mirror of
https://github.com/kuhyx/WUT_Computer_Science.git
synced 2026-11-10 19:47:32 +01:00
Krzysztof kuhy Rudnicki
8798105aa1
The pre-commit workflow already installs `python-version: "3.x"`, but ruff was still grading the archive against py310 -- so it kept withholding the pyupgrade and type-checking fixes that only become safe at the newer target. `pin_ci_python.py` rewrites the knob. The 17 findings that appeared are what the stale target was hiding: `datetime.timezone.utc` where `datetime.UTC` is now correct (UP017), and 14 imports used only in annotations that belong under `TYPE_CHECKING` (TC002, TC003) -- safe here precisely because 3.14's PEP 649 defers annotation evaluation, which is the same reason ruff would not offer them at py310. `ruff check` and `ruff format --check` are clean again, so the zero-findings state this repo reached earlier today holds at the new target. Still contradictory and deliberately untouched: the four `Programming/ERSMS-project/*/Dockerfile` files say `FROM python:3.10-slim`. They build an archived coursework service against its own era's pinned dependencies, and nothing in this repo runs them; bumping the base image would be a rewrite of that project, not a lint-target change. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01SfnJmxxsZ534AKk16rC2Gr
76 lines
2.4 KiB
Python
Executable File
76 lines
2.4 KiB
Python
Executable File
#!/usr/bin/env python3
|
|
"""Get chunk alignment from ChatGPT.
|
|
|
|
The two prompts live in ``prompts/`` rather than in this file: they are 2 kB
|
|
of instructions each, on lines far past any line-length limit, and wrapping
|
|
them here would put newlines into the middle of the prompt itself.
|
|
"""
|
|
|
|
import os
|
|
from pathlib import Path
|
|
from typing import TYPE_CHECKING
|
|
|
|
from openai import OpenAI
|
|
|
|
if TYPE_CHECKING:
|
|
import pandas as pd
|
|
|
|
PROMPTS = Path(__file__).parent / "prompts"
|
|
THE_PROMPT = (PROMPTS / "alignment.txt").read_text(encoding="utf-8")
|
|
THE_PROMPT_WITH_COMMENT = (PROMPTS / "alignment_with_comment.txt").read_text(
|
|
encoding="utf-8"
|
|
)
|
|
|
|
|
|
def create_gpt() -> OpenAI:
|
|
"""Build an OpenAI client from OPENAI_API_KEY.
|
|
|
|
The key used to be a placeholder string literal here, which meant the
|
|
script could not actually run as committed.
|
|
"""
|
|
key = os.environ.get("OPENAI_API_KEY")
|
|
if not key:
|
|
msg = "set OPENAI_API_KEY to run the alignment scripts"
|
|
raise RuntimeError(msg)
|
|
return OpenAI(api_key=key)
|
|
|
|
|
|
def call_api(client: OpenAI, chunks: str) -> str:
|
|
"""Ask the model to align one pair of chunked sentences."""
|
|
response = client.chat.completions.create(
|
|
model="gpt-3.5-turbo-1106",
|
|
messages=[
|
|
{"role": "user", "content": THE_PROMPT},
|
|
{"role": "user", "content": chunks},
|
|
],
|
|
)
|
|
return response.choices[0].message.content.strip()
|
|
|
|
|
|
def call_api_examples(client: OpenAI, examples: pd.DataFrame, alignment: str) -> str:
|
|
"""Ask the model to align, few-shot prompted with worked examples."""
|
|
user_input = []
|
|
assistant_output = []
|
|
for _index, row in examples.iterrows():
|
|
chunks = ""
|
|
for chunk in row["chunked_sentance1"]:
|
|
chunks = chunks + "[ " + chunk + " ] "
|
|
chunks = chunks + "\n"
|
|
for chunk in row["chunked_sentance2"]:
|
|
chunks = chunks + "[ " + chunk + " ] "
|
|
user_input.append(chunks)
|
|
assistant_output.append(row["alignment_text"])
|
|
|
|
messages = []
|
|
messages.append({"role": "user", "content": THE_PROMPT})
|
|
for u, a in zip(user_input, assistant_output, strict=True):
|
|
messages.append({"role": "user", "content": u})
|
|
messages.append({"role": "assistant", "content": a})
|
|
|
|
messages.append({"role": "user", "content": alignment})
|
|
|
|
response = client.chat.completions.create(
|
|
model="gpt-3.5-turbo-1106", messages=messages
|
|
)
|
|
return response.choices[0].message.content.strip()
|