Create app.py

app.py

@@ -0,0 +1,40 @@
+import gradio as gr
+import tensorflow as tf
+import numpy as np
+from keras.utils import normalize
+
+def dice_coef(y_true, y_pred):
+    smooth = 1e-5
+    intersection = K.sum(y_true * y_pred, axis=[1, 2, 3])
+    union = K.sum(y_true, axis=[1, 2, 3]) + K.sum(y_pred, axis=[1, 2, 3])
+    return K.mean((2.0 * intersection + smooth) / (union + smooth), axis=0)
+
+def predict_segmentation(image):
+    SIZE_X = 128
+    SIZE_Y = 128
+
+    img = cv2.resize(image, (SIZE_Y, SIZE_X))
+    img = np.expand_dims(img, axis=2)
+    img = normalize(img, axis=1)
+
+    # Prepare image for prediction
+    img = np.expand_dims(img, axis=0)
+
+    # Predict
+    prediction = model.predict(img)
+    predicted_img = np.argmax(prediction, axis=3)[0, :, :]
+
+    return predicted_img
+
+# Load the model
+model = tf.keras.models.load_model("path_to_your_model_directory", custom_objects={'dice_coef': dice_coef})
+
+# Gradio Interface
+iface = gr.Interface(
+    fn=predict_segmentation,
+    inputs="image",
+    outputs="image",
+    live=False
+)
+
+iface.launch()