Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Prize qualification baselines #586

Merged
merged 10 commits into from
Nov 22, 2023
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
83 changes: 83 additions & 0 deletions reference_algorithms/prize_qualification_baselines/README.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,83 @@
# Prize Qualification Baselines
This directory contains the baseine(s) that submissions that must beat to qualify for prizes.

TODO: link back to section in rules.

## Externally Tuned Ruleset

### JAX

The prize qualification baseline submissions for jax are:
- `reference_algorithms/prize_qualification_baselines/external_tuning/jax_nadamw_target_setting.py`
- `feference_algorithms/prize_qualification_baselines/external_tuning/jax_nadamw_full_budget.py`

Example command:

```bash
python3 submission_runner.py \
--framework=jax \
--data_dir=<data_dir> \
--experiment_dir=<experiment_dir> \
--experiment_name=<experiment_name> \
--workload=<workload> \
--submission_path=reference_algorithms/prize_qualification_baselines/external_tuning/jax_nadamw_target_setting.py \
--tuning_search_space=reference_algorithms/prize_qualification_baselines/external_tuning/tuning_search_space.json
```

### PyTorch

The prize qualification baseline submissionss for PyTorch are:
- `reference_algorithms/prize_qualification_baselines/external_tuning/pytorch_nadamw_target_setting.py`
- `feference_algorithms/prize_qualification_baselines/external_tuning/pytorch_nadamw_full_budget.py`


Example command:

```bash
torchrun --redirects 1:0,2:0,3:0,4:0,5:0,6:0,7:0 --standalone --nnodes=1 --nproc_per_node=8 submission_runner.py \
--framework=pytorch \
--data_dir=<data_dir> \
--experiment_dir=<experiment_dir> \
--experiment_name=t<experiment_name> \
--workload=<workload>\
--submission_path=reference_algorithms/prize_qualification_baselines/external_tuning/pytorch_nadamw_target_setting.py \
--tuning_search_space=reference_algorithms/prize_qualification_baselines/external_tuning/tuning_search_space.json
```

## Self-tuning Ruleset

### JAX

The prize qualification baseline submissionss for jax are:
- `reference_algorithms/prize_qualification_baselines/external_tuning/jax_nadamw_target_setting.py`
- `feference_algorithms/prize_qualification_baselines/external_tuning/jax_nadamw_full_budget.py`

Example command:
```bash
python3 submission_runner.py \
--framework=jax \
--data_dir=<data_dir> \
--experiment_dir=<experiment_dir> \
--experiment_name=<experiment_name> \
--workload=<workload> \
--submission_path=reference_algorithms/prize_qualification_baselines/external_tuning/jax_nadamw_target_setting.py \
--tuning_ruleset=self
```

### PyTorch

The prize qualification baseline submissionss for PyTorch are:
- `reference_algorithms/prize_qualification_baselines/external_tuning/pytorch_nadamw_target_setting.py`
- `feference_algorithms/prize_qualification_baselines/external_tuning/pytorch_nadamw_full_budget.py`

Example command:
```bash
torchrun --redirects 1:0,2:0,3:0,4:0,5:0,6:0,7:0 --standalone --nnodes=1 --nproc_per_node=8 submission_runner.py \
--framework=pytorch \
--data_dir=<data_dir> \
--experiment_dir=<experiment_dir> \
--experiment_name=t<experiment_name> \
--workload=<workload>\
--submission_path=reference_algorithms/prize_qualification_baselines/external_tuning/pytorch_nadamw_target_setting.py \
--tuning_ruleset=self
```
Loading
Loading