from keras.legacy import models as legacy_models ImportError: cannot import name models

[senaj]

After running the shuffle_uniprot.py command. I attempted to run train_dspace but received the following error.

Using TensorFlow backend.
Traceback (most recent call last):
File "/grid/sw/ncgr/dspace/TENSORFLOW_py2_ENV/bin/train_dspace", line 11, in
load_entry_point('sgidspace', 'console_scripts', 'train_dspace')()
File "/grid/sw/ncgr/dspace/TENSORFLOW_py2_ENV/lib/python2.7/site-packages/pkg_resources/init.py", line 565, in load_entry_point
return get_distribution(dist).load_entry_point(group, name)
File "/grid/sw/ncgr/dspace/TENSORFLOW_py2_ENV/lib/python2.7/site-packages/pkg_resources/init.py", line 2631, in load_entry_point
return ep.load()
File "/grid/sw/ncgr/dspace/TENSORFLOW_py2_ENV/lib/python2.7/site-packages/pkg_resources/init.py", line 2291, in load
return self.resolve()
File "/grid/sw/ncgr/dspace/TENSORFLOW_py2_ENV/lib/python2.7/site-packages/pkg_resources/init.py", line 2297, in resolve
module = import(self.module_name, fromlist=['name'], level=0)
File "/grid/sw/ncgr/dspace/sgidspace/sgidspace/train_dspace.py", line 31, in
from sgidspace.sgikeras.models import patch_all, load_model
File "/grid/sw/ncgr/dspace/sgidspace/sgidspace/sgikeras/models.py", line 35, in
from keras.legacy import models as legacy_models
ImportError: cannot import name models

keras version 2.2.4
Python 2.7.5
pydot 1.4.1
numpy 1.16.1

Please, can you give me advice as to what my problem might be?