keras-applications

keras-applications copied to clipboard

### Summary This is a feature request to see if it is possible to remove hard dependency of h5py from pip install? At the moment, keras_applications only have two dependencies...

yongtang

I am using a NASNetMobile. `nvidia-smi` somehow shows that GPU utilization is jumping between 12% and 95%. With mean somewhere between. I know from other trainings that GPU utilization is...

PhilipMay

Error coming when creating the Object of Class ImageDataGenerator from keras.preprocessing.image

4

When I am creating an object of Class ImageDataGenerator I am getting an error as follows: TypeError Traceback (most recent call last) in 11 height_shift_range=0.05, # randomly shift images vertically...

manojrustagi79

Hi, I'm setting up my keras layer for glove embeddings as follows: X = X[:set_size] y = y[:set_size] yy = [] for i in range(0,len(X)): yy = np.concatenate((yy,X[i,0:set_size]), axis=0) X...

csharma

``` pretrain_vgg = tf.keras.applications.vgg16.VGG16(include_top=False, weights='imagenet', input_tensor=None, input_shape=(224,224,3),#

cuongth95

I'm creating my embedding from glove.6B file as follows embedding_layer = Embedding(input_dim=embedding_matrix.shape[0], output_dim=embedding_matrix.shape[1], input_length=max_len, weights=[embedding_matrix], trainable=False, name='embedding_layer') sequence_input = Input(shape=(max_len,), dtype='int32') tf.squeeze(tf.cast(sequence_input, tf.string)) embedded_sequences = embedding_layer(sequence_input) return embedded_sequences This gets...

csharma

Hi, [I have seen that](https://keras.io/applications/#xception) many pre-trained models are released on licenses which allow people to use the models commercially. If you have used public datasets like ImageNet for training...

ssrp

Trained Depthwise mobilenet, but size is h5 file is 24 MB, need to reduce the h5 file size.

1

Hello Team, I trained Depthwise Mobilenet and got h5 file of size around `24 MB. ` I converted this h5 file to tensorflow_js supported file format(JSON), and used technique of...

vasanthhr

MobileNetV2 with undefined backend, layers and models

8

When instantiating `MobileNetV2` without special `kwargs` I get the error: ``` Using TensorFlow backend. Traceback (most recent call last): [...] File "/usr/local/lib/python2.7/dist-packages/keras_applications/mobilenet_v2.py", line 267, in MobileNetV2 if backend.image_data_format() == 'channels_first':...

christian-rauch

Data augmentation with noise resulted in unreasonable performance improvement on ResNet50

3

I trained a ResNet50 using the GTSRB data set (http://benchmark.ini.rub.de/?section=gtsrb&subsection=news), then I "augmented" the previously employed GTSRB data set by randomly generated images of the appropriate dimension and repeated the...

szlge