mirror of
https://github.com/deepset-ai/haystack.git
synced 2025-10-31 09:49:48 +00:00
738e008020
* Add run_batch methods for batch querying * Update Documentation & Code Style * Fix mypy * Update Documentation & Code Style * Fix mypy * Fix linter * Fix tests * Update Documentation & Code Style * Fix tests * Update Documentation & Code Style * Fix mypy * Fix rest api test * Update Documentation & Code Style * Add Doc strings * Update Documentation & Code Style * Add batch_size as attribute to nodes supporting batching * Adapt error messages * Adapt type of filters in retrievers * Revert change about truncation_warning in summarizer * Unify multiple_doc_lists tests * Use smaller models in extractor tests * Add return types to JoinAnswers and RouteDocuments * Adapt return statements in reader's run_batch method * Allow list of filters * Adapt error messages * Update Documentation & Code Style * Fix tests * Fix mypy * Adapt print_questions * Remove disabling warning about too many public methods * Add flag for pylint to disable warning about too many public methods in pipelines/base.py and document_stores/base.py * Add type check * Update Documentation & Code Style * Adapt tutorial 11 * Update Documentation & Code Style * Add query_batch method for DCDocStore * Update Documentation & Code Style Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
161 lines
6.7 KiB
Python
161 lines
6.7 KiB
Python
import logging
|
|
|
|
import pandas as pd
|
|
import pytest
|
|
|
|
from haystack.schema import Document, Answer
|
|
from haystack.pipelines.base import Pipeline
|
|
|
|
|
|
def test_table_reader(table_reader):
|
|
data = {
|
|
"actors": ["brad pitt", "leonardo di caprio", "george clooney"],
|
|
"age": ["58", "47", "60"],
|
|
"number of movies": ["87", "53", "69"],
|
|
"date of birth": ["18 december 1963", "11 november 1974", "6 may 1961"],
|
|
}
|
|
table = pd.DataFrame(data)
|
|
|
|
query = "When was Di Caprio born?"
|
|
prediction = table_reader.predict(query=query, documents=[Document(content=table, content_type="table")])
|
|
assert prediction["answers"][0].answer == "11 november 1974"
|
|
assert prediction["answers"][0].offsets_in_context[0].start == 7
|
|
assert prediction["answers"][0].offsets_in_context[0].end == 8
|
|
|
|
|
|
def test_table_reader_batch_single_query_single_doc_list(table_reader):
|
|
data = {
|
|
"actors": ["brad pitt", "leonardo di caprio", "george clooney"],
|
|
"age": ["58", "47", "60"],
|
|
"number of movies": ["87", "53", "69"],
|
|
"date of birth": ["18 december 1963", "11 november 1974", "6 may 1961"],
|
|
}
|
|
table = pd.DataFrame(data)
|
|
|
|
query = "When was Di Caprio born?"
|
|
prediction = table_reader.predict_batch(queries=query, documents=[Document(content=table, content_type="table")])
|
|
# Expected output: List of lists of answers
|
|
assert isinstance(prediction["answers"], list)
|
|
assert isinstance(prediction["answers"][0], list)
|
|
assert isinstance(prediction["answers"][0][0], Answer)
|
|
assert len(prediction["answers"]) == 1 # Predictions for 5 docs
|
|
|
|
|
|
def test_table_reader_batch_single_query_multiple_doc_lists(table_reader):
|
|
data = {
|
|
"actors": ["brad pitt", "leonardo di caprio", "george clooney"],
|
|
"age": ["58", "47", "60"],
|
|
"number of movies": ["87", "53", "69"],
|
|
"date of birth": ["18 december 1963", "11 november 1974", "6 may 1961"],
|
|
}
|
|
table = pd.DataFrame(data)
|
|
|
|
query = "When was Di Caprio born?"
|
|
prediction = table_reader.predict_batch(queries=query, documents=[[Document(content=table, content_type="table")]])
|
|
# Expected output: List of lists of answers
|
|
assert isinstance(prediction["answers"], list)
|
|
assert isinstance(prediction["answers"][0], list)
|
|
assert isinstance(prediction["answers"][0][0], Answer)
|
|
assert len(prediction["answers"]) == 1 # Predictions for 1 collection of docs
|
|
|
|
|
|
def test_table_reader_batch_multiple_queries_single_doc_list(table_reader):
|
|
data = {
|
|
"actors": ["brad pitt", "leonardo di caprio", "george clooney"],
|
|
"age": ["58", "47", "60"],
|
|
"number of movies": ["87", "53", "69"],
|
|
"date of birth": ["18 december 1963", "11 november 1974", "6 may 1961"],
|
|
}
|
|
table = pd.DataFrame(data)
|
|
|
|
query = "When was Di Caprio born?"
|
|
prediction = table_reader.predict_batch(
|
|
queries=[query, query], documents=[Document(content=table, content_type="table")]
|
|
)
|
|
# Expected output: List of lists of lists of answers
|
|
assert isinstance(prediction["answers"], list)
|
|
assert isinstance(prediction["answers"][0], list)
|
|
assert isinstance(prediction["answers"][0][0], list)
|
|
assert isinstance(prediction["answers"][0][0][0], Answer)
|
|
assert len(prediction["answers"]) == 2 # Predictions for 2 queries
|
|
|
|
|
|
def test_table_reader_batch_multiple_queries_multiple_doc_lists(table_reader):
|
|
data = {
|
|
"actors": ["brad pitt", "leonardo di caprio", "george clooney"],
|
|
"age": ["58", "47", "60"],
|
|
"number of movies": ["87", "53", "69"],
|
|
"date of birth": ["18 december 1963", "11 november 1974", "6 may 1961"],
|
|
}
|
|
table = pd.DataFrame(data)
|
|
|
|
query = "When was Di Caprio born?"
|
|
prediction = table_reader.predict_batch(
|
|
queries=[query, query],
|
|
documents=[[Document(content=table, content_type="table")], [Document(content=table, content_type="table")]],
|
|
)
|
|
# Expected output: List of lists answers
|
|
assert isinstance(prediction["answers"], list)
|
|
assert isinstance(prediction["answers"][0], list)
|
|
assert isinstance(prediction["answers"][0][0], Answer)
|
|
assert len(prediction["answers"]) == 2 # Predictions for 2 collections of documents
|
|
|
|
|
|
def test_table_reader_in_pipeline(table_reader):
|
|
pipeline = Pipeline()
|
|
pipeline.add_node(table_reader, "TableReader", ["Query"])
|
|
data = {
|
|
"actors": ["brad pitt", "leonardo di caprio", "george clooney"],
|
|
"age": ["58", "47", "60"],
|
|
"number of movies": ["87", "53", "69"],
|
|
"date of birth": ["18 december 1963", "11 november 1974", "6 may 1961"],
|
|
}
|
|
|
|
table = pd.DataFrame(data)
|
|
query = "When was Di Caprio born?"
|
|
|
|
prediction = pipeline.run(query=query, documents=[Document(content=table, content_type="table")])
|
|
|
|
assert prediction["answers"][0].answer == "11 november 1974"
|
|
assert prediction["answers"][0].offsets_in_context[0].start == 7
|
|
assert prediction["answers"][0].offsets_in_context[0].end == 8
|
|
|
|
|
|
@pytest.mark.parametrize("table_reader", ["tapas"], indirect=True)
|
|
def test_table_reader_aggregation(table_reader):
|
|
data = {
|
|
"Mountain": ["Mount Everest", "K2", "Kangchenjunga", "Lhotse", "Makalu"],
|
|
"Height": ["8848m", "8,611 m", "8 586m", "8 516 m", "8,485m"],
|
|
}
|
|
table = pd.DataFrame(data)
|
|
|
|
query = "How tall are all mountains on average?"
|
|
prediction = table_reader.predict(query=query, documents=[Document(content=table, content_type="table")])
|
|
assert prediction["answers"][0].answer == "8609.2 m"
|
|
assert prediction["answers"][0].meta["aggregation_operator"] == "AVERAGE"
|
|
assert prediction["answers"][0].meta["answer_cells"] == ["8848m", "8,611 m", "8 586m", "8 516 m", "8,485m"]
|
|
|
|
query = "How tall are all mountains together?"
|
|
prediction = table_reader.predict(query=query, documents=[Document(content=table, content_type="table")])
|
|
assert prediction["answers"][0].answer == "43046.0 m"
|
|
assert prediction["answers"][0].meta["aggregation_operator"] == "SUM"
|
|
assert prediction["answers"][0].meta["answer_cells"] == ["8848m", "8,611 m", "8 586m", "8 516 m", "8,485m"]
|
|
|
|
|
|
def test_table_without_rows(caplog, table_reader):
|
|
# empty DataFrame
|
|
table = pd.DataFrame()
|
|
document = Document(content=table, content_type="table", id="no_rows")
|
|
with caplog.at_level(logging.WARNING):
|
|
predictions = table_reader.predict(query="test", documents=[document])
|
|
assert "Skipping document with id 'no_rows'" in caplog.text
|
|
assert len(predictions["answers"]) == 0
|
|
|
|
|
|
def test_text_document(caplog, table_reader):
|
|
document = Document(content="text", id="text_doc")
|
|
with caplog.at_level(logging.WARNING):
|
|
predictions = table_reader.predict(query="test", documents=[document])
|
|
assert "Skipping document with id 'text_doc'" in caplog.text
|
|
assert len(predictions["answers"]) == 0
|