microsoft/nni

No error raise for wrong type of result

Open

#3,641 opened on May 14, 2021

View on GitHub
 (3 comments) (0 reactions) (1 assignee)Python (1,830 forks)batch import
help wanteduser raised

Repository metrics

Stars
 (13,504 stars)
PR merge metrics
 (No merged PRs in 30d)

Description

Describe the issue:

When parsing value in torch.tensor with dtype=torch.float to nni.report_intermediate_result and nni.report_final_result, there is no Exception raised and no result recorded.

It is desired to raise a ValueError, or auto-convert to python float (raise a ValueError when cannot be converted).

Environment:

  • NNI version:
  • Training service (local|remote|pai|aml|etc):
  • Client OS:
  • Server OS (for remote mode only):
  • Python version:
  • PyTorch/TensorFlow version:
  • Is conda/virtualenv/venv used?:
  • Is running in Docker?:

Configuration:

  • Experiment config (remember to remove secrets!):
  • Search space:

Log message:

  • nnimanager.log:
  • dispatcher.log:
  • nnictl stdout and stderr:

How to reproduce it?:

Contributor guide

Research direction
Locate the code for report intermediate result and report final result in the NNI source, specifically the validation logic for the result value. Add a type check to raise ValueError if the result is not a Python float or can't be converted, and ensure proper conversion or error handling.
Tech stack
pythonpytorch
Domain
backendmachine learningai
Issue type
Bug
DifficultyEstimated implementation difficulty for a new contributor, from 1 for very small changes to 5 for expert-level work.
2
Estimated timeA rough time range for an experienced contributor to investigate, implement, test, and prepare a pull request.
1-3 hours
Activity statusHow available the issue appears right now: fresh, active, stale, blocked, or waiting on maintainer input.
Active
ClarityHow clearly the issue explains the expected change, acceptance criteria, and next step.
Clear
Prerequisites
PythonPyTorch
Newbie friendlinessA 1-100 score estimating how approachable this issue is for first-time contributors.
50