Skip to content
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
38 changes: 37 additions & 1 deletion datasets/nemotron_terminal_corpus/atif_to_std.py
Original file line number Diff line number Diff line change
@@ -1,6 +1,42 @@
# ruff: noqa: E402, I001

from scripts.atif_to_std_common import main
import json
import sys

from schema.atif import ATIFTrajectory, Step, normalize_atif_trajectory
from scripts.atif_to_std_common import (
split_terminal_task_description_prompt,
standardize_tools,
structure_terminal_completion_step,
)
from scripts.raw_to_atif_common import renumber_steps, text_from_content


def is_empty_turn(step: Step) -> bool:
return (
step.observation is None
and not step.tool_calls
and not text_from_content(step.message).strip()
)


def normalize_nemotron(trajectory: ATIFTrajectory) -> ATIFTrajectory:
normalized = standardize_tools(normalize_atif_trajectory(trajectory))
split_terminal_task_description_prompt(normalized)
for step in normalized.steps:
structure_terminal_completion_step(step)
normalized.steps = renumber_steps(
[step for step in normalized.steps if not is_empty_turn(step)]
)
return normalized


def main(script_file: str | None = None) -> None: # noqa: ARG001
for line in sys.stdin:
if not line.strip():
continue
trajectory = ATIFTrajectory(**json.loads(line))
print(normalize_nemotron(trajectory).model_dump_json(exclude_none=True))


if __name__ == "__main__":
Expand Down
226 changes: 185 additions & 41 deletions datasets/nemotron_terminal_corpus/sample_sft/openhands_sdk.json

Large diffs are not rendered by default.

Loading
Loading