This is a toy reinforcement learning agent trained on the google's stock data from march 2014 to november 2017. The data is available here: https://www.kaggle.com/datasets/borismarjanovic/price-volume-data-for-all-us-stocks-etfs
I have used raw Deep Q learning algorithm and a custom Environment along side Chainer for the neural network layers.
https://drive.google.com/file/d/1YaWKXp_5Jn3sQG1PXXpybU80V6meWDup/view?usp=sharing