bogdankostic
8c63e295f4
fix: Allow filtering on list fields in InMemoryDocumentStore with all operators ( #5208 )
...
* Add support for list fields
* Unskip tests
2023-06-29 12:10:39 +02:00
Massimiliano Pippi
6373e2ea66
refactor: prepare support to Elasticsearch 8 ( #5226 )
...
* make a package
* Update haystack/document_stores/elasticsearch/es7.py
Co-authored-by: Silvano Cerza <3314350+silvanocerza@users.noreply.github.com>
* do not expose ES types from the package
---------
Co-authored-by: Silvano Cerza <3314350+silvanocerza@users.noreply.github.com>
2023-06-29 11:06:20 +02:00
bogdankostic
ed1bad1155
fix: Use add_isolated_node_eval of eval_batch in run_batch ( #5223 )
...
* Fix isolated node eval in eval_batch
* Add unit test
2023-06-28 16:51:23 +02:00
Vladimir Blagojevic
bc86f57715
feat: BM25 retrieval for MemoryDocumentStore ( #5151 )
2023-06-27 17:42:23 +02:00
Massimiliano Pippi
c068e34954
Remove deprecated param return_table_cell ( #5218 )
...
* remove deprecated param
* Update haystack/nodes/reader/table.py
Co-authored-by: Stefano Fiorucci <44616784+anakin87@users.noreply.github.com>
* try
* remove unused functions and ignore mypy error
---------
Co-authored-by: Stefano Fiorucci <44616784+anakin87@users.noreply.github.com>
2023-06-27 16:14:29 +02:00
ZanSara
462f3a5c99
feat: globally disable progress bars ( #5207 )
...
* add SilenceableTqdm and update usage
* pylint
* rename module
* add tests
2023-06-27 11:45:17 +02:00
Vladimir Blagojevic
5ee393226d
fix: Support all SageMaker HF text generation models (other than Falcon) ( #5205 )
...
* Create SageMaker base class and two implementation subclasses
---------
Co-authored-by: Stefano Fiorucci <44616784+anakin87@users.noreply.github.com>
2023-06-26 19:59:16 +02:00
bogdankostic
82291b56ad
fix: Send batches of query-doc pairs to inference_from_objects ( #5125 )
...
* Send batches of query-doc pairs to inference_from_objects
* Use absolute import path
* Add separate preprocessing_batch_size parameter
2023-06-26 14:26:26 +02:00
Vladimir Blagojevic
eb2255c0dd
Rename SageMakerInvocationLayer -> SageMakerHFTextGenerationInvocationLayer ( #5204 )
2023-06-26 11:03:30 +02:00
Stefano Fiorucci
25d5dedb46
Fix: FARMReader - Consider the max number of labels/answers during training ( #5197 )
...
* first draft
* improve it a bit
* unit tests
* PR review, improved tests
* PR review, improved tests 2
2023-06-26 10:14:21 +02:00
Sebastian
f1932492f1
feat: Add CohereRanker node using Cohere reranking endpoint ( #5152 )
...
* Started to add CohereRanker node
* Small refactoring of SentenceTransformersRanker node
* Started to add predict_batch method
* Simplified predict_batch code
* Added missing imports
* Undoing a change
* Fix mypy
* Adding unit tests using mocking
* Updated truncation warning message.
* Update doc strings
* Update to docs
* Update haystack/nodes/ranker/cohere.py
Co-authored-by: bogdankostic <bogdankostic@web.de>
* Update haystack/nodes/ranker/cohere.py
Co-authored-by: bogdankostic <bogdankostic@web.de>
* Update haystack/nodes/ranker/cohere.py
Co-authored-by: bogdankostic <bogdankostic@web.de>
* Update haystack/nodes/ranker/cohere.py
Co-authored-by: bogdankostic <bogdankostic@web.de>
* Update haystack/nodes/ranker/cohere.py
Co-authored-by: bogdankostic <bogdankostic@web.de>
* Update haystack/nodes/ranker/cohere.py
Co-authored-by: bogdankostic <bogdankostic@web.de>
* Updating docs to reflect PR discussion
* Update haystack/nodes/ranker/cohere.py
Co-authored-by: Daria Fokina <daria.f93@gmail.com>
---------
Co-authored-by: bogdankostic <bogdankostic@web.de>
Co-authored-by: Daria Fokina <daria.f93@gmail.com>
2023-06-23 16:46:46 +02:00
Malte Pietsch
c9179ed0eb
feat: enable LLMs hosted via AWS SageMaker in PromptNode ( #5155 )
...
* Add SageMakerInvocationLayer
---------
Co-authored-by: oryx1729 <78848855+oryx1729@users.noreply.github.com>
Co-authored-by: Vladimir Blagojevic <dovlex@gmail.com>
2023-06-23 15:33:20 +02:00
ZanSara
31664627eb
feat: hard document length limit at max_chars_check ( #5191 )
...
* implement hard cut at max_chars_check
* regenerate ids
* black
* docstring
* black
2023-06-23 12:34:19 +02:00
ZanSara
36192eca72
feat: current_datetime shaper function ( #5195 )
...
* current_datetime shaper
* explicitly add current_datetime to the functions allowed in a prompt template
2023-06-23 10:33:34 +02:00
bogdankostic
612c5cd005
chore: Remove add_tool from ToolsManager ( #5192 )
...
* Remove add_tool from ToolsManager
* Fix tests
2023-06-23 09:26:06 +02:00
Sebastian
1602f3abdd
test: Adding unit tests to Ranker ( #5167 )
...
* adding unit tests for sentence transformers ranker
* Adding more unit tests
* Remove empty line
* Undo static method
* Revert change
* Updated indentation and added match message
* Remove unneeded paranthesis
2023-06-22 15:23:23 +02:00
Michael Feil
cfd703fa3e
fix: model_tokenizer in openai text completion tokenization details ( #5104 )
...
* fix: model_tokenizer
* Update test
---------
Co-authored-by: Sebastian Husch Lee <sjrl423@gmail.com>
2023-06-22 14:23:19 +02:00
Stefano Fiorucci
637433841e
chore: remove deprecated Seq2SeqGenerator and RAGenerator ( #5180 )
...
* first draft of removal
* more removals
* don't download unused models
2023-06-21 16:38:45 +02:00
Sebastian
7a140c1524
feat: add ensure token limit for direct prompting of ChatGPT ( #5166 )
...
* Add support for prompt truncation when using chatgpt if direct prompting is used
* Update tests for test token limit for prompt node
* Update warning message to be correct
* Minor cleanup
* Mark back to integration
* Update count_openai_tokens_messages to reflect changes shown in tiktoken
* Use mocking to avoid request call
* Fix test to make it comply with unit test requirements
* Move tests to respective invocation layers
* Moved fixture to one spot
2023-06-21 15:41:28 +02:00
Vladimir Blagojevic
089187ac8b
fix: Check Agent's prompt template variables and prompt resolver parameters are aligned ( #5163 )
...
* Check Agent's prompt template parameters and prompt resolver parameters are aligned
* Lower the logger warning
* Automatically append transcript if needed
* Amend flaky test
2023-06-21 14:34:41 +02:00
Bilge Yücel
6a1b6b1ae3
feat: Update ConversationalAgent ( #5065 )
...
* feat: Update ConversationalAgent
* Add Tools
* Add test
* Change default params
* fix tests
* Fix circular import error
* Update conversational-agent prompt
* Add conversational-agent-without-tools to legacy list
* Add warning to add tools to conversational agent
* Add callable tools
* Add example script
* Fix linter errors
* Update ConversationalAgent depending on the existance of tools
* Initialize the base Agent with different arguments when there's tool
* Inject memory to the prompt in both cases, update prompts accordingly
* Override the add_tools method to prevent adding tools to ConversationalAgent without tools
* Update test
* Fix linter error
* Remove unused import
* Update docstrings and api reference
* Fix imports and doc string code snippet
* docstrings update
* Update conversational.py
* Mock PromptNode
* Prevent circular import error
* Add max_steps to the ConversationalAgent
* Update resolver description
* Add prompt_template as parameter
* Change docstring
---------
Co-authored-by: Darja Fokina <daria.f93@gmail.com>
2023-06-20 13:09:21 +03:00
Shukri
916e8452f5
feat!: simplify weaviate auth ( #5115 )
...
* feat!: simplify weaviate auth
* docs: explain param precedence
* refactor: simplify _get_embedded_options
2023-06-19 15:46:58 +02:00
Ben Heckmann
1318ac5074
feat: Optional Content Moderation for OpenAI PromptNode & OpenAIAnswerGenerator ( #5017 )
...
* #4071 implemented optional content moderation for OpenAI PromptNode
* added two simple integration tests
* improved documentation & renamed _invoke method to _execute_openai_request
* added a flag to check_openai_policy_violation that will return a full dict of all text violations and their categories
* re-implemented the tests as unit tests & without use of the OpenAI APIs
* removed unused patch
* changed check_openai_policy_violation back to only return a bool
* fixed pylint and test error
---------
Co-authored-by: Julian Risch <julian.risch@deepset.ai>
2023-06-19 13:27:11 +02:00
ZanSara
f52477d31b
fix: small improvement to pipeline v2 tests ( #5153 )
...
* add missing return
* improve test
* docstring
2023-06-16 12:07:00 +02:00
Vladimir Blagojevic
8d8de65492
Add AgentToolLogger, unit test, and example usage ( #5087 )
2023-06-15 08:43:20 +02:00
bogdankostic
7731713a1e
test: Add benchmark config files ( #5093 )
...
* Add config files
* Add top-k and batch size to configs
* Add batch size to configs
* Add batch size to configs
* Remove configs using 1m docs
2023-06-14 18:15:50 +02:00
Ben Heckmann
60e5d73424
fix: changing document scores ( #5090 )
...
* #4653 fix changing scores by returning new document objects from document store queries
* added integration test for InMemoryDocumentStore demonstrating the desired behavior
* Update test/document_stores/test_memory.py
2023-06-14 17:35:46 +02:00
Julian Risch
ce1c9c9ddb
fix: Relax ChatGPT model name check to support gpt-3.5-turbo-0613 ( #5142 )
...
* relax model name checking for chatgpt
* add unit tests
2023-06-14 09:53:00 +02:00
Julian Risch
4c8e0b9d4a
fix: PromptNode falls back to empty list of documents if none are provided but expected ( #5132 )
...
* add warning, default to empty docs list, tests
* pylint
2023-06-13 16:35:19 +02:00
Silvano Cerza
3b8992968d
test: Skip flaky PromptNode test ( #5039 )
...
* Skip flaky PromptNode test
* Add skip reason
* Update test/prompt/test_prompt_node.py
Co-authored-by: bogdankostic <bogdankostic@web.de>
---------
Co-authored-by: bogdankostic <bogdankostic@web.de>
2023-06-13 16:24:29 +02:00
ZanSara
65cdf36d72
chore: block all HTTP requests in CI ( #5088 )
2023-06-13 14:52:24 +02:00
ZanSara
3c71f0ae3d
chore: mark some unit tests under test/pipeline ( #5124 )
...
* mark some unit tests as such
* remove marker
2023-06-12 17:58:31 +02:00
ZanSara
49e037a055
fix: rename requests.py into requests_utils.py ( #5099 )
...
* requests.py -> requests_utils.py
* fix tests
* reimport requrests
* fix more tests
* review feedback
2023-06-12 12:40:21 +02:00
Vladimir Blagojevic
0cc9ce7522
fix: WebRetriever top_k is ignored in a pipeline ( #5106 )
...
* Initial changes
* Add WebSearch, WebRetriever top_k unit tests
* Add exact integration test that failed Tuana
* PR review
2023-06-09 10:42:37 +02:00
Julian Risch
d8a4f20379
feat: Consider prompt_node's default_prompt_template in agent ( #5095 )
...
* consider prompt_node's default_prompt_template in agent
* make test a unit test via mocking
* updated docstring
2023-06-08 13:42:28 +02:00
Vladimir Blagojevic
e3b069620b
feat: pass model parameters to HFLocalInvocationLayer via model_kwargs, enabling direct model usage ( #4956 )
...
* Simplify HFLocalInvocationLayer, move/add unit tests
* PR feedback
* Better pipeline invocation, add mocked tests
* Minor improvements
* Mock pipeline directly, unit test updates
* PR feedback, change pytest type to integration
* Mock supports unit test
* add full stop
* PR feedback, improve unit tests
* Add mock_get_task fixture
* Further improve unit tests
* Minor unit test improvement
* Add unit tests, increase coverage
* Add unit tests, increase test coverage
* Small optimization, improve _ensure_token_limit unit test
---------
Co-authored-by: Darja Fokina <daria.f93@gmail.com>
2023-06-07 13:34:45 +02:00
Silvano Cerza
a2156ee8fb
fix: Fix handling of streaming response in AnthropicClaudeInvocationLayer ( #4993 )
...
* Fix handling of streaming response in AnthropicClaudeInvocationLayer
---------
Co-authored-by: Vladimir Blagojevic <dovlex@gmail.com>
Co-authored-by: Darja Fokina <daria.f93@gmail.com>
2023-06-07 10:57:36 +02:00
bogdankostic
da1f245a84
feat: Add batch_size parameter and cast timeout_config value to tuple for WeaviateDocumentStore ( #5079 )
...
* Add batch_size parameter and cast timeout_config to tuple
* Add unit test
* Remove debug tqdm
* Remove debug tqdm introduced in #5063
2023-06-06 17:06:10 +02:00
Sebastian
1777b22fcb
fix: Ensure eval mode for farm and transformer models for predictions ( #3791 )
...
Co-authored-by: Massimiliano Pippi <mpippi@gmail.com>
2023-06-06 13:06:30 +02:00
Michael Feil
6ea8ae01a2
feat: Allow setting custom api_base for OpenAI nodes ( #5033 )
...
* add changes for api_base
* format retriever
* Update haystack/nodes/retriever/dense.py
Co-authored-by: bogdankostic <bogdankostic@web.de>
* Update haystack/nodes/audio/whisper_transcriber.py
Co-authored-by: bogdankostic <bogdankostic@web.de>
* Update haystack/preview/components/audio/whisper_remote.py
Co-authored-by: bogdankostic <bogdankostic@web.de>
* Update haystack/nodes/answer_generator/openai.py
Co-authored-by: bogdankostic <bogdankostic@web.de>
* Update test_retriever.py
* Update test_whisper_remote.py
* Update test_generator.py
* Update test_retriever.py
* reformat with black
* Update haystack/nodes/prompt/invocation_layer/chatgpt.py
Co-authored-by: Daria Fokina <daria.f93@gmail.com>
* Add unit tests
* apply docstring suggestions
---------
Co-authored-by: bogdankostic <bogdankostic@web.de>
Co-authored-by: michaelfeil <me@michaelfeil.eu>
Co-authored-by: Daria Fokina <daria.f93@gmail.com>
2023-06-05 11:32:06 +02:00
bogdankostic
a9a49e2c0a
feat: Add batching for querying in ElasticsearchDocumentStore and OpenSearchDocumentStore ( #5063 )
...
* Include benchmark config in output
* Use queries from aggregated labels
* Introduce batching for querying in ElasticsearchDocStore and OpenSearchDocStore
* Fix mypy
* Use self.batch_size in write_documents
* Use 10_000 as default batch size
* Add unit tests for write documents
2023-06-01 18:47:24 +02:00
bogdankostic
c3e59914da
refactor: Delete outdated benchmark files ( #5008 )
2023-06-01 13:59:12 +02:00
bogdankostic
6774e0ae58
fix: Use queries from aggregated labels in benchmarks ( #5054 )
...
* Include benchmark config in output
* Use queries from aggregated labels
2023-06-01 10:49:54 +02:00
ZanSara
89de76d5fe
feat: move cli out from preview ( #5055 )
...
* move cli from preview
* readme
* review feedback
* test mocks & import paths
* import path
2023-05-31 18:34:14 +02:00
Silvano Cerza
3fd9e0fd89
feat: Add CLI prompt cache command ( #5050 )
...
* Add CLI prompt cache command
* Rename prompt cache to prompt fetch
2023-05-30 18:04:52 +02:00
ZanSara
6249e65bc8
feat: prompts caching from PromptHub ( #5048 )
...
* split up prompttemplate init
* caching
* docstring
* add platformdirs
* use user_data_dir
* fix tests
* add tests
* pylint
* mypy
2023-05-30 16:55:48 +02:00
Silvano Cerza
37518c8b8c
chore: Simplify DefaultPromptHandler logic and add tests ( #4979 )
...
* Simplify DefaultPromptHandler logic and add tests
Co-authored-by: Vladimir Blagojevic <dovlex@gmail.com>
* Remove commented code
* Split single unit test into multiple tests
---------
Co-authored-by: Vladimir Blagojevic <dovlex@gmail.com>
2023-05-29 12:13:32 +02:00
ZanSara
7e5fa0dd94
fix: Move check for default PromptTemplates in PromptTemplate itself ( #5018 )
...
* make prompttemplate load the defaults instead of promptnode
* add test
* fix tenacity decorator
* fix tests
* fix error handling
* mypy
---------
Co-authored-by: Silvano Cerza <3314350+silvanocerza@users.noreply.github.com>
2023-05-27 18:05:05 +02:00
bogdankostic
b8ff1052d4
refactor: Adapt running benchmarks ( #5007 )
...
* Generate eval result in separate method
* Adapt benchmarking utils
* Adapt running retriever benchmarks
* Adapt error message
* Adapt running reader benchmarks
* Adapt retriever reader benchmark script
* Adapt running benchmarks script
* Adapt README.md
* Raise error if file doesn't exist
* Raise error if path doesn't exist or is a directory
* minor readme update
* Create separate methods for checking if pipeline contains reader or retriever
* Fix reader pipeline case
---------
Co-authored-by: Darja Fokina <daria.f93@gmail.com>
2023-05-26 18:48:11 +02:00
bogdankostic
5633446173
refactor: Add reader-retriever benchmark script ( #5006 )
...
* Generate eval result in separate method
* Adapt benchmarking utils
* Adapt running retriever benchmarks
* Adapt error message
* Adapt running reader benchmarks
* Adapt retriever reader benchmark script
* Raise error if file doesn't exist
* Raise error if path doesn't exist or is a directory
* Remove unused line
* Create separate method for getting reader config
* Make use of get_reader_config
* Create separate method for retriever config
2023-05-26 13:54:52 +02:00
