Add README

README.md

@@ -1,61 +1,51 @@
 ---
+license: apache-2.0
 base_model: llava-hf/llava-onevision-qwen2-7b-ov-hf
-library_name: peft
-model_name: llava_qlora_weather_model
 tags:
-- base_model:adapter:llava-hf/llava-onevision-qwen2-7b-ov-hf
-- lora
-- sft
-- trl
-licence: license
-pipeline_tag: text-generation
+- llava
+- llava-onevision
+- weather
+- satellite
+- morocco
+- meteorology
+- qlora
+- fine-tuned
 ---
 
-# Model Card for llava_qlora_weather_model
+# LLaVA-OneVision Weather Analysis - QLoRA
 
-This model is a fine-tuned version of [llava-hf/llava-onevision-qwen2-7b-ov-hf](https://huggingface.co/llava-hf/llava-onevision-qwen2-7b-ov-hf).
-It has been trained using [TRL](https://github.com/huggingface/trl).
+Fine-tuned using **QLoRA** technique for weather satellite imagery analysis.
 
-## Quick start
+## Model Details
 
-```python
-from transformers import pipeline
-
-question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
-generator = pipeline("text-generation", model="None", device="cuda")
-output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
-print(output["generated_text"])
-```
-
-## Training procedure
+- **Base Model:** llava-hf/llava-onevision-qwen2-7b-ov-hf
+- **Technique:** QLoRA
+- **Domain:** Weather satellite imagery analysis
+- **Dataset:** Weather satellite images with meteorological metadata
 
-[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/azdinsahir11-university-mohamed-v/llava-onevision-qlora-weather/runs/7e65t9r9) 
+## Usage
 
+```python
+from transformers import LlavaOnevisionForConditionalGeneration, AutoProcessor
+import torch
 
-This model was trained with SFT.
-
-### Framework versions
-
-- PEFT 0.16.0
-- TRL: 0.20.0
-- Transformers: 4.53.3
-- Pytorch: 2.6.0+cu124
-- Datasets: 4.0.0
-- Tokenizers: 0.21.2
+# Load base model
+model = LlavaOnevisionForConditionalGeneration.from_pretrained(
+    "llava-hf/llava-onevision-qwen2-7b-ov-hf",
+    torch_dtype=torch.bfloat16,
+    device_map="auto"
+)
+processor = AutoProcessor.from_pretrained("llava-hf/llava-onevision-qwen2-7b-ov-hf")
 
-## Citations
+# Load fine-tuned adapter
+model.load_adapter("azdin/llava-onevision-weather-qlora")
 
+# Use for weather analysis...
+```
 
+## Training Details
 
-Cite TRL as:
-    
-```bibtex
-@misc{vonwerra2022trl,
-	title        = {{TRL: Transformer Reinforcement Learning}},
-	author       = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin Gallou{\'e}dec},
-	year         = 2020,
-	journal      = {GitHub repository},
-	publisher    = {GitHub},
-	howpublished = {\url{https://github.com/huggingface/trl}}
-}
-```
+- **Technique:** QLoRA
+- **Quantization:** 4-bit NF4
+- **Training Data:** Weather satellite imagery with metadata
+- **Target Modules:** Attention and projection layers