caching updated
Browse files
app.py
CHANGED
|
@@ -31,7 +31,7 @@ cnn_model = 'CNN_model_weight/model_weights.weights.h5'
|
|
| 31 |
|
| 32 |
# CNN model
|
| 33 |
|
| 34 |
-
|
| 35 |
def run_cnn(img_arr):
|
| 36 |
my_model = Sequential()
|
| 37 |
my_model.add(Conv2D(
|
|
@@ -65,7 +65,7 @@ def run_cnn(img_arr):
|
|
| 65 |
prediction = my_model.predict(img_arr)
|
| 66 |
return prediction
|
| 67 |
|
| 68 |
-
|
| 69 |
def run_effNet(img_arr):
|
| 70 |
try:
|
| 71 |
resolver = tf.distribute.cluster_resolver.TPUClusterResolver()
|
|
@@ -78,6 +78,7 @@ def run_effNet(img_arr):
|
|
| 78 |
prediction = eff_net_model.predict(img_arr)
|
| 79 |
return prediction
|
| 80 |
|
|
|
|
| 81 |
def run_effNet_Art(img_arr):
|
| 82 |
try:
|
| 83 |
resolver = tf.distribute.cluster_resolver.TPUClusterResolver()
|
|
@@ -90,7 +91,6 @@ def run_effNet_Art(img_arr):
|
|
| 90 |
prediction = eff_net_art_model.predict(img_arr)
|
| 91 |
return prediction
|
| 92 |
|
| 93 |
-
@st.cache_resource
|
| 94 |
def pre_process_img_effNet(image):
|
| 95 |
img = load_img(image, target_size=(300, 300)) # Resize image to model input size
|
| 96 |
img_arr = img_to_array(img) # Convert to array
|
|
@@ -98,7 +98,6 @@ def pre_process_img_effNet(image):
|
|
| 98 |
result = run_effNet(img_arr)
|
| 99 |
return result
|
| 100 |
|
| 101 |
-
@st.cache_resource
|
| 102 |
def pre_process_img_effNetArt(image):
|
| 103 |
img = load_img(image, target_size=(224, 224)) # Resize image to model input size
|
| 104 |
img_arr = img_to_array(img) # Convert to array
|
|
@@ -107,7 +106,6 @@ def pre_process_img_effNetArt(image):
|
|
| 107 |
return result
|
| 108 |
|
| 109 |
# preprocess image for cnn
|
| 110 |
-
@st.cache_resource
|
| 111 |
def pre_process_img(image):
|
| 112 |
# Load and preprocess the image
|
| 113 |
input_picture = load_img(image, target_size=(256, 256))
|
|
|
|
| 31 |
|
| 32 |
# CNN model
|
| 33 |
|
| 34 |
+
@st.cache_resource
|
| 35 |
def run_cnn(img_arr):
|
| 36 |
my_model = Sequential()
|
| 37 |
my_model.add(Conv2D(
|
|
|
|
| 65 |
prediction = my_model.predict(img_arr)
|
| 66 |
return prediction
|
| 67 |
|
| 68 |
+
@st.cache_resource
|
| 69 |
def run_effNet(img_arr):
|
| 70 |
try:
|
| 71 |
resolver = tf.distribute.cluster_resolver.TPUClusterResolver()
|
|
|
|
| 78 |
prediction = eff_net_model.predict(img_arr)
|
| 79 |
return prediction
|
| 80 |
|
| 81 |
+
@st.cache_resource
|
| 82 |
def run_effNet_Art(img_arr):
|
| 83 |
try:
|
| 84 |
resolver = tf.distribute.cluster_resolver.TPUClusterResolver()
|
|
|
|
| 91 |
prediction = eff_net_art_model.predict(img_arr)
|
| 92 |
return prediction
|
| 93 |
|
|
|
|
| 94 |
def pre_process_img_effNet(image):
|
| 95 |
img = load_img(image, target_size=(300, 300)) # Resize image to model input size
|
| 96 |
img_arr = img_to_array(img) # Convert to array
|
|
|
|
| 98 |
result = run_effNet(img_arr)
|
| 99 |
return result
|
| 100 |
|
|
|
|
| 101 |
def pre_process_img_effNetArt(image):
|
| 102 |
img = load_img(image, target_size=(224, 224)) # Resize image to model input size
|
| 103 |
img_arr = img_to_array(img) # Convert to array
|
|
|
|
| 106 |
return result
|
| 107 |
|
| 108 |
# preprocess image for cnn
|
|
|
|
| 109 |
def pre_process_img(image):
|
| 110 |
# Load and preprocess the image
|
| 111 |
input_picture = load_img(image, target_size=(256, 256))
|