Why mixing 'closing prices' of 3 different companies?

FutureSharks / ml-finance

💰 A collection of finance related machine learning examples

GNU General Public License v3.0

97 stars 21 forks source link

Why mixing 'closing prices' of 3 different companies? #18

Open hellojinwoo opened 3 years ago

hellojinwoo commented 3 years ago

First of all, I would like to thank you for organizing and sharing such a nice repo, Max williams. I would appreciate it if you can answer one small question.

Question

What is the point of using the .flatten() method in the deep-q-learning-agent.ipynb file in the Example 0? (The screenshot is provided below)
It mixes the 'closing prices' of 3 different companies (msft, ibm, qcom), so I cannot see why you flattened the train_data, which is (3*3526) numpy.ndarray and plotted it.

FutureSharks commented 3 years ago

Hi @hellojinwoo 🙂

What is the point of using the .flatten() method

You're right, flattening these 3 columns was probably not a good idea. I was just to get an idea of what the data looks like roughly, not used in training AFAIK.

Feel free to make a PR!