All Projects → bshaw2019 → Crypto_trader

bshaw2019 / Crypto_trader

Licence: mit
Q-Learning Based Cryptocurrency Trader and Portfolio Optimizer for the Poloniex Exchange

Programming Languages

python
139335 projects - #7 most used programming language

Projects that are alternatives of or similar to Crypto trader

crypto-database
Database for crypto data, supporting several exchanges. Can be used for TA, bots, backtest, realtime trading, etc.
Stars: ✭ 72 (-60.87%)
Mutual labels:  technical-analysis, poloniex
Chatbot cn
基于金融-司法领域(兼有闲聊性质)的聊天机器人,其中的主要模块有信息抽取、NLU、NLG、知识图谱等,并且利用Django整合了前端展示,目前已经封装了nlp和kg的restful接口
Stars: ✭ 791 (+329.89%)
Mutual labels:  reinforcement-learning, sentiment-analysis
Machine Learning From Scratch
Succinct Machine Learning algorithm implementations from scratch in Python, solving real-world problems (Notebooks and Book). Examples of Logistic Regression, Linear Regression, Decision Trees, K-means clustering, Sentiment Analysis, Recommender Systems, Neural Networks and Reinforcement Learning.
Stars: ✭ 42 (-77.17%)
Mutual labels:  reinforcement-learning, sentiment-analysis
Jericho
A learning environment for man-made Interactive Fiction games.
Stars: ✭ 173 (-5.98%)
Mutual labels:  reinforcement-learning
Pytorch sac
PyTorch implementation of Soft Actor-Critic (SAC)
Stars: ✭ 174 (-5.43%)
Mutual labels:  reinforcement-learning
Gail Tf
Tensorflow implementation of generative adversarial imitation learning
Stars: ✭ 179 (-2.72%)
Mutual labels:  reinforcement-learning
Awesome Game Ai
Awesome Game AI materials of Multi-Agent Reinforcement Learning
Stars: ✭ 185 (+0.54%)
Mutual labels:  reinforcement-learning
Elf
An End-To-End, Lightweight and Flexible Platform for Game Research
Stars: ✭ 2,057 (+1017.93%)
Mutual labels:  reinforcement-learning
Promp
ProMP: Proximal Meta-Policy Search
Stars: ✭ 181 (-1.63%)
Mutual labels:  reinforcement-learning
Pytorch Sentiment Neuron
Stars: ✭ 178 (-3.26%)
Mutual labels:  sentiment-analysis
Tensorflow Rl
Implementations of deep RL papers and random experimentation
Stars: ✭ 176 (-4.35%)
Mutual labels:  reinforcement-learning
Atari
AI research environment for the Atari 2600 games 🤖.
Stars: ✭ 174 (-5.43%)
Mutual labels:  reinforcement-learning
Simple rl
A simple framework for experimenting with Reinforcement Learning in Python.
Stars: ✭ 179 (-2.72%)
Mutual labels:  reinforcement-learning
Multimodal Sentiment Analysis
Attention-based multimodal fusion for sentiment analysis
Stars: ✭ 172 (-6.52%)
Mutual labels:  sentiment-analysis
Deeprl Agents
A set of Deep Reinforcement Learning Agents implemented in Tensorflow.
Stars: ✭ 2,149 (+1067.93%)
Mutual labels:  reinforcement-learning
Weibospider sentimentanalysis
借助Python抓取微博数据,并对抓取的数据进行情绪分析
Stars: ✭ 173 (-5.98%)
Mutual labels:  sentiment-analysis
Retro Learning Environment
The Retro Learning Environment (RLE) -- a learning framework for AI
Stars: ✭ 180 (-2.17%)
Mutual labels:  reinforcement-learning
Deep Algotrading
A resource for learning about deep learning techniques from regression to LSTM and Reinforcement Learning using financial data and the fitness functions of algorithmic trading
Stars: ✭ 173 (-5.98%)
Mutual labels:  reinforcement-learning
Machine Learning And Reinforcement Learning In Finance
Machine Learning and Reinforcement Learning in Finance New York University Tandon School of Engineering
Stars: ✭ 173 (-5.98%)
Mutual labels:  reinforcement-learning
Reinforce.jl
Abstractions, algorithms, and utilities for reinforcement learning in Julia
Stars: ✭ 178 (-3.26%)
Mutual labels:  reinforcement-learning

Crypto_Trader

A Live Machine-Learning based Cryptocurrency Trader for the Poloniex Exchange

If you would like to follow our progress or reach out to developers, feel free to join our discord channel at: https://discord.gg/UPNV2fH

The goal of the project is to create an open sourced, machine learning based cryptocurrency portfolio optimizer including as many relevant variables as possible.

These will include:

  • Community Sentiment (such as forum and twitter sentiment analysis)
  • Analyst Opinions (notable twitter account/analyst feeds with a history of winning strategies - ex. "Haejin" a successful elliot wave trader with an established history)
  • Global Economic Indexes (DOW, Nikkei225, FOREX data, etc.)
  • Open/Close/High/Low/Volume
  • Live Orderbook Data (spread, bid, ask, order size)
  • Technical Indicators (MACD's, MA's, VWAPS, etc.)
  • Any Other Variable of interest found to be important (Please feel free to brainstorm and send suggestions)

A Boruta Analysis (variation of Random forest) will be used to reduce these input variables by removing "unimportant" features.

Using this input dataset, a machine learning Binary Classifier will be created to assign trading pairs on Poloniex a confidence score from 0 to 1. This value will represent the confidence that the trading pair will increase over a certain timeframe. Multiple machines may be constructed to score for a 5 minute window, 10 minute, 15 minute, 30 minute, 1 hour, etc.

Using these scores, a Q-Learning Bot ("reinforcement learning") will be created that will optimize a trading strategy based on the binary classifier scores. The machine will read the amount of capital in the users Poloniex Account, and automatically place trades to optimize the portfolios holdings. These strategies will use stop losses and sell limits. Because it is q learning based, the machine will receive, and use live data to make decisions in a live auto-updating context with a reward that optimizes profit. This will allow the machine to continue to train and optimize itself over time while feeding in live data and placing trades.

Ultimately, the program will have the ability to be run 24 hours a day while optimizing a portfolio using live data, while taking into account fees in order to identify and take advantage of all trading opportunities accross the entire cryptocurrency market on Poloniex. Because of the decimal based system of cryptocurrencies, in theory a portfolio of any size should be able to be used as long as the user has the minimum trade size on Poloniex.

Major Design Features Purpose
Identified Variables Related to Price Action
Data Scraper Live Variables to Array
Binary Classifers Score Trading Pairs Live
Q-Learning Bot Optimize Trade Strategy
Poloniex API Link Allow Bot to Make Trades

If you would like to donate to the project, please do so at the following Bitcoin/Litecoin/Ethereum addresses. All donations appreciated :)

DONATIONS

Currency Address
Bitcoin 1GjVgMUDfKHzhxgeauRagVfp1GCrSJXijb
Litecoin 0x9852389Bd431A90A9AEcb48EdA50Da1ac05Bd4d8
Ethereum M9oJaUnCB6Soistk3wSETziFDzz8gAaJCU
Note that the project description data, including the texts, logos, images, and/or trademarks, for each open source project belongs to its rightful owner. If you wish to add or remove any projects, please contact us at [email protected].