yujunjun/haystack
1
0
Fork 0
mirror of https://github.com/deepset-ai/haystack.git synced 2025-12-12 07:17:41 +00:00
Code Issues Packages Projects Releases Wiki Activity

Update tutorials (torch versions, ES version, replace Finder with Pipeline) (#814)

Browse Source 
* remove manual torch install on colab

* update elasticsearch version everywhere to 7.9.2

* fix FAQPipeline

* update tutorials with new pipelines

* Add latest docstring and tutorial changes

* revert faqpipeline change. fix field names in tutorial 4

* Add latest docstring and tutorial changes

Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
This commit is contained in:
Malte Pietsch 2021-02-09 14:56:54 +01:00 committed by GitHub
parent ac9f92466f
commit e91518ee00
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
26 changed files with 200 additions and 223 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

28
docs/_src/tutorials/tutorials/1.md
View File

@ -44,13 +44,11 @@ Make sure you enable the GPU runtime to experience decent speed in this tutorial
# Install the latest master of Haystack
!pip install git+https://github.com/deepset-ai/haystack.git
!pip install urllib3==1.25.4
!pip install torch==1.6.0+cu101 torchvision==0.6.1+cu101 -f https://download.pytorch.org/whl/torch_stable.html
```
```python
from haystack import Finder
from haystack.preprocessor.cleaning import clean_wiki_text
from haystack.preprocessor.utils import convert_files_to_dicts, fetch_archive_from_http
from haystack.reader.farm import FARMReader
@ -74,19 +72,19 @@ You can start Elasticsearch on your local machine instance using Docker. If Dock
```python
# Recommended: Start Elasticsearch using Docker
#! docker run -d -p 9200:9200 -e "discovery.type=single-node" elasticsearch:7.6.2
#! docker run -d -p 9200:9200 -e "discovery.type=single-node" elasticsearch:7.9.2
```
```python
# In Colab / No Docker environments: Start Elasticsearch from source
! wget https://artifacts.elastic.co/downloads/elasticsearch/elasticsearch-7.6.2-linux-x86_64.tar.gz -q
! tar -xzf elasticsearch-7.6.2-linux-x86_64.tar.gz
! chown -R daemon:daemon elasticsearch-7.6.2
! wget https://artifacts.elastic.co/downloads/elasticsearch/elasticsearch-7.9.2-linux-x86_64.tar.gz -q
! tar -xzf elasticsearch-7.9.2-linux-x86_64.tar.gz
! chown -R daemon:daemon elasticsearch-7.9.2
import os
from subprocess import Popen, PIPE, STDOUT
es_server = Popen(['elasticsearch-7.6.2/bin/elasticsearch'],
es_server = Popen(['elasticsearch-7.9.2/bin/elasticsearch'],
stdout=PIPE, stderr=STDOUT,
preexec_fn=lambda: os.setuid(1) # as daemon
)
@ -206,13 +204,17 @@ reader = FARMReader(model_name_or_path="deepset/roberta-base-squad2", use_gpu=Tr
# reader = TransformersReader(model_name_or_path="distilbert-base-uncased-distilled-squad", tokenizer="distilbert-base-uncased", use_gpu=-1)
```
### Finder
### Pipeline
The Finder sticks together reader and retriever in a pipeline to answer our actual questions.
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 `ExtractiveQAPipeline` that combines a retriever and a reader to answer our questions.
You can learn more about `Pipelines` in the [docs](https://haystack.deepset.ai/docs/latest/pipelinesmd).
```python
finder = Finder(reader, retriever)
from haystack.pipeline import ExtractiveQAPipeline
pipe = ExtractiveQAPipeline(reader, retriever)
```
## Voilà! Ask a question!
@ -221,13 +223,13 @@ finder = Finder(reader, retriever)
```python
# You can configure how many candidates the reader and retriever shall return
# The higher top_k_retriever, the better (but also the slower) your answers.
prediction = finder.get_answers(question="Who is the father of Arya Stark?", top_k_retriever=10, top_k_reader=5)
prediction = pipe.run(query="Who is the father of Arya Stark?", top_k_retriever=10, top_k_reader=5)
```
```python
# prediction = finder.get_answers(question="Who created the Dothraki vocabulary?", top_k_reader=5)
# prediction = finder.get_answers(question="Who is the sister of Sansa?", 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)
```

2
docs/_src/tutorials/tutorials/2.md
View File

@ -39,8 +39,6 @@ Make sure you enable the GPU runtime to experience decent speed in this tutorial
# Install the latest master of Haystack
!pip install git+https://github.com/deepset-ai/haystack.git
!pip install urllib3==1.25.4
!pip install torch==1.6.0+cu101 torchvision==0.6.1+cu101 -f https://download.pytorch.org/whl/torch_stable.html
```

20
docs/_src/tutorials/tutorials/3.md
View File

@ -39,8 +39,6 @@ Make sure you enable the GPU runtime to experience decent speed in this tutorial
# Install the latest master of Haystack
!pip install git+https://github.com/deepset-ai/haystack.git
!pip install urllib3==1.25.4
!pip install torch==1.6.0+cu101 torchvision==0.6.1+cu101 -f https://download.pytorch.org/whl/torch_stable.html
```
@ -153,13 +151,17 @@ reader = FARMReader(model_name_or_path="deepset/roberta-base-squad2", use_gpu=Tr
# reader = TransformersReader(model_name_or_path="distilbert-base-uncased-distilled-squad", tokenizer="distilbert-base-uncased", use_gpu=-1)
```
### Finder
### Pipeline
The Finder sticks together reader and retriever in a pipeline to answer our actual questions.
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 `ExtractiveQAPipeline` that combines a retriever and a reader to answer our questions.
You can learn more about `Pipelines` in the [docs](https://haystack.deepset.ai/docs/latest/pipelinesmd).
```python
finder = Finder(reader, retriever)
from haystack.pipeline import ExtractiveQAPipeline
pipe = ExtractiveQAPipeline(reader, retriever)
```
## Voilà! Ask a question!
@ -167,14 +169,14 @@ finder = Finder(reader, retriever)
```python
# You can configure how many candidates the reader and retriever shall return
# The higher top_k_retriever, the better (but also the slower) your answers.
prediction = finder.get_answers(question="Who is the father of Arya Stark?", top_k_retriever=10, top_k_reader=5)
# The higher top_k_retriever, the better (but also the slower) your answers.
prediction = pipe.run(query="Who is the father of Arya Stark?", top_k_retriever=10, top_k_reader=5)
```
```python
# prediction = finder.get_answers(question="Who created the Dothraki vocabulary?", top_k_reader=5)
# prediction = finder.get_answers(question="Who is the sister of Sansa?", 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)
```

19
docs/_src/tutorials/tutorials/4.md
View File

@ -47,8 +47,6 @@ Make sure you enable the GPU runtime to experience decent speed in this tutorial
# Install the latest master of Haystack
!pip install git+https://github.com/deepset-ai/haystack.git
!pip install urllib3==1.25.4
!pip install torch==1.6.0+cu101 torchvision==0.6.1+cu101 -f https://download.pytorch.org/whl/torch_stable.html
```
@ -136,7 +134,7 @@ print(df.head())
# Get embeddings for our questions from the FAQs
questions = list(df["question"].values)
df["question_emb"] = retriever.embed_queries(texts=questions)
df = df.rename(columns={"answer": "text"})
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")
@ -144,11 +142,18 @@ document_store.write_documents(docs_to_index)
```
### Ask questions
Initialize a Finder (this time without a reader) and ask questions
Initialize a Pipeline (this time without a reader) and ask questions
```python
finder = Finder(reader=None, retriever=retriever)
prediction = finder.get_answers_via_similar_questions(question="How is the virus spreading?", top_k_retriever=10)
print_answers(prediction, details="all")
from haystack.pipeline import FAQPipeline
pipe = FAQPipeline(retriever=retriever)
```
```python
prediction = pipe.run(query="How is the virus spreading?", top_k_retriever=10)
print_answers(prediction, details="all")
```

3
docs/_src/tutorials/tutorials/5.md
View File

@ -38,8 +38,6 @@ You can start Elasticsearch on your local machine instance using Docker. If Dock
# Install the latest master of Haystack
!pip install git+https://github.com/deepset-ai/haystack.git
!pip install urllib3==1.25.4
!pip install torch==1.6.0+cu101 torchvision==0.6.1+cu101 -f https://download.pytorch.org/whl/torch_stable.html
```
@ -177,7 +175,6 @@ print("Reader F1-Score:", reader_eval_results["f1"])
```python
# Evaluate combination of Reader and Retriever through Finder
# Evaluate combination of Reader and Retriever through Finder
finder_eval_results = finder.eval(top_k_retriever=1, top_k_reader=10, label_index=label_index, doc_index=doc_index)
finder.print_eval_results(finder_eval_results)
```

24
docs/_src/tutorials/tutorials/6.md
View File

@ -80,8 +80,6 @@ Make sure you enable the GPU runtime to experience decent speed in this tutorial
# Install the latest master of Haystack
!pip install git+https://github.com/deepset-ai/haystack.git
!pip install urllib3==1.25.4
!pip install torch==1.6.0+cu101 torchvision==0.6.1+cu101 -f https://download.pytorch.org/whl/torch_stable.html
```
@ -179,33 +177,31 @@ Here we use a FARMReader with the *deepset/roberta-base-squad2* model (see: http
reader = FARMReader(model_name_or_path="deepset/roberta-base-squad2", use_gpu=True)
```
#### Finder
### Pipeline
The Finder sticks together reader and retriever in a pipeline to answer our actual questions.
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 `ExtractiveQAPipeline` that combines a retriever and a reader to answer our questions.
You can learn more about `Pipelines` in the [docs](https://haystack.deepset.ai/docs/latest/pipelinesmd).
```python
finder = Finder(reader, retriever)
from haystack.pipeline import ExtractiveQAPipeline
pipe = ExtractiveQAPipeline(reader, retriever)
```
### Voilà! Ask a question!
## Voilà! Ask a question!
```python
# You can configure how many candidates the reader and retriever shall return
# The higher top_k_retriever, the better (but also the slower) your answers.
prediction = finder.get_answers(question="Who created the Dothraki vocabulary?", top_k_retriever=10, top_k_reader=5)
#prediction = finder.get_answers(question="Who is the father of Arya Stark?", top_k_retriever=10, top_k_reader=5)
#prediction = finder.get_answers(question="Who is the sister of Sansa?", top_k_retriever=10, top_k_reader=5)
# The higher top_k_retriever, the better (but also the slower) your answers.
prediction = pipe.run(query="Who created the Dothraki vocabulary?", top_k_retriever=10, top_k_reader=5)
```
```python
print_answers(prediction, details="minimal")
```
```python
```

13
docs/_src/tutorials/tutorials/7.md
View File

@ -37,8 +37,6 @@ Here are the packages and imports that we'll need:
```python
!pip install git+https://github.com/deepset-ai/haystack.git
!pip install urllib3==1.25.4
!pip install torch==1.6.0+cu101 torchvision==0.6.1+cu101 -f https://download.pytorch.org/whl/torch_stable.html
```
@ -188,3 +186,14 @@ for question in QUESTIONS:
answers = predicted_result["answers"]
print(f'Generated answer is \'{answers[0]["answer"]}\' for the question = \'{question}\'')
```
```python
# Or alternatively use the Pipeline class
from haystack.pipeline import GenerativeQAPipeline
pipe = GenerativeQAPipeline(generator=generator, retriever=retriever)
for question in QUESTIONS:
res = pipe.run(query=question, top_k_generator=1, top_k_retriever=5)
print(res)
```

5
docs/_src/tutorials/tutorials/8.md
View File

@ -36,9 +36,8 @@ This tutorial will show you all the tools that Haystack provides to help you cas
# Install the latest master of Haystack
!pip install git+https://github.com/deepset-ai/haystack.git
!pip install torch==1.6.0+cu101 torchvision==0.6.1+cu101 -f https://download.pytorch.org/whl/torch_stable.html
!wget --no-check-certificate https://dl.xpdfreader.com/xpdf-tools-linux-4.02.tar.gz
!tar -xvf xpdf-tools-linux-4.02.tar.gz && sudo cp xpdf-tools-linux-4.02/bin64/pdftotext /usr/local/bin
!wget --no-check-certificate https://dl.xpdfreader.com/xpdf-tools-linux-4.03.tar.gz
!tar -xvf xpdf-tools-linux-4.03.tar.gz && sudo cp xpdf-tools-linux-4.03/bin64/pdftotext /usr/local/bin
```

1
docs/_src/tutorials/tutorials/9.md
View File

@ -21,7 +21,6 @@ This tutorial will guide you through the steps required to create a retriever th
# Install the latest master of Haystack
!pip install git+https://github.com/deepset-ai/haystack.git
!pip install torch==1.6.0+cu101 torchvision==0.6.1+cu101 -f https://download.pytorch.org/whl/torch_stable.html
```

3
haystack/pipeline.py
View File

@ -434,7 +434,7 @@ class FAQPipeline(BaseStandardPipeline):
results: Dict = {"query": query, "answers": []}
for doc in documents:
# TODO proper calibratation of pseudo probabilities
# TODO proper calibration of pseudo probabilities
cur_answer = {
"query": doc.text,
"answer": doc.meta["answer"],
@ -448,7 +448,6 @@ class FAQPipeline(BaseStandardPipeline):
}
results["answers"].append(cur_answer)
return results

2
haystack/retriever/squad_to_dpr.py
View File

@ -98,7 +98,7 @@ class HaystackDocumentStore:
if not es.ping():
logging.info("Starting Elasticsearch ...")
status = subprocess.run(
['docker run -d -p 9200:9200 -e "discovery.type=single-node" elasticsearch:7.6.2'], shell=True
['docker run -d -p 9200:9200 -e "discovery.type=single-node" elasticsearch:7.9.2'], shell=True
)
if status.returncode:
raise Exception(

2
run_docker_gpu.sh
View File

@ -6,7 +6,7 @@
#
# To use GPU with Docker, ensure nvidia-docker(https://github.com/NVIDIA/nvidia-docker) is installed.
docker run -d -p 9200:9200 -e "discovery.type=single-node" elasticsearch:7.6.1
docker run -d -p 9200:9200 -e "discovery.type=single-node" elasticsearch:7.9.2
# alternative: for a demo you can also use this elasticsearch image with already indexed GoT articles
#docker run -d -p 9200:9200 -e "discovery.type=single-node" deepset/elasticsearch-game-of-thrones

30
tutorials/Tutorial1_Basic_QA_Pipeline.ipynb
View File

@ -60,8 +60,7 @@
"\n",
"# Install the latest master of Haystack\n",
"!pip install git+https://github.com/deepset-ai/haystack.git\n",
"!pip install urllib3==1.25.4\n",
"!pip install torch==1.6.0+cu101 torchvision==0.6.1+cu101 -f https://download.pytorch.org/whl/torch_stable.html\n"
"!pip install urllib3==1.25.4\n"
]
},
{
@ -70,7 +69,6 @@
"metadata": {},
"outputs": [],
"source": [
"from haystack import Finder\n",
"from haystack.preprocessor.cleaning import clean_wiki_text\n",
"from haystack.preprocessor.utils import convert_files_to_dicts, fetch_archive_from_http\n",
"from haystack.reader.farm import FARMReader\n",
@ -111,7 +109,7 @@
],
"source": [
"# Recommended: Start Elasticsearch using Docker\n",
"#! docker run -d -p 9200:9200 -e \"discovery.type=single-node\" elasticsearch:7.6.2"
"#! docker run -d -p 9200:9200 -e \"discovery.type=single-node\" elasticsearch:7.9.2"
]
},
{
@ -121,13 +119,13 @@
"outputs": [],
"source": [
"# In Colab / No Docker environments: Start Elasticsearch from source\n",
"! wget https://artifacts.elastic.co/downloads/elasticsearch/elasticsearch-7.6.2-linux-x86_64.tar.gz -q\n",
"! tar -xzf elasticsearch-7.6.2-linux-x86_64.tar.gz\n",
"! chown -R daemon:daemon elasticsearch-7.6.2\n",
"! wget https://artifacts.elastic.co/downloads/elasticsearch/elasticsearch-7.9.2-linux-x86_64.tar.gz -q\n",
"! tar -xzf elasticsearch-7.9.2-linux-x86_64.tar.gz\n",
"! chown -R daemon:daemon elasticsearch-7.9.2\n",
"\n",
"import os\n",
"from subprocess import Popen, PIPE, STDOUT\n",
"es_server = Popen(['elasticsearch-7.6.2/bin/elasticsearch'],\n",
"es_server = Popen(['elasticsearch-7.9.2/bin/elasticsearch'],\n",
" stdout=PIPE, stderr=STDOUT,\n",
" preexec_fn=lambda: os.setuid(1) # as daemon\n",
" )\n",
@ -359,9 +357,12 @@
"cell_type": "markdown",
"metadata": {},
"source": [
"### Finder\n",
"### Pipeline\n",
"\n",
"The Finder sticks together reader and retriever in a pipeline to answer our actual questions. "
"With a Haystack `Pipeline` you can stick together your building blocks to a search pipeline.\n",
"Under the hood, `Pipelines` are Directed Acyclic Graphs (DAGs) that you can easily customize for your own use cases.\n",
"To speed things up, Haystack also comes with a few predefined Pipelines. One of them is the `ExtractiveQAPipeline` that combines a retriever and a reader to answer our questions.\n",
"You can learn more about `Pipelines` in the [docs](https://haystack.deepset.ai/docs/latest/pipelinesmd)."
]
},
{
@ -374,7 +375,8 @@
},
"outputs": [],
"source": [
"finder = Finder(reader, retriever)"
"from haystack.pipeline import ExtractiveQAPipeline\n",
"pipe = ExtractiveQAPipeline(reader, retriever)"
]
},
{
@ -406,7 +408,7 @@
"source": [
"# You can configure how many candidates the reader and retriever shall return\n",
"# The higher top_k_retriever, the better (but also the slower) your answers. \n",
"prediction = finder.get_answers(question=\"Who is the father of Arya Stark?\", top_k_retriever=10, top_k_reader=5)"
"prediction = pipe.run(query=\"Who is the father of Arya Stark?\", top_k_retriever=10, top_k_reader=5)"
]
},
{
@ -415,8 +417,8 @@
"metadata": {},
"outputs": [],
"source": [
"# prediction = finder.get_answers(question=\"Who created the Dothraki vocabulary?\", top_k_reader=5)\n",
"# prediction = finder.get_answers(question=\"Who is the sister of Sansa?\", top_k_reader=5)"
"# prediction = pipe.run(query=\"Who created the Dothraki vocabulary?\", top_k_reader=5)\n",
"# prediction = pipe.run(query=\"Who is the sister of Sansa?\", top_k_reader=5)"
]
},
{

28
tutorials/Tutorial1_Basic_QA_Pipeline.py
View File

@ -48,7 +48,7 @@ def tutorial1_basic_qa_pipeline():
if LAUNCH_ELASTICSEARCH:
logging.info("Starting Elasticsearch ...")
status = subprocess.run(
['docker run -d -p 9200:9200 -e "discovery.type=single-node" elasticsearch:7.6.2'], shell=True
['docker run -d -p 9200:9200 -e "discovery.type=single-node" elasticsearch:7.9.2'], shell=True
)
if status.returncode:
raise Exception("Failed to launch Elasticsearch. If you want to connect to an existing Elasticsearch instance"
@ -138,20 +138,20 @@ def tutorial1_basic_qa_pipeline():
# reader = TransformersReader(
# model_name_or_path="distilbert-base-uncased-distilled-squad", tokenizer="distilbert-base-uncased", use_gpu=-1)
# ### Finder
#
# The Finder sticks together reader and retriever in a pipeline to answer our actual questions.
# ### 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 `ExtractiveQAPipeline` that combines a retriever and a reader to answer our questions.
# You can learn more about `Pipelines` in the [docs](https://haystack.deepset.ai/docs/latest/pipelinesmd).
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)
finder = Finder(reader, retriever)
# ## Voilà! Ask a question!
# You can configure how many candidates the reader and retriever shall return
# The higher top_k_retriever, the better (but also the slower) your answers.
prediction = finder.get_answers(question="Who is the father of Sansa Stark?", top_k_retriever=10, top_k_reader=5)
# prediction = finder.get_answers(question="Who created the Dothraki vocabulary?", top_k_reader=5)
# prediction = finder.get_answers(question="Who is the sister of Sansa?", 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")

3
tutorials/Tutorial2_Finetune_a_model_on_your_data.ipynb
View File

@ -56,8 +56,7 @@
"\n",
"# Install the latest master of Haystack\n",
"!pip install git+https://github.com/deepset-ai/haystack.git\n",
"!pip install urllib3==1.25.4\n",
"!pip install torch==1.6.0+cu101 torchvision==0.6.1+cu101 -f https://download.pytorch.org/whl/torch_stable.html\n"
"!pip install urllib3==1.25.4"
]
},
{

21
tutorials/Tutorial3_Basic_QA_Pipeline_without_Elasticsearch.ipynb
View File

@ -56,8 +56,7 @@
"\n",
"# Install the latest master of Haystack\n",
"!pip install git+https://github.com/deepset-ai/haystack.git\n",
"!pip install urllib3==1.25.4\n",
"!pip install torch==1.6.0+cu101 torchvision==0.6.1+cu101 -f https://download.pytorch.org/whl/torch_stable.html\n"
"!pip install urllib3==1.25.4"
]
},
{
@ -278,9 +277,12 @@
"cell_type": "markdown",
"metadata": {},
"source": [
"### Finder\n",
"### Pipeline\n",
"\n",
"The Finder sticks together reader and retriever in a pipeline to answer our actual questions. "
"With a Haystack `Pipeline` you can stick together your building blocks to a search pipeline.\n",
"Under the hood, `Pipelines` are Directed Acyclic Graphs (DAGs) that you can easily customize for your own use cases.\n",
"To speed things up, Haystack also comes with a few predefined Pipelines. One of them is the `ExtractiveQAPipeline` that combines a retriever and a reader to answer our questions.\n",
"You can learn more about `Pipelines` in the [docs](https://haystack.deepset.ai/docs/latest/pipelinesmd)."
]
},
{
@ -293,7 +295,8 @@
},
"outputs": [],
"source": [
"finder = Finder(reader, retriever)"
"from haystack.pipeline import ExtractiveQAPipeline\n",
"pipe = ExtractiveQAPipeline(reader, retriever)"
]
},
{
@ -340,8 +343,8 @@
],
"source": [
"# You can configure how many candidates the reader and retriever shall return\n",
"# The higher top_k_retriever, the better (but also the slower) your answers. \n",
"prediction = finder.get_answers(question=\"Who is the father of Arya Stark?\", top_k_retriever=10, top_k_reader=5)"
"# The higher top_k_retriever, the better (but also the slower) your answers.\n",
"prediction = pipe.run(query=\"Who is the father of Arya Stark?\", top_k_retriever=10, top_k_reader=5)"
]
},
{
@ -350,8 +353,8 @@
"metadata": {},
"outputs": [],
"source": [
"# prediction = finder.get_answers(question=\"Who created the Dothraki vocabulary?\", top_k_reader=5)\n",
"# prediction = finder.get_answers(question=\"Who is the sister of Sansa?\", top_k_reader=5)"
"# prediction = pipe.run(query=\"Who created the Dothraki vocabulary?\", top_k_reader=5)\n",
"# prediction = pipe.run(query=\"Who is the sister of Sansa?\", top_k_reader=5)"
]
},
{

20
tutorials/Tutorial3_Basic_QA_Pipeline_without_Elasticsearch.py
View File

@ -91,18 +91,20 @@ def tutorial3_basic_qa_pipeline_without_elasticsearch():
# Alternative:
# reader = TransformersReader(model_name_or_path="distilbert-base-uncased-distilled-squad", tokenizer="distilbert-base-uncased", use_gpu=-1)
# ### Finder
# ### Pipeline
#
# The Finder sticks together reader and retriever in a pipeline to answer our actual questions.
finder = Finder(reader, retriever)
# 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 `ExtractiveQAPipeline` that combines a retriever and a reader to answer our questions.
# You can learn more about `Pipelines` in the [docs](https://haystack.deepset.ai/docs/latest/pipelinesmd).
from haystack.pipeline import ExtractiveQAPipeline
pipe = ExtractiveQAPipeline(reader, retriever)
# ## Voilà! Ask a question!
## Voilà! Ask a question!
prediction = pipe.run(query="Who is the father of Arya Stark?", top_k_retriever=10, top_k_reader=5)
# You can configure how many candidates the reader and retriever shall return
# The higher top_k_retriever, the better (but also the slower) your answers.
prediction = finder.get_answers(question="Who is the father of Arya Stark?", top_k_retriever=10, top_k_reader=5)
# prediction = finder.get_answers(question="Who created the Dothraki vocabulary?", top_k_reader=5)
# prediction = finder.get_answers(question="Who is the sister of Sansa?", 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")

28
tutorials/Tutorial4_FAQ_style_QA.ipynb
View File

@ -64,8 +64,7 @@
"\n",
"# Install the latest master of Haystack\n",
"!pip install git+https://github.com/deepset-ai/haystack.git\n",
"!pip install urllib3==1.25.4\n",
"!pip install torch==1.6.0+cu101 torchvision==0.6.1+cu101 -f https://download.pytorch.org/whl/torch_stable.html\n"
"!pip install urllib3==1.25.4"
]
},
{
@ -221,7 +220,7 @@
"# Get embeddings for our questions from the FAQs\n",
"questions = list(df[\"question\"].values)\n",
"df[\"question_emb\"] = retriever.embed_queries(texts=questions)\n",
"df = df.rename(columns={\"answer\": \"text\"})\n",
"df = df.rename(columns={\"question\": \"text\"})\n",
"\n",
"# Convert Dataframe to list of dicts and index them in our DocumentStore\n",
"docs_to_index = df.to_dict(orient=\"records\")\n",
@ -238,7 +237,7 @@
"cell_type": "markdown",
"source": [
"### Ask questions\n",
"Initialize a Finder (this time without a reader) and ask questions"
"Initialize a Pipeline (this time without a reader) and ask questions"
],
"metadata": {
"collapsed": false
@ -249,9 +248,8 @@
"execution_count": null,
"outputs": [],
"source": [
"finder = Finder(reader=None, retriever=retriever)\n",
"prediction = finder.get_answers_via_similar_questions(question=\"How is the virus spreading?\", top_k_retriever=10)\n",
"print_answers(prediction, details=\"all\")"
"from haystack.pipeline import FAQPipeline\n",
"pipe = FAQPipeline(retriever=retriever)"
],
"metadata": {
"collapsed": false,
@ -259,6 +257,22 @@
"name": "#%%\n"
}
}
},
{
"cell_type": "code",
"source": [
"prediction = pipe.run(query=\"How is the virus spreading?\", top_k_retriever=10)\n",
"print_answers(prediction, details=\"all\")\n",
"\n"
],
"metadata": {
"collapsed": false,
"pycharm": {
"name": "#%%\n"
}
},
"execution_count": null,
"outputs": []
}
],
"metadata": {

14
tutorials/Tutorial4_FAQ_style_QA.py
View File

@ -24,7 +24,7 @@ def tutorial4_faq_style_qa():
# - Generalizability: We can only answer questions that are similar to existing ones in FAQ
#
# In some use cases, a combination of extractive QA and FAQ-style can also be an interesting option.
LAUNCH_ELASTICSEARCH=True
LAUNCH_ELASTICSEARCH=False
if LAUNCH_ELASTICSEARCH:
logging.info("Starting Elasticsearch ...")
@ -34,7 +34,7 @@ def tutorial4_faq_style_qa():
if status.returncode:
raise Exception("Failed to launch Elasticsearch. If you want to connect to an existing Elasticsearch instance"
"then set LAUNCH_ELASTICSEARCH in the script to False.")
time.sleep(15)
time.sleep(30)
### Init the DocumentStore
# In contrast to Tutorial 1 (extractive QA), we:
@ -71,16 +71,18 @@ def tutorial4_faq_style_qa():
# Get embeddings for our questions from the FAQs
questions = list(df["question"].values)
df["question_emb"] = retriever.embed_queries(texts=questions)
df = df.rename(columns={"answer": "text"})
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.write_documents(docs_to_index)
# Initialize a Pipeline (this time without a reader) and ask questions
# Init reader & and use Finder to get answer (same as in Tutorial 1)
finder = Finder(reader=None, retriever=retriever)
prediction = finder.get_answers_via_similar_questions(question="How is the virus spreading?", top_k_retriever=10)
from haystack.pipeline import FAQPipeline
pipe = FAQPipeline(retriever=retriever)
prediction = pipe.run(query="How is the virus spreading?", top_k_retriever=10)
print_answers(prediction, details="all")

4
tutorials/Tutorial5_Evaluation.ipynb
View File

@ -77,8 +77,7 @@
"\n",
"# Install the latest master of Haystack\n",
"!pip install git+https://github.com/deepset-ai/haystack.git\n",
"!pip install urllib3==1.25.4\n",
"!pip install torch==1.6.0+cu101 torchvision==0.6.1+cu101 -f https://download.pytorch.org/whl/torch_stable.html\n"
"!pip install urllib3==1.25.4"
]
},
{
@ -469,7 +468,6 @@
},
"outputs": [],
"source": [
"# Evaluate combination of Reader and Retriever through Finder\n",
"# Evaluate combination of Reader and Retriever through Finder\n",
"finder_eval_results = finder.eval(top_k_retriever=1, top_k_reader=10, label_index=label_index, doc_index=doc_index)\n",
"finder.print_eval_results(finder_eval_results)"

81
tutorials/Tutorial6_Better_Retrieval_via_DPR.ipynb
View File

@ -286,8 +286,7 @@
"\n",
"# Install the latest master of Haystack\n",
"!pip install git+https://github.com/deepset-ai/haystack.git\n",
"!pip install urllib3==1.25.4\n",
"!pip install torch==1.6.0+cu101 torchvision==0.6.1+cu101 -f https://download.pytorch.org/whl/torch_stable.html\n"
"!pip install urllib3==1.25.4"
]
},
{
@ -647,9 +646,12 @@
"id": "unhLD18yA6OF"
},
"source": [
"#### Finder\n",
"### Pipeline\n",
"\n",
"The Finder sticks together reader and retriever in a pipeline to answer our actual questions. "
"With a Haystack `Pipeline` you can stick together your building blocks to a search pipeline.\n",
"Under the hood, `Pipelines` are Directed Acyclic Graphs (DAGs) that you can easily customize for your own use cases.\n",
"To speed things up, Haystack also comes with a few predefined Pipelines. One of them is the `ExtractiveQAPipeline` that combines a retriever and a reader to answer our questions.\n",
"You can learn more about `Pipelines` in the [docs](https://haystack.deepset.ai/docs/latest/pipelinesmd)."
]
},
{
@ -662,7 +664,8 @@
},
"outputs": [],
"source": [
"finder = Finder(reader, retriever)"
"from haystack.pipeline import ExtractiveQAPipeline\n",
"pipe = ExtractiveQAPipeline(reader, retriever)"
]
},
{
@ -672,7 +675,7 @@
"id": "bXlBBxKXA6OL"
},
"source": [
"### Voilà! Ask a question!"
"## Voilà! Ask a question!"
]
},
{
@ -712,65 +715,8 @@
],
"source": [
"# You can configure how many candidates the reader and retriever shall return\n",
"# The higher top_k_retriever, the better (but also the slower) your answers. \n",
"prediction = finder.get_answers(question=\"Who created the Dothraki vocabulary?\", top_k_retriever=10, top_k_reader=5)\n",
"\n",
"#prediction = finder.get_answers(question=\"Who is the father of Arya Stark?\", top_k_retriever=10, top_k_reader=5)\n",
"#prediction = finder.get_answers(question=\"Who is the sister of Sansa?\", top_k_retriever=10, top_k_reader=5)"
]
},
{
"cell_type": "code",
"execution_count": 9,
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/",
"height": 561
},
"colab_type": "code",
"id": "N70FgfkwA6OQ",
"outputId": "9419c75d-181c-4ef6-cea8-b328a503f19a",
"pycharm": {
"name": "#%%\n"
}
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"[ { 'answer': 'David J. Peterson',\n",
" 'context': '\\n'\n",
" '===Valyrian===\\n'\n",
" 'David J. Peterson, who created the Dothraki language for '\n",
" 'the first season of the show, was entrusted by the '\n",
" 'producers to design a new '},\n",
" { 'answer': 'David Peterson',\n",
" 'context': '\\n'\n",
" '==Phonology and romanization==\\n'\n",
" 'David Peterson has said, \"You know, most people probably '\n",
" \"don't really know what Arabic actually sounds like, so to \"\n",
" 'an '},\n",
" { 'answer': 'books',\n",
" 'context': 'ints. First, the language had to match the uses already '\n",
" 'put down in the books. Secondly, it had to be easily '\n",
" 'pronounceable or learnable by the actors'},\n",
" { 'answer': \"'''Nevakhi vekha ha maan: Rekke, m'aresakea norethi fitte.'\",\n",
" 'context': '\\n'\n",
" '==Sample==\\n'\n",
" \": '''Nevakhi vekha ha maan: Rekke, m'aresakea norethi \"\n",
" \"fitte.'''\\n\"\n",
" ': seat. exist. for there. with.coward. hair. short\\n'\n",
" \": ''There is a place f\"},\n",
" { 'answer': 'Tyrion',\n",
" 'context': 'ding, as well as his nephew Joffrey, the new king, as '\n",
" 'civil war begins. Tyrion struggles to strengthen and '\n",
" 'protect the city and family who hate him an'}]\n"
]
}
],
"source": [
"print_answers(prediction, details=\"minimal\")"
"# The higher top_k_retriever, the better (but also the slower) your answers.\n",
"prediction = pipe.run(query=\"Who created the Dothraki vocabulary?\", top_k_retriever=10, top_k_reader=5)"
]
},
{
@ -778,7 +724,10 @@
"execution_count": null,
"metadata": {},
"outputs": [],
"source": []
"source": [
"print_answers(prediction, details=\"minimal\")\n",
"\n"
]
}
],
"metadata": {

17
tutorials/Tutorial6_Better_Retrieval_via_DPR.py
View File

@ -52,18 +52,15 @@ def tutorial6_better_retrieval_via_dpr():
# Hugging Face's model hub (https://huggingface.co/models)
reader = FARMReader(model_name_or_path="deepset/roberta-base-squad2", use_gpu=True)
### Finder
# The Finder sticks together reader and retriever in a pipeline to answer our actual questions.
finder = Finder(reader, retriever)
### Pipeline
from haystack.pipeline import ExtractiveQAPipeline
pipe = ExtractiveQAPipeline(reader, retriever)
### Voilà! Ask a question!
# You can configure how many candidates the reader and retriever shall return
# The higher top_k_retriever, the better (but also the slower) your answers.
prediction = finder.get_answers(question="Who is the father of Arya Stark?", top_k_retriever=10, top_k_reader=5)
## Voilà! Ask a question!
prediction = pipe.run(query="Who is the father of Arya Stark?", top_k_retriever=10, top_k_reader=5)
# prediction = finder.get_answers(question="Who created the Dothraki vocabulary?", top_k_reader=5)
# prediction = finder.get_answers(question="Who is the sister of Sansa?", 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")

23
tutorials/Tutorial7_RAG_Generator.ipynb
View File

@ -62,8 +62,7 @@
"outputs": [],
"source": [
"!pip install git+https://github.com/deepset-ai/haystack.git\n",
"!pip install urllib3==1.25.4\n",
"!pip install torch==1.6.0+cu101 torchvision==0.6.1+cu101 -f https://download.pytorch.org/whl/torch_stable.html\n"
"!pip install urllib3==1.25.4"
],
"metadata": {
"collapsed": false,
@ -322,6 +321,26 @@
"name": "#%%\n"
}
}
},
{
"cell_type": "code",
"execution_count": null,
"outputs": [],
"source": [
"# Or alternatively use the Pipeline class\n",
"from haystack.pipeline import GenerativeQAPipeline\n",
"\n",
"pipe = GenerativeQAPipeline(generator=generator, retriever=retriever)\n",
"for question in QUESTIONS:\n",
" res = pipe.run(query=question, top_k_generator=1, top_k_retriever=5)\n",
" print(res)"
],
"metadata": {
"collapsed": false,
"pycharm": {
"name": "#%%\n"
}
}
}
],
"metadata": {

6
tutorials/Tutorial7_RAG_Generator.py
View File

@ -111,6 +111,12 @@ def tutorial7_rag_generator():
answers = predicted_result["answers"]
print(f'Generated answer is \'{answers[0]["answer"]}\' for the question = \'{question}\'')
# Or alternatively use the Pipeline class
from haystack.pipeline import GenerativeQAPipeline
pipe = GenerativeQAPipeline(generator=generator, retriever=retriever)
for question in QUESTIONS:
res = pipe.run(query=question, top_k_generator=1, top_k_retriever=5)
print(res)
if __name__ == "__main__":
tutorial7_rag_generator()

14
tutorials/Tutorial8_Preprocessing.ipynb
View File

@ -59,9 +59,8 @@
"\n",
"# Install the latest master of Haystack\n",
"!pip install git+https://github.com/deepset-ai/haystack.git\n",
"!pip install torch==1.6.0+cu101 torchvision==0.6.1+cu101 -f https://download.pytorch.org/whl/torch_stable.html\n",
"!wget --no-check-certificate https://dl.xpdfreader.com/xpdf-tools-linux-4.02.tar.gz\n",
"!tar -xvf xpdf-tools-linux-4.02.tar.gz && sudo cp xpdf-tools-linux-4.02/bin64/pdftotext /usr/local/bin"
"!wget --no-check-certificate https://dl.xpdfreader.com/xpdf-tools-linux-4.03.tar.gz\n",
"!tar -xvf xpdf-tools-linux-4.03.tar.gz && sudo cp xpdf-tools-linux-4.03/bin64/pdftotext /usr/local/bin"
],
"metadata": {
"collapsed": false,
@ -497,15 +496,6 @@
"nbconvert_exporter": "python",
"pygments_lexer": "ipython2",
"version": "2.7.6"
},
"pycharm": {
"stem_cell": {
"cell_type": "raw",
"source": [],
"metadata": {
"collapsed": false
}
}
}
},
"nbformat": 4,

12
tutorials/Tutorial9_DPR_training.ipynb
View File

@ -26,8 +26,7 @@
"#! pip install farm-haystack\n",
"\n",
"# Install the latest master of Haystack\n",
"!pip install git+https://github.com/deepset-ai/haystack.git\n",
"!pip install torch==1.6.0+cu101 torchvision==0.6.1+cu101 -f https://download.pytorch.org/whl/torch_stable.html"
"!pip install git+https://github.com/deepset-ai/haystack.git"
],
"metadata": {
"collapsed": false,
@ -403,15 +402,6 @@
"nbconvert_exporter": "python",
"pygments_lexer": "ipython2",
"version": "2.7.6"
},
"pycharm": {
"stem_cell": {
"cell_type": "raw",
"source": [],
"metadata": {
"collapsed": false
}
}
}
},
"nbformat": 4,