A man's journey through machine learning
The second set of tutorials are titled “Research and Experimentation”. They cover eager execution (which is default in TFv2), Automatic differentiation, custom training, and custom layers.
Eager Execution and Automatic Differentiation: Research-EagerExecution.ipynb
Custom Training, Layers, and Testing: Research-EagerExecution.ipynb
I am going to create a series of myself going through the TensorFlow training. My ultimate goal is to be better using the framework and expand beyond RL.
Here are my notebooks going through classifications from the TensorFlow learning page.
Overfitting: TensorFlow/Overfitting.ipynb
After getting shot down for a video creation job I decided the best way to fix that was to start creating videos for YouTube. They won’t be as professional but they will help me get used to being on camera and using the editing tools.
For my tools, I used PowerPoint for my slides, OBS Studio for video capture, and Lightworks to create the videos. I have never used OBS or Lightworks before so there was a slight learning curve.
OBS was a little clunky because I wasn’t able to use my secondary monitor. Had my primary video card had both monitors plugged in it wouldn’t have been an issue. So, my work around was to swap my screens in PowerPoint and hide the start bar.
I had used Camtasia for my interview process and Lightworks was very similar. At least enough that I could figure it out. Plus, (big surprise) there are TONS of online videos on how to use it.
What a great time to be alive. With very little money I am able to create blog posts, publish code, create videos, interact online, and many more things to demonstrate my abilities. I don’t need to have an actual job to get experience.
I am going to create a series of myself going through the TensorFlow training. My ultimate goal is to be better using the framework and expand beyond RL.
Here are my notebooks going through classifications from the TensorFlow learning page.
Basic Classification: TensorFlow/BasicClassification.ipynb
Text Classification: TensorFlow/TextClassification.ipynb
Regression: TensorFlow/Regression.ipynb
I am going to create a series of myself going through the TensorFlow training. My ultimate goal is to be better using the framework and expand beyond RL.
First, I had to create a new environment (tf_current) so that I won’t break my other projects that are NOT using TensorFlow 1.12. To do that I ran the following commands from the Anaconda 3.3 prompt:
conda create -n tf_current python=3.6
After that I had to download the latest TensorFlow version:
pip install tensorflow
I had to install a plugin that would allow the Juypter Notebook to change kernels
conda install nb_conda
Once I did that I had new entries in the start menu for both (root) and (tf_current). But, since Windows hates me the links don’t work because the length of the shortcut text. It doesn’t matter because I can load the normal Jupyter Notebook and when I select “New” I can select the tf_current environment.
In my previous blog post I showed how I upgraded my Black Scholes/Monte Carlo notebook to use TensorFlow v2. Today, I am going to show how I was EXTREMELY easily able to convert DDQN to the pre release of TensorFlow v2.
The notebook is located here: DDQN-TFv2.ipynb
Since I was mostly using Keras there were a few library changes but the code ran pretty much as is.
In my previous blog post I showed how to use TensorFlow to price options using Black Scholes and running a Monte Carlo Simulation. Today, I am going to show how I was able to convert that code to the pre release of TensorFlow v2.
The notebook is located here: MonteCarloBlackScholes-TF2.ipynb
The largest change was moving from using Sessions to using Functions with the addition of eager execution. There were some stumbling blocks but once I went through a few changes it all made sense and should be easy for any future upgrades.
