1/ @werat asked about whether the debugger was using the C# compiler or language service in VS 2002. It was not. The debugger has a component called an ‘expression evaluator’ that is provided per language and is responsible for parsing and evaluating expressions when stopped at a

2/ breakpoint. For example, if you type into the immediate window, hover over a variable, type into the watch window, etc. the expression evaluator is involved. The debugger and the language service are actually deeply integrated in a number of scenarios in VS, which may

3/ initially seem surprising. I may talk about more of these scenarios in the future, but to give a flavor, when you set a breakpoint at design time the language service is involved, when you are using Edit and Continue the LS is involved, the range of what is being evaluated

4/ when you get a debugger data tip, completion lists when typing in the data windows or immediate window, etc. Prior to VS 2002, there were distinct debug environments tailored to each target customer (VB/VBA, C++, VJ, VBScript, TSQL, FoxPro, Fortran, etc.). The goals of the

5/ VS 2002 debugger included unifying the debugger backend components and providing a single UI stack that would enable new scenarios (for example, a callstack across languages). Expression evaluators are the architecture that separate the language specific pieces from the

Ever heard of Autoencoders?

The first time I saw a Neural Network with more output neurons than in the hidden layers, I couldn't figure how it would work?!

#DeepLearning #MachineLearning
Here's a little something about them: 🧵👇

Autoencoders are unsupervised neural networks whose architecture you can picture as two funnels connect from the narrow ends.

These networks are primary focus for compression tasks of data in Machine Learning.

We feed them the data so that they can learn the most important features, a smaller representation while keep the integrity of the data.

Later when someone needs, can just take that small representation and recreate the original, just like a zip file.📥

Being unsupervised, they require no labels.
Our inputs and outputs are same and a simple euclidean distance can be used as a loss function for measuring the reconstruction.

Of course, we wouldn't expect a perfect reconstruction.

We can think of an autoencoder having two components, encoder and decoder, represented by the below equations:

We are just trying to minimize the L here. All the backpropagation rules still hold.