mirror of
https://github.com/hwchase17/langchain.git
synced 2025-05-08 16:48:49 +00:00
a0c2281540
```python
"""python scripts/update_mypy_ruff.py"""
import glob
import tomllib
from pathlib import Path
import toml
import subprocess
import re
ROOT_DIR = Path(__file__).parents[1]
def main():
for path in glob.glob(str(ROOT_DIR / "libs/**/pyproject.toml"), recursive=True):
print(path)
with open(path, "rb") as f:
pyproject = tomllib.load(f)
try:
pyproject["tool"]["poetry"]["group"]["typing"]["dependencies"]["mypy"] = (
"^1.10"
)
pyproject["tool"]["poetry"]["group"]["lint"]["dependencies"]["ruff"] = (
"^0.5"
)
except KeyError:
continue
with open(path, "w") as f:
toml.dump(pyproject, f)
cwd = "/".join(path.split("/")[:-1])
completed = subprocess.run(
"poetry lock --no-update; poetry install --with typing; poetry run mypy . --no-color",
cwd=cwd,
shell=True,
capture_output=True,
text=True,
)
logs = completed.stdout.split("\n")
to_ignore = {}
for l in logs:
if re.match("^(.*)\:(\d+)\: error:.*\[(.*)\]", l):
path, line_no, error_type = re.match(
"^(.*)\:(\d+)\: error:.*\[(.*)\]", l
).groups()
if (path, line_no) in to_ignore:
to_ignore[(path, line_no)].append(error_type)
else:
to_ignore[(path, line_no)] = [error_type]
print(len(to_ignore))
for (error_path, line_no), error_types in to_ignore.items():
all_errors = ", ".join(error_types)
full_path = f"{cwd}/{error_path}"
try:
with open(full_path, "r") as f:
file_lines = f.readlines()
except FileNotFoundError:
continue
file_lines[int(line_no) - 1] = (
file_lines[int(line_no) - 1][:-1] + f" # type: ignore[{all_errors}]\n"
)
with open(full_path, "w") as f:
f.write("".join(file_lines))
subprocess.run(
"poetry run ruff format .; poetry run ruff --select I --fix .",
cwd=cwd,
shell=True,
capture_output=True,
text=True,
)
if __name__ == "__main__":
main()
```
90 lines
3.1 KiB
Python
90 lines
3.1 KiB
Python
# flake8: noqa
|
|
"""Test Llama.cpp wrapper."""
|
|
|
|
import os
|
|
from typing import Generator
|
|
from urllib.request import urlretrieve
|
|
|
|
import pytest
|
|
|
|
from langchain_community.llms import LlamaCpp
|
|
|
|
from tests.unit_tests.callbacks.fake_callback_handler import FakeCallbackHandler
|
|
|
|
|
|
def get_model() -> str:
|
|
"""Download model. f
|
|
From https://huggingface.co/Sosaka/Alpaca-native-4bit-ggml/,
|
|
convert to new ggml format and return model path."""
|
|
model_url = "https://huggingface.co/Sosaka/Alpaca-native-4bit-ggml/resolve/main/ggml-alpaca-7b-q4.bin"
|
|
tokenizer_url = "https://huggingface.co/decapoda-research/llama-7b-hf/resolve/main/tokenizer.model"
|
|
conversion_script = "https://github.com/ggerganov/llama.cpp/raw/master/convert-unversioned-ggml-to-ggml.py"
|
|
local_filename = model_url.split("/")[-1]
|
|
|
|
if not os.path.exists("convert-unversioned-ggml-to-ggml.py"):
|
|
urlretrieve(conversion_script, "convert-unversioned-ggml-to-ggml.py")
|
|
if not os.path.exists("tokenizer.model"):
|
|
urlretrieve(tokenizer_url, "tokenizer.model")
|
|
if not os.path.exists(local_filename):
|
|
urlretrieve(model_url, local_filename)
|
|
os.system(f"python convert-unversioned-ggml-to-ggml.py . tokenizer.model")
|
|
|
|
return local_filename
|
|
|
|
|
|
def test_llamacpp_inference() -> None:
|
|
"""Test valid llama.cpp inference."""
|
|
model_path = get_model()
|
|
llm = LlamaCpp(model_path=model_path)
|
|
output = llm.invoke("Say foo:")
|
|
assert isinstance(output, str)
|
|
assert len(output) > 1
|
|
|
|
|
|
def test_llamacpp_streaming() -> None:
|
|
"""Test streaming tokens from LlamaCpp."""
|
|
model_path = get_model()
|
|
llm = LlamaCpp(model_path=model_path, max_tokens=10)
|
|
generator = llm.stream("Q: How do you say 'hello' in German? A:'", stop=["'"])
|
|
stream_results_string = ""
|
|
assert isinstance(generator, Generator)
|
|
|
|
for chunk in generator:
|
|
assert not isinstance(chunk, str)
|
|
# Note that this matches the OpenAI format:
|
|
assert isinstance(chunk["choices"][0]["text"], str)
|
|
stream_results_string += chunk["choices"][0]["text"]
|
|
assert len(stream_results_string.strip()) > 1
|
|
|
|
|
|
def test_llamacpp_streaming_callback() -> None:
|
|
"""Test that streaming correctly invokes on_llm_new_token callback."""
|
|
MAX_TOKENS = 5
|
|
OFF_BY_ONE = 1 # There may be an off by one error in the upstream code!
|
|
|
|
callback_handler = FakeCallbackHandler()
|
|
llm = LlamaCpp(
|
|
model_path=get_model(),
|
|
callbacks=[callback_handler],
|
|
verbose=True,
|
|
max_tokens=MAX_TOKENS,
|
|
)
|
|
llm.invoke("Q: Can you count to 10? A:'1, ")
|
|
assert callback_handler.llm_streams <= MAX_TOKENS + OFF_BY_ONE
|
|
|
|
|
|
def test_llamacpp_model_kwargs() -> None:
|
|
llm = LlamaCpp(model_path=get_model(), model_kwargs={"n_gqa": None})
|
|
assert llm.model_kwargs == {"n_gqa": None}
|
|
|
|
|
|
def test_llamacpp_invalid_model_kwargs() -> None:
|
|
with pytest.raises(ValueError):
|
|
LlamaCpp(model_path=get_model(), model_kwargs={"n_ctx": 1024})
|
|
|
|
|
|
def test_llamacpp_incorrect_field() -> None:
|
|
with pytest.warns(match="not default parameter"):
|
|
llm = LlamaCpp(model_path=get_model(), n_gqa=None)
|
|
llm.model_kwargs == {"n_gqa": None}
|