NameError: name 'activations' is not defined

[ShivamPanchal]

I am using the latest installation as per this - farizrahman4u/seq2seq#276 (comment)

Now, I am trying to call you Seq2Seq model from recurrentshop, but it gives me an import error.

51 model.add(Embedding(NUM_WORDS, 200, input_length=MAX_LEN)) ---> 52 attn = AttentionSeq2Seq(batch_input_shape=(None, MAX_LEN, 200), hidden_dim=10, output_length=MAX_LEN, output_dim=NUM_WORDS, depth=1)

NameError: name 'activations' is not defined

Its note from keras, I tried installing other versions also.

[ShivamPanchal]

Aye! I used this piece of code and it started working.

'
!pip install keras==2.2.4
!git clone https://www.github.com/farizrahman4u/recurrentshop.git
%cd recurrentshop
!python setup.py install
!sudo pip install git+https://github.com/farizrahman4u/seq2seq.git
'

Thanks, I am closing the issue

[kp-444]

ran this code snippet on my Jupyter notebook(was desperate to solve the issue), got unexpected errors, pd.read_csv could not access the local file uploaded and there were errors with numpy module as well.

Original code :
layers = [[20],[40,20],[45,30,15]]
activation = ['sigmoid','relu']
param_grid = dict(layers = layers,activation=activations,batch_size=[128,256],epochs=[30])
grid = GridSearchCV(estimator=model, param_grid=param_grid,cv=5)

Error :
NameError Traceback (most recent call last)
in
1 layers = [[20],[40,20],[45,30,15]]
2 activation = ['sigmoid','relu']
----> 3 param_grid = dict(layers = layers,activation=activations,batch_size=[128,256],epochs=[30])
4 grid = GridSearchCV(estimator=model, param_grid=param_grid,cv=5)

NameError: name 'activations' is not defined