# Comprehensive introduction to active learning with Tensorflow and Keras

**URL:** <https://discuss.ai.google.dev/t/comprehensive-introduction-to-active-learning-with-tensorflow-and-keras/32647>\
**Category:** Show and Tell\
**Tags:** education, keras\
**Created:** [November 12, 2021, 3:29am UTC](https://discuss.ai.google.dev/t/comprehensive-introduction-to-active-learning-with-tensorflow-and-keras/32647 "2021-11-12T03:29:08Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![darshan\_deshpande](https://d1dlmcr85iqnpo.cloudfront.net/user_avatar/discuss.ai.google.dev/darshan_deshpande/32/6856_2.png) [@darshan\_deshpande](https://discuss.ai.google.dev/u/darshan_deshpande)\
**Post date:** [November 12, 2021, 3:29am UTC](https://discuss.ai.google.dev/t/comprehensive-introduction-to-active-learning-with-tensorflow-and-keras/32647/1 "2021-11-12T03:29:08Z")

</div>

Active Learning is an incremental process of learning. In this process, we initially annotate and train on a small subset of the unlabled data pool and then query the model for what data it would want to train on in the future. This is done iteratively till business metrics are met.

 ![dmNKusp](https://d3qe71uytubmmx.cloudfront.net/original/2X/a/ac8c348b5a62b8bde87c43dcce1612c2bf130e17.png)

In my recent tutorial on Keras, I implement a ratio-based sampling technique and demonstrate its usefulness on a toy, IMDB dataset. Some salient features of this tutorial include:

1. The active learning method of training achieved similar results as compared to a standard training loop while eliminating the unnecessary annotation of ~10,000 labels!
2. This method of sampling slightly balances out the false positives and false negatives which is beneficial for businesses that require balanced performance for both labels.

This tutorial serves as a comprehensive introduction to active learning with plenty of resources for beginners. If you’re interested then check it out.  
[Review Classification using Active Learning](https://keras.io/examples/nlp/active_learning_review_classification/)

---

<div class="post-metadata">

**Author:** ![Bhack](https://d1dlmcr85iqnpo.cloudfront.net/user_avatar/discuss.ai.google.dev/bhack/32/6388_2.png) [@Bhack](https://discuss.ai.google.dev/u/Bhack)\
**Post date:** [May 6, 2022, 2:18pm UTC](https://discuss.ai.google.dev/t/comprehensive-introduction-to-active-learning-with-tensorflow-and-keras/32647/4 "2022-05-06T14:18:14Z")

</div>

Thanks,  
if you are interested we have a CV thread at:

> **[Active learning and OOD · keras-team/keras-cv · Discussion #416](https://github.com/keras-team/keras-cv/discussions/416)**
>
> Is there already some mature component that we want to pick from the Active learning/OOD resesrch. This is a quite recent survey (not only CV): A Comparative Survey of Deep Active Learning /cc @qlz...