Skip to content

Conversation

@rikhuijzer
Copy link
Collaborator

@rikhuijzer rikhuijzer commented Mar 5, 2022

Current output:

 ────────────────────────────────────────────────────────────────────
                                          Time          Allocations
                                    ───────────────   ───────────────
           Total measured:                401s             190GiB

 Section                    ncalls     time    %tot     alloc    %tot
 ────────────────────────────────────────────────────────────────────
 data_constructors               1    14.5s    3.7%   1.87GiB    1.0%
 utils                           1    2.82s    0.7%    414MiB    0.2%
   standardize_predictors        1    527ms    0.1%   71.6MiB    0.0%
   center_predictors             1    2.29s    0.6%    342MiB    0.2%
 priors                          1    134ms    0.0%   11.3MiB    0.0%
 turing_model                    1     379s   95.6%    187GiB   98.8%
   Hierarchical Model            1    44.3s   11.2%   21.8GiB   11.5%
   NegBin Model                  1    35.0s    8.8%   6.58GiB    3.5%
   Pois Model                    1    66.9s   16.9%   30.5GiB   16.1%
   Logistic Model                1    86.6s   21.9%   4.48GiB    2.4%
   Student Model                 1    44.3s   11.2%   53.3GiB   28.2%
   Gaussian Model                1    75.7s   19.1%   67.3GiB   35.6%
 ────────────────────────────────────────────────────────────────────

This PR is largely based on nice suggestions by David Widmann and Kristoffer Carlsson in TuringLang/Turing.jl#1787.

The ordering of the subelements is currently wrong. The PR to fix it is at KristofferC/TimerOutputs.jl#144.

@rikhuijzer rikhuijzer requested a review from storopoli March 5, 2022 14:16
Copy link
Member

@storopoli storopoli left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Looks great thanks! Let's also capitalize on this PR and change the timed_testsets names.

@storopoli storopoli merged commit b7c7b32 into main Mar 5, 2022
@storopoli storopoli deleted the rh/measure-tests branch March 5, 2022 21:59
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

Track performance of turing_model

3 participants