Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
There are two files modified: docqa.ipynb and docqa/nn/recurrent_layers.py
Details:
conda create --name Python=3.6 (if not specify this, many other stuff will be installed, cause issue)
pip install tensorflow-gpu (I removed the part "==1.2" as I am currently use most recent version 1.10).
In a NC6 DLVM, I first created a new conda env with Python 3.6. In this env, I then pip installed tensorflow-gpu==1.2 successfully. When I check if the tensorflow works or not by "import tensorflow", it shows error msg. It is because DLVM has CUDA version 9.0, which is incompatible with tensorflow 1.2.
I would suggest to include both tensorflow version and corresponding CUDA version. If DLVM keeps upgrading, the user is likely to encounter issue again.
===========================================
in file docqa/nn/recurrent_layers.py,
ModuleNotFoundError: No module named 'tensorflow.contrib.keras.python'
from tensorflow.contrib.keras.python.keras.initializers import TruncatedNormal
I changed it to:
from tensorflow.keras.initializers import TruncatedNormal