FlexEval
Flexible evaluation tool for language models. Easy to extend, highly customizable!
English | 日本語 |
With FlexEval, you can evaluate language models with:
- Zero/few-shot in-context learning tasks
- Open-ended text-generation benchmarks such as MT-Bench with automatic evaluation using GPT-4
- Log-probability-based multiple-choice tasks
- Computing perplexity of text data
For more use cases, see the documentation.
Key Features
- Flexibility:
flexevalis flexible in terms of the evaluation setup and the language model to be evaluated. - Modularity: The core components of
flexevalare easily extensible and replaceable. - Clarity: The results of evaluation are clear and all the details are saved.
- Reproducibility:
flexevalshould be reproducible, with the ability to save and load configurations and results.
Installation
pip install flexeval
Quick Start
The following minimal example evaluates the hugging face model sbintuitions/sarashina2.2-0.5b with the commonsense_qa task.
Run Command
flexeval_lm \
--language_model HuggingFaceLM \
--language_model.model "sbintuitions/sarashina2.2-0.5b" \
--eval_setup "commonsense_qa" \
--save_dir "results/commonsense_qa"
Output
...
2025-09-03 16:22:58.434 | INFO | flexeval.core.evaluate_generation:evaluate_generation:92 - {'exact_match': 0.3185913185913186, 'finish_reason_ratio-stop': 1.0, 'avg_output_length': 9.095004095004095, 'max_output_length': 69, 'min_output_length': 2}
...
The results saved in --saved_dir contain:
config.json: The configuration of the evaluation, which can be used to replicate the evaluation.metrics.json: The evaluation metrics.outputs.jsonl: The outputs of the language model that comes with instance-level metrics.
You can flexibly customize the evaluation by specifying command-line arguments or configuration files. Besides the Transformers model, you can also evaluate models via OpenAI ChatGPT and vLLM, and other models can be readily added!
Next Steps
- Run
flexeval_presetsto check the list of off-the-shelf presets in addition tocommonsense_qa. You can find the details in the Preset Configs section. - See Getting Started to check the tutorial examples for other kinds of tasks.
- See the Configuration Guide to set up your evaluation.
1 maintainer
Metadata
1 maintainer
Release files for flexeval 0.18.3
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | |
|---|---|---|---|---|
| flexeval-0.18.3.tar.gz | 318.3 kB | Without provenance | Details |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Reset |
|---|---|---|---|---|
| flexeval-0.18.3-py3-none-any.whl | Python 3 | none | any | Details |
Total release size: 747.2 kB
Release files / flexeval-0.18.3.tar.gz
| Download URL | flexeval-0.18.3.tar.gz |
|---|---|
| Size | 318.3 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
c28a1dd9aaca690308fb268bd660c1aee64bd9a390182b2a2de7427cab48b59d
|
|
BLAKE2b-256 checksum How to use checksums |
dbd03c54d2f36d8a16e3b7782bf8b657f92c6d96032eea8bf9e6e81419dee90e
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
poetry/2.5.1 CPython/3.10.22 Linux/6.17.0-1022-azure
|
Release files / flexeval-0.18.3-py3-none-any.whl
| Download URL | flexeval-0.18.3-py3-none-any.whl |
|---|---|
| Size | 428.9 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
b5a90e334221d700c510002aacdee653d09b30e4354dc308633498108d8ae94a
|
|
BLAKE2b-256 checksum How to use checksums |
e32cef63e0619dc72151a9905b80819e413c3064e52a6786bcb6d982834bb181
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
poetry/2.5.1 CPython/3.10.22 Linux/6.17.0-1022-azure
|
Security for flexeval 0.18.3
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
No provenance found
None of the files in this release were uploaded with provenance metadata.
A note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
To learn more about protecting yourself from malicious actors, visit our package security documentation.
