Found a great #deeplearning #ai community at facebook.com/groups/DeepNet… lot of advanced users out there who are (also) very supportive. has none of of the toxicity and gatekeeping of reddit/r/MachineLearning

Tutorials:Michael Nielsen ⿻ Andrew Trask Emil Wallner Projects:Hamel Husain Min Sang Kim @harvardnlp harshvardhan Gupta Alvira Swalin Ed Rushton Blogs:@EmmanuelAmeisen rtriangle Debarko De 🦁

Today I decided to make a symbolic #neuralnetwork framework inspired from TF + pytorch. Learned a lot. I hope to release something based on it as a tool to understand the inner workings of backprop + computational graphs

my very own automatic differentiation library. I decided to make this to get a deep understanding of neural networks. Will release to the public very soon #deeplearning #ai #neuralnetworks

most common neural net mistakes: 1) you didn't try to overfit a single batch first. 2) you forgot to toggle train/eval mode for the net. 3) you forgot to .zero_grad() (in pytorch) before .backward(). 4) you passed softmaxed outputs to a loss that expects raw logits. ; others? :)

As a team of programmers aiming to help #machinelearning engineers, this book in progress on ML from Jeremy Howard (fast.ai) + Terence Parr (USF) was a great refresher for us! Both Jeremy and Terence are looking for comments + suggestions! mlbook.explained.ai

was helped by fast.ai student Radek Osmulski about backprop. Got the solution barely a few minutes after first posting question. Thanks a ton and as always, the forums were of great help!

starting to write tutorials for my simple computation graph library github.com/harveyslash/sy… A learning aid to explain inner workings of neural networks.its like a 'code version' of Terence Parr, Jeremy Howard awesome "The matrix calculus you need for DL". contributors welcome

Wrote basic logistic regression tutorial harveyslash.github.io/sympyle/tutori… using my own nn library. feedback appreciated. #MachineLearning Also, matplotlib rocks:

We are investigating reports of service unavailability.

I have started experimenting with interactive plots using Plotly for my automatic differentiation library. quite cool. Draft is harveyslash.github.io/sympyle/tutori…

Working on an animation to visualise learning rates for simple functions. This is the progress so far. Planning to document popular ops (e.g. softmax,cross entropy) and maybe also different optimisation algorithms with my library #NeuralNetworks harveyslash.github.io/sympyle/

Visualising a 2d surface optimisation. using #d3js Looking for contributors to write interesting and new perspectives for #ML / #DL harveyslash.github.io/sympyle/

Really proud to share "What is torch.nn, really?", which takes you from a neural net written from scratch, refactored step by step using all the key concepts in `torch.nn`. If you want to really understand how neural nets work in PyTorch, start here! pytorch.org/tutorials/begi…

Here's a quick method to calculate the square root of a number (√y) by hand

Want tab-completion of (and direct attribute access to) your database tables and fields in Jupyter? OK then, here you go! :) github.com/jph00/fastsql

#drivezy Your zone members harrased me and forced me to enter paytm pin code to make payment. The agreed amount was 6000. After agreeing, they made an extra payment by forcefully using my fingerprint. I will take you to court for harassment

#drivezy members surrounded me and snatched my phone away. I was helpless and alone in a new city. They would not let me call my parents or even the car service center.

How can our book be out of stock before its release date?!?

矢印の方向に動いて見えますね～？ 人間の脳は単純なので矢印で惑わされてしまうのでーす