AgriPath-Qwen2-VL-2B-LoRA64

Fine-tuned Qwen2-VL-2B on AgriPath-LF16-30k for crop and disease classification. Uses LoRA (rank=64) to adapt vision and language layers.

---

Model Details

• Base Model: Qwen2-VL-2B
• Fine-tuned on: AgriPath-LF16-30k
• Fine-tuning Method: LoRA (Rank=64, Alpha=64 Dropout=0)
• Layers Updated: Vision, Attention, Language, MLP Modules
• Optimiser: AdamW (8-bit)
• Batch Size: 2 per device (Gradient Accumulation = 4)
• Learning Rate: 2e-4
• Training Time: 176.78 minutes (~3 hours)
• Peak GPU Usage: 13.6GB (RTX 4080 Super)

---

Dataset

AgriPath-LF16-30k

• 30,000 images across 16 crops and 65 (crop, disease) pairs
• 50% lab images, 50% field images
• Preprocessing:
  • Images resized: max_pixels = 512x512, min_pixels = 224x224
  • No additional augmentation

---

Training Performance

Step	Training Loss	Validation Loss
500	0.038800	0.087705
1000	0.014300	0.058995
1500	0.014200	0.030874
2000	0.002800	0.026959
2500	0.045300	0.018349

✅ Best validation loss: 0.018349 at step 2500
✅ Stable training with low overfitting

---

Uploaded model

• Developed by: hamzamooraj99
• License: apache-2.0
• Finetuned from model : unsloth/Qwen2-VL-2B-Instruct-unsloth-bnb-4bit

This qwen2_vl model was trained 2x faster with Unsloth and Huggingface's TRL library.