Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/craftworksgmbh/mldocs
Documentation for the Machine Learning Process
https://github.com/craftworksgmbh/mldocs
documentation keras machine-learning sklearn
Last synced: about 1 month ago
JSON representation
Documentation for the Machine Learning Process
- Host: GitHub
- URL: https://github.com/craftworksgmbh/mldocs
- Owner: craftworksgmbh
- License: mit
- Archived: true
- Created: 2018-08-14T13:50:56.000Z (about 6 years ago)
- Default Branch: master
- Last Pushed: 2018-08-23T12:42:14.000Z (about 6 years ago)
- Last Synced: 2024-09-19T07:10:54.573Z (about 2 months ago)
- Topics: documentation, keras, machine-learning, sklearn
- Language: Python
- Homepage:
- Size: 22.5 KB
- Stars: 3
- Watchers: 2
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# Documenting the Machine Learning Process
Supports models following the Sklearn or Keras structure.
## Example Usage
```
from mldocs.documentation import Documentation
from sklearn.datasets import load_iris
from sklearn.model_selection import train_test_split
from sklearn.linear_model import LogisticRegression
from sklearn import metrics
df_x, df_y = load_iris(return_X_y=True)
x_train, x_test, y_train, y_test = train_test_split(df_x, df_y, random_state=0)
lg = LogisticRegression()
lg.fit(x_train, y_train)
doc = Documentation(x_train=x_train, y_train=y_train, x_test=x_test, y_test=y_test, model=lg, random_state=0,
metrics={'accuracy': (metrics.accuracy_score, {}), 'precision': (metrics.precision_score, {'average':'micro'})},
save_dir=SAVE_PATH, comment='this is an example usage', problem_kind='classification')
doc.document()
```
## Output
- Saves train and test data to specified directory as snapshot of training.
- Saves trained model.
- Saves settings/parameters/... to json (see below).
```
{
"timestamp": "14-08-2018_14-04-34",
"dataset": {
"test": {
"y_test": {
"one_hot_encoding": false,
"features": [
0
],
"class_frequencies": {
"0": 13,
"2": 9,
"1": 16
},
"n_classes": 3,
"n_rows": 38,
"n_features": 1
},
"x_test": {
"n_rows": 38,
"n_features": 4,
"features": [
0,
1,
2,
3
]
}
},
"train": {
"y_train": {
"one_hot_encoding": false,
"features": [
0
],
"class_frequencies": {
"0": 37,
"2": 41,
"1": 34
},
"n_classes": 3,
"n_rows": 112,
"n_features": 1
},
"x_train": {
"n_rows": 112,
"n_features": 4,
"features": [
0,
1,
2,
3
]
}
}
},
"model": {
"kind": "LogisticRegression",
"parameters": {
"dual": false,
"fit_intercept": true,
"verbose": 0,
"class_weight": null,
"max_iter": 100,
"random_state": null,
"solver": "liblinear",
"n_jobs": 1,
"intercept_scaling": 1,
"multi_class": "ovr",
"tol": 0.0001,
"C": 1.0,
"warm_start": false,
"penalty": "l2"
}
},
"performance": {
"metrics": {
"precision": 0.868421052631579,
"accuracy": 0.868421052631579
},
"pred_time_per_sample_in_sec": 3.9276323820415294e-06
},
"comment": "this is an example usage",
"problem_kind": "classification",
"random_state": 0
}
```