GitLab

* Initial support for Unitxt datasets in LM Eval Harness

See  https://github.com/IBM/unitxt



The script 'generate_yamls.py' creates LM Eval Harness yaml files corresponding to Unitxt datasets specified in the 'unitxt_datasets' file.

The glue code required to register Unitxt metrics is in 'unitxt_wrapper.py'.

* Added dataset loading check to generate_yaml

Improved error messages.

* Speed up generate_yaml

Added printouts and improved error message

* Added output printout

* Simplified integration of unitxt datasets

Store all the common yaml configuration in a yaml include shared by all datasets of the same task.

* Post code review comments - part 1

1. Made sure include files don't end wth 'yaml' so they won't be marked as tasks
2. Added more datasets and tasks (NER, GEC)
3. Added README

* Post code review comments - part 2

1. Added install unitxt install option in pyproject.toml:
pip install 'lm_eval[unitxt]'
2. Added a check that unitxt is installed and print a clear error message if not

* Commited missing pyproject change

* Added documentation on adding datasets

* More doc changes

* add unitxt extra to readme

* run precommit

---------
Co-authored-by: haileyschoelkopf <hailey@eleuther.ai>

* fix auto-batch size bug for seq2seq models

* alphabetize task + group tables ; fix eval tracker bug

* fix eval tracker bug

* add Hendrycks MATH (no sympy checking) variant

* add readmes for MATH tasks

* Added fewshot sampling seeds to evaluator.simple_evaluate signature

Way to control seed of fewshot sampling
may help with #1591

* Added ability for custom sampler for ConfigurableTask

May be set in config like
```
fewshot_config:
  sampler: !function utils.MyFewshotSampler
```

* explicitly set fewshot random generator seed for HFLM generate_until_task test

* add backward compatibility for three args seed setup

* save seeds info to logs/reports

fix `----hf_hub_log_args` to `--hf_hub_log_args`

* remove echo parameter in OpenAI completions API

* remove context length parameter doc string

* evaluation tracker implementation

* OVModelForCausalLM test fix

* typo fix

* moved methods args

* multiple args in one flag

* loggers moved to dedicated dir

* improved filename sanitization

* Add option to set OpenVINO config

* Use utils.eval_logger for logging

* vllm lora support

* remove print

* version check, rename lora kwarg

* upload new tasks

* add readmes

* run linters

---------
Co-authored-by: haileyschoelkopf <hailey@eleuther.ai>

* Update utils.py

This is a 4-choice task, option_e is null for all but 3 samples

* Fix options

Adaptive choices

* add option e

* bump multilingual arc version

---------
Co-authored-by: Hailey Schoelkopf <65563625+haileyschoelkopf@users.noreply.github.com>

* Add Pile-10k readme

* Add Pile-10k task configuration file

* Add register_filter decorator

* Add register_filter docs

* Support individual scrolls datasets

* Add qmsum context

* Fix formatting

* Update task.py

* Update __init__.py

* add xnli_eu tasks

* update tasks readme

* update readme

https://github.com/EleutherAI/lm-evaluation-harness/issues/1698

* Add neuralmagic models for SparseML and DeepSparse

* Update to latest and add test

* Format

* Fix list to List

* Format

* Add deepsparse/sparseml to automated testing

* Update pyproject.toml

* Update pyproject.toml

* Update README

* Fixes for dtype and device

* Format

* Fix test

* Apply suggestions from code review
Co-authored-by: Hailey Schoelkopf <65563625+haileyschoelkopf@users.noreply.github.com>

* Address review comments!

---------
Co-authored-by: Hailey Schoelkopf <65563625+haileyschoelkopf@users.noreply.github.com>

* added delta weights

* removed debug

* readme update

* better error handling

* autogptq warn

* warn update

* peft and delta error, explicitly deleting _model_delta

* linter fix

* correction bug EleutherAI#1664

* add any invalid characters for Windows filenames and Unix-like systems

see:
https://gist.github.com/doctaphred/d01d05291546186941e1b7ddc02034d3?permalink_comment_id=3958715



* Update lm_eval/__main__.py

* Update scripts/zeno_visualize.py

* fix format

---------
Co-authored-by: Hailey Schoelkopf <65563625+haileyschoelkopf@users.noreply.github.com>
Co-authored-by: haileyschoelkopf <hailey@eleuther.ai>

* claude3

* supply for anthropic claude3

* supply for anthropic claude3

* anthropic config changes

* add callback options on anthropic

* line passed

* claude3 tiny change

* help anthropic installation

* mention sysprompt / being careful with format in readme

---------
Co-authored-by: haileyschoelkopf <hailey@eleuther.ai>

* implementation of TMMLU+

* implemented: TMMLU+

****TMMLU+ : large-scale Traditional chinese Massive Multitask language Understanding****

- 4 categories
    - STEM
    - Social Science
    - Humanities
    - Other

The TMMLU+ dataset, encompassing over 67 subjects and 20160 tasks, is six times larger and more balanced than its predecessor, TMMLU, and includes benchmark results from both closed-source and 20 open-weight Chinese large language models with 1.8B to 72B parameters. However, Traditional Chinese variants continue to underperform compared to major Simplified Chinese models.

```markdown
Total number of tasks in the 'test' sets: 20160
Total number of tasks in the 'validation' sets: 2247
Total number of tasks in the 'train' sets: 335
```

* Remove print from __init__.py

There was my mistake in forgetting to remove the debug print from the code.

* update: move TMMLU+ config generation program into default

* fix: we should use training set as few shots example

* update: README for TMMLU+

* update: a small changes of TMMLU+ README file

* pre-commit run thought

* Add README for TMMLU+ dataset

* run precommit

* trigger precommit again

* trigger precommit again

* isort is fussy

* isort is fussy

* format, again

* oops

* oops

---------
Co-authored-by: lintang <lintang@eleuther.ai>
Co-authored-by: haileyschoelkopf <hailey@eleuther.ai>