Commit d252638b authored by Jacky Lin's avatar Jacky Lin
Browse files


parent d8a21dd7
......@@ -16,18 +16,18 @@ For long, researchers have been interested in finding ways for AI to predict sto
The project can be broken down into 2 generic parts: stock prediction with Neural Network (LSTM), and an agent designed to operate on a simulated stock market (currently only 1 stock) to achieve max profit.
If you are interested in our source code, 3 folders that are worth looking into are :
1. [price_pred/](price_pred/), which focuses on using neural network to predict future prices. Especially the [](price_pred/, which contains the most important class that deals with price prediction NN. Here’s a graph of its result:
2. [stock_env/](stock_env/), which is mainly for creating a stock market environment for the q-learning agent. Especially the [](stock_env/, which defines a stock trading environment that works with the gym api.
3. [agent/](agent/), which deals with creating a q-learning agent to operate on the existing environment and try to achieve max profit. Especially the [](agent/, which is the main class that use the gym api to utilize reinforcement learning for operating and profiting in the designed stock environment. Here’s a graph of the result achieved by the agent.
Other folders includes:
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment