aws / aws/sagemaker-experiments
Training job as a trial component is not accessible/visible in Sagemaker Studio pane
- Dominant language
- Python
- Stars
- 129
- Forks
- 36
- PR merge metrics
- No merged PRs in 30d
Description
**Describe the bug**
mnist-handwritten-digits-classification-experiment.ipynb script is run but the training jobs which are also configured to be a trial component is not visible in the Sagemaker Studio pane.
```
search_expression = {
"Filters": [
{
"Name": "DisplayName",
"Operator": "Equals",
"Value": "Training",
}
],
}
trial_component_analytics = ExperimentAnalytics(
sagemaker_session=Session(sess, sm),
experiment_name=mnist_experiment.experiment_name,
search_expression=search_expression,
sort_by="metrics.test:accuracy.max",
sort_order="Descending",
metric_names=["test:accuracy"],
parameter_names=["hidden_channels", "epochs", "dropout", "optimizer"],
)
trial_component_analytics.dataframe()
```
The above script **gives no output**
**To Reproduce**
To run the script as is mnist-handwritten-digits-classification-experiment.ipynb using Kernel Python 3 (Data Science) in Sagemaker Studio.
**Expected behavior**
To view training jobs as trial components. Only the preprocessing trial component are visible.
**Screenshots**
If applicable, add screenshots to help explain your problem.

**Environment:**
mnist-handwritten-digits-classification-experiment.ipynb (from https://docs.aws.amazon.com/sagemaker/latest/dg/experiments-mnist.html) ran as it is in Sagemaker studio
Contributor guide
Assessment
This issue has not been assessed yet.