Upload folder using huggingface_hub

.gitattributes

@@ -1,35 +1,2 @@
-*.7z filter=lfs diff=lfs merge=lfs -text
-*.arrow filter=lfs diff=lfs merge=lfs -text
-*.bin filter=lfs diff=lfs merge=lfs -text
-*.bz2 filter=lfs diff=lfs merge=lfs -text
-*.ckpt filter=lfs diff=lfs merge=lfs -text
-*.ftz filter=lfs diff=lfs merge=lfs -text
-*.gz filter=lfs diff=lfs merge=lfs -text
-*.h5 filter=lfs diff=lfs merge=lfs -text
-*.joblib filter=lfs diff=lfs merge=lfs -text
-*.lfs.* filter=lfs diff=lfs merge=lfs -text
-*.mlmodel filter=lfs diff=lfs merge=lfs -text
-*.model filter=lfs diff=lfs merge=lfs -text
-*.msgpack filter=lfs diff=lfs merge=lfs -text
-*.npy filter=lfs diff=lfs merge=lfs -text
-*.npz filter=lfs diff=lfs merge=lfs -text
-*.onnx filter=lfs diff=lfs merge=lfs -text
-*.ot filter=lfs diff=lfs merge=lfs -text
-*.parquet filter=lfs diff=lfs merge=lfs -text
-*.pb filter=lfs diff=lfs merge=lfs -text
-*.pickle filter=lfs diff=lfs merge=lfs -text
-*.pkl filter=lfs diff=lfs merge=lfs -text
-*.pt filter=lfs diff=lfs merge=lfs -text
-*.pth filter=lfs diff=lfs merge=lfs -text
-*.rar filter=lfs diff=lfs merge=lfs -text
-*.safetensors filter=lfs diff=lfs merge=lfs -text
-saved_model/**/* filter=lfs diff=lfs merge=lfs -text
-*.tar.* filter=lfs diff=lfs merge=lfs -text
-*.tar filter=lfs diff=lfs merge=lfs -text
-*.tflite filter=lfs diff=lfs merge=lfs -text
-*.tgz filter=lfs diff=lfs merge=lfs -text
-*.wasm filter=lfs diff=lfs merge=lfs -text
-*.xz filter=lfs diff=lfs merge=lfs -text
-*.zip filter=lfs diff=lfs merge=lfs -text
-*.zst filter=lfs diff=lfs merge=lfs -text
-*tfevents* filter=lfs diff=lfs merge=lfs -text
+*.safetensors filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text

.gitignore

@@ -0,0 +1,7 @@
+train.py
+test.py
+test.jpeg
+vit_trainer_output
+__pycache__
+data
+venv

README.md

@@ -1,3 +1,65 @@
----
-license: mit
----
+# Waste Classifier AI
+
+## Dataset Structure
+
+Place your dataset in a directory like:
+
+```
+waste_dataset/
+├── biodegradable/
+│   ├── img1.jpg
+│   └── ...
+└── non_biodegradable/
+    ├── img2.jpg
+    └── ...
+```
+
+## Training the Model
+
+1. Install dependencies:
+   ```
+   pip install -r requirements.txt
+   ```
+2. Run the training script:
+   ```
+   python train_waste_classifier.py
+   ```
+   This will save the model and processor to `models/waste_classifier_model/`.
+
+## Running the API
+
+1. Start the Flask API:
+   ```
+   python waste_classification_api.py
+   ```
+2. The API will be available at `http://localhost:5000/classify`.
+
+## API Usage
+
+- **Endpoint:** `/classify` (POST)
+- **Payload:** Multipart form with one or more images (field name: `images`)
+- **Response:**
+  ```json
+  {
+    "results": [
+      {
+        "label": "biodegradable",
+        "confidence": 0.94,
+        "description": "Easily breaks down naturally. Good for composting.",
+        "recyclable": false,
+        "disposal": "Use compost or organic bin",
+        "example_items": ["banana peel", "food waste", "paper"]
+      },
+      ...
+    ]
+  }
+  ```
+
+## Frontend Integration
+
+- The React Native frontend can POST images to `/classify` and display the results using the modal.
+- No changes are needed to the modal if it expects the above JSON structure.
+
+## Notes
+- If your dataset folders are named differently (e.g., `R` and `O`), update the LABEL2INFO mapping and class names in the training script.
+- The model is based on `google/vit-base-patch16-224` and fine-tuned for binary classification. 

requirements.txt

@@ -0,0 +1,8 @@
+torch>=2.0.0
+torchvision>=0.15.0
+transformers>=4.30.0
+datasets>=2.10.0
+scikit-learn>=1.0.0
+flask>=2.0.0
+flask-cors>=3.0.10
+pillow>=9.0.0 

waste_classification_api.py

@@ -0,0 +1,84 @@
+from flask import Flask, request, jsonify
+from flask_cors import CORS
+from transformers import AutoImageProcessor, AutoModelForImageClassification
+from PIL import Image
+import torch
+import io
+import os
+from pathlib import Path
+
+app = Flask(__name__)
+CORS(app)
+
+MODEL_PATH = r"D:/Green_IQ/Green_IQ/AI/waste_classifier"
+
+LABEL2INFO = {
+    0: {
+        "label": "biodegradable",
+        "description": "Easily breaks down naturally. Good for composting.",
+        "recyclable": False,
+        "disposal": "Use compost or organic bin",
+        "example_items": ["banana peel", "food waste", "paper"],
+        "environmental_benefit": "Composting biodegradable waste returns nutrients to the soil, reduces landfill use, and lowers greenhouse gas emissions.",
+        "protection_tip": "Compost at home or use municipal organic waste bins. Avoid mixing with plastics or hazardous waste.",
+        "poor_disposal_effects": "If disposed of improperly, biodegradable waste can cause methane emissions in landfills and contribute to water pollution and eutrophication."
+    },
+    1: {
+        "label": "non_biodegradable",
+        "description": "Does not break down easily. Should be disposed of carefully.",
+        "recyclable": False,
+        "disposal": "Use general waste bin or recycling if possible",
+        "example_items": ["plastic bag", "styrofoam", "metal can"],
+        "environmental_benefit": "Proper disposal and recycling of non-biodegradable waste reduces pollution, conserves resources, and protects wildlife.",
+        "protection_tip": "Reduce use, reuse items, and recycle whenever possible. Never burn or dump in nature.",
+        "poor_disposal_effects": "Improper disposal leads to soil and water pollution, harms wildlife, and causes long-term environmental damage. Plastics can persist for hundreds of years."
+    }
+}
+
+# Check if the model path exists
+if not os.path.exists(MODEL_PATH):
+    raise FileNotFoundError(f"Model path does not exist: {MODEL_PATH}")
+
+# Load model and processor with local_files_only=True
+try:
+    model = AutoModelForImageClassification.from_pretrained(
+        MODEL_PATH, 
+        local_files_only=True
+    )
+    image_processor = AutoImageProcessor.from_pretrained(
+        MODEL_PATH, 
+        local_files_only=True
+    )
+    model.eval()
+    print("Model and processor loaded successfully!")
+except Exception as e:
+    print(f"Error loading model: {e}")
+    raise
+
+def predict_image(image_bytes, model, image_processor, device="cpu"):
+    image = Image.open(io.BytesIO(image_bytes)).convert("RGB")
+    inputs = image_processor(images=image, return_tensors="pt")
+    inputs = {k: v.to(device) for k, v in inputs.items()}
+    with torch.no_grad():
+        outputs = model(**inputs)
+        probs = torch.softmax(outputs.logits, dim=1)
+        conf, pred = torch.max(probs, dim=1)
+        label_id = pred.item()
+        confidence = conf.item()
+    info = LABEL2INFO[label_id].copy()
+    info["confidence"] = round(confidence, 2)
+    info["eco_points_earned"] = 10  # Dummy value
+    return info
+
+@app.route('/classify', methods=['POST'])
+def classify():
+    results = []
+    files = request.files.getlist('images')
+    for file in files:
+        image_bytes = file.read()
+        result = predict_image(image_bytes, model, image_processor)
+        results.append(result)
+    return jsonify({"results": results})
+
+if __name__ == '__main__':
+    app.run(debug=True, port=5000)

waste_classifier/config.json

@@ -0,0 +1,33 @@
+{
+  "architectures": [
+    "ViTForImageClassification"
+  ],
+  "attention_probs_dropout_prob": 0.0,
+  "encoder_stride": 16,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.0,
+  "hidden_size": 768,
+  "id2label": {
+    "0": "biodegradable",
+    "1": "non_biodegradable"
+  },
+  "image_size": 224,
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "label2id": {
+    "biodegradable": 0,
+    "non_biodegradable": 1
+  },
+  "layer_norm_eps": 1e-12,
+  "model_type": "vit",
+  "num_attention_heads": 12,
+  "num_channels": 3,
+  "num_hidden_layers": 12,
+  "patch_size": 16,
+  "pooler_act": "tanh",
+  "pooler_output_size": 768,
+  "problem_type": "single_label_classification",
+  "qkv_bias": true,
+  "torch_dtype": "float32",
+  "transformers_version": "4.53.2"
+}

waste_classifier/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:af8a93025faf3e4fd053b12c9825a286a22ec22f64a50a8f27728a73cd5c078b
+size 343223968

waste_classifier/preprocessor_config.json

@@ -0,0 +1,31 @@
+{
+  "crop_size": null,
+  "data_format": "channels_first",
+  "default_to_square": true,
+  "device": null,
+  "disable_grouping": null,
+  "do_center_crop": null,
+  "do_convert_rgb": null,
+  "do_normalize": true,
+  "do_rescale": true,
+  "do_resize": true,
+  "image_mean": [
+    0.5,
+    0.5,
+    0.5
+  ],
+  "image_processor_type": "ViTImageProcessorFast",
+  "image_std": [
+    0.5,
+    0.5,
+    0.5
+  ],
+  "input_data_format": null,
+  "resample": 2,
+  "rescale_factor": 0.00392156862745098,
+  "return_tensors": null,
+  "size": {
+    "height": 224,
+    "width": 224
+  }
+}