Upload folder using huggingface_hub

.dockerignore

@@ -1,3 +1,8 @@
+# Git
+.git
+.gitignore
+
+# Python
 __pycache__
 *.pyc
 *.pyo
@@ -18,9 +23,46 @@ coverage.xml
 .mypy_cache
 .pytest_cache
 .hypothesis
+
+# OS
 .DS_Store
+.DS_Store?
+._*
+.Spotlight-V100
+.Trashes
+ehthumbs.db
+Thumbs.db
+
+# IDE
+.vscode
+.idea
 *.swp
 *.swo
 *~
-results
-uploads
+
+# Documentation
+*.md
+!README.md
+
+# Temporary files
+*.tmp
+*.temp
+
+# Large files (if any)
+*.zip
+*.tar.gz
+*.tar
+*.rar
+
+# Model files that will be downloaded
+Models/*.keras
+Models/*.h5
+Models/*.pb
+
+# Upload directories
+uploads/*
+!uploads/.gitkeep
+
+# Local testing
+test_images/
+examples/

README.md

@@ -1,66 +1,58 @@
 ---
-title: WBC Classification API
+title: WBC Classification
 emoji: 🩸
 colorFrom: red
 colorTo: blue
 sdk: docker
 pinned: false
 license: mit
+app_port: 7860
 ---
 
-# White Blood Cell Classification API
+# WBC Classification System
 
-This is a FastAPI-based application for classifying white blood cells using deep learning.
+An AI-powered White Blood Cell (WBC) classification system that can identify different types of white blood cells from microscopic images.
+
+## Features
+
+- **Real-time Classification**: Upload images and get instant predictions
+- **Batch Processing**: Process multiple images simultaneously
+- **High Accuracy**: Deep learning model trained on medical imagery
+- **User-friendly Interface**: Simple web interface for easy interaction
 
 ## Supported Cell Types
+
 - Basophil
 - Eosinophil
 - Lymphocyte
 - Monocyte
 - Neutrophil
 
+## Usage
+
+1. **Single Image**: Upload a single microscopic image for classification
+2. **Batch Processing**: Upload multiple images for simultaneous processing
+3. **View Results**: Get detailed predictions with confidence scores
+
 ## API Endpoints
 
-### Health Check
-```
-GET /api/health
-```
+- `GET /` - Main web interface
+- `POST /api/predict` - Single image prediction
+- `POST /api/predict_batch` - Batch image prediction
+- `GET /api/health` - Health check
+- `GET /api/model_info` - Model information
 
-### Single Image Prediction
-```
-POST /api/predict
-```
+## Technical Details
 
-### Batch Prediction
-```
-POST /api/predict_batch
-```
+- **Framework**: FastAPI with TensorFlow
+- **Model**: Custom CNN architecture
+- **Input**: 128x128 RGB images
+- **Output**: Classification with confidence scores
 
-## Usage
+## Development
+
+Built for medical image analysis and research purposes. The model is trained on high-quality microscopic images of white blood cells.
+
+## License
 
-The API accepts image files and returns predictions with confidence scores.
-
-Example response:
-```json
-{
-  "predictions": [
-    {
-      "filename": "image.jpg",
-      "prediction": "Monocyte",
-      "confidence": 0.9876,
-      "all_probabilities": {
-        "Basophil": 0.0012,
-        "Eosinophil": 0.0045,
-        "Lymphocyte": 0.0023,
-        "Monocyte": 0.9876,
-        "Neutrophil": 0.0044
-      }
-    }
-  ],
-  "processing_time": 1.23
-}
-```
-
-## Model Information
-
-The model is trained to classify white blood cell images with high accuracy using convolutional neural networks.
+MIT License - See LICENSE file for details.

orchestration.py

@@ -530,7 +530,7 @@ def main():
     config = load_configuration()
     
     # Update these paths according to your setup
-    MODEL_PATH = "/home/amit/White_blood_application/Models/at_batch_size=32.D24E.keras"  # Update this path
+    MODEL_PATH = "Models/at_batch_size=32.D24E.keras"  # Update this path
     TEST_IMAGE_PATH = "/home/amit/White_blood_application/uploads"  # Update this path  
     TEST_FOLDER_PATH = "/home/amit/White_blood_application/uploads"  # Update this path
     
@@ -573,5 +573,3 @@ def main():
     print("Update the file paths above to test with your actual data.")
     print(f"{'='*60}")
 
-if __name__ == "__main__":
-    main()

requirements.txt

@@ -14,4 +14,5 @@ opencv-python==4.8.1.78
 
 # Data Science
 matplotlib==3.9.2
-pandas==2.1.3
+pandas==2.1.3
+huggingface-hub==0.17.3