Skip to content

feat: parrallel evaluations #238

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

Draft
wants to merge 14 commits into
base: dev
Choose a base branch
from
Draft
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
42 changes: 41 additions & 1 deletion ecosystem.config.js
Original file line number Diff line number Diff line change
Expand Up @@ -24,6 +24,46 @@ module.exports = {
},
args: `--wallet.name templar_test --wallet.hotkey M2 --device cuda:2 --subtensor.network local --netuid 2 --use_wandb --project "${PROJECT_NAME}"`
},
{
name: "TM3",
script: "neurons/miner.py",
interpreter: "python3",
env: {
...process.env,
PROJECT_NAME: PROJECT_NAME
},
args: `--wallet.name templar_test --wallet.hotkey M3 --device cuda:3 --subtensor.network local --netuid 2 --use_wandb --project "${PROJECT_NAME}"`
},
{
name: "TM4",
script: "neurons/miner.py",
interpreter: "python3",
env: {
...process.env,
PROJECT_NAME: PROJECT_NAME
},
args: `--wallet.name templar_test --wallet.hotkey M4 --device cuda:4 --subtensor.network local --netuid 2 --use_wandb --project "${PROJECT_NAME}"`
},
{
name: "TM5",
script: "neurons/miner.py",
interpreter: "python3",
env: {
...process.env,
PROJECT_NAME: PROJECT_NAME
},
args: `--wallet.name templar_test --wallet.hotkey M5 --device cuda:5 --subtensor.network local --netuid 2 --use_wandb --project "${PROJECT_NAME}"`
},
{
name: "TM6",
script: "neurons/miner.py",
interpreter: "python3",
env: {
...process.env,
PROJECT_NAME: PROJECT_NAME
},
args: `--wallet.name templar_test --wallet.hotkey M6 --device cuda:6 --subtensor.network local --netuid 2 --use_wandb --project "${PROJECT_NAME}"`
},
{
name: "TV1",
script: "neurons/validator.py",
Expand All @@ -42,7 +82,7 @@ module.exports = {
...process.env,
PROJECT_NAME: PROJECT_NAME
},
args: "--netuid 3 --device cuda:4 --project templar"
args: "--netuid 3 --device cuda:7 --project templar"
}
]
}
7 changes: 4 additions & 3 deletions hparams.json
Original file line number Diff line number Diff line change
Expand Up @@ -31,7 +31,7 @@
"checkpoint_frequency": 100,
"topk_peers": 20,
"max_topk_peers": 15,
"minimum_peers": 5,
"minimum_peers": 4,
"peers_to_replace": 1,
"peer_replacement_frequency": 5,
"peer_list_window_margin": 2,
Expand All @@ -42,9 +42,10 @@
"catch_up_threshold": 15,
"catch_up_batch_size": 5,
"catch_up_timeout": 300,
"uids_per_window": 2,
"uids_per_window": 4,
"hparams.parallel_eval_uids": 4,
Copy link
Collaborator

Choose a reason for hiding this comment

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

Why hparams before?

Copy link
Collaborator

Choose a reason for hiding this comment

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

Also curious what the reason to lower minimum peers was?

Copy link
Member Author

Choose a reason for hiding this comment

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

testing locally , should have put this in draft it wasnt ready

Copy link
Member Author

Choose a reason for hiding this comment

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

Why hparams before? i dont understand this , can you please rephrase your question

Copy link
Collaborator

Choose a reason for hiding this comment

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

My bad. Why prefix parallel eval uids with hparams.

Copy link
Member Author

Choose a reason for hiding this comment

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

its how it currently works , we have a max uids to evals in the window, this tell us how many of them should be parrallel.

Copy link
Collaborator

Choose a reason for hiding this comment

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

Why not call it parallel_eval_uids instead of hparams.parallel_eval_uids?

"time_window_delta_seconds": 10,
"reset_inactivity_windows": 25,
"max_gradient_score": 0.05,
"sync_max_steps_behind": 3
}
}
Loading