ktrain 0.1.4

ktrain is a lightweight wrapper for Keras to help train neural networks

ktrain

ktrain is a lightweight wrapper for the deep learning library Keras to help build, train, and deploy neural networks. With only a few lines of code, ktrain allows you to easily and quickly:

• estimate an optimal learning rate for your model given your data using a Learning Rate Finder
• utilize learning rate schedules such as the triangular policy, the 1cycle policy, and SGDR to effectively minimize loss and improve generalization
• employ fast and easy-to-use pre-canned models for both text classification (e.g., NBSVM, fastText, logreg) and image classification (e.g., ResNet, Wide ResNet, Inception)
• load and preprocess text and image data from a variety of formats
• inspect data points that were misclassified to help improve your model
• leverage a simple prediction API for saving and deploying both models and data-preprocessing steps to make predictions on new raw data

Tutorial Notebooks

Please see the following tutorial notebooks for a guide on how to use ktrain on your projects:

• Tutorial 1: Introduction
• Tutorial 2: Tuning Learning Rates
• Tutorial 3: Image Classification
• Tutorial 4: Text Classification
• Tutorial A1: Additional tricks, which covers topics such as examining misclassifications, inspecting intermediate output of Keras models for debugging, and built-in callbacks.

Tasks such as text classification and image classification can be accomplished easily with only a few lines of codes.

Example: Classifying Images of Dogs and Cats Using ktrain

import ktrain
from ktrain import vision as vis
(train_data, val_data, preproc) = vis.images_from_folder(
                                              datadir='data/dogscats',
                                              data_aug = vis.get_data_aug(horizontal_flip=True),
                                              train_test_names=['train', 'valid'], 
                                              target_size=(224,224), color_mode='rgb')
model = vis.image_classifier('pretrained_resnet50', train_data, val_data, freeze_layers=80)
learner = ktrain.get_learner(model=model, train_data=train_data, val_data=val_data, 
                             workers=8, use_multiprocessing=False, batch_size=64)
learner.lr_find()             # briefly simulate training to find good learning rate
learner.lr_plot()             # visually identify best learning rate
# implicitly employs triangular learning rate policy with ReduceLROnPlateau, ModelCheckpoint, and EarlyStopping
learner.autofit(1e-4, checkpoint_folder='/tmp') 

Requirements

The following software/libraries should be installed:

• Python 3.6+ (tested on 3.6.7)
• Keras (tested on 2.2.4)
• TensorFlow (tested on 1.10.1)
• scikit-learn (tested on 0.20.0)
• matplotlib (tested on 3.0.0)
• pandas (tested on 0.24.2)

This code was tested on Ubuntu 18.04 LTS using Keras 2.2.4 with a TensorFlow 1.10 backend. There are a few portions of the code that may explicitly depend on TensorFlow, but such dependencies are kept to a minimum.

---

Creator: Arun S. Maiya

Email: arun [at] maiya [dot] net