Skip to content

Conversation

ShirApp
Copy link
Collaborator

@ShirApp ShirApp commented Jan 22, 2025

there are a few large tables, that cannot be added to the prompts (context window)

@ShirApp ShirApp requested a review from csrajmohan January 22, 2025 16:21
@ShirApp ShirApp self-assigned this Jan 22, 2025
Copy link
Collaborator

@csrajmohan csrajmohan left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Looks good to me..

@ShirApp ShirApp merged commit fa24fd7 into main Jan 22, 2025
17 of 18 checks passed
@ShirApp ShirApp deleted the filter_wikitq branch January 22, 2025 18:48
tejaswini pushed a commit that referenced this pull request Jan 24, 2025
elronbandel added a commit that referenced this pull request Jan 26, 2025
* Renamed criterias in LLM-as-a-Judge metrics to criteria.

* Reintroduced imports that were removed from llm_as_judge.py

* Updated the examples documentation for LLM-as-a-judge

* Added missing import

* Fixed formatting using ruff

* add a filter to wikitq (#1547)

* Add text2sql tasks to unitxt (#1414)

* add text2sql templates

Signed-off-by: Yotam-Perlitz <y.perlitz@ibm.com>

* add data managment utility for text2sql

Signed-off-by: Yotam-Perlitz <y.perlitz@ibm.com>

* add basic template

Signed-off-by: Yotam-Perlitz <y.perlitz@ibm.com>

* add sql execution accuracy metric

Signed-off-by: Yotam-Perlitz <y.perlitz@ibm.com>

* add text2sql execution accuracy metric

Signed-off-by: Yotam-Perlitz <y.perlitz@ibm.com>

* add text2sql task

Signed-off-by: Yotam-Perlitz <y.perlitz@ibm.com>

* condition download in presence of a cache dir

Signed-off-by: Yotam-Perlitz <y.perlitz@ibm.com>

* add init fille

Signed-off-by: Yotam-Perlitz <y.perlitz@ibm.com>

* add processors

Signed-off-by: Yotam-Perlitz <y.perlitz@ibm.com>

* add processors

Signed-off-by: Yotam-Perlitz <y.perlitz@ibm.com>

* add basic template

Signed-off-by: Yotam-Perlitz <y.perlitz@ibm.com>

* change id to int

Signed-off-by: Yotam-Perlitz <y.perlitz@ibm.com>

* change notations in templates

Signed-off-by: Yotam-Perlitz <y.perlitz@ibm.com>

* push to catalog

Signed-off-by: Yotam-Perlitz <y.perlitz@ibm.com>

* add evidence, remove SL

Signed-off-by: Yotam-Perlitz <y.perlitz@ibm.com>

* remove unued function, fix

Signed-off-by: Yotam-Perlitz <y.perlitz@ibm.com>

* fix imports from unitxt.text2sql

Signed-off-by: Yotam-Perlitz <y.perlitz@ibm.com>

* push to catalog

Signed-off-by: Yotam-Perlitz <y.perlitz@ibm.com>

* fix cache location

Signed-off-by: Yotam-Perlitz <y.perlitz@ibm.com>

* add example

Signed-off-by: Yotam-Perlitz <y.perlitz@ibm.com>

* fix imports

Signed-off-by: Yotam-Perlitz <y.perlitz@ibm.com>

* add func_timeout to test reqs

Signed-off-by: Yotam-Perlitz <y.perlitz@ibm.com>

* fix typing

Signed-off-by: Yotam-Perlitz <y.perlitz@ibm.com>

* change template name

Signed-off-by: Yotam-Perlitz <y.perlitz@ibm.com>

* push to catalog

Signed-off-by: Yotam-Perlitz <y.perlitz@ibm.com>

* add req

Signed-off-by: Yotam-Perlitz <y.perlitz@ibm.com>

* add local model option

Signed-off-by: Yotam-Perlitz <y.perlitz@ibm.com>

* fix databases download

Signed-off-by: Yotam-Perlitz <y.perlitz@ibm.com>

* fix databases download

Signed-off-by: Yotam-Perlitz <y.perlitz@ibm.com>

* add loader limit ot make example faster

Signed-off-by: Yotam-Perlitz <y.perlitz@ibm.com>

* fix cache paths, avoid re-download

Signed-off-by: Yotam-Perlitz <y.perlitz@ibm.com>

* add type schema

Signed-off-by: Yotam-Perlitz <y.perlitz@ibm.com>

* remove inports from inits

Signed-off-by: Yotam-Perlitz <y.perlitz@ibm.com>

* add text2sql to inits

Signed-off-by: Yotam-Perlitz <y.perlitz@ibm.com>

* update card to use serializers

Signed-off-by: Yotam-Perlitz <y.perlitz@ibm.com>

* add schema serializer

Signed-off-by: Yotam-Perlitz <y.perlitz@ibm.com>

* add text2sql serializer to default template

Signed-off-by: Yotam-Perlitz <y.perlitz@ibm.com>

* add schema to task

Signed-off-by: Yotam-Perlitz <y.perlitz@ibm.com>

* adjust templates to using serializer

Signed-off-by: Yotam-Perlitz <y.perlitz@ibm.com>

* adjust templates to using serializer

Signed-off-by: Yotam-Perlitz <y.perlitz@ibm.com>

* fix processor

Signed-off-by: Yotam-Perlitz <y.perlitz@ibm.com>

* remove target prefix from template

Signed-off-by: Yotam-Perlitz <y.perlitz@ibm.com>

* add shuffle to bird

Signed-off-by: Yotam-Perlitz <y.perlitz@ibm.com>

* add shuffle to bird

Signed-off-by: Yotam-Perlitz <y.perlitz@ibm.com>

* edit template

Signed-off-by: Yotam-Perlitz <y.perlitz@ibm.com>

* remove comment from init

Signed-off-by: Yotam-Perlitz <y.perlitz@ibm.com>

* clear processors code

Signed-off-by: Yotam-Perlitz <y.perlitz@ibm.com>

* add option with ticks

Signed-off-by: Yotam-Perlitz <y.perlitz@ibm.com>

* add anls metric

Signed-off-by: Yotam-Perlitz <y.perlitz@ibm.com>

* add template

Signed-off-by: Yotam-Perlitz <y.perlitz@ibm.com>

* drop comment

Signed-off-by: Yotam-Perlitz <y.perlitz@ibm.com>

* remove recursion limit

Signed-off-by: Yotam-Perlitz <y.perlitz@ibm.com>

* add loader_limit to example

Signed-off-by: Yotam-Perlitz <y.perlitz@ibm.com>

* fix recursion error

Signed-off-by: Yotam-Perlitz <y.perlitz@ibm.com>

* move import to withing metric

Signed-off-by: Yotam-Perlitz <y.perlitz@ibm.com>

* remove catalog files wo prepare

Signed-off-by: Yotam-Perlitz <y.perlitz@ibm.com>

* fix typing

Signed-off-by: Yotam-Perlitz <y.perlitz@ibm.com>

* change template im example

Signed-off-by: Yotam-Perlitz <y.perlitz@ibm.com>

* moving text2sql implementaion to the main src dir

Signed-off-by: Yotam-Perlitz <y.perlitz@ibm.com>

* fix imports

Signed-off-by: Yotam-Perlitz <y.perlitz@ibm.com>

* fix imports

Signed-off-by: Yotam-Perlitz <y.perlitz@ibm.com>

* fix imports

Signed-off-by: Yotam-Perlitz <y.perlitz@ibm.com>

* fix imports

Signed-off-by: Yotam-Perlitz <y.perlitz@ibm.com>

* import data_utils

Signed-off-by: Yotam-Perlitz <y.perlitz@ibm.com>

* fix formatting

Signed-off-by: Yotam-Perlitz <y.perlitz@ibm.com>

* refactor names

Signed-off-by: Yotam-Perlitz <y.perlitz@ibm.com>

* add processors tests

Signed-off-by: Yotam-Perlitz <y.perlitz@ibm.com>

* add more tests

Signed-off-by: Yotam-Perlitz <y.perlitz@ibm.com>

* add tests

Signed-off-by: Yotam-Perlitz <y.perlitz@ibm.com>

* refactor: allow more data sources

Signed-off-by: Yotam-Perlitz <y.perlitz@ibm.com>

* allow db source input

Signed-off-by: Yotam-Perlitz <y.perlitz@ibm.com>

* organize imports

Signed-off-by: Yotam-Perlitz <y.perlitz@ibm.com>

* update example

Signed-off-by: Yotam-Perlitz <y.perlitz@ibm.com>

* add db_type to task

Signed-off-by: Yotam-Perlitz <y.perlitz@ibm.com>

* format

Signed-off-by: Yotam-Perlitz <y.perlitz@ibm.com>

* add db_type to task

Signed-off-by: Yotam-Perlitz <y.perlitz@ibm.com>

* add local db definition ability

Signed-off-by: Yotam-Perlitz <y.perlitz@ibm.com>

* add EE tests

Signed-off-by: Yotam-Perlitz <y.perlitz@ibm.com>

* add tests

Signed-off-by: Yotam-Perlitz <y.perlitz@ibm.com>

* rename file

Signed-off-by: Yotam-Perlitz <y.perlitz@ibm.com>

* rename file

Signed-off-by: Yotam-Perlitz <y.perlitz@ibm.com>

* update sql metric

Signed-off-by: Yotam-Perlitz <y.perlitz@ibm.com>

* rename file

Signed-off-by: Yotam-Perlitz <y.perlitz@ibm.com>

* refactor types, serializers and metric

Signed-off-by: Yotam-Perlitz <y.perlitz@ibm.com>

---------

Signed-off-by: Yotam-Perlitz <y.perlitz@ibm.com>

* Add deduplicate operator (#1549)

* Add deduplicate operator

Signed-off-by: elronbandel <elronbandel@gmail.com>

* Deduplicate MMLU

Signed-off-by: elronbandel <elronbandel@gmail.com>

* Update Deduplicate example in documentation for clarity

Signed-off-by: elronbandel <elronbandel@gmail.com>

* Deduplicate social iqa

Signed-off-by: elronbandel <elronbandel@gmail.com>

---------

Signed-off-by: elronbandel <elronbandel@gmail.com>

* Fix the authentication problem (#1550)

* Attach assitant answers to their origins with url link (#1528)

Signed-off-by: elronbandel <elronbandel@gmail.com>

* Add mtrag benchmark (#1548)

* Add mtrag benchmark

Signed-off-by: elronbandel <elronbandel@gmail.com>

* Add multi_type_serializer for references and prediction fields in various JSON metrics

Signed-off-by: elronbandel <elronbandel@gmail.com>

* Remove unused TempOperator class and delete obsolete multi_turn.json task file

Signed-off-by: elronbandel <elronbandel@gmail.com>

---------

Signed-off-by: elronbandel <elronbandel@gmail.com>

* Update end of year summary blog (#1552)

Signed-off-by: elronbandel <elronbandel@gmail.com>

* Update strategic focus section in 2024 summary blog to emphasize usability

Signed-off-by: elronbandel <elronbandel@gmail.com>

* Added missing import

* Fix llm as judge example

Signed-off-by: Martín Santillán Cooper <msantillancooper@ibm.com>

* apply 'pre-commit run --all-files'

Signed-off-by: Martín Santillán Cooper <msantillancooper@ibm.com>

---------

Signed-off-by: Yotam-Perlitz <y.perlitz@ibm.com>
Signed-off-by: elronbandel <elronbandel@gmail.com>
Signed-off-by: Martín Santillán Cooper <msantillancooper@ibm.com>
Co-authored-by: Tejaswini Pedapati <tejaswinip@us.ibm.com>
Co-authored-by: ShirApp <58909189+ShirApp@users.noreply.github.com>
Co-authored-by: Yotam Perlitz <perlitz@gmail.com>
Co-authored-by: Elron Bandel <elronbandel@gmail.com>
Co-authored-by: Elad <eladv@il.ibm.com>
Co-authored-by: Martín Santillán Cooper <msantillancooper@ibm.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants