[FEAT] Time elapsed for training and inference #101

Open
opened 2025-07-25 12:38:47 +00:00 by nuluh · 0 comments
nuluh commented 2025-07-25 12:38:47 +00:00 (Migrated from github.com)

Problem Statement

Currently, there is no systematic way to report or compare the time taken by different algorithms/models during both training and inference. This makes it difficult to evaluate and select models based on their efficiency.

Proposed Solution

Integrate timing measurement utilities in both the training and inference stages. Automatically log and report the elapsed time for each model/algorithm, and present the results in a comparative format (such as a table or chart).

Alternatives Considered

No response

Component

Jupyter Notebook

Priority

High (significantly improves workflow)

Implementation Ideas

No response

Expected Benefits

This will allow for objective and consistent comparison between algorithms/models, making it easier to select the most efficient option for deployment and reporting in the thesis.

Additional Context

No response

### Problem Statement Currently, there is no systematic way to report or compare the time taken by different algorithms/models during both training and inference. This makes it difficult to evaluate and select models based on their efficiency. ### Proposed Solution Integrate timing measurement utilities in both the training and inference stages. Automatically log and report the elapsed time for each model/algorithm, and present the results in a comparative format (such as a table or chart). ### Alternatives Considered _No response_ ### Component Jupyter Notebook ### Priority High (significantly improves workflow) ### Implementation Ideas _No response_ ### Expected Benefits This will allow for objective and consistent comparison between algorithms/models, making it easier to select the most efficient option for deployment and reporting in the thesis. ### Additional Context _No response_
Sign in to join this conversation.
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: nuluh/thesis#101