Upload 3 files

DenseNet121v2_95.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0a90907a1a27a8d477f847702ec685b468f02037d9fc15d9bd6d384696983610
+size 29765928

app.py

@@ -0,0 +1,20 @@
+import gradio as gr
+import numpy as np
+import cv2
+from model import classify_image
+
+def main(img):
+    img = cv2.resize(img, (224,224))
+    img = img/255.0
+    img = np.expand_dims(img, axis=0)
+    label, accuracy = classify_image(img)
+    print(label)
+    out = {label: accuracy}
+    return out
+
+demo = gr.Interface(fn=main,
+             inputs=gr.Image(),
+             outputs=gr.Label(num_top_classes=1), allow_flagging='never')
+
+if __name__ == "__main__":
+    demo.launch(share=True)

model.py

@@ -0,0 +1,20 @@
+from keras.models import load_model
+import numpy as np
+
+labels = ['Chalky Soil', 'Mary Soil', 'Sand', 'Slit SOil', 'Alluvial Soil', 'Black Soil', 'Clay Soil', 'Red Soil']
+MODEL_PATH = r"D:\Official\Reseach Projects\Official Project\soil-type-classification\models\DenseNet121v2_95.h5"
+
+def load_CNN_model():
+    model = load_model(MODEL_PATH)
+    print("model loaded")
+    return model
+
+def classify_image(img):
+    model = load_CNN_model()
+    prediction = model.predict(img)
+    predicted_class = np.argmax(prediction)
+
+    predicted_label = labels[predicted_class]
+    accuracy = prediction[0][predicted_class]
+
+    return predicted_label, accuracy