aws / aws/amazon-sagemaker-examples
mxnet-realtime-analysis.ipynb failed CI
- Dominant language
- Jupyter Notebook
- Stars
- 11k
- Forks
- 7k
- Avg merge
- 8h 29m
- Merged PRs (30d)
- 8
Description
Link to the notebook:
https://github.com/aws/amazon-sagemaker-examples/blob/master/sagemaker-debugger/mxnet_realtime_analysis/mxnet-realtime-analysis.ipynb
Error:
---------------------------------------------------------------------------
Exception encountered at "In [9]":
---------------------------------------------------------------------------
TensorUnavailableForStep Traceback (most recent call last)
in
31 steps_to_render = list(set(steps).symmetric_difference(set(rendered_steps)))
32 # plot only tensors from newer chunk
---> 33 plot_tensors(trial, layer, batch_index, steps_to_render)
34 rendered_steps.extend(steps_to_render)
35 time.sleep(5)
in plot_tensors(trial, layer, batch_index, steps_range)
18 if len(steps_range) > 0:
19 fig, axs = create_plots(steps_range)
---> 20 vals = get_data(trial, layer, batch_index, steps_range)
21
22 for ax, image, step in
[...]
TensorUnavailableForStep: Value for tensor conv0_output_0 is not available for step 1000 with mode GLOBAL as it was not saved.
Contributor guide
Research direction
Open sagemaker-debugger/mxnet_realtime_analysis/mxnet-realtime-analysis.ipynb and inspect In [9], especially plot_tensors and get_data. Run the notebook through the repository's CI process to reproduce the TensorUnavailableForStep error. Done means the notebook completes without requesting tensor data that was not saved.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- aws, jupyter-notebook
- Domain
- machine-learning, testing
- Issue type
- Bug
- Difficulty
- 3/5
- Estimated time
- 1-2 days
- Activity status
- Stale
- Clarity
- Mostly clear
- Newbie friendliness
- 35/100