The Keras progress bars look nice if you are training 20 epochs, but no one wants an infinite scroll in their logs of 300 epochs progress bars (I find it disgusting). It makes it difficult to get a sense of the progress of training, and it’s just bad practice (at least if you’re training from a Jupyter Notebook).
The alternative is to have a simple plot, with train and test loss, that updates every epoch or every n steps. It’s an extremely simple implementation and it’s much more useful and insightful.
So let’s get into it.
A Keras Callback is…
When coding machine learning, the same patterns usually emerge for every application, especially when everyone took the same courses and you can’t take much liberty with what library to use. That’s why these libraries include a multitude of features and options that may be unknown to us but solve traditional issues.
Keras callbacks are classes that execute functions at different steps of your epoch — either to log data, to plot it, or to save your model. These callbacks can be extremely useful, allowing you to code less and improving the effectiveness of your code.
Let’s dive in.
If you are like me, you spend a lot of time looking at stats, curves, and good moments to invest. Or maybe you are deciding what strategy to use that takes the minimum time possible.
Using the strategy implemented in this article, we will set up a notification system for Telegram, that takes this information and generates a curated message whenever it is time to invest or move assets.
Disclaimer: I set up a notification system because the broker I use for buying funds is not automatable. Otherwise, I would automate the investment of the money too.
Do you spend your time looking at your stocks, looking for opportunities or anomalies? Would you like to be at peace and forget about it and just get a notification when this things happen?
I don’t know about you, but I want to be at peace. I waste a lot of time looking at my curves, stats, predictions, and it makes me waste valuable time.
So I decided to make a notification bot that tells me when I should buy or sell (I don’t sell, because I believe in long term, but the strategy includes selling indicators)
In this article…
With the stock market “crash” in May 2020, I thought “Hey, I should be able to benefit from this situation”. So I decided to learn a bit about investing and the stock market to squeeze it from all its potential.
And so the research begins.
A lot of the information out there tries to tell you that you by yourself can beat the market. That you can learn how to value a company. That you can set up a day-trading strategy that will guarantee you earnings. …
How I developed a Streamlit Dashboard to calculate savings and rentability of new investments.
I recently made a case study of buying a new coffee machine to substitute my capsule one. I wanted a whole grain machine because it tastes better, and the results I found was that I would save a lot of money if I bought it.
So I bought it.
Talking with many people about our experience with the pandemic and the quarantine, it has been a recurring theme how a microscopic parasite has been able to shock our world as it has. The fact that millions of people have suddenly found themselves without the job they had worked so hard for, most of their plans for the future shaken or even destroyed.
But is COVID-19 an exception? Do we really have the control over our lives that we think we have?
The answer I have reached from my experience is that it is all an illusion. Most of the…
I am a Data Scientist. Let's get nerdy.