Kashyap Raval

Try this : tf.contrib.legacy_seq2seq.model_with_buckets

It'll work with the latest versions of keras and Tensorflow.

You can download it manually from here: https://github.com/fchollet/deep-learning-models/releases/download/v0.1/vgg16_weights_tf_dim_ordering_tf_kernels_notop.h5 and paste it on `~/.keras/models`

changing a batch size can improve accuracy . try it with 1024. And can you share me model.save file mail - [email protected] Thanks!