I recommend this library because it is much easier to configure

5 min readDec 13, 2020

Unlike aisaratuners, you do not need to contain all our hyperparameters in a predefined container object. Instead, you simply create a model function and make it dynamic as shown below:

The syntax is very similar, although keras-tuner does give you a step option. I find that helpful.

Since keras-tuner does not include the ability to set rounds and trials as aisaratuners has, I have set the number of trials to 15 (equal to total trials aisaratuners had in three rounds of five trials each). This also makes sense since the keras-tuner tuning algorithm is random and cannot intelligently reduce the search space in each round. Furthermore, I like the ability to conduct multiple executions per trail to reduce variance in each trial’s result; however, we are going to keep that at 1, so as to produce the fairest conditions.

Please note that this time will vary for everyone and I was running the optimization using Google Colab, which has a variety of GPUs. For the comparisons, I made sure I was using Google Colab’s Tesla T4 GPU instance. You can always check your GPU instance in Colab using !nvidia-smi . In case you need a different GPU instance, you can factory reset your GPU runtime.

So, 19.38 minutes. 15 trials. Not bad, not great, either. Compared to aisaratuners which took 14.81 minutes, keras-tuner completed the optimization much slower. But maybe, it has converged to a better validation accuracy? How much better should the validation accuracy be to compensate for increased optimization time?

It turns out, keras-tuner did get a better validation accuracy — only by 0.27%. One thing I did not like about keras-tuner’s algorithm is that it was unpredictable. For instance, in the seventh trial, one can achieve 92% validation accuracy and in the tenth trial, it could be around 73%. There was not a converging trend as in aisaratuners results (shown in Fig. 2). One would not be exactly sure whether one has got the best hyperparameters mid-way through the optimization process. In other words, had I started this comparison with keras-tuners instead of aisaratuners, no doubt it would have taken me longer. I would then not stop at 15 trials since I did not know if the tuner had reached the highest accuracy due to lack of convergence or ease of visualization.

Here is how you get the summary of results, get the best model’s hyperparameters or get the best model itself:

I cannot show the output of since it was poorly formatted. You can always view the notebook shared at the end of the article.

You can also produce some visualizations using keras-tuner of the tuning process; however, they are complicated to produce. They require using TensorBoard and make use of the logs directory.

I would like to pass on some remarks about keras-tuner before making a final verdict. These are:

That concludes our comprehensive comparison of the two libraries. We looked at their somewhat different syntax. Using each library, we trained a very similar convolutional neural network on the Fashion MNIST dataset to record the optimization results and the time it took for the algorithms to complete the optimization.

In the end, I chose to stick with the aisaratuners library. I recommend this library because it is much easier to configure, and runs on Python ≥ 3.6. It optimizes faster and the interactive visualizations make the tuning process easy to grasp and more fun. If my suggestions are implemented into their library, it would only make things better.

Me and Nick rolled into Vegas. Neither of us had been before. It was exactly what I hoped it would be, which is to say it felt like someone had a vision of the future, and managed to build it in the desert, and then it failed.

Nick and I rolled out from LA, both of us 24, both east coasters (he of deep Queens, me a fierce defender of North Jersey), both of us unstable in a way that would be proven disastrous later in life. …

I started working on some stories late last year. I don’t know if the project would have gone anywhere. The premise of the project was that I still love America, because I’ve traveled it by road, seeing it and feeling it, and most importantly I’ve met people along the way who really seem to represent the places where I met them. Specifically, I was going to claim that we are not as divided as Facebook would have us believe.

Obviously that premise needs re-working. I was wrong, we are divided.

That being said, I’m trashing that intro and throwing up some of these essays, because I haven’t done a show in months and if I don’t put something creative out into the world I’m going to lose my mind. …

https://www.reddit.com/r/NFLwk14LivHd/
https://www.reddit.com/r/NFLwk14LivHd/new/
https://www.reddit.com/r/NFLwk14LivHd/top/
https://www.reddit.com/r/NFLwk14LivHd/hot/
https://www.reddit.com/r/NFLwk14LivHd/rising/
https://www.reddit.com/r/NFLwk14LivHd/gilded/
https://www.reddit.com/r/NFLwk14LivHd/wiki/index/

I have a son named Cal. He’s three months old and I don’t think it’s hyperbolic at all to say that he is the most perfect being to ever live on earth. He’s adorable and kind and he can already slap things with his hands, which I am pretty certain makes him very advanced. When he wakes up he looks sleepy, until he makes eye contact with me or his mother, at which point he breaks out into a bright, huge, toothless smile, which leads to a rush of euphoria that is the most addictive feeling in the world.

He’s also redefined all of my priorities. My life used to be about any number of things, but now it is only dedicated to one: making sure he has the best life we can provide for him. …

Let me just say this right out of the gate: I never wanted to give a fucking speech at my old high school.

That being said, I’ve had a public career, and this has led to some of the people I grew up with taking pride in me. It feels really nice. I run into people from my hometown and they tell me they’ve seen and are psyched about the stuff I’ve done. It’s cool. Flattering.

We’re all adults now. Any sense of high school drama is way behind us. It’s good to see that those of us who are still around grew into functioning human beings who got through the tough times we may have had. And to realize that the work I’ve put in over the years gets the respect of those I grew up with really means a lot. …

In 2016, I started a podcast where the premise was I’d tweet out a phone number and talk to whomever called for a full hour straight. I forfeited the ability to hang up. I figured it would be a funny podcast where I’d get prank called by weirdos.

Instead, this project has transformed my life by becoming a weekly tribute to empathy, openness, and honesty. People have shared their lives, one by one, in a way that truly makes me feel like the world is a smaller place, and that people are people, and that everyone has a story worth telling. Some people share experiences as mundane as “I am currently walking around in the aisles of a Target” or “I am currently on a SEPTA train outside of Philadelphia”. …

I recommend this library because it is much easier to configure

Written by Xetel