Update model card with full training details and usage

README.md

@@ -3,76 +3,178 @@ library_name: transformers
 license: apache-2.0
 base_model: PekingU/rtdetr_r18vd_coco_o365
 tags:
-- generated_from_trainer
+  - object-detection
+  - pill-detection
+  - medical
+  - rt-detr
+  - computer-vision
+  - counting
+  - healthcare
+datasets:
+  - Francesco/pills-sxdht
+pipeline_tag: object-detection
 model-index:
-- name: rtdetr-pill-detector
-  results: []
+  - name: rtdetr-pill-detector
+    results:
+      - task:
+          type: object-detection
+          name: Object Detection
+        dataset:
+          name: pills-sxdht (Roboflow 100)
+          type: Francesco/pills-sxdht
+        metrics:
+          - type: loss
+            value: 3.528
+            name: Best Validation Loss
 ---
 
-<!-- This model card has been generated automatically according to the information the Trainer had access to. You
-should probably proofread and complete it, then remove this comment. -->
+# 💊 RT-DETR Pill Detector & Counter
 
-# rtdetr-pill-detector
+A real-time medicine pill detection model that can **detect and count pills, capsules, and specific medications** in images.
 
-This model is a fine-tuned version of [PekingU/rtdetr_r18vd_coco_o365](https://huggingface.co/PekingU/rtdetr_r18vd_coco_o365) on an unknown dataset.
-It achieves the following results on the evaluation set:
-- Loss: 3.5814
+**🎮 Try it live:** [**Pill Detector Demo**](https://huggingface.co/spaces/SARANGx/pill-detector-demo)
 
-## Model description
+## Model Description
 
-More information needed
+This model is a fine-tuned [RT-DETR R18](https://huggingface.co/PekingU/rtdetr_r18vd_coco_o365) (Real-Time DEtection TRansformer with ResNet-18 backbone) for detecting medicine pills in images.
 
-## Intended uses & limitations
-
-More information needed
-
-## Training and evaluation data
-
-More information needed
-
-## Training procedure
-
-### Training hyperparameters
-
-The following hyperparameters were used during training:
-- learning_rate: 5e-05
-- train_batch_size: 8
-- eval_batch_size: 8
-- seed: 42
-- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
-- lr_scheduler_type: cosine
-- lr_scheduler_warmup_steps: 50
-- num_epochs: 20
-
-### Training results
-
-| Training Loss | Epoch | Step | Validation Loss |
-|:-------------:|:-----:|:----:|:---------------:|
-| 36.6133       | 1.0   | 40   | 25.6964         |
-| 16.9363       | 2.0   | 80   | 10.3644         |
-| 13.2576       | 3.0   | 120  | 7.4604          |
-| 11.2193       | 4.0   | 160  | 6.0646          |
-| 8.5718        | 5.0   | 200  | 5.2125          |
-| 8.3696        | 6.0   | 240  | 4.8302          |
-| 7.5242        | 7.0   | 280  | 4.4670          |
-| 7.0528        | 8.0   | 320  | 4.0657          |
-| 6.3887        | 9.0   | 360  | 3.8813          |
-| 6.3234        | 10.0  | 400  | 3.7910          |
-| 5.9830        | 11.0  | 440  | 3.6750          |
-| 5.8632        | 12.0  | 480  | 3.5891          |
-| 6.7105        | 13.0  | 520  | 3.7327          |
-| 5.4882        | 14.0  | 560  | 3.5292          |
-| 5.5451        | 15.0  | 600  | 3.5306          |
-| 5.3121        | 16.0  | 640  | 3.5427          |
-| 5.2448        | 17.0  | 680  | 3.5641          |
-| 5.3653        | 18.0  | 720  | 3.6307          |
-| 5.3143        | 19.0  | 760  | 3.5280          |
-| 5.7803        | 20.0  | 800  | 3.5814          |
+### Why RT-DETR?
+- **NMS-free architecture** — Each pill is detected exactly once, making it ideal for **accurate counting**
+- **End-to-end detection** — No post-processing like Non-Maximum Suppression needed
+- **Real-time capable** — 20M parameters, fast inference even on CPU
+- **Strong baseline** — Pre-trained on Objects365 (365 categories, 2M images)
+
+### What it detects (9 classes)
+
+| Class | Description |
+|-------|-------------|
+| `pills` | Generic pill detection |
+| `Cipro 500` | Ciprofloxacin 500mg |
+| `Ibuphil 600 mg` | Ibuprofen 600mg |
+| `Ibuphil Cold 400-60` | Ibuprofen/Pseudoephedrine combination |
+| `Xyzall 5mg` | Levocetirizine 5mg |
+| `blue` | Blue-colored pills |
+| `pink` | Pink-colored pills |
+| `red` | Red-colored pills |
+| `white` | White-colored pills |
+
+## Usage
+
+### Quick Start with Pipeline
+
+```python
+from transformers import pipeline
+from PIL import Image
+
+detector = pipeline("object-detection", model="SARANGx/rtdetr-pill-detector")
+image = Image.open("pills.jpg")
+
+results = detector(image, threshold=0.5)
+for r in results:
+    print(f"{r['label']}: {r['score']:.2%} at {r['box']}")
+
+print(f"Total pills: {len(results)}")
+```
+
+### Manual Inference (more control)
+
+```python
+import torch
+from transformers import RTDetrForObjectDetection, RTDetrImageProcessor
+from PIL import Image
+from collections import Counter
+
+model_id = "SARANGx/rtdetr-pill-detector"
+device = "cuda" if torch.cuda.is_available() else "cpu"
+
+image_processor = RTDetrImageProcessor.from_pretrained(model_id)
+model = RTDetrForObjectDetection.from_pretrained(model_id).to(device).eval()
+
+image = Image.open("pills.jpg").convert("RGB")
+inputs = image_processor(images=image, return_tensors="pt").to(device)
+
+with torch.no_grad():
+    outputs = model(**inputs)
+
+# Post-process — boxes in original image coordinates
+target_sizes = torch.tensor([(image.height, image.width)], device=device)
+results = image_processor.post_process_object_detection(
+    outputs, target_sizes=target_sizes, threshold=0.5
+)[0]
+
+# Count pills by class
+counts = Counter()
+for score, label_id, box in zip(results["scores"], results["labels"], results["boxes"]):
+    label = model.config.id2label[label_id.item()]
+    counts[label] += 1
+    x1, y1, x2, y2 = box.tolist()
+    print(f"  {label}: {score:.2%} at [{x1:.0f}, {y1:.0f}, {x2:.0f}, {y2:.0f}]")
+
+print(f"\nTotal pills detected: {sum(counts.values())}")
+for label, count in sorted(counts.items(), key=lambda x: -x[1]):
+    print(f"  {label}: {count}")
+```
+
+## Training Details
+
+### Dataset
+- **[Francesco/pills-sxdht](https://huggingface.co/datasets/Francesco/pills-sxdht)** from Roboflow 100
+- 316 training / 45 validation / 90 test images
+- All images 640×640 with COCO-format bounding box annotations
+- 9 object classes (pills + specific medications + color categories)
+
+### Training Configuration
+| Parameter | Value |
+|-----------|-------|
+| Base model | PekingU/rtdetr_r18vd_coco_o365 (Objects365 pretrained) |
+| Image size | 480×480 |
+| Epochs | 20 |
+| Batch size | 8 |
+| Learning rate | 5e-5 |
+| LR scheduler | Cosine with 50 warmup steps |
+| Optimizer | AdamW (fused) |
+| Max grad norm | 0.1 |
+| Augmentations | HorizontalFlip, ColorJitter, RandomBrightnessContrast, GaussNoise, Blur |
+
+### Training Loss Curve
+
+| Epoch | Train Loss | Eval Loss |
+|-------|-----------|-----------|
+| 1 | 36.61 | 25.70 |
+| 5 | 8.57 | 5.21 |
+| 10 | 6.32 | 3.79 |
+| 15 | 5.55 | 3.53 |
+| **19** | **5.31** | **3.53 (best)** |
+| 20 | 5.78 | 3.58 |
+
+Best validation loss: **3.528** at epoch 19 (loaded as final checkpoint).
+
+### Technical Notes
+- Classification head re-initialized from 80 COCO classes → 9 pill classes
+- Auxiliary loss enabled for training stability
+- `freeze_backbone_batch_norms=True` to preserve pretrained backbone statistics
+- Focal loss (α=0.75, γ=2.0) for handling class imbalance
+
+## Limitations
+- Trained on a small dataset (316 images) — may not generalize well to all pill types
+- Best on images similar to training data (top-down views, clean backgrounds)
+- Color-based classes (blue, pink, red, white) may overlap with medication-specific classes
+- Not intended for medical decision-making — for counting/inventory purposes only
+
+## Framework Versions
+- Transformers 5.6.1
+- PyTorch 2.11.0
+- Datasets 4.8.4
 
+## Citation
 
-### Framework versions
+If you use this model, please cite the underlying RT-DETR architecture:
 
-- Transformers 5.6.1
-- Pytorch 2.11.0+cu130
-- Datasets 4.8.4
-- Tokenizers 0.22.2
+```bibtex
+@article{zhao2024detrs,
+  title={DETRs Beat YOLOs on Real-time Object Detection},
+  author={Zhao, Yian and Lv, Wenyu and Xu, Shangliang and Wei, Jinman and Wang, Guanzhong and Dang, Qingqing and Liu, Yi and Chen, Jie},
+  journal={CVPR},
+  year={2024}
+}
+```