Skip to content

Expose rollout-batch-size, n-samples-per-prompt, global-batch-size as CLI args in swe-agent-v2 #954

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.

Sign up for GitHub

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

Merged
guapisolo merged 1 commit into from
Apr 7, 2026
Merged

Expose rollout-batch-size, n-samples-per-prompt, global-batch-size as CLI args in swe-agent-v2 #954

Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
13 changes: 9 additions & 4 deletions examples/experimental/swe-agent-v2/run.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -38,9 +38,14 @@ class ScriptArgs(U.ExecuteTrainConfig):
hf_checkpoint: str = "zai-org/GLM-4.7-Flash"
ref_load: str = "/root/GLM-4.7-Flash_torch_dist"
save_dir: str = "/root/GLM-4.7-Flash_agent_v2/"
max_seq_len: int = 16384
prompt_data: str = "/root/swe_train.jsonl"

# Training settings
max_seq_len: int = 16384
rollout_batch_size: int = 2
n_samples_per_prompt: int = 4
global_batch_size: int = 8
Comment on lines +45 to +47

@gemini-code-assist gemini-code-assist Bot Apr 7, 2026

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

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

medium

For consistency and improved configurability, consider also moving other hardcoded training parameters from the "execute" function to this "ScriptArgs" class. Specifically, "num_rollout", "rollout_temperature", and "rollout_max_response_len" are currently hardcoded in the "rollout_args" string. Moving them here would centralize all related training settings.

References
  1. Model parameters should be retrieved from the model configuration rather than being hardcoded.


# Agent settings
agent_server_url: str = os.environ.get(
"AGENT_SERVER_URL", os.environ.get("SWE_AGENT_URL", "http://agent_env:11000")
Expand Down Expand Up @@ -104,12 +109,12 @@ def execute(args: ScriptArgs):
"--metadata-key metadata "
"--rollout-shuffle "
"--num-rollout 3000 "
"--rollout-batch-size 2 "
"--n-samples-per-prompt 4 "
f"--rollout-batch-size {args.rollout_batch_size} "
f"--n-samples-per-prompt {args.n_samples_per_prompt} "
"--rollout-temperature 0.8 "
"--rollout-max-response-len 8192 "
f"--max-seq-len {args.max_seq_len} "
"--global-batch-size 8 "
f"--global-batch-size {args.global_batch_size} "
"--balance-data "
)

Expand Down
Loading