Without @tf.function OOM

def train_step(self, x, step, grad_clip=True, clip_value=1.0):
    print('tracing', x, step, grad_clip, clip_value)
    with tf.name_scope("input_data"):
      ...

Source: github.com

Similar codes

2 votes223 views1 answer
Django group with permission
Python Django
62 votes207 views6 answers
Class python
Python
0 vote188 views1 answer
Get the mean of all not nan values
Python
0 vote211 views1 answer
How to get location using python
Python
0 vote187 views1 answer
Combine int and object columns into one
Python
1 vote196 views1 answer
Example of __init__ self
Python
5 votes210 views2 answers
Encrypt and decrypt python
Python
0 vote214 views1 answer
Read list from txt python
Python
4 votes189 views1 answer
How to save a neural network pytorch
Python
1 vote208 views1 answer
Append attribute ofpython
Python