Switch model to ResNet18, update app and dependencies

.gitattributes

@@ -33,4 +33,6 @@ saved_model/**/* 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
+*.data filter=lfs diff=lfs merge=lfs -text
+models/resnet18.onnx.data filter=lfs diff=lfs merge=lfs -text
 *.onnx.data filter=lfs diff=lfs merge=lfs -text

Dockerfile

@@ -1,13 +1,26 @@
+# PV Defect Classifier — HuggingFace Spaces
 FROM python:3.11-slim
+
+# HuggingFace requirement: UID 1000 non-root user
 RUN useradd -m -u 1000 user
+
 WORKDIR /app
+
+# Install dependencies (cache layer)
 COPY requirements.txt .
 RUN pip install --no-cache-dir -r requirements.txt
 RUN pip install --no-cache-dir gunicorn
+
+# Copy application code (--chown avoids permission issues)
 COPY --chown=user app.py .
 COPY --chown=user templates/ templates/
 COPY --chown=user models/ models/
+
+# Switch to non-root user
 USER user
 ENV HOME=/home/user \
     PATH=/home/user/.local/bin:$PATH
+
+# HuggingFace Spaces requires port 7860
+# 1 worker = 1 model copy in memory; timeout 120s for cold start
 CMD ["gunicorn", "--bind", "0.0.0.0:7860", "--workers", "1", "--timeout", "120", "app:app"]

app.py

@@ -4,11 +4,6 @@ PV Defect Classification — Flask Demo
 Loads the best ONNX model and serves a web interface for
 real-time photovoltaic panel defect classification.
 
-Usage:
-  1. Put your .onnx model file in the /models folder
-  2. pip install flask onnxruntime pillow numpy
-  3. python app.py
-  4. Open http://localhost:7860 in your browser
 """
 
 import os
@@ -18,7 +13,7 @@ from PIL import Image
 from flask import Flask, render_template, request, jsonify
 import onnxruntime as ort
 
-# ── Config ────────────────────────────────────────────────────
+#  Config 
 MODEL_DIR = os.path.join(os.path.dirname(__file__), "models")
 CLASS_NAMES = ["DEFECTIVE", "NORMAL"]
 IMG_SIZE = 224
@@ -27,7 +22,7 @@ STD  = np.array([0.229, 0.224, 0.225], dtype=np.float32)
 
 app = Flask(__name__)
 
-# ── Load ONNX model ──────────────────────────────────────────
+# Load ONNX model 
 def find_onnx_model():
     """Auto-detect the first .onnx file in /models."""
     for f in os.listdir(MODEL_DIR):
@@ -39,13 +34,13 @@ model_path = find_onnx_model()
 if model_path:
     session = ort.InferenceSession(model_path)
     input_name = session.get_inputs()[0].name
-    print(f"✅ Loaded model: {os.path.basename(model_path)}")
+    print(f"Loaded model: {os.path.basename(model_path)}")
 else:
     session = None
-    print("⚠️  No .onnx file found in /models — place your model there and restart.")
+    print(" No .onnx file found in /models — place your model there and restart.")
 
 
-# ── Preprocessing (same as val_tf in your notebook) ──────────
+#  Preprocessing
 def preprocess(image: Image.Image) -> np.ndarray:
     """Resize, normalise, and convert PIL image to ONNX input tensor."""
     img = image.convert("RGB").resize((IMG_SIZE, IMG_SIZE))
@@ -60,7 +55,7 @@ def softmax(x):
     return e / e.sum()
 
 
-# ── Routes ────────────────────────────────────────────────────
+# Routes 
 @app.route("/")
 def index():
     model_name = os.path.basename(model_path) if model_path else "No model loaded"
@@ -117,4 +112,4 @@ def health():
 
 
 if __name__ == "__main__":
-    app.run(debug=False, host="0.0.0.0", port=7860)
+    app.run(debug=True, host="0.0.0.0", port=5000)

models/resnet18.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0c17d9def594a5c33a89d6331b6497f9c145e40dd4aa44a728b8a82bd9119c46
+size 88415

models/resnet18.onnx.data

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:498153c0a1da770809f354285351ccdc58a36818d0e5dbfe13127e4cac4c0b6d
+size 44695552

requirements.txt

@@ -1,4 +1,4 @@
 flask==3.0.0
-onnxruntime>=1.18.1
+onnxruntime>=1.20.0
 Pillow==10.2.0
 numpy==1.26.3