johnkorn / distillation
Keras + tensorflow experiments with knowledge distillation on EMNIST dataset
☆35Updated 7 years ago
Related projects ⓘ
Alternatives and complementary repositories for distillation
- A machine learning experiment☆182Updated 7 years ago
- Keras implementation of temporal ensembling(semi-supervised learning)☆22Updated 5 years ago
- Transfer knowledge from a large DNN or an ensemble of DNNs into a small DNN☆29Updated 7 years ago
- Knowledge Distillation using Tensorflow☆142Updated 5 years ago
- Implementation of Learning Rate Finder, SGDR and Cyclical Learning Rate in Keras☆29Updated 6 years ago
- Focal Loss of multi-classification in tensorflow☆79Updated 5 years ago
- Bilinear CNN implementation using keras☆35Updated 6 years ago
- resnet_cifar10_cifar100_imagenet☆13Updated 6 years ago
- YOLO-v2, ResNet-32, GoogLeNet-lite☆36Updated 5 years ago
- Focal Loss implementation by Keras with TensorFlow backend☆42Updated 6 years ago
- Cyclic learning rate TensorFlow implementation.☆66Updated 5 years ago
- wrapping a keras optimizer to implement gradient accumulation☆119Updated 4 years ago
- It contains the Attention-56 and Attention-96 models built from scratch in keras. Residual Attention Networks are described in the paper …☆38Updated 6 years ago
- pytorch implement of Label Smoothing☆32Updated 4 years ago
- Keras implementation of AutoAugment.☆30Updated 5 years ago
- Multi heads attention for image classification☆81Updated 6 years ago
- Implementation of RetinaNet in TensorFlow☆25Updated 6 years ago
- Keras implementation of Octave Convolutions☆53Updated 5 years ago
- Try to use tf.estimator and tf.data together to train a cnn model.☆79Updated 6 years ago
- RAdam optimizer for keras☆71Updated 5 years ago
- The codes for recent knowledge distillation algorithms and benchmark results via TF2.0 low-level API☆108Updated 2 years ago
- Implementation in Keras of: Snapshot Ensembles: Train 1, get M for free (https://arxiv.org/abs/1704.00109)☆25Updated 6 years ago
- This repo is an implementation of quantized CNN for both weights (1-bit compression) and feature maps (2-bit compression).☆18Updated 6 years ago
- Focal Loss for multi-class classification☆55Updated 6 years ago
- ShuffleNet Implementation using Keras Functional Framework 2.0☆77Updated 4 years ago
- Model Compression Based on Geoffery Hinton's Logit Regression Method in Keras applied to MNIST 16x compression over 0.95 percent accuracy…☆62Updated 5 years ago
- Tensorflow version implementation of focal loss for binary and multi classification☆108Updated 5 years ago
- Multi-class classification with focal loss for imbalanced datasets☆82Updated 5 years ago
- keras ImageDataGenerator example☆61Updated 7 years ago
- Knowledge distillation implemented in TensorFlow☆19Updated 7 years ago