mirror of
https://github.com/hwchase17/langchain
synced 2024-11-10 01:10:59 +00:00
f92006de3c
0.2rc migrations - [x] Move memory - [x] Move remaining retrievers - [x] graph_qa chains - [x] some dependency from evaluation code potentially on math utils - [x] Move openapi chain from `langchain.chains.api.openapi` to `langchain_community.chains.openapi` - [x] Migrate `langchain.chains.ernie_functions` to `langchain_community.chains.ernie_functions` - [x] migrate `langchain/chains/llm_requests.py` to `langchain_community.chains.llm_requests` - [x] Moving `langchain_community.cross_enoders.base:BaseCrossEncoder` -> `langchain_community.retrievers.document_compressors.cross_encoder:BaseCrossEncoder` (namespace not ideal, but it needs to be moved to `langchain` to avoid circular deps) - [x] unit tests langchain -- add pytest.mark.community to some unit tests that will stay in langchain - [x] unit tests community -- move unit tests that depend on community to community - [x] mv integration tests that depend on community to community - [x] mypy checks Other todo - [x] Make deprecation warnings not noisy (need to use warn deprecated and check that things are implemented properly) - [x] Update deprecation messages with timeline for code removal (likely we actually won't be removing things until 0.4 release) -- will give people more time to transition their code. - [ ] Add information to deprecation warning to show users how to migrate their code base using langchain-cli - [ ] Remove any unnecessary requirements in langchain (e.g., is SQLALchemy required?) --------- Co-authored-by: Erick Friis <erick@langchain.dev>
76 lines
2.3 KiB
Python
76 lines
2.3 KiB
Python
"""Test LLM chain."""
|
|
from tempfile import TemporaryDirectory
|
|
from typing import Dict, List, Union
|
|
from unittest.mock import patch
|
|
|
|
import pytest
|
|
from langchain.chains.llm import LLMChain
|
|
from langchain_core.output_parsers import BaseOutputParser
|
|
from langchain_core.prompts import PromptTemplate
|
|
|
|
from tests.unit_tests.llms.fake_llm import FakeLLM
|
|
|
|
|
|
class FakeOutputParser(BaseOutputParser):
|
|
"""Fake output parser class for testing."""
|
|
|
|
def parse(self, text: str) -> Union[str, List[str], Dict[str, str]]:
|
|
"""Parse by splitting."""
|
|
return text.split()
|
|
|
|
|
|
@pytest.fixture
|
|
def fake_llm_chain() -> LLMChain:
|
|
"""Fake LLM chain for testing purposes."""
|
|
prompt = PromptTemplate(input_variables=["bar"], template="This is a {bar}:")
|
|
return LLMChain(prompt=prompt, llm=FakeLLM(), output_key="text1")
|
|
|
|
|
|
@patch(
|
|
"langchain_community.llms.loading.get_type_to_cls_dict",
|
|
lambda: {"fake": lambda: FakeLLM},
|
|
)
|
|
def test_serialization(fake_llm_chain: LLMChain) -> None:
|
|
"""Test serialization."""
|
|
from langchain.chains.loading import load_chain
|
|
|
|
with TemporaryDirectory() as temp_dir:
|
|
file = temp_dir + "/llm.json"
|
|
fake_llm_chain.save(file)
|
|
loaded_chain = load_chain(file)
|
|
assert loaded_chain == fake_llm_chain
|
|
|
|
|
|
def test_missing_inputs(fake_llm_chain: LLMChain) -> None:
|
|
"""Test error is raised if inputs are missing."""
|
|
with pytest.raises(ValueError):
|
|
fake_llm_chain({"foo": "bar"})
|
|
|
|
|
|
def test_valid_call(fake_llm_chain: LLMChain) -> None:
|
|
"""Test valid call of LLM chain."""
|
|
output = fake_llm_chain({"bar": "baz"})
|
|
assert output == {"bar": "baz", "text1": "foo"}
|
|
|
|
# Test with stop words.
|
|
output = fake_llm_chain({"bar": "baz", "stop": ["foo"]})
|
|
# Response should be `bar` now.
|
|
assert output == {"bar": "baz", "stop": ["foo"], "text1": "bar"}
|
|
|
|
|
|
def test_predict_method(fake_llm_chain: LLMChain) -> None:
|
|
"""Test predict method works."""
|
|
output = fake_llm_chain.predict(bar="baz")
|
|
assert output == "foo"
|
|
|
|
|
|
def test_predict_and_parse() -> None:
|
|
"""Test parsing ability."""
|
|
prompt = PromptTemplate(
|
|
input_variables=["foo"], template="{foo}", output_parser=FakeOutputParser()
|
|
)
|
|
llm = FakeLLM(queries={"foo": "foo bar"})
|
|
chain = LLMChain(prompt=prompt, llm=llm)
|
|
output = chain.predict_and_parse(foo="foo")
|
|
assert output == ["foo", "bar"]
|