mirror of
https://github.com/Aider-AI/aider.git
synced 2025-05-29 08:44:59 +00:00
feat: Add --junior-model and --junior-edit-format flags to the benchmark
This commit is contained in:
parent
15cc709322
commit
24c959af2d
1 changed files with 16 additions and 1 deletions
|
@ -110,6 +110,8 @@ def main(
|
||||||
graphs: bool = typer.Option(False, "--graphs", help="Generate graphs"),
|
graphs: bool = typer.Option(False, "--graphs", help="Generate graphs"),
|
||||||
model: str = typer.Option("gpt-3.5-turbo", "--model", "-m", help="Model name"),
|
model: str = typer.Option("gpt-3.5-turbo", "--model", "-m", help="Model name"),
|
||||||
edit_format: str = typer.Option(None, "--edit-format", "-e", help="Edit format"),
|
edit_format: str = typer.Option(None, "--edit-format", "-e", help="Edit format"),
|
||||||
|
junior_model: str = typer.Option(None, "--junior-model", help="Junior model name"),
|
||||||
|
junior_edit_format: str = typer.Option(None, "--junior-edit-format", help="Junior edit format"),
|
||||||
replay: str = typer.Option(
|
replay: str = typer.Option(
|
||||||
None,
|
None,
|
||||||
"--replay",
|
"--replay",
|
||||||
|
@ -221,6 +223,8 @@ def main(
|
||||||
commit_hash,
|
commit_hash,
|
||||||
replay,
|
replay,
|
||||||
max_apply_update_errors,
|
max_apply_update_errors,
|
||||||
|
junior_model,
|
||||||
|
junior_edit_format,
|
||||||
)
|
)
|
||||||
|
|
||||||
all_results.append(results)
|
all_results.append(results)
|
||||||
|
@ -240,6 +244,8 @@ def main(
|
||||||
commit_hash,
|
commit_hash,
|
||||||
replay,
|
replay,
|
||||||
max_apply_update_errors,
|
max_apply_update_errors,
|
||||||
|
junior_model,
|
||||||
|
junior_edit_format,
|
||||||
)
|
)
|
||||||
all_results = run_test_threaded.gather(tqdm=True)
|
all_results = run_test_threaded.gather(tqdm=True)
|
||||||
|
|
||||||
|
@ -496,6 +502,8 @@ def run_test_real(
|
||||||
commit_hash,
|
commit_hash,
|
||||||
replay,
|
replay,
|
||||||
max_apply_update_errors,
|
max_apply_update_errors,
|
||||||
|
junior_model,
|
||||||
|
junior_edit_format,
|
||||||
):
|
):
|
||||||
if not os.path.isdir(testdir):
|
if not os.path.isdir(testdir):
|
||||||
print("Not a dir:", testdir)
|
print("Not a dir:", testdir)
|
||||||
|
@ -552,7 +560,12 @@ def run_test_real(
|
||||||
# weak_model_name = model_name
|
# weak_model_name = model_name
|
||||||
weak_model_name = None
|
weak_model_name = None
|
||||||
|
|
||||||
main_model = models.Model(model_name, weak_model=weak_model_name)
|
main_model = models.Model(
|
||||||
|
model_name,
|
||||||
|
weak_model=weak_model_name,
|
||||||
|
junior_model=junior_model,
|
||||||
|
junior_edit_format=junior_edit_format,
|
||||||
|
)
|
||||||
edit_format = edit_format or main_model.edit_format
|
edit_format = edit_format or main_model.edit_format
|
||||||
|
|
||||||
dump(main_model)
|
dump(main_model)
|
||||||
|
@ -644,6 +657,8 @@ def run_test_real(
|
||||||
testcase=testdir.name,
|
testcase=testdir.name,
|
||||||
model=main_model.name,
|
model=main_model.name,
|
||||||
edit_format=edit_format,
|
edit_format=edit_format,
|
||||||
|
junior_model=junior_model,
|
||||||
|
junior_edit_format=junior_edit_format,
|
||||||
tests_outcomes=test_outcomes,
|
tests_outcomes=test_outcomes,
|
||||||
cost=coder.total_cost,
|
cost=coder.total_cost,
|
||||||
duration=dur,
|
duration=dur,
|
||||||
|
|
Loading…
Add table
Add a link
Reference in a new issue