Update pages/Entorno de Ejecución.py

pages/Entorno de Ejecución.py

@@ -20,6 +20,22 @@ st.sidebar.write("contact@patacotron.tech")
 
 cnn, autoencoder, svm, iforest, gan, vit, zero_shot= st.tabs(["CNN", "Autoencoder", "OC-SVM", 'iForest', 'GAN', 'ViT', 'Zero-Shot'])
 
+def predict(model_list, weights, img): #for non-supported formats
+    y_gorrito = 0
+    raw_img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)
+    img = cv2.resize(img, (IMAGE_WIDTH, IMAGE_HEIGHT))
+    for model, weight in zip(model_list, weights):
+        y_gorrito += tf.cast(model(tf.expand_dims(img/255., 0)), dtype=tf.float32)*weight
+    return [y_gorrito / sum(weights), raw_img]
+
+def preprocess(file_uploader, module = 'cv2'): #makes the uploaded image readable
+    img = np.frombuffer(uploaded_file.read(), np.uint8)
+    if module == 'cv2':
+        img = cv2.imdecode(img, cv2.IMREAD_COLOR)
+    elif module == 'pil':
+        img = Image.open(file_uploader.read())
+    return img
+
 with cnn:
 
     col_a, col_b, = st.columns(2)
@@ -62,23 +78,6 @@ with cnn:
         threshold = st.slider('¿Cuál va a ser el límite donde se considere patacón? (el valor recomendado para Ultra-Patacotrón es 50%, para los demás, 75%-80%)', 0, 100, 50)
         
         selected_models = []
-    
-        def predict(model_list, weights, img): #for non-supported formats
-            y_gorrito = 0
-            raw_img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)
-            img = cv2.resize(img, (IMAGE_WIDTH, IMAGE_HEIGHT))
-            for model, weight in zip(model_list, weights):
-                y_gorrito += tf.cast(model(tf.expand_dims(img/255., 0)), dtype=tf.float32)*weight
-            return [y_gorrito / sum(weights), raw_img]
-
-        def preprocess(file_uploader, module = 'cv2'): #makes the uploaded image readable
-            img = np.frombuffer(uploaded_file.read(), np.uint8)
-            if module == 'cv2':
-                img = cv2.imdecode(img, cv2.IMREAD_COLOR)
-            elif module == 'pil':
-                img = Image.frombuffer(data = uploaded_file.read())
-                #img = Image.open(io.BytesIO(file_uploader.read()))
-            return img
                 
         # Set the image dimensions
         IMAGE_WIDTH = IMAGE_HEIGHT = 224
@@ -146,9 +145,11 @@ with vit:
                     img = preprocess(uploaded_file, module = 'pil')
                     raw_img = img
                     img = cv2.resize(img, (IMAGE_WIDTH, IMAGE_HEIGHT))     
-                    classifier(img)
             else:
                 st.write("Asegúrate de haber subido correctamente la imagen.")
+    with col_b:
+        st.write(classifier(img))
+        st.image(raw_img)
 
 with zero_shot:
     st.write('Próximamente')