Igor Ilic
902979c1de
refactor: Refactor get source code chunks based on tokenizer rework
2025-01-24 13:40:10 +01:00
hajdul88
49e10832d6
chore: cleans the modal entrypoint
2025-01-24 10:17:48 +01:00
hajdul88
d5ad370b1d
Merge branch 'dev' into feature/cog-1084-modal-example-for-cognee
2025-01-24 09:53:22 +01:00
alekszievr
bc1b05437e
Merge branch 'dev' into cog-1069-update-notebooks-evals
2025-01-23 20:32:30 +01:00
Boris Arzentar
d1ce7e531c
fix: mcp server lint errors
2025-01-23 20:03:31 +01:00
Igor Ilic
844d99cb72
docs: Remove commented code
2025-01-23 18:24:26 +01:00
Igor Ilic
7dea1d54d7
refactor: Add specific max token values to embedding models
2025-01-23 18:18:45 +01:00
Igor Ilic
6d5679f9d2
Merge branch 'dev' into COG-970-refactor-tokenizing
2025-01-23 18:14:49 +01:00
hajdul88
2410feea4f
feat: implements modal wrapper + dockerfile for modal containers
2025-01-23 18:06:09 +01:00
Igor Ilic
1319944dcd
docs: Update .env.template to include llm and embedding options
2025-01-23 18:05:45 +01:00
Igor Ilic
b25a82e206
chore: Add google-generativeai as gemini optional dependency to Cognee
2025-01-23 17:56:56 +01:00
Igor Ilic
b686376c54
feat: Add gemini tokenizer to cognee
2025-01-23 17:55:04 +01:00
Vasilije
d50af60b59
Merge pull request #465 from topoteretes/feat/COG-1058-fastmcp
...
feat: use fastmcp for mcp server
2025-01-23 17:38:55 +01:00
Igor Ilic
294ed1d960
feat: Add HuggingFace Tokenizer support
2025-01-23 16:52:35 +01:00
Igor Ilic
2e1a48e22c
docs: Add usage example of function
2025-01-23 15:13:46 +01:00
Igor Ilic
de19016494
fix: Add flag to allow SQLite to use foreign keys
2025-01-23 15:10:27 +01:00
Igor Ilic
d4453e4a1d
fix: Add support for SQLite and PostgreSQL for inserting data in SQLAlchemyAdapter
2025-01-23 14:59:02 +01:00
Boris Arzentar
eb22dc9889
fix: remove unnecessary dot
2025-01-23 11:47:20 +01:00
Boris Arzentar
e577276d91
Merge remote-tracking branch 'origin/dev' into feat/COG-1058-fastmcp
2025-01-23 11:46:25 +01:00
Boris Arzentar
00f302c37a
feat: use fastmcp for mcp server
2025-01-23 11:45:40 +01:00
hande-k
cdecf5fb8f
add short decsription in md
2025-01-23 11:17:48 +01:00
hande-k
343de01d5a
update notebooks with latest eval
2025-01-23 11:11:51 +01:00
Igor Ilic
9f6a0ba783
Merge branch 'dev' into pgvector-add-normalization
2025-01-23 11:11:43 +01:00
Vasilije
90657a262c
Merge pull request #460 from topoteretes/COG-793-metadata-rework
...
Cog 793 metadata rework
2025-01-23 11:00:07 +01:00
alekszievr
4e3a666b33
Feat: Save and load contexts and answers for eval ( #462 )
...
* feat: make tasks a configurable argument in the cognify function
* fix: add data points task
* eval on random samples instead of first couple
* Save and load contexts and answers
* Fix random seed usage and handle empty descriptions
* include insights search in cognee option
* create output dir if doesnt exist
---------
Co-authored-by: lxobr <122801072+lxobr@users.noreply.github.com>
2025-01-22 16:17:01 +01:00
Igor Ilic
40c0279ec5
Merge branch 'COG-793-metadata-rework' of github.com:topoteretes/cognee into COG-793-metadata-rework
2025-01-22 16:13:11 +01:00
Igor Ilic
80e67b0619
refactor: Rename foreign to external metadata
...
Rename foreign metadata to external metadata for metadata coming outside of Cognee
2025-01-22 16:07:35 +01:00
Rita Aleksziev
b2f7f733d9
create output dir if doesnt exist
2025-01-22 10:58:44 +01:00
alekszievr
5b6fe00576
Merge branch 'dev' into feat/save_and_load_answers
2025-01-22 10:52:05 +01:00
Rita Aleksziev
e0980361a1
include insights search in cognee option
2025-01-22 10:51:32 +01:00
Igor Ilic
93249c72c5
fix: Initial commit to resolve issue with using tokenizer based on LLMs
...
Currently TikToken is used for tokenizing by default which is only supported by OpenAI,
this is an initial commit in an attempt to add Cognee tokenizing support for multiple LLMs
2025-01-21 19:53:22 +01:00
Igor Ilic
655ab0b8cc
Merge branch 'dev' into COG-793-metadata-rework
2025-01-21 18:20:49 +01:00
Rita Aleksziev
9fec8fd322
Fix random seed usage and handle empty descriptions
2025-01-21 17:04:00 +01:00
Vasilije
c9536f97a5
Merge pull request #451 from topoteretes/add_docstrings
...
chore: add docstrings any typing to cognee tasks
2025-01-21 14:07:19 +01:00
Rita Aleksziev
1c16a1744c
Save and load contexts and answers
2025-01-20 18:42:09 +01:00
Igor Ilic
bd3a5a758c
Merge branch 'dev' into COG-793-metadata-rework
2025-01-20 18:06:21 +01:00
Igor Ilic
77f0b45a0d
refactor: Resolve issue with notebook after metadata refactor
...
Resolve issue with LlamaIndex notebook after refactor
2025-01-20 18:02:57 +01:00
Igor Ilic
4196a4ce89
refactor: Update test to be up to date with current metadata refactor effort
2025-01-20 17:53:54 +01:00
Rita Aleksziev
015f0084c8
eval on random samples instead of first couple
2025-01-20 17:47:52 +01:00
lxobr
1a73779353
fix: add data points task
2025-01-20 17:43:21 +01:00
lxobr
c747a05717
feat: make tasks a configurable argument in the cognify function
2025-01-20 17:43:21 +01:00
Igor Ilic
5c17501bb8
refactor: add missing foreing_metadata attr to tests
2025-01-20 17:38:28 +01:00
Igor Ilic
ab8d95cc30
refactor: As neo4j can't support dictionaries, add foreign metadata as string
2025-01-20 17:28:14 +01:00
Igor Ilic
49ad292592
refactor: Reduce complexity of metadata handling
...
Have foreign metadata be a table column in data instead of it's own table to reduce complexity
Refactor COG-793
2025-01-20 16:39:05 +01:00
Igor Ilic
0c7c1d7503
refactor: Refactor ingestion to only have one ingestion task
2025-01-20 14:33:47 +01:00
hajdul88
813a03c6e2
Merge branch 'dev' into pgvector-add-normalization
2025-01-20 13:46:50 +01:00
Igor Ilic
2546844787
feat: Add normalization to PGVector search
...
Add normalization to PGVector search results
2025-01-20 13:42:39 +01:00
Vasilije
bbb8e8951c
Merge pull request #458 from topoteretes/feature/cog-1034-implementing-windows-test-into-cicd-pipeline
...
Adds windows test + fixes networkx file loading issue
2025-01-20 13:16:07 +01:00
hajdul88
957ab81879
chore: renaming test
2025-01-20 12:27:22 +01:00
hajdul88
bf70705ed0
Fix: fixes networkx failed to load graph from file error
2025-01-20 12:19:34 +01:00
