1.8 KiB
| name | about | labels |
|---|---|---|
| Performance Issue | Use this template for reporting a performance issue | type:performance |
Please make sure that this is an issue related to performance of TensorFlow. As per our GitHub Policy, we only address code/doc bugs, performance issues, feature requests and build/installation issues on GitHub. tag:performance_template
System information
- Have I written custom code (as opposed to using a stock example script provided in TensorFlow):
- OS Platform and Distribution (e.g., Linux Ubuntu 16.04):
- Mobile device (e.g. iPhone 8, Pixel 2, Samsung Galaxy) if the issue happens on mobile device:
- TensorFlow installed from (source or binary):
- TensorFlow version (use command below):
- Python version:
- Bazel version (if compiling from source):
- GCC/Compiler version (if compiling from source):
- CUDA/cuDNN version:
- GPU model and memory:
You can collect some of this information using our environment capture script You can also obtain the TensorFlow version with:
- TF 1.0:
python -c "import tensorflow as tf; print(tf.GIT_VERSION, tf.VERSION)" - TF 2.0:
python -c "import tensorflow as tf; print(tf.version.GIT_VERSION, tf.version.VERSION)"
Describe the current behavior
Describe the expected behavior
Standalone code to reproduce the issue Provide a reproducible test case that is the bare minimum necessary to generate the problem. If possible, please share a link to Colab/Jupyter/any notebook.
Other info / logs Include any logs or source code that would be helpful to diagnose the problem. If including tracebacks, please include the full traceback. Large logs and files should be attached.