def test_faiss_index_save_and_load(tmp_path):
document_store = FAISSDocumentStore(
sql_url=f"sqlite:////{tmp_path/'haystack_test.db'}",
index="haystack_test",
)
document_store.write_documents(DOCUMENTS)
# test saving the index
document_store.save(tmp_path / "haystack_test_faiss")
# clear existing faiss_index
document_store.faiss_indexes[document_store.index].reset()
# test faiss index is cleared
assert document_store.faiss_indexes[document_store.index].ntotal == 0
# test loading the index
new_document_store = FAISSDocumentStore.load(
sql_url=f"sqlite:////{tmp_path/'haystack_test.db'}",
faiss_file_path=tmp_path / "haystack_test_faiss",
index=document_store.index)
# check faiss index is restored
assert new_document_store.faiss_indexes[
document_store.index].ntotal == len(DOCUMENTS)
# check if documents are restored
assert len(new_document_store.get_all_documents()) == len(DOCUMENTS)
def test_faiss_retrieving(index_factory):
document_store = FAISSDocumentStore(
sql_url="sqlite:///haystack_test_faiss.db",
faiss_index_factory_str=index_factory)
document_store.delete_all_documents(index="document")
if "ivf" in index_factory.lower():
document_store.train_index(DOCUMENTS)
document_store.write_documents(DOCUMENTS)
retriever = EmbeddingRetriever(document_store=document_store,
embedding_model="deepset/sentence_bert",
use_gpu=False)
result = retriever.retrieve(query="How to test this?")
assert len(result) == len(DOCUMENTS)
assert type(result[0]) == Document
def Find_answer(text_file_path, data_folder_path, symbol, question):
document_store = FAISSDocumentStore(faiss_index_factory_str="Flat")
with open(text_file_path, 'r', encoding='utf-8') as f:
data = f.read()
for i, line in enumerate(data.split(symbol)):
with open(f'{data_folder_path}/data{i+1}.txt', 'w') as f:
print(f'writing file no.{i+1}')
f.write(line)
test_dicts = convert_files_to_dicts(dir_path=data_folder_path,
clean_func=clean_wiki_text,
split_paragraphs=True)
document_store.write_documents(test_dicts)
retriever = DensePassageRetriever(
document_store=document_store,
query_embedding_model="facebook/dpr-question_encoder-single-nq-base",
passage_embedding_model="facebook/dpr-ctx_encoder-single-nq-base",
max_seq_len_query=64,
max_seq_len_passage=256,
batch_size=16,
use_gpu=True,
embed_title=True,
use_fast_tokenizers=True)
document_store.update_embeddings(retriever)
reader = FARMReader(model_name_or_path="deepset/roberta-base-squad2",
use_gpu=True,
context_window_size=300)
pipe = ExtractiveQAPipeline(reader, retriever)
prediction = pipe.run(query=question, top_k_retriever=10, top_k_reader=3)
doc_with_ans = []
for i in range(len(prediction['answers'])):
if prediction['answers'][i]['context'] not in doc_with_ans:
doc_with_ans.append(prediction['answers'][i]['context'])
answer = ' '.join(doc_with_ans)
return answer
def test_faiss_passing_index_from_outside():
d = 768
nlist = 2
quantizer = faiss.IndexFlatIP(d)
faiss_index = faiss.IndexIVFFlat(quantizer, d, nlist,
faiss.METRIC_INNER_PRODUCT)
faiss_index.nprobe = 2
document_store = FAISSDocumentStore(
sql_url="sqlite:///haystack_test_faiss.db", faiss_index=faiss_index)
document_store.delete_all_documents(index="document")
# as it is a IVF index we need to train it before adding docs
document_store.train_index(DOCUMENTS)
document_store.write_documents(documents=DOCUMENTS, index="document")
documents_indexed = document_store.get_all_documents(index="document")
# test document correctness
check_data_correctness(documents_indexed, DOCUMENTS)
def test_faiss_passing_index_from_outside():
d = 768
nlist = 2
quantizer = faiss.IndexFlatIP(d)
faiss_index = faiss.IndexIVFFlat(quantizer, d, nlist,
faiss.METRIC_INNER_PRODUCT)
faiss_index.set_direct_map_type(faiss.DirectMap.Hashtable)
faiss_index.nprobe = 2
document_store = FAISSDocumentStore(
sql_url="sqlite:///haystack_test_faiss.db", faiss_index=faiss_index)
document_store.delete_all_documents(index="document")
# as it is a IVF index we need to train it before adding docs
document_store.train_index(DOCUMENTS)
document_store.write_documents(documents=DOCUMENTS, index="document")
documents_indexed = document_store.get_all_documents(index="document")
# test if vectors ids are associated with docs
for doc in documents_indexed:
assert 0 <= int(doc.meta["vector_id"]) <= 7
# Initialize RAG Generator
generator = RAGenerator(
model_name_or_path="facebook/rag-token-nq",
use_gpu=False,
top_k_answers=1,
max_length=200,
min_length=2,
embed_title=True,
num_beams=2,
)
# Delete existing documents in documents store
document_store.delete_all_documents()
# Write documents to document store
document_store.write_documents(documents)
# Add documents embeddings to index
document_store.update_embeddings(retriever=retriever)
# Now ask your questions
# We have some sample questions
QUESTIONS = [
"who got the first nobel prize in physics",
"when is the next deadpool movie being released",
"which mode is used for short wave broadcast service",
"who is the owner of reading football club",
"when is the next scandal episode coming out",
"when is the last time the philadelphia won the superbowl",
"what is the most current adobe flash player version",
"how many episodes are there in dragon ball z",
"what is the first step in the evolution of the eye",
def tutorial7_rag_generator():
# Add documents from which you want generate answers
# Download a csv containing some sample documents data
# Here some sample documents data
temp = requests.get("https://raw.githubusercontent.com/deepset-ai/haystack/master/tutorials/small_generator_dataset.csv")
open('small_generator_dataset.csv', 'wb').write(temp.content)
# Get dataframe with columns "title", and "text"
df = pd.read_csv("small_generator_dataset.csv", sep=',')
# Minimal cleaning
df.fillna(value="", inplace=True)
print(df.head())
titles = list(df["title"].values)
texts = list(df["text"].values)
# Create to haystack document format
documents: List[Document] = []
for title, text in zip(titles, texts):
documents.append(
Document(
text=text,
meta={
"name": title or ""
}
)
)
# Initialize FAISS document store to documents and corresponding index for embeddings
# Set `return_embedding` to `True`, so generator doesn't have to perform re-embedding
document_store = FAISSDocumentStore(
faiss_index_factory_str="Flat",
return_embedding=True
)
# Initialize DPR Retriever to encode documents, encode question and query documents
retriever = DensePassageRetriever(
document_store=document_store,
query_embedding_model="facebook/dpr-question_encoder-single-nq-base",
passage_embedding_model="facebook/dpr-ctx_encoder-single-nq-base",
use_gpu=True,
embed_title=True,
)
# Initialize RAG Generator
generator = RAGenerator(
model_name_or_path="facebook/rag-token-nq",
use_gpu=True,
top_k_answers=1,
max_length=200,
min_length=2,
embed_title=True,
num_beams=2,
)
# Delete existing documents in documents store
document_store.delete_all_documents()
# Write documents to document store
document_store.write_documents(documents)
# Add documents embeddings to index
document_store.update_embeddings(
retriever=retriever
)
# Now ask your questions
# We have some sample questions
QUESTIONS = [
"who got the first nobel prize in physics",
"when is the next deadpool movie being released",
"which mode is used for short wave broadcast service",
"who is the owner of reading football club",
"when is the next scandal episode coming out",
"when is the last time the philadelphia won the superbowl",
"what is the most current adobe flash player version",
"how many episodes are there in dragon ball z",
"what is the first step in the evolution of the eye",
"where is gall bladder situated in human body",
"what is the main mineral in lithium batteries",
"who is the president of usa right now",
"where do the greasers live in the outsiders",
"panda is a national animal of which country",
"what is the name of manchester united stadium",
]
# Now generate answer for question
for question in QUESTIONS:
# Retrieve related documents from retriever
retriever_results = retriever.retrieve(
query=question
)
# Now generate answer from question and retrieved documents
predicted_result = generator.predict(
query=question,
documents=retriever_results,
top_k=1
)
# Print you answer
answers = predicted_result["answers"]
print(f'Generated answer is \'{answers[0]["answer"]}\' for the question = \'{question}\'')
def tutorial6_better_retrieval_via_dpr():
# FAISS is a library for efficient similarity search on a cluster of dense vectors.
# The FAISSDocumentStore uses a SQL(SQLite in-memory be default) document store under-the-hood
# to store the document text and other meta data. The vector embeddings of the text are
# indexed on a FAISS Index that later is queried for searching answers.
# The default flavour of FAISSDocumentStore is "Flat" but can also be set to "HNSW" for
# faster search at the expense of some accuracy. Just set the faiss_index_factor_str argument in the constructor.
# For more info on which suits your use case: https://github.com/facebookresearch/faiss/wiki/Guidelines-to-choose-an-index
document_store = FAISSDocumentStore(faiss_index_factory_str="Flat")
# ## Preprocessing of documents
# Let's first get some documents that we want to query
doc_dir = "data/article_txt_got"
s3_url = "https://s3.eu-central-1.amazonaws.com/deepset.ai-farm-qa/datasets/documents/wiki_gameofthrones_txt.zip"
fetch_archive_from_http(url=s3_url, output_dir=doc_dir)
# convert files to dicts containing documents that can be indexed to our datastore
dicts = convert_files_to_dicts(dir_path=doc_dir,
clean_func=clean_wiki_text,
split_paragraphs=True)
# Now, let's write the docs to our DB.
document_store.write_documents(dicts)
### Retriever
retriever = DensePassageRetriever(
document_store=document_store,
query_embedding_model="facebook/dpr-question_encoder-single-nq-base",
passage_embedding_model="facebook/dpr-ctx_encoder-single-nq-base",
max_seq_len_query=64,
max_seq_len_passage=256,
batch_size=2,
use_gpu=True,
embed_title=True,
use_fast_tokenizers=True)
# Important:
# Now that after we have the DPR initialized, we need to call update_embeddings() to iterate over all
# previously indexed documents and update their embedding representation.
# While this can be a time consuming operation (depending on corpus size), it only needs to be done once.
# At query time, we only need to embed the query and compare it the existing doc embeddings which is very fast.
document_store.update_embeddings(retriever)
### Reader
# Load a local model or any of the QA models on
# Hugging Face's model hub (https://huggingface.co/models)
reader = FARMReader(model_name_or_path="deepset/roberta-base-squad2",
use_gpu=True)
### Pipeline
from haystack.pipeline import ExtractiveQAPipeline
pipe = ExtractiveQAPipeline(reader, retriever)
## Voilà! Ask a question!
prediction = pipe.run(query="Who is the father of Arya Stark?",
top_k_retriever=10,
top_k_reader=5)
# prediction = pipe.run(query="Who created the Dothraki vocabulary?", top_k_reader=5)
# prediction = pipe.run(query="Who is the sister of Sansa?", top_k_reader=5)
print_answers(prediction, details="minimal")
# max_seq_len_query=64,
# max_seq_len_passage=256,
# batch_size=16,
# use_gpu=True,
# embed_title=True,
# use_fast_tokenizers=True)
# Get dataframe with columns "question", "answer" and some custom metadata
df = pd.read_csv("faq.csv")
# Minimal cleaning
df.fillna(value="", inplace=True)
df["question"] = df["question"].apply(lambda x: x.strip())
print(df.head())
# Get embeddings for our questions from the FAQs
# questions = list(df["question"].values)
# df["question_emb"] = retriever2.embed_queries(texts=questions)
# text is the field to be converted to embeddings
df = df.rename(columns={"question": "text"})
# Convert Dataframe to list of dicts and index them in our DocumentStore
docs_to_index = df.to_dict(orient="records")
document_store.delete_all_documents()
document_store.write_documents(docs_to_index)
retriever2 = EmbeddingRetriever(
document_store=document_store, embedding_model="sentence_bert-saved", use_gpu=False)
document_store.update_embeddings(retriever2)
document_store.save('faiss2')
def tutorial12_lfqa():
"""
Document Store:
FAISS is a library for efficient similarity search on a cluster of dense vectors.
The `FAISSDocumentStore` uses a SQL(SQLite in-memory be default) database under-the-hood
to store the document text and other meta data. The vector embeddings of the text are
indexed on a FAISS Index that later is queried for searching answers.
The default flavour of FAISSDocumentStore is "Flat" but can also be set to "HNSW" for
faster search at the expense of some accuracy. Just set the faiss_index_factor_str argument in the constructor.
For more info on which suits your use case: https://github.com/facebookresearch/faiss/wiki/Guidelines-to-choose-an-index
"""
from haystack.document_store.faiss import FAISSDocumentStore
document_store = FAISSDocumentStore(vector_dim=128,
faiss_index_factory_str="Flat")
"""
Cleaning & indexing documents:
Similarly to the previous tutorials, we download, convert and index some Game of Thrones articles to our DocumentStore
"""
# Let's first get some files that we want to use
doc_dir = "data/article_txt_got"
s3_url = "https://s3.eu-central-1.amazonaws.com/deepset.ai-farm-qa/datasets/documents/wiki_gameofthrones_txt.zip"
fetch_archive_from_http(url=s3_url, output_dir=doc_dir)
# Convert files to dicts
dicts = convert_files_to_dicts(dir_path=doc_dir,
clean_func=clean_wiki_text,
split_paragraphs=True)
# Now, let's write the dicts containing documents to our DB.
document_store.write_documents(dicts)
"""
Initalize Retriever and Reader/Generator:
We use a `RetribertRetriever` and we invoke `update_embeddings` to index the embeddings of documents in the `FAISSDocumentStore`
"""
from haystack.retriever.dense import EmbeddingRetriever
retriever = EmbeddingRetriever(
document_store=document_store,
embedding_model="yjernite/retribert-base-uncased",
model_format="retribert")
document_store.update_embeddings(retriever)
"""Before we blindly use the `RetribertRetriever` let's empirically test it to make sure a simple search indeed finds the relevant documents."""
from haystack.utils import print_answers, print_documents
from haystack.pipeline import DocumentSearchPipeline
p_retrieval = DocumentSearchPipeline(retriever)
res = p_retrieval.run(query="Tell me something about Arya Stark?",
top_k_retriever=5)
print_documents(res, max_text_len=512)
"""
Similar to previous Tutorials we now initalize our reader/generator.
Here we use a `Seq2SeqGenerator` with the *yjernite/bart_eli5* model (see: https://huggingface.co/yjernite/bart_eli5)
"""
generator = Seq2SeqGenerator(model_name_or_path="yjernite/bart_eli5")
"""
Pipeline:
With a Haystack `Pipeline` you can stick together your building blocks to a search pipeline.
Under the hood, `Pipelines` are Directed Acyclic Graphs (DAGs) that you can easily customize for your own use cases.
To speed things up, Haystack also comes with a few predefined Pipelines. One of them is the `GenerativeQAPipeline` that combines a retriever and a reader/generator to answer our questions.
You can learn more about `Pipelines` in the [docs](https://haystack.deepset.ai/docs/latest/pipelinesmd).
"""
from haystack.pipeline import GenerativeQAPipeline
pipe = GenerativeQAPipeline(generator, retriever)
"""Voilà! Ask a question!"""
query_1 = "Why did Arya Stark's character get portrayed in a television adaptation?"
result_1 = pipe.run(query=query_1, top_k_retriever=1)
print(f"Query: {query_1}")
print(f"Answer: {result_1['answers'][0]}")
print()
query_2 = "What kind of character does Arya Stark play?"
result_2 = pipe.run(query=query_2, top_k_retriever=1)
print(f"Query: {query_2}")
print(f"Answer: {result_2['answers'][0]}")
print()
pipe.run(query=query_2, top_k_retriever=1)
clean_header_footer=True,
split_by="passage",
split_length=1,
split_respect_sentence_boundary=False,
split_overlap=0
)
as4Docs = processor.process(as4)
for i in range(len(as4Docs)):
as4Docs[i]["meta"]["table"] = False
document_store_ES.delete_all_documents()
document_store_ES.write_documents(as4Docs)
document_store_FAISS.delete_all_documents()
document_store_FAISS.write_documents(as4Docs)
#backagain = document_store.get_all_documents();
for i in range(0, len(as4Docs)):
print(str(i) + ":", end=" ")
print(as4Docs[i])
# Update table content and table description - csv for table content, txt for table description
# The files are under the /tables directory, the name does not matter, as long as the csv and the txt match each other.
# Need to rerun this section when new table is uploaded.
import csv
import os
data = []
