doteval-datasets
Standard datasets for dotevals LLM evaluations.
Installation
pip install dotevals-datasets
Usage
Once installed, the datasets are automatically available in doteval:
from dotevals import foreach
@foreach.bfcl("simple")
def eval_bfcl(question: str, schema: list, answer: list):
# Your evaluation logic here
pass
@foreach.gsm8k("test")
def eval_gsm8k(question: str, reasoning: str, answer: str):
# Your evaluation logic here
pass
@foreach.humaneval()
def eval_humaneval(prompt: str, canonical_solution: str, test: str, entry_point: str):
# Your evaluation logic here
pass
@foreach.mmlu("test")
def eval_mmlu_all(question: str, subject: str, choices: list, answer: int):
# Your evaluation logic here
pass
@foreach.mmlu["college_mathematics"]("test")
def eval_mmlu_math(question: str, choices: list, answer: int):
# Your evaluation logic here
pass
@foreach.sroie("test")
def eval_sroie(image: Image, entities: dict):
# Your evaluation logic here
pass
Available Datasets
-
BFCL (Berkeley Function Calling Leaderboard): Tests function calling capabilities
- Variants:
simple,multiple,parallel - Columns:
question,schema,answer
- Variants:
-
GSM8K: Grade school math word problems
- Splits:
train,test - Columns:
question,reasoning,answer
- Splits:
-
HumanEval: Hand-written programming problems for code generation evaluation
- Columns:
prompt,canonical_solution,test,entry_point
- Columns:
-
MMLU: Massive Multitask Language Understanding across 57 academic subjects
- All subjects:
mmlu("test")- Columns:question,subject,choices,answer - Specific subject:
mmlu["college_mathematics"]("test")- Columns:question,choices,answer - Splits:
test,validation,dev
- All subjects:
-
SROIE: Scanned receipts OCR and information extraction
- Splits:
train,test - Columns:
image,address,company,date
- Splits:
Metadata
Release files for dotevals-datasets 0.8.0
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 | Actions |
|---|---|---|---|---|
| dotevals_datasets-0.8.0.tar.gz | 15.2 kB | Without provenance | Details for dotevals_datasets-0.8.0.tar.gz |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Actions Reset |
|---|---|---|---|---|
| dotevals_datasets-0.8.0-py3-none-any.whl | Python 3 | none | any | Details for dotevals_datasets-0.8.0-py3-none-any.whl |
Total release size: 24.8 kB
Release files / dotevals_datasets-0.8.0.tar.gz
| Download URL | dotevals_datasets-0.8.0.tar.gz |
|---|---|
| Size | 15.2 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
74b419258b82de9e1bf48aad2f7e2b8ac98e1fd753a1ee620d522d73cfe724f0
|
|
BLAKE2b-256 checksum How to use checksums |
83ff7a891484a0a95f18ce53278c554cc52eab4c85c6b103f7a8dcc6666f6c86
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
uv/0.8.14
|
Release files / dotevals_datasets-0.8.0-py3-none-any.whl
| Download URL | dotevals_datasets-0.8.0-py3-none-any.whl |
|---|---|
| Size | 9.6 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
6e075a76d419a76d786f5b3d5b59a6d7e23a29a92e6ae01d6877dbeb8f712c33
|
|
BLAKE2b-256 checksum How to use checksums |
5c0bf0ebac6632e00223a1229a0c28894f6cc9b47acc2aeae4823871d3c82b5d
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
uv/0.8.14
|
Security for dotevals-datasets 0.8.0
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.