weaklysupervised

Weakly Supervised Learning

Description

This is a Python framework for weakly supervised learning. This package can be used for weak supervised learning classification tasks.

Installation

You can install it by pip method.

pip install weaklysupervised

API

We have implemented following weakly-supervised learning algorithm.

bootstrapping

Examples

from examples_utils import get_data, DNN
from sklearn import metrics
from weaklysupervised import BootstrappingNeuralNetworkClassifier

if __name__ == "__main__":
    X_train, X_test, y_train, y_test = get_data()
    DNN = DNN()
    clf = DNN.build_model(input_dim=30, output_dim=2)
    model = BootstrappingNeuralNetworkClassifier(clf, batch_size=128, epochs=40, bootstrapping_type="soft",
    beta=0.95, patience=5, best_model_name="model_check_point_best_model")
    model.fit(X_train, y_train, validation_data=(X_test, y_test), )
    predict = model.predict(X_test)
    acc = metrics.accuracy_score(y_test, predict)
    print("bootstrapping accuracy", acc)

Please see examples folder for more examples.

Metadata

Release files for weaklysupervised 0.0.5

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for weaklysupervised 0.0.5
File	Size	Uploaded
weaklysupervised-0.0.5.tar.gz	4.5 kB	Nov 23, 2020	Details

Release files / weaklysupervised-0.0.5.tar.gz

Download URL	weaklysupervised-0.0.5.tar.gz
Size	4.5 kB
Tags	Source
SHA-256 checksum How to use checksums	`aa8905c096f37c9449def0c569dd8b37e65e71dbad267750e993cce672168a3a`
BLAKE2b-256 checksum How to use checksums	`baa459ba38f40607988e731b584d2548e8d21a94c2431d0b13de0bcc795a0d96`
Upload date	Nov 23, 2020
Uploaded using Trusted Publishing? What is trusted publishing?	No
Uploaded via	`Python-urllib/3.6`