me2too

Uh, no - it's precisely what happened. Being banned from US tech companies AND US payment networks implies all of those issues

Is this 2015?

Great write-up

Hey, thanks for the link - I added it to my reading list. From the introduction, it seems to fit the current situation (and that's not good).

From the outside, it looks like he's doing whatever he wants without control.

He starts to tell that everyone is maintained by the US - lol what?

He starts blaming Europe for being a parasite (thanks!) and treating Zelensky like everyone has seen live.

No one is telling him that his actions will have consequences for the US in their international credibility (gone) and relationships with partners (ruined)?

One day, he does a totally broken linear regression and applies no sense tariffs to every country in the world (no Russia - weird enough).

It's well known that tariffs are not a way to attract investment in your country - there's a lot of theory and historical examples that demonstrate that.

Moreover, it's just absurd to think that these kinds of actions are going to make American (!) corporations, now producing elsewhere, back in the US. No supply chain ready, lack of resources, ...

I don't know - it looks like a single man has too much power and no one around him is helping him make better decisions (for the country) in the long term. On the contrary, it looks like all the decisions are pretty much market manipulations in order to make some "friend" get some money.

But again, this is what it looks like from the outside - and it's quite worrying

He literally created the emergency himself. What the hell. I can't understand how one of the most powerful countries in the world doesn't have a system in place to prevent the president to act like a dictator

This man's absurdity never ceases to amaze me. How can his supporters appreciate all this - constant - chaos?

Moreover, from the outside (European here), it looks like the US president can literally do whatever he wants without being subject to regulations or facing any consequences - pretty much like a dictatorship. No discussion, no parliament, no opposition taken into account. Crazy.

He's just right

That's great

Good read, thanks. I found myself in this article having started to repair an old bicycle and bringing it back to life. I always thought that manual labor is more satisfactory than mental labor because you can immediately see the outcome and the value produced. Being a computer engineer I feel the need to disconnect myself from what I call "the abstract world" and get in touch with something real. Something I can touch.

That's what I try to do with my personal blog. When I find learn or discover something interesting I try to write an article about it. It helps me understanding if I do really understood the topic (and frequently it's not true), and it also is something really useful for my future self

After reading "Why we sleep" (Matthew Walker) and "Spark" (John Ratey) I went from drinking a couple of times a week (some beers, sometimes a cocktail) to zero.

Knowing the damage I was doing to myself, all the side effects and bad impact on my health (sleep, hearth, overall physical performance) literally stopped me from continuing with this. Moreover, in the same period the report the article "THE IMPACT OF ALCOHOL CONSUMPTION ON CARDIOVASCULAR HEALTH: MYTHS AND MEASURES"[1] from World Health Federation came out. TL; DR: the right amount of alcohol for anyone is precisely zero.

In short, a better knowledge of what I was doing to my body helped my in stopping a bad habit.

[1] https://world-heart-federation.org/wp-content/uploads/WHF-Po...

Great suggestion! I guess I'll write a followup with the convolution-based solution

rofl, high change I'll end up this way. But right now I'm excited about this approach :D

So far I solved 1,2, and 3. I start with the 4 today while I write the article on the solution for the puzzle #2.

Let's see if I'm able to face them all (that's also my first year that I join the AoC - so it's totally new for me)

That's Google AdSense auto ads (ML based placing of the ads in the page). I've set the level to "low" but I guess the number and positioning of ads are modulated by google using the info it has on you. I see only 3 ads, but maybe different users see different numbers of ads.

That's unfortunate

Author here. The goal of this challenges (and related articles) is to demonstrate how TensorFlow can be used as any other programming language. I am totally against of the usage of machine learning where's not required at all, and I hope to be able to solve all the puzzles without having to rely upon deep learning based solutions.

If in the other puzzles there's some optimization problem that can be expressed in a differentialble way, then I'd use ML for sure. But until there exist a deterministic solution, ML is just a waste (I say this as a ML researcher :) )

Author here. The goal of this challenges (and related articles) is to demonstrate how TensorFlow can be used as any other programming language. I am totally against of the usage of machine learning where's not required at all, and I hope to be able to solve all the puzzles without having to rely upon deep learning based solutions (but who knows, if there's something that can be easily expressed as an optimization problem having a language that's differentiable can help a lot).

wtf also a reddit user contacted me for telling me that. Do you have any idea on the reason and what can I do to avoid this censorship? The reddit user guessed because of the word "leone" in the domain name, but it is part of my surname, I can't change it :<

Yup! I realized it. I wrote that because in the past there was a huge problem with the operators, as I explained in this talk: https://pgaleone.eu/tf-function-talk/#slide=28

Even if now autograph is able to convert them correctly, I still prefer to have every operator explicitly converted whenever possibile. The loop, luckily, never had this transpilation problems

This is the website I used to learn Prolog and all I can do is to highly recommend it

https://www.tensorflow.org/api_docs/python/tf/contrib/distri...

Computer Vision tasks: I train models to to object detection, image classification, semantic segmentation... export the model, load the trained model in C++ and run inference.

Or generative models: same workflow (python definition and train. Export and use in other languages).

I use Tensorflow basically everyday. I use pytorch only when I read a model implemented in this framework to reimplement it in Tensorflow (in order to use all the tools I developed to simplify the train to shipping to production phases)

Point 1 is a good point, although tfdebug exists and it works.

For point 2: I don't see the problem. I use almost daily a wrapper (defined inside tf.contrib, that in version 2.x will go in core [I hope]) around the optimizer that in 2 lines allows me to distribute the training on multiple GPUs on the same machine

But the API is strictly statically typed in reality. The core is written in C++ and every operation requires a well-known type in input. As you can see from the error raised in this example: https://pgaleone.eu/tensorflow/go/2017/05/29/understanding-t...

But why? I do use Tensorflow form the 0.x release and I'm still using it right now, migrating all the codebase to the 2.x version.

I can do everything in both pytorch and tensorflow and when I have to define really efficient input pipelines (tf.data is a great thing), parallelize and distribute the training and export a trained model to production... With tensorflow everything is easier.

Moreover, pytorch in 1.x will have static-graph too, exactly like Tensorflow.

Both frameworks are converging to something really similar. I don't see a reason to switch (right now)

The fact that I have all the computation described in a coherent manner, in something that's agnostic to the language I'm using.

The same description (the graph) can be taken and used in every other language. I can move a trained model to production just picking a single file (a `tf.train.SavedModel` IIRC), give it a tag and I'm ready to go, with different models with a native support for the "tagging" (hence the model versioning).

At your contrary, I love the graph.

However, migrating to a keras-like approach, you can work thinking about "objects with variables inside" and then the graph will be built for you by the abstraction introduced by `tf.keras.Model`.

However, for automatic differentiation, graph is always required (as you can see from the example that uses eager)