Experimental support for optimization using multiple devices #763
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| benchmark_name | dt(%) | dt(s) | t_new(s) | t_old(s) |
| -------------------------------------- | ---------------------- | ---------------------- | ---------------------- | ---------------------- |
test_build_transform_fft_lowres | -0.39 +/- 1.28 | -4.87e-05 +/- 1.60e-04 | 1.24e-02 +/- 1.2e-04 | 1.25e-02 +/- 1.1e-04 |
test_build_transform_fft_midres | +1.22 +/- 0.70 | +1.10e-03 +/- 6.32e-04 | 9.15e-02 +/- 4.8e-04 | 9.04e-02 +/- 4.1e-04 |
test_build_transform_fft_highres | +0.85 +/- 0.77 | +3.89e-03 +/- 3.54e-03 | 4.61e-01 +/- 3.0e-03 | 4.57e-01 +/- 1.9e-03 |
test_equilibrium_init_lowres | -1.13 +/- 1.17 | -8.95e-03 +/- 9.26e-03 | 7.85e-01 +/- 7.0e-03 | 7.94e-01 +/- 6.1e-03 |
test_equilibrium_init_medres | -0.52 +/- 1.09 | -7.31e-03 +/- 1.54e-02 | 1.41e+00 +/- 1.0e-02 | 1.41e+00 +/- 1.1e-02 |
test_equilibrium_init_highres | +0.31 +/- 0.79 | +1.28e-02 +/- 3.29e-02 | 4.17e+00 +/- 1.9e-02 | 4.16e+00 +/- 2.7e-02 |
test_objective_compile_dshape_current | +0.19 +/- 7.35 | +7.78e-03 +/- 3.00e-01 | 4.09e+00 +/- 2.1e-01 | 4.09e+00 +/- 2.2e-01 |
test_objective_compile_atf | +0.45 +/- 6.54 | +4.12e-02 +/- 5.95e-01 | 9.13e+00 +/- 4.2e-01 | 9.09e+00 +/- 4.3e-01 |
test_objective_compute_dshape_current | -1.22 +/- 1.98 | -2.60e-05 +/- 4.21e-05 | 2.10e-03 +/- 3.2e-05 | 2.13e-03 +/- 2.7e-05 |
test_objective_compute_atf | +1.04 +/- 1.09 | +7.87e-05 +/- 8.24e-05 | 7.66e-03 +/- 5.9e-05 | 7.58e-03 +/- 5.7e-05 |
test_objective_jac_dshape_current | -3.88 +/- 12.78 | -1.80e-03 +/- 5.93e-03 | 4.46e-02 +/- 3.7e-03 | 4.64e-02 +/- 4.6e-03 |
test_objective_jac_atf | -4.56 +/- 5.00 | -1.06e-01 +/- 1.16e-01 | 2.21e+00 +/- 6.1e-02 | 2.31e+00 +/- 9.8e-02 |
test_perturb_1 | -1.02 +/- 13.21 | -8.70e-02 +/- 1.12e+00 | 8.42e+00 +/- 8.0e-01 | 8.51e+00 +/- 7.9e-01 |
test_perturb_2 | -1.19 +/- 4.13 | -1.70e-01 +/- 5.88e-01 | 1.41e+01 +/- 4.4e-01 | 1.42e+01 +/- 3.9e-01 | |
| benchmark_name | dt(%) | dt(s) | t_new(s) | t_old(s) |
| -------------------------------------- | ---------------------- | ---------------------- | ---------------------- | ---------------------- |
test_build_transform_fft_lowres | -1.14 +/- 0.85 | -1.42e-04 +/- 1.06e-04 | 1.23e-02 +/- 8.7e-05 | 1.24e-02 +/- 6.0e-05 |
test_build_transform_fft_midres | -1.88 +/- 1.55 | -1.74e-03 +/- 1.43e-03 | 9.07e-02 +/- 1.1e-03 | 9.24e-02 +/- 9.3e-04 |
test_build_transform_fft_highres | -0.13 +/- 1.04 | -5.95e-04 +/- 4.77e-03 | 4.59e-01 +/- 4.3e-03 | 4.60e-01 +/- 2.1e-03 |
test_equilibrium_init_lowres | -1.44 +/- 1.81 | -1.14e-02 +/- 1.43e-02 | 7.81e-01 +/- 9.3e-03 | 7.92e-01 +/- 1.1e-02 |
test_equilibrium_init_medres | -1.09 +/- 1.43 | -1.54e-02 +/- 2.02e-02 | 1.39e+00 +/- 1.0e-02 | 1.41e+00 +/- 1.8e-02 |
test_equilibrium_init_highres | -0.96 +/- 0.72 | -4.02e-02 +/- 3.00e-02 | 4.14e+00 +/- 2.0e-02 | 4.18e+00 +/- 2.3e-02 |
test_objective_compile_dshape_current | -1.16 +/- 8.11 | -4.82e-02 +/- 3.36e-01 | 4.09e+00 +/- 2.3e-01 | 4.14e+00 +/- 2.4e-01 |
test_objective_compile_atf | -0.26 +/- 8.14 | -2.40e-02 +/- 7.46e-01 | 9.14e+00 +/- 4.8e-01 | 9.16e+00 +/- 5.7e-01 |
test_objective_compute_dshape_current | -0.46 +/- 4.23 | -9.63e-06 +/- 8.88e-05 | 2.09e-03 +/- 8.4e-05 | 2.10e-03 +/- 3.0e-05 |
test_objective_compute_atf | +0.93 +/- 1.11 | +6.82e-05 +/- 8.11e-05 | 7.40e-03 +/- 6.0e-05 | 7.33e-03 +/- 5.5e-05 |
test_objective_jac_dshape_current | +3.95 +/- 15.19 | +1.84e-03 +/- 7.07e-03 | 4.84e-02 +/- 5.9e-03 | 4.65e-02 +/- 3.9e-03 |
test_objective_jac_atf | +3.41 +/- 5.96 | +7.63e-02 +/- 1.33e-01 | 2.31e+00 +/- 9.9e-02 | 2.24e+00 +/- 8.9e-02 |
test_perturb_1 | -1.35 +/- 12.34 | -1.15e-01 +/- 1.06e+00 | 8.45e+00 +/- 7.3e-01 | 8.56e+00 +/- 7.6e-01 |
test_perturb_2 | +0.62 +/- 4.91 | +8.84e-02 +/- 7.05e-01 | 1.44e+01 +/- 5.2e-01 | 1.44e+01 +/- 4.8e-01 | |
| benchmark_name | dt(%) | dt(s) | t_new(s) | t_old(s) |
| -------------------------------------- | ---------------------- | ---------------------- | ---------------------- | ---------------------- |
test_build_transform_fft_lowres | -0.68 +/- 1.22 | -8.46e-05 +/- 1.51e-04 | 1.24e-02 +/- 1.0e-04 | 1.24e-02 +/- 1.1e-04 |
test_build_transform_fft_midres | -1.25 +/- 1.32 | -1.16e-03 +/- 1.22e-03 | 9.15e-02 +/- 1.0e-03 | 9.26e-02 +/- 6.3e-04 |
test_build_transform_fft_highres | +1.49 +/- 0.76 | +6.83e-03 +/- 3.48e-03 | 4.64e-01 +/- 2.6e-03 | 4.57e-01 +/- 2.3e-03 |
test_equilibrium_init_lowres | +2.67 +/- 1.36 | +2.11e-02 +/- 1.07e-02 | 8.11e-01 +/- 8.3e-03 | 7.90e-01 +/- 6.8e-03 |
test_equilibrium_init_medres | +1.63 +/- 1.82 | +2.31e-02 +/- 2.58e-02 | 1.44e+00 +/- 2.0e-02 | 1.41e+00 +/- 1.6e-02 |
test_equilibrium_init_highres | +0.08 +/- 1.04 | +3.30e-03 +/- 4.35e-02 | 4.20e+00 +/- 3.5e-02 | 4.20e+00 +/- 2.6e-02 |
test_objective_compile_dshape_current | -0.88 +/- 7.16 | -3.64e-02 +/- 2.98e-01 | 4.12e+00 +/- 2.1e-01 | 4.15e+00 +/- 2.1e-01 |
test_objective_compile_atf | +0.19 +/- 5.83 | +1.74e-02 +/- 5.39e-01 | 9.26e+00 +/- 3.9e-01 | 9.24e+00 +/- 3.7e-01 |
test_objective_compute_dshape_current | +2.57 +/- 4.95 | +5.43e-05 +/- 1.04e-04 | 2.16e-03 +/- 9.5e-05 | 2.11e-03 +/- 4.2e-05 |
test_objective_compute_atf | +2.97 +/- 1.55 | +2.23e-04 +/- 1.17e-04 | 7.73e-03 +/- 8.4e-05 | 7.50e-03 +/- 8.1e-05 |
test_objective_jac_dshape_current | +4.14 +/- 6.79 | +1.86e-03 +/- 3.05e-03 | 4.68e-02 +/- 2.0e-03 | 4.49e-02 +/- 2.3e-03 |
test_objective_jac_atf | -1.27 +/- 4.15 | -2.97e-02 +/- 9.73e-02 | 2.31e+00 +/- 5.2e-02 | 2.34e+00 +/- 8.2e-02 |
test_perturb_1 | -4.78 +/- 12.27 | -4.33e-01 +/- 1.11e+00 | 8.62e+00 +/- 7.3e-01 | 9.05e+00 +/- 8.4e-01 |
test_perturb_2 | -5.39 +/- 3.81 | -8.30e-01 +/- 5.87e-01 | 1.46e+01 +/- 4.0e-01 | 1.54e+01 +/- 4.3e-01 | |
Codecov ReportAttention: Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## master #763 +/- ##
==========================================
+ Coverage 94.85% 94.95% +0.10%
==========================================
Files 80 80
Lines 19779 19799 +20
==========================================
+ Hits 18762 18801 +39
+ Misses 1017 998 -19
|
| benchmark_name | dt(%) | dt(s) | t_new(s) | t_old(s) |
| -------------------------------------- | ---------------------- | ---------------------- | ---------------------- | ---------------------- |
test_build_transform_fft_lowres | +0.37 +/- 2.77 | +4.63e-05 +/- 3.51e-04 | 1.27e-02 +/- 1.7e-04 | 1.27e-02 +/- 3.1e-04 |
test_build_transform_fft_midres | +0.16 +/- 2.12 | +1.50e-04 +/- 1.96e-03 | 9.28e-02 +/- 1.8e-03 | 9.26e-02 +/- 7.1e-04 |
test_build_transform_fft_highres | -1.89 +/- 1.62 | -8.84e-03 +/- 7.57e-03 | 4.59e-01 +/- 6.7e-03 | 4.68e-01 +/- 3.5e-03 |
test_equilibrium_init_lowres | +0.25 +/- 2.45 | +2.02e-03 +/- 2.00e-02 | 8.18e-01 +/- 1.3e-02 | 8.16e-01 +/- 1.5e-02 |
test_equilibrium_init_medres | +0.83 +/- 2.16 | +1.20e-02 +/- 3.11e-02 | 1.45e+00 +/- 2.6e-02 | 1.44e+00 +/- 1.7e-02 |
test_equilibrium_init_highres | +0.28 +/- 1.22 | +1.21e-02 +/- 5.18e-02 | 4.26e+00 +/- 4.1e-02 | 4.25e+00 +/- 3.2e-02 |
test_objective_compile_dshape_current | +3.13 +/- 9.20 | +1.35e-01 +/- 3.95e-01 | 4.43e+00 +/- 2.7e-01 | 4.30e+00 +/- 2.9e-01 |
test_objective_compile_atf | +1.73 +/- 5.84 | +1.65e-01 +/- 5.57e-01 | 9.71e+00 +/- 3.2e-01 | 9.54e+00 +/- 4.6e-01 |
test_objective_compute_dshape_current | +16.88 +/- 7.46 | +3.68e-04 +/- 1.63e-04 | 2.55e-03 +/- 1.5e-04 | 2.18e-03 +/- 5.7e-05 |
test_objective_compute_atf | +1.48 +/- 3.25 | +1.15e-04 +/- 2.52e-04 | 7.86e-03 +/- 2.2e-04 | 7.75e-03 +/- 1.2e-04 |
test_objective_jac_dshape_current | +3.64 +/- 10.43 | +1.71e-03 +/- 4.89e-03 | 4.86e-02 +/- 3.9e-03 | 4.69e-02 +/- 3.0e-03 |
test_objective_jac_atf | +0.59 +/- 8.21 | +1.42e-02 +/- 1.97e-01 | 2.42e+00 +/- 1.7e-01 | 2.40e+00 +/- 9.2e-02 |
test_perturb_1 | -0.67 +/- 12.69 | -5.90e-02 +/- 1.12e+00 | 8.80e+00 +/- 8.6e-01 | 8.86e+00 +/- 7.2e-01 |
test_perturb_2 | +2.26 +/- 6.89 | +3.39e-01 +/- 1.04e+00 | 1.54e+01 +/- 9.0e-01 | 1.50e+01 +/- 5.2e-01 | |
| benchmark_name | dt(%) | dt(s) | t_new(s) | t_old(s) |
| -------------------------------------- | ---------------------- | ---------------------- | ---------------------- | ---------------------- |
test_build_transform_fft_lowres | +0.08 +/- 1.22 | +1.03e-05 +/- 1.52e-04 | 1.25e-02 +/- 1.0e-04 | 1.25e-02 +/- 1.1e-04 |
test_build_transform_fft_midres | +0.21 +/- 1.01 | +1.88e-04 +/- 9.25e-04 | 9.16e-02 +/- 7.0e-04 | 9.14e-02 +/- 6.0e-04 |
test_build_transform_fft_highres | -0.08 +/- 0.88 | -3.87e-04 +/- 4.05e-03 | 4.62e-01 +/- 3.1e-03 | 4.62e-01 +/- 2.6e-03 |
test_equilibrium_init_lowres | +0.66 +/- 1.07 | +5.17e-03 +/- 8.44e-03 | 7.95e-01 +/- 4.8e-03 | 7.89e-01 +/- 7.0e-03 |
test_equilibrium_init_medres | -0.37 +/- 0.91 | -5.26e-03 +/- 1.29e-02 | 1.41e+00 +/- 9.6e-03 | 1.42e+00 +/- 8.7e-03 |
test_equilibrium_init_highres | -0.39 +/- 0.68 | -1.62e-02 +/- 2.84e-02 | 4.17e+00 +/- 2.2e-02 | 4.19e+00 +/- 1.8e-02 |
test_objective_compile_dshape_current | +1.87 +/- 8.48 | +8.18e-02 +/- 3.71e-01 | 4.46e+00 +/- 2.9e-01 | 4.38e+00 +/- 2.3e-01 |
test_objective_compile_atf | +1.70 +/- 6.50 | +1.57e-01 +/- 5.99e-01 | 9.37e+00 +/- 4.2e-01 | 9.21e+00 +/- 4.2e-01 |
test_objective_compute_dshape_current | +1.16 +/- 2.86 | +2.49e-05 +/- 6.10e-05 | 2.16e-03 +/- 5.0e-05 | 2.14e-03 +/- 3.6e-05 |
test_objective_compute_atf | -1.68 +/- 1.18 | -1.27e-04 +/- 8.91e-05 | 7.46e-03 +/- 6.3e-05 | 7.58e-03 +/- 6.3e-05 |
test_objective_jac_dshape_current | +1.14 +/- 10.28 | +5.30e-04 +/- 4.79e-03 | 4.72e-02 +/- 4.0e-03 | 4.66e-02 +/- 2.6e-03 |
test_objective_jac_atf | -0.75 +/- 3.39 | -1.69e-02 +/- 7.65e-02 | 2.24e+00 +/- 5.4e-02 | 2.26e+00 +/- 5.4e-02 |
test_perturb_1 | +0.73 +/- 13.35 | +6.30e-02 +/- 1.15e+00 | 8.65e+00 +/- 8.7e-01 | 8.58e+00 +/- 7.5e-01 |
test_perturb_2 | +0.60 +/- 5.58 | +8.48e-02 +/- 7.89e-01 | 1.42e+01 +/- 6.7e-01 | 1.41e+01 +/- 4.2e-01 | |
![github-actions[bot]](https://avatars.githubusercontent.com/in/15368?s=60&v=4){kind=small}
|
|
||
| import psutil | ||
|
|
||
| cpu_mem = psutil.virtual_memory().available / 1024**2 # RAM in MB |
There was a problem hiding this comment.
[flake8] <841> reported by reviewdog 🐶
local variable 'cpu_mem' is assigned to but never used
|
Closing this in favor of #1071 |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This doesn't work as is, will likely need to wait until we get MPI working.