Update classification.py

classification.py

@@ -3,18 +3,18 @@ import time
 from tensorflow.keras.preprocessing import image
 # from tensorflow.keras.preprocessing.image import ImageDataGenerator
 import tensorflow as tf
-# gpus = tf.config.experimental.list_physical_devices('GPU')
-# if gpus:
-#     try:
-#         for gpu in gpus:
-#             tf.config.experimental.set_memory_growth(gpu, True)
-#     except RuntimeError as e:
-#         # Memory growth must be set before GPUs have been initialized
-#         print(e)
+gpus = tf.config.experimental.list_physical_devices('GPU')
+if gpus:
+    try:
+        for gpu in gpus:
+            tf.config.experimental.set_memory_growth(gpu, True)
+    except RuntimeError as e:
+        # Memory growth must be set before GPUs have been initialized
+        print(e)
 import streamlit as st
-with tf.device('/cpu:0'):
+# with tf.device('/cpu:0'):
 # Load the saved model
-    model = tf.keras.models.load_model('best_resnet152_model.h5')
+model = tf.keras.models.load_model('best_resnet152_model.h5')
 
 class_names = {0: '1099_Div', 1: '1099_Int', 2: 'Non_Form', 3: 'w_2', 4: 'w_3'}
 # print(class_names)
@@ -29,16 +29,16 @@ def predict(pil_img):
     img_array /= 255.0  # Rescale pixel values
 
     # Predict the class
-    with tf.device('/cpu:0'):
-        start_time = time.time()
-        predictions = model.predict(img_array)
-        end_time = time.time()
-        predicted_class_index = np.argmax(predictions, axis=1)[0]
+    # with tf.device('/cpu:0'):
+    start_time = time.time()
+    predictions = model.predict(img_array)
+    end_time = time.time()
+    predicted_class_index = np.argmax(predictions, axis=1)[0]
 
-        # Get the predicted class name
-        predicted_class_name = class_names[predicted_class_index]
-        print("Predicted class:", predicted_class_name)
-        print("Execution time: ", end_time - start_time)
+    # Get the predicted class name
+    predicted_class_name = class_names[predicted_class_index]
+    print("Predicted class:", predicted_class_name)
+    print("Execution time: ", end_time - start_time)
     return predicted_class_name
 # import numpy as np
 # import time