MichelBartels e80771f839
Adding yaml functionality to standard pipelines (save/load...) (#1735)
* adding yaml functionality to BaseStandardPipeline

fixes #1681

* Add latest docstring and tutorial changes

* Update API Reference Pages for v1.0 (#1729)

* Create new API pages and update existing ones

* Create query classifier page

* Remove Objects suffix

* Change answer aggregation key to doc_id, query instead of label_id, query (#1726)

* Add debugging example to tutorial (#1731)

* Add debugging example to tutorial

* Add latest docstring and tutorial changes

* Remove Objects suffix

* Add latest docstring and tutorial changes

* Revert "Remove Objects suffix"

This reverts commit 6681cb06510b080775994effe6a50bae42254be4.

* Revert unintentional commit

* Add third debugging option

* Add latest docstring and tutorial changes

Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>

* Fix another self.device/s typo (#1734)

* Fix yet another self.device(s) typo

* Add typing to 'initialize_device_settings' to try prevent future issues

* Fix bug in Tutorial5

* Fix the same bug in the notebook

Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>

* added test for saving and loading prebuilt pipelines

* fixed typo, changed variable name and added comments

* Add latest docstring and tutorial changes

* Fix a few details of some tutorials (#1733)

* Make Tutorial10 use print instead of logs and fix a typo in Tutoria15

* Add a type check in 'print_answers'

* Add same checks to print_documents and print_questions

* Make RAGenerator return Answers instead of dictionaries

* Fix RAGenerator tests

Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>

* Fix `print_answers` (#1743)

* Fix a specific path of print_answers that was assuming answers are dictionaries

Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>

* Split pipeline tests into three suites (#1755)

* Split pipeline tests into three suites

* Will this trigger the CI?

* Rename duplicate test into test_most_similar_documents_pipeline

* Fixing a bug that was probably never noticed

* Capitalize starting letter in params (#1750)

* Capitalize starting letter in params

Capitalized the starting letter in code examples for params in keeping with the latest names for nodes where first letter is capitalized. 
Refer: https://github.com/deepset-ai/haystack/issues/1748

* Update standard_pipelines.py

Capitalized some starting letters in the docstrings in keeping with the updated node names for standard pipelines

* Multi query eval (#1746)

* add eval() to pipeline

* Add latest docstring and tutorial changes

* support multiple queries in eval()

* Add latest docstring and tutorial changes

* keep single query test

* fix EvaluationResult node_results default

* adjust docstrings

* Add latest docstring and tutorial changes

* minor improvements from comments

* Add latest docstring and tutorial changes

* move EvaluationResult and calculate_metrics to schema

* Add latest docstring and tutorial changes

Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>

* Split summarizer tests in order to make windows CI work again (#1757)

* separate testfile for summarizer with translation

* Add latest docstring and tutorial changes

* import SPLIT_DOCS from test_summarizer

* add workflow_dispatch to windows_ci

* add worflow_dispatch to linux_ci

Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>

* fix import of EvaluationResult in test case

* exclude test_summarizer_translation.py for windows_ci (#1759)

* Pipelines now tolerate custom _debug content (#1756)

* Pipelines now tolerate custom _debug content

* Support Tables in all DocumentStores (#1744)

* Add support for tables in SQLDocumentStore, FAISSDocumentStore and MilvuDocumentStore

* Add support for WeaviateDocumentStore

* Make sure that embedded meta fields are strings + add embedding_dim to WeaviateDocStore in test config

* Add latest docstring and tutorial changes

* Represent tables in WeaviateDocumentStore as nested lists

* Fix mypy

Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>

* Allow TableReader models without aggregation classifier (#1772)

* Fix usage of filters in `/query` endpoint in REST API (#1774)

* WIP filter refactoring

* fix filter formatting

* remove inplace modification of filters

* Public demo (#1747)

* Queries now run only when pressing RUN. File upload hidden. Question is not sent if the textbox is empty.

* Add latest docstring and tutorial changes

* Tidy up: remove needless state, add comments, fix minor bugs

* Had to add results to the status to avoid some bugs in eval mode

* Added 'credits'

* Add footers, update requirements, some random questions for the evaluation

* Add requested changes

* Temporary rollback the UI to the old GoT dataset

Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>

* Facilitate concurrent query / indexing in Elasticsearch with dense retrievers (new `skip_missing_embeddings` param) (#1762)

* Filtering records not having embeddings

* Added support for skip_missing_embeddings Flag. Default behavior is throw error when embeddings are missing. If skip_missing_embeddings=True then documents without embeddings are ignored for vector similarity

* Fix for below error:
haystack/document_stores/elasticsearch.py:852: error: Need type annotation for "script_score_query"

* docstring for skip_missing_embeddings parameter

* Raise exception where no documents with embeddings is found for Embedding retriever.

* Default skip_missing_embeddings to True

* Explicitly check if embeddings are present if no results are returned by EmbeddingRetriever for Elasticsearch

* Added test case for based on Julian's input

* Added test case for based on Julian's input. Fix pytest error on the testcase

* Added test case for based on Julian's input. Fix pytest error on the testcase

* Added test case for based on Julian's input. Fix pytest error on the testcase

* Simplify code by using get_embed_count

* Adjust docstring & error msg slightly

* Revert error msg

Co-authored-by: Malte Pietsch <malte.pietsch@deepset.ai>

* Huggingface private model support via API tokens (FARMReader) (#1775)

* passed kwargs to model loading

* Pass Auth token explicitly

* add use_auth_token to get_language_model_class

* added use_auth_token parameter at FARMReader

* Add latest docstring and tutorial changes

* added docs for parameter `use_auth_token`

* Add latest docstring and tutorial changes

* adding docs link

* Add latest docstring and tutorial changes

Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>

* private hugging face models for retrievers (#1785)

* private dpr

* Add latest docstring and tutorial changes

* added parameters to child functions

* Add latest docstring and tutorial changes

* added tableextractor

* Add latest docstring and tutorial changes

Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>

* ignore empty filters parameter (#1783)

* ignore empty filters parameter

* Add latest docstring and tutorial changes

Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>

* initialize doc store with doc and label index in tutorial 5 (#1730)

* initialize doc store with doc and label index

* change ipynb according to py for tutorial 5

* Add latest docstring and tutorial changes

Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>

* Small fixes to the public demo (#1781)

* Make strealit tolerant to haystack not knowing its version, and adding special error for docstore issues

* Add workaround for a Streamlit bug

* Make default filters value an empty dict

* Return more context for each answer in the rest api

* Make the hs_version call not-blocking by adding a very quick timeout

* Add disclaimer on low confidence answer

* Use the no-answer feature of the reader to highlight questions with no good answer

* Upgrade torch to v1.10.0 (#1789)

* Upgrade torch to v1.10.0

* Adapt torch version for torch-scatter in TableQA tutorial

* Add latest docstring and tutorial changes

* Make torch version more flexible

Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>

* adding yaml functionality to BaseStandardPipeline

fixes #1681

* Add latest docstring and tutorial changes

* added test for saving and loading prebuilt pipelines

* fixed typo, changed variable name and added comments

* Add latest docstring and tutorial changes

* fix code rendering for example

* Add latest docstring and tutorial changes

Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
Co-authored-by: Branden Chan <33759007+brandenchan@users.noreply.github.com>
Co-authored-by: Julian Risch <julian.risch@deepset.ai>
Co-authored-by: Sara Zan <sara.zanzottera@deepset.ai>
Co-authored-by: nishanthcgit <5066268+nishanthcgit@users.noreply.github.com>
Co-authored-by: tstadel <60758086+tstadel@users.noreply.github.com>
Co-authored-by: bogdankostic <bogdankostic@web.de>
Co-authored-by: Malte Pietsch <malte.pietsch@deepset.ai>
Co-authored-by: C V Goudar <cvgoudar@users.noreply.github.com>
Co-authored-by: Kristof Herrmann <37148029+ArzelaAscoIi@users.noreply.github.com>
2021-11-23 17:01:39 +01:00
..
2021-04-13 09:45:04 +02:00
2021-04-13 09:45:04 +02:00
2021-04-13 09:45:04 +02:00
2021-04-13 09:45:04 +02:00
2020-09-18 12:57:32 +02:00
2020-09-18 12:57:32 +02:00

📒 Looking for the docs?

You find them here here:

https://haystack.deepset.ai/overview/intro

💻 How to update docs?

Overview and Usage

We move the Overview and Usage docs to the haystack-website repository. You will find the docs in the folder docs. Please make sure to only edit the newest version of the docs. We will release the docs together with the Haystack version. We are open for contibutions to our documentation. Please make sure to check our Contribution Guidelines. You will find a step by step introduction to our docs here.

Tutorials

The Tutorials live in the folder tutorials. They are created as colab notebooks which can be used by users to explore new haystack features. To include tutorials into the docs website, markdowns files need to be generated from the notebook. This can be done by running the script /docs/_src/tutorials/tutorials/convert_ipynb.py. Just run python convert_ipynb.py and the script will update all existing notebooks. Furthermore, plaese make sure to update the headers.py file with headers for the new tutorials. These headers are important for the docs website workflow. After the markdown files are generated successfully, you can raise a PR. We will review it and as soons as the markdown file is merged to master, it can be added to our website. Please follow the steps described here under Tutorial & Reference Docs.

API Reference

We use Pydoc-Markdown to create markdown files from the docstrings in our code.

Update docstrings

Execute the following commands in /haystack/docs/_src/api/api:

pip install 'pydoc-markdown==3.11.0'
./generate_docstrings.sh

If you want to generate a new markdown file for a new haystack module, please create a .yml which is inline with the following configuration and a a new line to generate_docstrings.sh for the module. After you ran the generate_docstrings.sh again, there should be a new markdown file for the module. To include it into the docs website, push it to master and follow the steps described here under Tutorial & Reference Docs.

Configuration

Pydoc will read the configuration from a .yml file which is located in the current working directory. Our files contains three main sections:

  • loader: A list of plugins that load API objects from python source files.
    • type: Loader for python source files
    • search_path: Location of source files
    • modules: Module which are used for generating the markdown file
    • ignore_when_discovered: Define which files should be ignored
  • processor: A list of plugins that process API objects to modify their docstrings (e.g. to adapt them from a documentation format to Markdown or to remove items that should not be rendered into the documentation).
    • type: filter: Filter for specific modules
    • documented_only: Only documented API objects
    • do_not_filter_modules: Do not filter module objects
    • skip_empty_modules: Skip modules without content
  • renderer: A plugin that produces the output files.
    • type: Define the renderer which you want to use. We are using the Markdown renderer as it can be configured in very detail.
    • descriptive_class_title: Remove the word "Object" from class titles.
    • descriptive_module_title: Adding the word “Module” before the module name
    • add_method_class_prefix: Add the class name as a prefix to method names
    • add_member_class_prefix: Add the class name as a prefix to member names
    • filename: file name of the generated file