doem

app.py

@@ -1,7 +1,63 @@
 import gradio as gr
+import numpy as np
+import pickle
+from PIL import Image
 
-def greet(name):
-    return "Hello " + name + "!!"
+# Load the model
+with open('model.pkl', 'rb') as f:
+    model_params = pickle.load(f)
 
-demo = gr.Interface(fn=greet, inputs="text", outputs="text")
-demo.launch()
+W1 = model_params['W1']
+b1 = model_params['b1']
+W2 = model_params['W2']
+b2 = model_params['b2']
+
+def ReLu(Z):
+    return np.maximum(Z, 0)
+
+def softmax(Z):
+    return np.exp(Z) / sum(np.exp(Z))
+
+def forward_prop(W1, b1, W2, b2, X):
+    Z1 = W1.dot(X) + b1
+    A1 = ReLu(Z1)
+    Z2 = W2.dot(A1) + b2
+    A2 = softmax(Z2)
+    return Z1, Z2, A1, A2
+
+def get_predictions(A2):
+    return np.argmax(A2, 0)
+
+def preprocess_image(image):
+    # Convert to grayscale
+    img = image.convert('L')
+    
+    # Resize the image
+    img = img.resize((28, 28))
+    
+    # Convert to numpy array and normalize
+    img_array = np.array(img).reshape(1, 28*28) / 255.0
+    
+    return img_array.T  # Transpose to match the shape (784, 1)
+
+def predict_digit(image):
+    X = preprocess_image(image)
+    
+    # Forward propagation
+    _, _, _, A2 = forward_prop(W1, b1, W2, b2, X)
+    
+    # Get the prediction
+    prediction = get_predictions(A2)
+    
+    return int(prediction[0])
+
+# Gradio interface
+iface = gr.Interface(
+    fn=predict_digit,
+    inputs=gr.Image(type="pil"),
+    outputs=gr.Label(num_top_classes=1),
+    title="Handwritten Digit Recognition",
+    description="Upload an image of a handwritten digit (0-9) and the model will predict which digit it is."
+)
+
+iface.launch()

model.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:508fef3cdc4adbc5402d9694b9519b70df7cd22bd231bb6dd469338e1dda3679
+size 63956

requirements.txt

@@ -0,0 +1,3 @@
+gradio
+numpy
+Pillow