AutoRes Evaluator

※ 追加で実装して欲しい機能や質問などがあれば以下のissuesから投稿をお願いします

https://github.com/auto-res/autores-evaluator/issues/new

Examples

Prediction by logistic regression on the Titanic dataset
Prediction by CNN on CIFAR10

Architecture

ロゴ1

How to use

Install
- https://pypi.org/project/autoresevaluator/

pip install autoresevaluator

from autoresevaluator import AutoResEvaluator

Setting

# Hyperparameter setting
# Specify "type" and "args" for items to be searched in optuna.
params = {
    'lambda_l1': {'type': 'log_float', 'args': [1e-8, 10.0]},
    'lambda_l2': {'type': 'log_float', 'args': [1e-8, 10.0]},
    'num_leaves': {'type': 'int', 'args': [2, 256]},
    'feature_fraction': {'type': 'float', 'args': [0.4, 1.0]},
    'bagging_fraction': {'type': 'float', 'args': [0.4, 1.0]},
    'verbosity': -1
}

are = AutoResEvaluator(
    # dataset name
    dataset_name='titanic',
    # model file path
    model_path='/content/example.py',
    # parameter
    params=params,
    # Metrics you want to maximize/minimize
    valuation_index='roc_auc',
    # Where to store data
    datasave_path=None
    )

Execution

are.exec()

Output

result.log
- File to output the results
model_error.log
- File to write errors in model files

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

autoresevaluator-0.1.11.tar.gz (9.5 kB view details)

Uploaded Mar 5, 2024 Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

The dropdown lists show the available interpreters, ABIs, and platforms. Enable javascript to be able to filter the list of wheel files.

autoresevaluator-0.1.11-py3-none-any.whl (13.5 kB view details)

Uploaded Mar 5, 2024 Python 3

File details

Details for the file autoresevaluator-0.1.11.tar.gz.

File metadata

Download URL: autoresevaluator-0.1.11.tar.gz
Upload date: Mar 5, 2024
Size: 9.5 kB
Tags: Source
Uploaded using Trusted Publishing? No
Uploaded via: poetry/1.5.1 CPython/3.9.18 Darwin/23.2.0

File hashes

Hashes for autoresevaluator-0.1.11.tar.gz
Algorithm	Hash digest
SHA256	`0fcbc48d2464980c3b00c483e9e41dcae9f1ae25ea3feb7a1452c9d2b068d43d`
MD5	`5220d63c1c236469aa22c4d513a8ddf5`
BLAKE2b-256	`1d3197d4758fa5de16a3f5d92579b7964013ec16b203ba1550d45e689f785519`

See more details on using hashes here.

File details

Details for the file autoresevaluator-0.1.11-py3-none-any.whl.

File metadata

Download URL: autoresevaluator-0.1.11-py3-none-any.whl
Upload date: Mar 5, 2024
Size: 13.5 kB
Tags: Python 3
Uploaded using Trusted Publishing? No
Uploaded via: poetry/1.5.1 CPython/3.9.18 Darwin/23.2.0

File hashes

Hashes for autoresevaluator-0.1.11-py3-none-any.whl
Algorithm	Hash digest
SHA256	`bab7bcb02634c87117ef2c658db60776cba23636530dd9d4d8bc9ab115a3dd94`
MD5	`e1ac37d784bc16f5bf560808bc30bed8`
BLAKE2b-256	`9af5adb340cfab41b6bab2e4430f81a5b363296ef11ff6978de896292d7867bf`