Skip to content

Fix per-domain eval args misalignment when polyglot is combined with other domains #40

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

Open
rootkiller6788 wants to merge 1 commit into
base: main
Choose a base branch
from
Open

Fix per-domain eval args misalignment when polyglot is combined with other domains #40

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
26 changes: 20 additions & 6 deletions generate_loop.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -740,6 +740,20 @@ def generate_loop(
# Initialization
docker_client = docker.DockerClient()
parent_selection = "latest" if run_baseline == "no_archive" else parent_selection

# Polyglot is evaluated through its own harness (run_harness_polyglot), so
# `generate` only handles the other domains. Filter the per-domain
# evaluation arguments alongside the domain list; otherwise, when polyglot
# is not the last domain in `domains`, every domain after it would receive
# polyglot's eval_subset / eval_samples due to the positional zip inside
# generate().
non_polyglot_domains = [d for d in domains if d != "polyglot"]
non_polyglot_eval_subsets = [
s for d, s in zip(domains, eval_subsets) if d != "polyglot"
]
non_polyglot_eval_samples = [
n for d, n in zip(domains, eval_samples) if d != "polyglot"
]
if resume_from:
output_dir = os.path.normpath(os.path.abspath(resume_from))
run_id = os.path.basename(output_dir).split("generate_")[-1]
Expand Down Expand Up @@ -821,16 +835,16 @@ def generate_loop(
archive = update_and_save_archive(output_dir, [], new_node=0)
metadata = generate(
docker_client,
[d for d in domains if d != "polyglot"],
non_polyglot_domains,
output_dir,
run_id,
current_genid=0,
parent_genid=None,
root_dir=root_dir,
root_commit=root_commit,
eval_samples=eval_samples,
eval_samples=non_polyglot_eval_samples,
eval_workers=eval_workers,
eval_subsets=eval_subsets,
eval_subsets=non_polyglot_eval_subsets,
meta_patch_files=meta_patch_files,
run_meta_agent=False,
run_baseline=run_baseline,
Expand Down Expand Up @@ -900,16 +914,16 @@ def generate_loop(
for current_genid in range(start_genid, max_generation + 1):
metadata = generate(
docker_client,
[d for d in domains if d != "polyglot"],
non_polyglot_domains,
output_dir,
run_id,
current_genid,
parent_genid=parent_genid,
root_dir=root_dir,
root_commit=root_commit,
eval_samples=eval_samples,
eval_samples=non_polyglot_eval_samples,
eval_workers=eval_workers,
eval_subsets=eval_subsets,
eval_subsets=non_polyglot_eval_subsets,
meta_patch_files=meta_patch_files,
run_meta_agent=True,
run_baseline=run_baseline,
Expand Down