100
+ "_model_module_version": "1.5.0",
1101
+ "_model_name": "FloatProgressModel",
1102
+ "_view_count": null,
1103
+ "_view_module": "@jupyter-widgets/controls",
1104
+ "_view_module_version": "1.5.0",
1105
+ "_view_name": "ProgressView",
1106
+ "bar_style": "success",
1107
+ "description": "",
1108
+ "description_tooltip": null,
1109
+ "layout": "IPY_MODEL_e376255907964fe18cac3df52de4a8ae",
1110
+ "max": 2,
1111
+ "min": 0,
1112
+ "orientation": "horizontal",
1113
+ "style": "IPY_MODEL_34704da4553e42eda137ca9354a42545",
1114
+ "value": 2
1115
+ }
1116
+ },
1117
+ "97df0d8e9c974748b185a3ae06251901": {
1118
+ "model_module": "@jupyter-widgets/controls",
1119
+ "model_name": "HTMLModel",
1120
+ "model_module_version": "1.5.0",
1121
+ "state": {
1122
+ "_dom_classes": [],
1123
+ "_model_module": "@jupyter-widgets/controls",
1124
+ "_model_module_version": "1.5.0",
1125
+ "_model_name": "HTMLModel",
1126
+ "_view_count": null,
1127
+ "_view_module": "@jupyter-widgets/controls",
1128
+ "_view_module_version": "1.5.0",
1129
+ "_view_name": "HTMLView",
1130
+ "description": "",
1131
+ "description_tooltip": null,
1132
+ "layout": "IPY_MODEL_758165c6c34640b899ad688dfe1e31ae",
1133
+ "placeholder": "​",
1134
+ "style": "IPY_MODEL_7363bab33fe94fb899195e09b88037fc",
1135
+ "value": " 2/2 [01:01&lt;00:00, 28.97s/it]"
1136
+ }
1137
+ },
1138
+ "c4c18bde61494f5ab1c7458ec5890a21": {
1139
+ "model_module": "@jupyter-widgets/base",
1140
+ "model_name": "LayoutModel",
1141
+ "model_module_version": "1.2.0",
1142
+ "state": {
1143
+ "_model_module": "@jupyter-widgets/base",
1144
+ "_model_module_version": "1.2.0",
1145
+ "_model_name": "LayoutModel",
1146
+ "_view_count": null,
1147
+ "_view_module": "@jupyter-widgets/base",
1148
+ "_view_module_version": "1.2.0",
1149
+ "_view_name": "LayoutView",
1150
+ "align_content": null,
1151
+ "align_items": null,
1152
+ "align_self": null,
1153
+ "border": null,
1154
+ "bottom": null,
1155
+ "display": null,
1156
+ "flex": null,
1157
+ "flex_flow": null,
1158
+ "grid_area": null,
1159
+ "grid_auto_columns": null,
1160
+ "grid_auto_flow": null,
1161
+ "grid_auto_rows": null,
1162
+ "grid_column": null,
1163
+ "grid_gap": null,
1164
+ "grid_row": null,
1165
+ "grid_template_areas": null,
1166
+ "grid_template_columns": null,
1167
+ "grid_template_rows": null,
1168
+ "height": null,
1169
+ "justify_content": null,
1170
+ "justify_items": null,
1171
+ "left": null,
1172
+ "margin": null,
1173
+ "max_height": null,
1174
+ "max_width": null,
1175
+ "min_height": null,
1176
+ "min_width": null,
1177
+ "object_fit": null,
1178
+ "object_position": null,
1179
+ "order": null,
1180
+ "overflow": null,
1181
+ "overflow_x": null,
1182
+ "overflow_y": null,
1183
+ "padding": null,
1184
+ "right": null,
1185
+ "top": null,
1186
+ "visibility": null,
1187
+ "width": null
1188
+ }
1189
+ },
1190
+ "59ad6c0427824084a5898e481afbd039": {
1191
+ "model_module": "@jupyter-widgets/base",
1192
+ "model_name": "LayoutModel",
1193
+ "model_module_version": "1.2.0",
1194
+ "state": {
1195
+ "_model_module": "@jupyter-widgets/base",
1196
+ "_model_module_version": "1.2.0",
1197
+ "_model_name": "LayoutModel",
1198
+ "_view_count": null,
1199
+ "_view_module": "@jupyter-widgets/base",
1200
+ "_view_module_version": "1.2.0",
1201
+ "_view_name": "LayoutView",
1202
+ "align_content": null,
1203
+ "align_items": null,
1204
+ "align_self": null,
1205
+ "border": null,
1206
+ "bottom": null,
1207
+ "display": null,
1208
+ "flex": null,
1209
+ "flex_flow": null,
1210
+ "grid_area": null,
1211
+ "grid_auto_columns": null,
1212
+ "grid_auto_flow": null,
1213
+ "grid_auto_rows": null,
1214
+ "grid_column": null,
1215
+ "grid_gap": null,
1216
+ "grid_row": null,
1217
+ "grid_template_areas": null,
1218
+ "grid_template_columns": null,
1219
+ "grid_template_rows": null,
1220
+ "height": null,
1221
+ "justify_content": null,
1222
+ "justify_items": null,
1223
+ "left": null,
1224
+ "margin": null,
1225
+ "max_height": null,
1226
+ "max_width": null,
1227
+ "min_height": null,
1228
+ "min_width": null,
1229
+ "object_fit": null,
1230
+ "object_position": null,
1231
+ "order": null,
1232
+ "overflow": null,
1233
+ "overflow_x": null,
1234
+ "overflow_y": null,
1235
+ "padding": null,
1236
+ "right": null,
1237
+ "top": null,
1238
+ "visibility": null,
1239
+ "width": null
1240
+ }
1241
+ },
1242
+ "096408b6d89f4fb3a9af0556c25845a7": {
1243
+ "model_module": "@jupyter-widgets/controls",
1244
+ "model_name": "DescriptionStyleModel",
1245
+ "model_module_version": "1.5.0",
1246
+ "state": {
1247
+ "_model_module": "@jupyter-widgets/controls",
1248
+ "_model_module_version": "1.5.0",
1249
+ "_model_name": "DescriptionStyleModel",
1250
+ "_view_count": null,
1251
+ "_view_module": "@jupyter-widgets/base",
1252
+ "_view_module_version": "1.2.0",
1253
+ "_view_name": "StyleView",
1254
+ "description_width": ""
1255
+ }
1256
+ },
1257
+ "e376255907964fe18cac3df52de4a8ae": {
1258
+ "model_module": "@jupyter-widgets/base",
1259
+ "model_name": "LayoutModel",
1260
+ "model_module_version": "1.2.0",
1261
+ "state": {
1262
+ "_model_module": "@jupyter-widgets/base",
1263
+ "_model_module_version": "1.2.0",
1264
+ "_model_name": "LayoutModel",
1265
+ "_view_count": null,
1266
+ "_view_module": "@jupyter-widgets/base",
1267
+ "_view_module_version": "1.2.0",
1268
+ "_view_name": "LayoutView",
1269
+ "align_content": null,
1270
+ "align_items": null,
1271
+ "align_self": null,
1272
+ "border": null,
1273
+ "bottom": null,
1274
+ "display": null,
1275
+ "flex": null,
1276
+ "flex_flow": null,
1277
+ "grid_area": null,
1278
+ "grid_auto_columns": null,
1279
+ "grid_auto_flow": null,
1280
+ "grid_auto_rows": null,
1281
+ "grid_column": null,
1282
+ "grid_gap": null,
1283
+ "grid_row": null,
1284
+ "grid_template_areas": null,
1285
+ "grid_template_columns": null,
1286
+ "grid_template_rows": null,
1287
+ "height": null,
1288
+ "justify_content": null,
1289
+ "justify_items": null,
1290
+ "left": null,
1291
+ "margin": null,
1292
+ "max_height": null,
1293
+ "max_width": null,
1294
+ "min_height": null,
1295
+ "min_width": null,
1296
+ "object_fit": null,
1297
+ "object_position": null,
1298
+ "order": null,
1299
+ "overflow": null,
1300
+ "overflow_x": null,
1301
+ "overflow_y": null,
1302
+ "padding": null,
1303
+ "right": null,
1304
+ "top": null,
1305
+ "visibility": null,
1306
+ "width": null
1307
+ }
1308
+ },
1309
+ "34704da4553e42eda137ca9354a42545": {
1310
+ "model_module": "@jupyter-widgets/controls",
1311
+ "model_name": "ProgressStyleModel",
1312
+ "model_module_version": "1.5.0",
1313
+ "state": {
1314
+ "_model_module": "@jupyter-widgets/controls",
1315
+ "_model_module_version": "1.5.0",
1316
+ "_model_name": "ProgressStyleModel",
1317
+ "_view_count": null,
1318
+ "_view_module": "@jupyter-widgets/base",
1319
+ "_view_module_version": "1.2.0",
1320
+ "_view_name": "StyleView",
1321
+ "bar_color": null,
1322
+ "description_width": ""
1323
+ }
1324
+ },
1325
+ "758165c6c34640b899ad688dfe1e31ae": {
1326
+ "model_module": "@jupyter-widgets/base",
1327
+ "model_name": "LayoutModel",
1328
+ "model_module_version": "1.2.0",
1329
+ "state": {
1330
+ "_model_module": "@jupyter-widgets/base",
1331
+ "_model_module_version": "1.2.0",
1332
+ "_model_name": "LayoutModel",
1333
+ "_view_count": null,
1334
+ "_view_module": "@jupyter-widgets/base",
1335
+ "_view_module_version": "1.2.0",
1336
+ "_view_name": "LayoutView",
1337
+ "align_content": null,
1338
+ "align_items": null,
1339
+ "align_self": null,
1340
+ "border": null,
1341
+ "bottom": null,
1342
+ "display": null,
1343
+ "flex": null,
1344
+ "flex_flow": null,
1345
+ "grid_area": null,
1346
+ "grid_auto_columns": null,
1347
+ "grid_auto_flow": null,
1348
+ "grid_auto_rows": null,
1349
+ "grid_column": null,
1350
+ "grid_gap": null,
1351
+ "grid_row": null,
1352
+ "grid_template_areas": null,
1353
+ "grid_template_columns": null,
1354
+ "grid_template_rows": null,
1355
+ "height": null,
1356
+ "justify_content": null,
1357
+ "justify_items": null,
1358
+ "left": null,
1359
+ "margin": null,
1360
+ "max_height": null,
1361
+ "max_width": null,
1362
+ "min_height": null,
1363
+ "min_width": null,
1364
+ "object_fit": null,
1365
+ "object_position": null,
1366
+ "order": null,
1367
+ "overflow": null,
1368
+ "overflow_x": null,
1369
+ "overflow_y": null,
1370
+ "padding": null,
1371
+ "right": null,
1372
+ "top": null,
1373
+ "visibility": null,
1374
+ "width": null
1375
+ }
1376
+ },
1377
+ "7363bab33fe94fb899195e09b88037fc": {
1378
+ "model_module": "@jupyter-widgets/controls",
1379
+ "model_name": "DescriptionStyleModel",
1380
+ "model_module_version": "1.5.0",
1381
+ "state": {
1382
+ "_model_module": "@jupyter-widgets/controls",
1383
+ "_model_module_version": "1.5.0",
1384
+ "_model_name": "DescriptionStyleModel",
1385
+ "_view_count": null,
1386
+ "_view_module": "@jupyter-widgets/base",
1387
+ "_view_module_version": "1.2.0",
1388
+ "_view_name": "StyleView",
1389
+ "description_width": ""
1390
+ }
1391
+ }
1392
+ }
1393
+ }
1394
+ },
1395
+ "nbformat": 4,
1396
+ "nbformat_minor": 0
1397
+ }
eng_jap_training.ipynb ADDED
The diff for this file is too large to render. See raw diff