Upload 20 files

.gitattributes

@@ -33,3 +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
+examples/cat2.jpg filter=lfs diff=lfs merge=lfs -text
+examples/dog2.jpg filter=lfs diff=lfs merge=lfs -text
+Results/InceptionV3_Dogs_and_Cats_Classification.mp4 filter=lfs diff=lfs merge=lfs -text

Results/InceptionV3_Dogs_and_Cats_Classification.mp4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:40c4b5378681b9aea4108a5ee5db31cb026b110d6c1509cd9406b65cb8f845eb
+size 5820619

app.py

@@ -0,0 +1,7 @@
+import os
+from app.gradio_app import demo
+
+if __name__ == "__main__":
+    port = int(os.environ.get("PORT", 7860))
+    print(f"Launching Gradio demo on 0.0.0.0:{port}")
+    demo.launch(server_name="0.0.0.0", server_port=port, share=False, show_error=True)

app/gradio_app.py

@@ -0,0 +1,87 @@
+import gradio as gr
+from PIL import Image
+from .model import predict
+import os
+
+model_path = os.path.join(
+    os.path.dirname(os.path.dirname(__file__)), 
+    "saved_model", 
+    "InceptionV3_Dogs_and_Cats_Classification.h5"
+)
+
+def classify_image(image):
+    if image is None:
+        return None, {"error": "Please upload an image"}
+    
+    try:
+        label, confidence, probs = predict(image)
+        results = {
+            "Predicted Class": label,
+            "Confidence": f"{confidence * 100:.2f}%",
+            "Cat Probability": f"{probs['Cat'] * 100:.2f}%",
+            "Dog Probability": f"{probs['Dog'] * 100:.2f}%"
+        }
+        return image, results
+    
+    except Exception as e:
+        return image, {"error": f"Classification failed: {str(e)}"}
+
+with gr.Blocks(title="Cats vs Dogs Classifier", theme=gr.themes.Soft()) as demo:
+    gr.Markdown(
+        """
+        # Cats vs Dogs Classifier
+        
+        Upload an image of a cat or dog, and the InceptionV3 model will classify it!
+        
+        **Model:** InceptionV3 (Transfer Learning)  
+        **Classes:** Cat | Dog  
+        **Image Size:** 256x256 pixels
+
+        **NOTE:**
+            - You can upload an image from your device, just press "X" icon and start uploading.
+        """
+    )
+    
+    with gr.Row():
+        with gr.Column():
+            gr.Markdown("### Upload Image")
+            image_input = gr.Image(
+                type="pil",
+                label="Upload Image",
+                sources=["upload", "webcam"],
+                interactive=True
+            )
+        with gr.Column():
+            gr.Markdown("### Prediction Results")
+            output = gr.JSON(label="Classification Results")
+    
+    submit_btn = gr.Button("Classify Image", variant="primary", scale=1)
+    submit_btn.click(
+        fn=classify_image,
+        inputs=image_input,
+        outputs=[image_input, output]
+    )
+    
+    gr.Markdown("### Examples")
+    gr.Examples(
+        examples=[
+            ["examples/cat1.jpg"],
+            ["examples/cat2.jpg"],
+            ["examples/cat3.jpg"],
+            ["examples/dog1.jpg"],
+            ["examples/dog2.jpg"]
+        ],
+        inputs=image_input,
+        outputs=[image_input, output],
+        fn=classify_image,
+        run_on_click=True,
+        label="Example Images (Click to run)"
+    )
+
+if __name__ == "__main__":
+    demo.launch(
+        server_name="0.0.0.0",
+        server_port=7860,
+        share=False,
+        show_error=True
+    )

app/main.py

@@ -0,0 +1,26 @@
+from fastapi import FastAPI, UploadFile, File
+from fastapi.responses import JSONResponse
+from app.model import predict
+from PIL import Image
+import io
+
+app = FastAPI(title="Cats and Dogs Image Classifier")
+
+@app.post("/predict")
+async def predict_image(file: UploadFile = File(...)):
+    try:
+        # Read image from uploaded file
+        contents = await file.read()
+        img = Image.open(io.BytesIO(contents))
+
+        # Run prediction
+        label, confidence, probs = predict(img)
+
+        return JSONResponse(content={
+            "predicted_label": label,
+            "confidence": round(confidence, 3),
+            "probabilities": {k: round(v, 3) for k, v in probs.items()}
+        })
+
+    except Exception as e:
+        return JSONResponse(content={"error": str(e)}, status_code=500)

app/model.py

@@ -0,0 +1,42 @@
+import tensorflow as tf
+import numpy as np
+from PIL import Image
+
+# Load your trained CNN model
+model = tf.keras.models.load_model(
+    "saved_model/InceptionV3_Dogs_and_Cats_Classification.h5",
+    compile=False
+)
+
+# Same label order you used when training (from LabelEncoder)
+CLASS_NAMES = ["Cat", "Dog"]
+
+def preprocess_image(img: Image.Image, target_size=(256, 256)):
+
+    img = img.convert("RGB")  # ensure 3 channels
+    img = img.resize(target_size)
+    img = np.array(img).astype("float32") / 255.0  # normalize
+    img = np.expand_dims(img, axis=0)  # (1, 256, 256, 3)
+    return img
+
+def predict(img: Image.Image):
+    # Apply preprocessing
+    input_tensor = preprocess_image(img)  # (1, 256, 256, 3)
+
+    # Model prediction (sigmoid output)
+    prob = float(model.predict(input_tensor)[0][0])  # probability of class 1 (Dog) or class 0 (Cat)
+
+    # Determine label based on 0.5 threshold
+    if prob >= 0.5:
+        label = CLASS_NAMES[1]  # "Dog"
+    else:
+        label = CLASS_NAMES[0]  # "Cat"
+
+    # Confidence and probability dictionary
+    confidence = prob if label == CLASS_NAMES[1] else 1 - prob
+    prob_dict = {
+        CLASS_NAMES[0]: 1 - prob,
+        CLASS_NAMES[1]: prob
+    }
+
+    return label, confidence, prob_dict

requirements.txt

@@ -0,0 +1,7 @@
+fastapi
+uvicorn
+tensorflow
+numpy
+python-multipart
+pillow
+gradio

saved_model/InceptionV3_Dogs_and_Cats_Classification.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:78e103158acf66458a00ed9c90203be54e907f27adc2ee9fa9fdc059e944e2e8
+size 142060784