huggingface/lighteval

Pull Requests Commits

Merge branch 'main' into multichoice_continuations_start_space_fix

clefourrier committed 2 years ago

Verified 14f9c93a

Expose samples via the CLI (#228)

clefourrier committed 2 years ago

Verified cbae17dc

Fix inference endpoint config (#244)

clefourrier committed 2 years ago

Verified f047874a

Merge branch 'main' into multichoice_continuations_start_space_fix

clefourrier committed 2 years ago

Verified 513b9f0c

uses torch compile if provided (#248)

clefourrier committed 2 years ago

Verified 4751027a

Merge branch 'main' into multichoice_continuations_start_space_fix

NathanHB committed 2 years ago

Verified 737df937

clefourrier committed 2 years ago

bbf6e936

Update src/lighteval/models/model_config.py

clefourrier committed 2 years ago

Verified e3180579

udpated piqa (#222)

clefourrier committed 2 years ago

Verified 506abda3

clefourrier committed 2 years ago

Verified db502dde

Removes default bert scorer init (#234)

clefourrier committed 2 years ago

Verified 2b4b637e

remove latex writer since we don't use it (#231)

clefourrier committed 2 years ago

Verified 86fbe641

Update issue templates (#235)

clefourrier committed 2 years ago

Verified 003f05e2

Fix a tiny bug in DROP metric (#229)

sadra-barikbin committed 2 years ago

Verified 66ed7a28

fixing issues - was not parsed properly

clefourrier committed 2 years ago

5b2c1eb6

Quantization related issues (#224)

clefourrier committed 2 years ago

Verified 44f9a461

Make evaluator invariant of input request type order (#215)

sadra-barikbin committed 2 years ago

Verified 951cd5b5

Fix _init_max_length in base_model.py (#185)

gucci-j committed 2 years ago

Verified d43c9a32

launch lighteval using `lighteval --args` (#152)

NathanHB committed 2 years ago

Verified 4550cb7e

Add metrics as functions (#214)

clefourrier committed 2 years ago

Verified aaf7e8af

should fix most inference endpoints issues of version config (#226)

clefourrier committed 2 years ago

Verified 733257ff

Data split depending on eval params (#169)

clefourrier committed 2 years ago

Verified 66e6aaeb

Now only uses functions for prompt definition (#213)

clefourrier committed 2 years ago

Verified 4651531e

Use only dataclasses for task init (#212)

clefourrier committed 2 years ago

Verified 3aaec227

Fix a few typos in `metrics.py` (#218)

sadra-barikbin committed 2 years ago

Verified 3f909509

Homogeneize logging system (#150)

clefourrier committed 2 years ago

Verified ac57b78e

Adds a dummy/random model for baseline init (#220)

guipenedo committed 2 years ago

Verified 70f7fc6d

Fix the bug (#216)

sadra-barikbin committed 2 years ago

Verified 0528f297

[Bugfix] Avoid truncating the outputs based on string lengths (#201)

anton-l committed 2 years ago

Verified 60646959

Fix a few typos and do a tiny refactor (#187)

sadra-barikbin committed 2 years ago

Verified 843a0f8c

Older