Machine Learning

How skip connections changed Deep Learning

Skip connections revolutionized Deep Learning allowing for the first time to cross the 100 layer barrier.

Posted

#Machine Learning

Concise Lecture Notes - Lesson 7 | Fastai v3 (2019)

CNNs, U-nets, GANs, Feature losses and RNNs.

Posted

#Machine Learning #fastai notes

Concise Lecture Notes - Lesson 6 | Fastai v3 (2019)

Dropout, Data Augmentation, Batch Normalizations, Convolutions and most importantly data ethics.

Posted

#Machine Learning #fastai notes

Concise Lecture Notes - Lesson 5 | Fastai v3 (2019)

Embedding matrices, Collaborative Filtering, Regularization, weight decay and Adam Optimizer.

Posted

#Machine Learning #fastai notes

Concise Lecture Notes - Lesson 4 | Fastai v3 (2019)

Creating language models and text classifiers, fastai for tabular data, collaborative filtering, deeper dive into structure of Deep Neural Networks.

Posted

#Machine Learning #fastai notes

Concise Lecture Notes - Lesson 3 | Fastai v3 (2019)

Datablock API to prepare datasets, Image Segmentation, Image Regression and NLP.

Posted

#Machine Learning #fastai notes

Concise Lecture Notes - Lesson 2 | Fastai v3 (2019)

Creating Deep Learning datasets, learning rates, making model production ready, debugging general issues and SGD.

Posted

#Machine Learning #fastai notes

Concise Lecture Notes - Lesson 1 | Fastai v3 (2019)

how to use the fastai library to quickly get a model up and running with state of the art results.

Posted

#Machine Learning #fastai notes

Logistic Regression: Flag bearer of Classification

Logistic Regression is actually one of the many popular Classification Algorithms. Classification being the act of grouping an example or sample into one of two or more classes.

Posted

#Machine Learning

Method behind Madness: Linear Regression - Part 2

In Simple Linear Regression, where we had only two variables, made a 2D plot. The "best fit" linear model there was a line. In case of three variables, the points are gonna make a 3D plot. In this case, the "best fit" linear model is a plane! What about 'n' variables and hence 'n' dimensions?

Posted

#Machine Learning