hub / github.com/AI4QTrading/R2-SAC

github.com/AI4QTrading/R2-SAC @main

88 symbols 218 edges 13 files 27 documented · 31%

What it actually does AI analysis from the code graph — generated when you open this

loading…

README

R2-SAC

a novel framework for stock portfolio trading that employs a 'Relaxation and Refinement' strategy to boost the Soft Actor-Critic (SAC) agent 1. PLEASE NOTE: Find the basic version of the Hawkes scripts, which can be founded in the https://github.com/HongtengXu/PoPPy/

TCN_GAT_zz1000.py is the training program for TCN and GAT, which introduced the scripts model.py for TCN model and gat.py for GAT model.
SAC_zz1000.py is the traning program for SAC model, which introduced the scripts StockEnv_zz1000.py for trading environment and StcokAgent.py for agent model.
To run R2-SAC, you should rewrite the test procedure for your trading strategy and get the hawkes scripts from the https://github.com/HongtengXu/PoPPy/.
For some suggetions, we built AI4QTrading-patch-1 branch to help reproducing the strategy. And for commercial reason, some data need to be downloaded from the public data source.

Core symbols most depended-on inside this repo

append

called by 15

replay_memory_sacduiqi.py

to_tensor

called by 10

stock_dataset_havkes_zz1000.py

replay_memory_sacduiqi.py

save

called by 3

replay_memory_sacduiqi.py

recover_data

called by 2

stock_dataset_havkes_zz1000.py

Shape

Method 59

Class 15

Function 14

Languages

Python100%

Modules by API surface

stock_dataset_havkes_zz1000.py12 symbols

model.py12 symbols

StockModel.py12 symbols

replay_memory_sacduiqi.py11 symbols

gat.py9 symbols

StockEnv_zz1000.py9 symbols

utils.py7 symbols

util.py5 symbols

StockAgent.py5 symbols

TCN_GAT_zz1000.py4 symbols

SAC_zz1000.py2 symbols

For agents

$ claude mcp add R2-SAC \
  -- python -m otcore.mcp_server <graph>

⬇ download graph artifact