Traffic-Sign-Classification.keras
Traffic-Sign-Classification.keras copied to clipboard
chsasank
Deep leaning for traffic sign classification in keras and python
When running on Windows, the '/' will result in error, int() problem with string. Workaround is to change '/' to double backslash '\\\\', specifically: 1) def get_class(img_path): return int(img_path.split('\\\\')[-2]) 2)...
Epoch 1/30 31367/31367 [==============================] - 979s 31ms/step - loss: 0.2193 - acc: 0.9354 - val_loss: 0.0298 - val_acc: 0.9916 Epoch 2/30 31367/31367 [==============================] - 969s 31ms/step - loss: 5.1823 -...
`Loaded images from X.h5 _________________________________________________________________ Layer (type) Output Shape Param # ================================================================= conv2d_17 (Conv2D) (None, 48, 48, 32) 896 _________________________________________________________________ conv2d_18 (Conv2D) (None, 46, 46, 32) 9248 _________________________________________________________________ max_pooling2d_9 (MaxPooling2...
I got an error saying : ValueError: Error when checking input: expected conv2d_62_input to have 4 dimensions, but got array with shape (0, 1)
In test data section, the accuracy is computed as integer ... Changing to float is better. acc = np.sum(y_pred == y_test) / np.size(y_pred) in acc = np.float(np.sum(y_pred == y_test)) /...
Dear, please add a file with the result of the **_pip list_** command. Than you,
Hi Sasank Chilamkurthy I have a problem with your source code when trying to convert your model from channels_first to channels_last. I have deleted this line: "K.set_image_data_format('channels_first')" and changed the...
