数据分析和机器学习实验框架
项目描述
实验赛跑者（exp-runner）
exp-runner是一个简单且可扩展的 Python 数据分析和机器学习实验框架。
结构
该框架包括以下步骤：
- 数据加载
- 数据转换
- 模型训练和测试
- 绩效评估
- 结果保存
主要特点
- 可生成性：支持多种模型和方法，可用于多种任务（如预处理、降维、分类、回归、聚类、统计测试等）
- 灵活性：可以轻松跳过和/或包含步骤
- 动态加载：在运行时自动导入模块 - 不需要额外的行
安装
pip install exp-runner
用法
假设您的项目具有以下结构：
MyAwesomeProject/
main.py
my_custom_module.py
data/
data_00.npy
data_01.npy
...
data_NN.npy
protocols/
experiment_config.json
results/
只要给我一个代码！
您只需要在JSON配置文件中描述您的框架：
实验配置.json
{
"Setup": {
"description": "You can add detailed description of the experiment",
"random_seed": 42
},
"Dataset": {
"class": "my_custom_module.MyAwesomeDataLoader",
"args": {"path_to_data": "data/*.npy"}
},
"Transforms": [
{
"class": "sklearn.decomposition.PCA",
"args": {"n_components": 3, "whiten": true}
}
],
"Model": {
"class": "sklearn.cluster.KMeans",
"args": {"n_clusters": 3, "n_jobs": -1, "verbose": 0}
},
"Metric": {
"class": "my_custom_module.SklearnMetricWrapper",
"args": {"metric": "normalized_mutual_info_score"}
},
"Saver": {
"class": "my_custom_module.CSVReport",
"args": {"path_to_output": "results/evaluation_results.csv", "sep": ";"}
}
}
以下是前面提到的类（点击）：
my_custom_module.py
import os
import glob
import numpy as np
import sklearn.metrics
from exp_runner import Dataset, Metric, Saver
from collections import defaultdict
from typing import Any, Dict, List, Union, NoReturn, Iterable, Callable
from sklearn.model_selection import StratifiedShuffleSplit
class MyAwesomeDataLoader(Dataset):
def __init__(self, path_to_data: str, test_size: float = 0.1, training: bool = True):
super(MyAwesomeDataLoader, self).__init__()
self._samples = dict()
self._labels = dict()
self._splits = defaultdict(dict)
paths_to_data = glob.glob(path_to_data)
for path in paths_to_data:
fname = os.path.basename(path)
data = np.load(path)
X = data[:, :-1]
y = data[:, -1]
indices_train, indices_test = next(StratifiedShuffleSplit(
test_size=test_size
).split(X, y))
self._samples[fname] = X
self._labels[fname] = y
self._splits[fname]['train'] = indices_train
self._splits[fname]['test'] = indices_test
self._indices = list(self._samples.keys())
self._training = training
def __getitem__(self, index: int) -> Dict[str, Dict[str, Union[str, np.ndarray]]]:
if not (0 <= index < len(self._indices)):
raise IndexError
fname = self._indices[index]
item = {
'X': self._samples[fname][self._splits[fname]['train'] if self.training else self._splits[fname]['test']],
'y': self._labels[fname][self._splits[fname]['train'] if self.training else self._splits[fname]['test']]
}
item['desc'] = 'it is possible to add description for each data sample'
return {'filename': fname, 'item': item}
def __len__(self) -> int:
return len(self._indices)
@property
def training(self):
return self._training
class SklearnMetricWrapper(Metric):
def __init__(self, metric: str):
super(SklearnMetricWrapper, self).__init__()
metric = getattr(sklearn.metrics, metric)
self._metric: Callable[[Iterable[Union[float, int]], Iterable[Union[float, int]]], float] = metric
def __call__(self, y_true: Iterable[Union[float, int]], y_pred: Iterable[Union[float, int]]) -> float:
return self._metric(y_true, y_pred)
class CSVReport(Saver):
def __init__(self, path_to_output: str, sep: str = ';', append: bool = True):
super(CSVReport, self).__init__()
self.path_to_output = path_to_output
self.sep = sep
self.mode = 'a+' if append else 'w+'
def save(self, report: List[Dict[str, Any]]) -> NoReturn:
with open(self.path_to_output, self.mode) as csv:
for entry in report:
line = self.sep.join([
entry['filename'],
entry['desc'],
entry['perf']
]) + '\n'
csv.write(line)
最后，要在终端中运行您的实验类型：
cd /path/to/MyAwesomeProject
python main.py --config protocols/experiment_config.json
关
exp_runner -0.1.0b2-py3-none-any.whl 的哈希值
| 算法 | 哈希摘要 | |
|---|---|---|
| SHA256 | ed7118cada6a5f2dcc74ac84485bdddc0fc988514a18f7fac905da42e822b202 |
|
| MD5 | 5cbca2ebb141bb49036031448c596be6 |
|
| 布莱克2-256 | 3a5ca9af5166e6c2dbf84e945b248409494a0f33e59bb4ac0d500203ccb38b1d |