Update README: remove gated access, add model details and demo links

README.md

@@ -6,108 +6,198 @@ tags:
 - image-segmentation
 - pytorch
 - unet
+- fungal-colony
+- petri-dish
+- morphometry
+- magnaporthe
+- area-consistency
 language:
 - en
 ---
 
-# Gray Leaf Spot Segmentation Model
+# 🔬 Gray Leaf Spot Segmentation Model
 
-PyTorch U-Net segmentation model trained for gray leaf spot segmentation.
+PyTorch U-Net models for **gray leaf spot** (*Magnaporthe* and related fungal) colony segmentation on 90 mm petri-dish images.
 
+**[▶ Try the live demo](https://huggingface.co/spaces/rotsl/grayleafspot-segmentation-demo)** — upload images, run inference, see overlays & growth charts in your browser.
 
+---
+
+## Model Weights
+
+| File | Architecture | Params | Area-Consistency Weight | Description |
+|---|---|---|---|---|
+| `grayleafspot.pt` | smp.Unet (ResNet-34) | ~24.4 M | — | Main encoder–decoder model |
+| `best_area_w_0.1.pt` | SmallUNet | ~250 K | 0.1 | Light area regularisation |
+| `best_area_w_0.3.pt` | SmallUNet | ~250 K | 0.3 | Moderate area regularisation |
+| `best_area_w_0.5.pt` | SmallUNet | ~250 K | 0.5 | Balanced BCE + area |
+| **`best_area_w_0.7.pt`** | **SmallUNet** | **~250 K** | **0.7** | **Strong area consistency (used by demo)** |
+
+All SmallUNet variants share the same architecture:
+
+```
+Input (3 × 256 × 256)
+  │
+  ├─ enc1: ConvBlock(3 → 16)              ─── skip s1
+  ├─ enc2: MaxPool2d → ConvBlock(16 → 32)  ─── skip s2
+  ├─ enc3: MaxPool2d → ConvBlock(32 → 64)  ─── skip s3
+  ├─ enc4: MaxPool2d → ConvBlock(64 → 128) ─── skip s4
+  │
+  ├─ bottleneck: MaxPool2d → ConvBlock(128 → 256)
+  │
+  ├─ up4: Upsample + cat(s4) → ConvBlock(384 → 128)
+  ├─ up3: Upsample + cat(s3) → ConvBlock(192 → 64)
+  ├─ up2: Upsample + cat(s2) → ConvBlock(96 → 32)
+  ├─ up1: Upsample + cat(s1) → ConvBlock(48 → 16)
+  │
+  └─ head: Conv2d(16 → 1) → Sigmoid
+```
 
-- `grayleafspot.pt`: main U-Net model artifact
-- `best_area_w_0.1.pt`: area consistency U-Net (weight=0.1)
-- `best_area_w_0.3.pt`: area consistency U-Net (weight=0.3)
-- `best_area_w_0.5.pt`: area consistency U-Net (weight=0.5)
-- `best_area_w_0.7.pt`: area consistency U-Net (weight=0.7)
-- `requirements.txt`: minimal runtime dependencies
+Each `ConvBlock` = Conv3×3 (no bias) → ReLU → Conv3×3 (no bias) → ReLU.
 
-## Usage
+| Property | Value |
+|---|---|
+| **Input** | 256 × 256 RGB |
+| **Output** | 1-channel sigmoid probability mask |
+| **Training loss** | BCE + area-consistency loss |
+| **CPU compatible** | ✅ Pure PyTorch — no custom CUDA kernels |
+
+---
+
+## Quick Start
+
+### Download & Inference (SmallUNet)
 
 ```python
 import torch
+from huggingface_hub import hf_hub_download
+
+# Download weights
+path = hf_hub_download("rotsl/grayleafspot-segmentation", "best_area_w_0.7.pt")
+
+# Load checkpoint
+ckpt = torch.load(path, map_location="cpu", weights_only=False)
+
+# Build model (SmallUNet architecture — see demo repo for full class definition)
+# https://huggingface.co/rotsl/grayleafspot-segmentation-demo/blob/main/app.py
+from model import SmallUNet  # or copy the class from the demo app.py
 
-model = torch.load("grayleafspot.pt", map_location="cpu")
+model = SmallUNet(in_channels=3, out_channels=1, base_channels=16)
+model.load_state_dict(ckpt["model_state_dict"])
 model.eval()
+
+# Run inference on a 256×256 RGB tensor
+import numpy as np
+from PIL import Image
+
+img = Image.open("petri_dish.jpg").convert("RGB").resize((256, 256))
+x = torch.from_numpy(np.array(img).transpose(2, 0, 1)).float() / 255.0
+x = x.unsqueeze(0)
+
+with torch.no_grad():
+    prob = model(x)[0, 0].numpy()
+
+mask = (prob > 0.5).astype(np.uint8) * 255
+Image.fromarray(mask).save("colony_mask.png")
 ```
 
-## Training & Inference Pipeline
+### Download & Inference (Main U-Net)
 
-This repository follows the TrainPetri pipeline for image segmentation, adapted for gray leaf spot. Key steps:
+```python
+import torch
+
+path = hf_hub_download("rotsl/grayleafspot-segmentation", "grayleafspot.pt")
+model = torch.load(path, map_location="cpu", weights_only=False)
+model.eval()
+```
+
+---
+
+## Training & Inference Pipeline
 
 ### Environment Setup (Apple Silicon recommended)
 
-1. Create and activate a Python 3.10+ virtual environment:
-	 ```bash
-	 python3.10 -m venv trainenv
-	 source trainenv/bin/activate
-	 pip install --upgrade pip
-	 pip install -r requirements.txt
-	 ```
+```bash
+python3.10 -m venv trainenv
+source trainenv/bin/activate
+pip install --upgrade pip
+pip install -r requirements.txt
+```
 
 ### Dataset Preparation
 
-- Place your raw images in `raw/` and corresponding masks in `masks/` (matching filenames).
-- To expand the dataset with augmentations:
-	```bash
-	python src/build_augmented_dataset.py --copies-per-image 4 --clean
-	```
+Place raw images in `raw/` and corresponding masks in `masks/` (matching filenames). Expand the dataset with augmentations:
+
+```bash
+python src/build_augmented_dataset.py --copies-per-image 4 --clean
+```
 
 ### Training
 
-- Standard U-Net training:
-	```bash
-	python src/train.py --image-dir augmented_dataset/raw --mask-dir augmented_dataset/masks --epochs 40 --batch-size 4 --lr 1e-4 --image-size 256 --freeze-encoder-epochs 5
-	```
-- Area consistency U-Net (LabelMe JSON polygons):
-	```bash
-	./trainenv/bin/python src/area_consistency/train_area.py
-	```
-
-### Inference
-
-- On a single image:
-	```bash
-	python src/predict.py --input raw/your_image.jpg --weights models/best_finetuned.pt --output-dir predictions
-	```
-- On a folder:
-	```bash
-	python src/predict.py --input raw --weights models/best_finetuned.pt --output-dir predictions
-	```
+Standard U-Net training:
+
+```bash
+python src/train.py \
+    --image-dir augmented_dataset/raw \
+    --mask-dir augmented_dataset/masks \
+    --epochs 40 --batch-size 4 --lr 1e-4 \
+    --image-size 256 --freeze-encoder-epochs 5
+```
+
+Area-consistency U-Net (LabelMe JSON polygons):
+
+```bash
+./trainenv/bin/python src/area_consistency/train_area.py
+```
+
+### Inference (CLI)
+
+Single image:
+
+```bash
+python src/predict.py --input raw/your_image.jpg --weights models/best_finetuned.pt --output-dir predictions
+```
+
+Folder:
+
+```bash
+python src/predict.py --input raw --weights models/best_finetuned.pt --output-dir predictions
+```
 
 ### Best Practices
 
-- Always keep your trusted human-label masks unchanged.
-- Use augmentations and area consistency loss for improved generalization.
+- Keep trusted human-labelled masks unchanged.
+- Use augmentations and area-consistency loss for improved generalisation.
 - Inspect overlay outputs to verify mask quality.
-- For Apple Silicon, MPS acceleration is used automatically if available.
+- On Apple Silicon, MPS acceleration is used automatically if available.
 
+---
 
+## Demo Space
 
+The full analysis pipeline is available as a Gradio Space:
 
+**[`rotsl/grayleafspot-segmentation-demo`](https://huggingface.co/spaces/rotsl/grayleafspot-segmentation-demo)**
 
-## Citation
+Features: dish detection → colony segmentation → crack & hyphae analysis → 16 morphometric measurements → time-series growth charts → CSV/JSON export.
 
-If you use this model or code, please cite:
+Source code: [`rotsl/grayleafspot-segmentation-demo`](https://huggingface.co/rotsl/grayleafspot-segmentation-demo) (model repo)
 
-```
+---
+
+## Citation
+
+```bibtex
 @misc{rohan_r_2026,
-	author       = { rohan r },
-	title        = { grayleafspot-segmentation (Revision 0e85f71) },
-	year         = 2026,
-	url          = { https://huggingface.co/rotsl/grayleafspot-segmentation },
-	doi          = { 10.57967/hf/8416 },
-	publisher    = { Hugging Face }
+  author       = {rohan r},
+  title        = {grayleafspot-segmentation (Revision 0e85f71)},
+  year         = 2026,
+  url          = {https://huggingface.co/rotsl/grayleafspot-segmentation},
+  doi          = {10.57967/hf/8416},
+  publisher    = {Hugging Face}
 }
 ```
 
-
 ## License
 
-This project is licensed under the Apache License 2.0. See the LICENSE file for details.
-
-## Access
-
-This repository is gated with manual approval enabled on the Hugging Face Hub.
-Users must request access before they can download the model.
+Apache License 2.0 — see [LICENSE](LICENSE) for details.