FinRL 入门指南
Open-Source software developers and programmers are the backbone of today’s humanity and humanitarian technology.
FinRL provides an open-source ecosystem that features Deep Reinforcement Learning in finance for all-level users.
Entry-level users
FinRL provides demonstrative and educational materials to help beginners such as students and entry-level professionals to walk through the DRL for finance pipeline.
Intermediate-level users
FinRL provides lightweight and scalable DRL algorithms with finance-oriented optimizations for full-stack developers and professionals.
Advanced-level users
FinRL delivers cloud-native solutions with high performance and high scalability training for investment banks and hedge funds.
Cloud-Native Solution
FinRL-Podracer can obtain a profitable trading agent in 10 minutes on an NVIDIA DGX SuperPOD cloud with 80 A100 GPUs, for a stock trend prediction task on NASDAQ-100 constituent stocks with minute-level data over 5 years.
Market Simulations
Reduce data processing burden; Reduce simulation-to-reality gap; Provide benchmark performance
OUR COLLABORATORS
AI4Finance-Foundation
AI community has accumulated an open-source code ocean over the past decade. We believe applying these intellectual and engineering properties to finance will initiate a paradigm shift FinRL 入门指南 from the conventional trading routine to an automated machine learning approach, even RLOps in finance.
Welcome to FinRL Library!¶
AI4Finance community provides this demonstrative and educational resource, in order to efficiently automate trading. FinRL is the first open source FinRL 入门指南 framework for financial reinforcement learning.
Reinforcement learning (RL) trains an agent to solve tasks by trial and error, while DRL uses deep neural networks as function approximators. DRL balances exploration (of uncharted territory) and exploitation (of current knowledge), and has been recognized as a competitive edge for automated trading. DRL framework is powerful in solving dynamic decision making problems by learning through interactions with an unknown environment, thus exhibiting two major advantages: portfolio scalability and market model independence. Automated trading is FinRL 入门指南 essentially making dynamic decisions, namely to decide where to trade, at what price, and what quantity, over a highly stochastic and complex stock market. Taking many complex financial factors into account, DRL trading agents build a multi-factor model and provide algorithmic trading strategies, which are FinRL 入门指南 difficult for human traders.
FinRL provides a framework that supports various markets, SOTA DRL algorithms, benchmarks of many quant finance tasks, live trading, etc.
Join or discuss FinRL with us: AI4Finance mailing list.
Feel free to leave us feedback: report bugs using Github issues or discuss FinRL development in the Slack Channel.
手把手教你用强化学习框架做量化交易：FinRL
Logging to tensorboard_log/td3\td3_1
---------------------------------
| time/ | |
| episodes | 4 |
| fps | 44 |
| time_elapsed | 48 |
| total timesteps | 2160 |
| train/ | |
| actor_loss | 253 |
| critic_loss | 2.28e+04 |
| learning_rate | 0.001 |
| n_updates | 1620 |
---------------------------------
---------------------------------
| time/ | |
| episodes | 8 |
| fps | 32 |
| time_elapsed | 133 |
| total timesteps | 4320 |
| train/ | |
| actor_loss | 96.3 |
| critic_loss | 1.24e+04 |
| learning_rate | 0.001 |
| n_updates | 3780 |
---------------------------------
day: 539, episode: 10
begin_total_asset: 1000000.00
end_total_asset: 2336606.29
total_reward: 1336606.29
total_cost: 1024.01
total_trades: 7568
Sharpe: 1.453
量化投资的强化学习神器！FinRL 入门指南
(可选1) 如果你用Python的目的是数据分析，可以直接安装Anaconda：Python数据分析与挖掘好帮手—Anaconda，它内置了Python和pip.
(可选2) FinRL 入门指南 此外，推荐大家用VSCode编辑器来编写小型Python项目：Python 编程的最好搭档—VSCode 详细指南
请注意 Python 版本要大于等于 3.7。此外，如果你的当前Python环境下安装了 zipline，请 pip uninstall 掉 zipline，因为Zipline与FinRL不兼容。