Wassim Seifeddine

[ARCHIVE] Serving BERT Model with Pytorch using TorchServe

Author published on 2020-04-22

Finally

So finally Pytorch is getting a decent (?) production serving capabilities. TorchServe was introduced a couple of days ago along with other interesting things

I'm not in ANY way expert on putting pytorch in production environment. What I've been using is Flask. I have never tried ONNX or torchscript before to judge on.

So TorchServce was announced as a “industrial-grade path to deploying PyTorch models for inference at scale”. In this tutorial we will try to load a finetuned BERT model.

[ARCHIVE] Learning High School Physics with Tensorflow

Author published on 2018-05-28

## Introduction Learning to convert from Angular displacement to linear displacement was nothing fancy in highschool, but TEACHING a machine learning model how to do the translation is on a whole new level . Just kidding, it's also nothing fancy.

Task

What we want to do is learn the mapping between the angular rotation (complete circle) of a circular object given its diameter to the distance covered on a linear path

[ARCHIVE] Using Tensorflow Gradient Descent to minimize functions

Author published on 2016-10-02

You probably heard the hype around Gradient Descent & how it’s used to optimize Deep Learning Models. But you may never knew that it can be used to minimize ANY function you know.

In this blog post we’re going to see how to use Gradient Descent to optimize a simple quadratic function

we all know from highschool.

$ax^2 + bx + c$

with

a = 1
b = -40
c = 400

we get the following formula