[
https://issues.apache.org/jira/browse/BEAM-13985?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Anand Inguva updated BEAM-13985:
--------------------------------
Description: RunInference benchmarks will evaluate performance of
Pipelines, which represent common use cases of Beam + Dataflow in Pytorch,
sklearn and possibly TFX. These benchmarks would be the integration tests that
exercise several software components using Beam, PyTorch, Scikit learn and
TensorFlow extended. (was: The goal of the end-to-end it test is to check if
the code changes in RunInference are working as intended.
Make calls to the RunInference classes for TFX, Pytorch, and Scikit-learn.
* For TFX, need to use their proto
Process
* Read data from GCS bucket
* Use pre-trained model.
* Predict the output predictions
* Assert if output predictions match actual
Add task for using GPU container images)
> Implement end-to-end tests for RunInference classes
> ---------------------------------------------------
>
> Key: BEAM-13985
> URL: https://issues.apache.org/jira/browse/BEAM-13985
> Project: Beam
> Issue Type: Sub-task
> Components: sdk-py-core
> Reporter: Andy Ye
> Assignee: Anand Inguva
> Priority: P2
> Labels: run-inference
>
> RunInference benchmarks will evaluate performance of Pipelines, which
> represent common use cases of Beam + Dataflow in Pytorch, sklearn and
> possibly TFX. These benchmarks would be the integration tests that exercise
> several software components using Beam, PyTorch, Scikit learn and TensorFlow
> extended.
--
This message was sent by Atlassian Jira
(v8.20.1#820001)
