This site is built with fastpages, I use it as a page for paper summaries and hope it helps others as well.
Implementing an own LSTM RNN with PyTorch (Lesson 5 Part 6)
Apr 21, 2020
Using our originally Vision based methods to use them on NLP (Lesson 5 Part 5)
Imagenette and Training our classifier (Lesson 5 Part 3)
Transfer Learning, use a network pretrained on Imagenette on Imagewoof (Lesson 5 Part 4)
FP16 advanteges and GPU understanding (Lesson 5 Part 2)
Implementation of Mixup Augmentation and Label Smoothing as proposed in the Bag of Tricks for Image Classification paper (Lesson 5 Part 1)
Data Augmentation with Pillow and PyTorch (Lesson 4 Part 6)
Apr 20, 2020
Fastai Implementation of the Learner (Lesson 4 Part 5)
Apr 19, 2020
Fastai Progress Bar with the fastprrogress library, for better visualization during training (Lesson 4 Part 4)
Apr 18, 2020
Implementation of Optimizers from scratch, such as ADAM and friends as well as LAMB (Lesson 4 Part 3)
Apr 17, 2020
Fastai Datablock API from scratch (Lesson 4 Part 2)
Apr 16, 2020
LSUV, an initalization technique (Lesson 4 Part 1)
Apr 15, 2020
Normalizing Neural Networks to allow for better performance and faster convergence
Apr 14, 2020
Allowing the training of deeper Networks, than seen before.
Apr 13, 2020
Implementing Batch Norm, Layer Norm, Instance Norm, Group Norm and running batch norm (Lesson 3 Part 4)
Apr 12, 2020
