Skip to content

Conversation

@the-bumble
Copy link
Contributor

This will help reduce some of the noise from tensorflow. Most of the noise comes from Keras and there isn't a great way to limit it without redirecting all stderr from the library.

This PR also allows eyeballer to produce the --help message when no params are provided. This depends on PR #42

Unfortunately, this will only somewhat quiet down tensorflow, but it helps
@dan-bishopfox
Copy link
Member

I'm still getting pretty spammy output with this change. Looks like this:

eyeballer evaluate
Using TensorFlow backend.
WARNING:tensorflow:From /home/altf4/Code/eyeballer/venv/lib/python3.6/site-packages/tensorflow/python/framework/op_def_library.py:263: colocate_with (from tensorflow.python.framework.ops) is deprecated and will be removed in a future version.
Instructions for updating:
Colocations handled automatically by placer.
WARNING:tensorflow:From /home/altf4/Code/eyeballer/venv/lib/python3.6/site-packages/keras/backend/tensorflow_backend.py:3445: calling dropout (from tensorflow.python.ops.nn_ops) with keep_prob is deprecated and will be removed in a future version.

@the-bumble
Copy link
Contributor Author

That is correct. it is still very spammy. I reduced the output where I could without redirecting all stderr from the library, which would start to mask errors.

We could suppress all stderr from the library and only provide it with a verbose flag (-v), if you'd prefer.

@dan-bishopfox
Copy link
Member

Let's try that, because it's still super spammy

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants