Free, open source crypto trading bot
Go to file
2017-12-27 17:39:54 +02:00
bin force to python3 2017-11-08 23:39:29 +01:00
freqtrade more unit tests for backtesting 2017-12-27 17:39:54 +02:00
scripts More lint fixes (#198) 2017-12-18 17:36:00 +01:00
.coveragerc move plot_dataframe to scripts/ folder 2017-11-09 22:29:23 +01:00
.dockerignore Docker improvements (faster and more secure builds) 2017-10-13 15:47:13 +02:00
.gitignore Fix hyperopt when using MongoDB 2017-12-21 19:20:47 -08:00
.pylintrc whitelist TA-lib in pylint 2017-11-18 10:13:14 +02:00
.travis.yml More lint fixes (#198) 2017-12-18 17:36:00 +01:00
config.json.example update example coins to be from monthly max volume list 2017-12-10 11:16:28 +02:00
CONTRIBUTING.md Create contribution guideline 2017-12-22 14:29:31 +02:00
Dockerfile Use ENTRYPOINT instead of CMD so additional arguments can be supplied for docker run. (#184) 2017-12-14 18:41:40 +01:00
freqtrade.service add service file 2017-12-16 03:00:43 +01:00
install_ta-lib.sh pass follow redirects for curl to fix travis 2017-12-19 08:27:52 +02:00
LICENSE Initial commit 2017-05-18 02:13:02 +02:00
MANIFEST.in move qtpylib to vendor folder 2017-10-30 20:41:36 +01:00
README.md link to contribution guide from README.md 2017-12-22 14:31:08 +02:00
requirements.txt Update python-bittrex from 0.2.1 to 0.2.2 (#189) 2017-12-15 16:10:10 +01:00
setup.cfg fix pep8 warnings 2017-12-16 03:39:47 +01:00
setup.py upgrade python-bittrex to 0.2.1 2017-11-23 20:53:13 +01:00

freqtrade

Build Status Coverage Status

Simple High frequency trading bot for crypto currencies. Currently supports trading on Bittrex exchange.

This software is for educational purposes only. Don't risk money which you are afraid to lose.

The command interface is accessible via Telegram (not required). Just register a new bot on https://telegram.me/BotFather and enter the telegram token and your chat_id in config.json

Persistence is achieved through sqlite.

Telegram RPC commands:

  • /start: Starts the trader
  • /stop: Stops the trader
  • /status [table]: Lists all open trades
  • /count: Displays number of open trades
  • /profit: Lists cumulative profit from all finished trades
  • /forcesell <trade_id>|all: Instantly sells the given trade (Ignoring minimum_roi).
  • /performance: Show performance of each finished trade grouped by pair
  • /balance: Show account balance per currency
  • /daily : Shows profit or loss per day, over the last n days
  • /help: Show help message
  • /version: Show version

Config

minimal_roi is a JSON object where the key is a duration in minutes and the value is the minimum ROI in percent. See the example below:

"minimal_roi": {
    "40": 0.0,    # Sell after 40 minutes if the profit is not negative
    "30": 0.01,   # Sell after 30 minutes if there is at least 1% profit
    "20": 0.02,   # Sell after 20 minutes if there is at least 2% profit
    "0":  0.04    # Sell immediately if there is at least 4% profit
},

stoploss is loss in percentage that should trigger a sale. For example value -0.10 will cause immediate sell if the profit dips below -10% for a given trade. This parameter is optional.

initial_state is an optional field that defines the initial application state. Possible values are running or stopped. (default=running) If the value is stopped the bot has to be started with /start first.

ask_last_balance sets the bidding price. Value 0.0 will use ask price, 1.0 will use the last price and values between those interpolate between ask and last price. Using ask price will guarantee quick success in bid, but bot will also end up paying more then would probably have been necessary.

The other values should be self-explanatory, if not feel free to raise a github issue.

Prerequisites

  • python3.6
  • sqlite
  • TA-lib binaries
  • Minimal (advised) system requirements: 2GB RAM, 1GB data, 2vCPU

Install

Arch Linux

Use your favorite AUR helper and install python-freqtrade-git.

Manually

master branch contains the latest stable release.

develop branch has often new features, but might also cause breaking changes. To use it, you are encouraged to join our slack channel.

$ cd freqtrade/
# copy example config. Dont forget to insert your api keys
$ cp config.json.example config.json
$ python -m venv .env
$ source .env/bin/activate
$ pip install -r requirements.txt
$ pip install -e .
$ ./freqtrade/main.py

There is also an article about how to setup the bot (thanks @gurghet).*

* Note: that article was written for an earlier version, so it may be outdated

Docker

Building the image:

$ cd freqtrade
$ docker build -t freqtrade .

For security reasons, your configuration file will not be included in the image, you will need to bind mount it. It is also advised to bind mount a SQLite database file (see second example) to keep it between updates.

You can run a one-off container that is immediately deleted upon exiting with the following command (config.json must be in the current working directory):

$ docker run --rm -v `pwd`/config.json:/freqtrade/config.json -it freqtrade

To run a restartable instance in the background (feel free to place your configuration and database files wherever it feels comfortable on your filesystem):

$ cd ~/.freq
$ touch tradesv3.sqlite
$ docker run -d \
  --name freqtrade \
  -v ~/.freq/config.json:/freqtrade/config.json \
  -v ~/.freq/tradesv3.sqlite:/freqtrade/tradesv3.sqlite \
  freqtrade

If you are using dry_run=True it's not necessary to mount tradesv3.sqlite.

You can then use the following commands to monitor and manage your container:

$ docker logs freqtrade
$ docker logs -f freqtrade
$ docker restart freqtrade
$ docker stop freqtrade
$ docker start freqtrade

You do not need to rebuild the image for configuration changes, it will suffice to edit config.json and restart the container.

systemd service file

Copy ./freqtrade.service to your systemd user directory (usually ~/.config/systemd/user) and update WorkingDirectory and ExecStart to match your setup. After that you can start the daemon with:

$ systemctl --user start freqtrade

Usage

usage: main.py [-h] [-c PATH] [-v] [--version] [--dynamic-whitelist [INT]]
               [--dry-run-db]
               {backtesting,hyperopt} ...

Simple High Frequency Trading Bot for crypto currencies

positional arguments:
  {backtesting,hyperopt}
    backtesting         backtesting module
    hyperopt            hyperopt module

optional arguments:
  -h, --help            show this help message and exit
  -c PATH, --config PATH
                        specify configuration file (default: config.json)
  -v, --verbose         be verbose
  --version             show program's version number and exit
  --dynamic-whitelist [INT]
                        dynamically generate and update whitelist based on 24h
                        BaseVolume (Default 20 currencies)
  --dry-run-db          Force dry run to use a local DB
                        "tradesv3.dry_run.sqlite" instead of memory DB. Work
                        only if dry_run is enabled.

Dynamic whitelist example

Per default --dynamic-whitelist will retrieve the 20 currencies based on BaseVolume. This value can be changed when you run the script.

By Default
Get the 20 currencies based on BaseVolume.

freqtrade --dynamic-whitelist

Customize the number of currencies to retrieve
Get the 30 currencies based on BaseVolume.

freqtrade --dynamic-whitelist 30

Exception
--dynamic-whitelist must be greater than 0. If you enter 0 or a negative value (e.g -2), --dynamic-whitelist will use the default value (20).

Backtesting

Backtesting also uses the config specified via -c/--config.

usage: freqtrade backtesting [-h] [-l] [-i INT] [--realistic-simulation]
                             [-r]

optional arguments:
  -h, --help            show this help message and exit
  -l, --live            using live data
  -i INT, --ticker-interval INT
                        specify ticker interval in minutes (default: 5)
  --realistic-simulation
                        uses max_open_trades from config to simulate real
                        world limitations
  -r, --refresh-pairs-cached
                        refresh the pairs files in tests/testdata with 
                        the latest data from Bittrex. Use it if you want
                        to run your backtesting with up-to-date data.

How to use --refresh-pairs-cached parameter?

The first time your run Backtesting, it will take the pairs your have set in your config file and download data from Bittrex.

If for any reason you want to update your data set, you use --refresh-pairs-cached to force Backtesting to update the data it has. Use it only if you want to update your data set. You will not be able to come back to the previous version.

To test your strategy with latest data, we recommend to continue using
the parameter -l or --live.

Hyperopt

It is possible to use hyperopt for trading strategy optimization. Hyperopt uses an internal config named OPTIMIZE_CONFIG located in freqtrade/optimize/hyperopt.py.

usage: freqtrade hyperopt [-h] [-e INT] [--use-mongodb]

optional arguments:
  -h, --help            show this help message and exit
  -e INT, --epochs INT  specify number of epochs (default: 100)
  --use-mongodb         parallelize evaluations with mongodb (requires mongod
                        in PATH)

Execute tests

$ pytest freqtrade

Contributing

We welcome contributions. See our contribution guide for more details.