Merge branch 'develop' into order-book

This commit is contained in:
Matthias
2018-08-29 19:32:44 +02:00
40 changed files with 775 additions and 234 deletions

View File

@@ -151,7 +151,7 @@ cp freqtrade/tests/testdata/pairs.json user_data/data/binance
Then run:
```bash
python scripts/download_backtest_data --exchange binance
python scripts/download_backtest_data.py --exchange binance
```
This will download ticker data for all the currency pairs you defined in `pairs.json`.
@@ -238,6 +238,31 @@ On the other hand, if you set a too high `minimal_roi` like `"0": 0.55`
profit. Hence, keep in mind that your performance is a mix of your
strategies, your configuration, and the crypto-currency you have set up.
## Backtesting multiple strategies
To backtest multiple strategies, a list of Strategies can be provided.
This is limited to 1 ticker-interval per run, however, data is only loaded once from disk so if you have multiple
strategies you'd like to compare, this should give a nice runtime boost.
All listed Strategies need to be in the same folder.
``` bash
freqtrade backtesting --timerange 20180401-20180410 --ticker-interval 5m --strategy-list Strategy001 Strategy002 --export trades
```
This will save the results to `user_data/backtest_data/backtest-result-<strategy>.json`, injecting the strategy-name into the target filename.
There will be an additional table comparing win/losses of the different strategies (identical to the "Total" row in the first table).
Detailed output for all strategies one after the other will be available, so make sure to scroll up.
```
=================================================== Strategy Summary ====================================================
| Strategy | buy count | avg profit % | cum profit % | total profit ETH | avg duration | profit | loss |
|:-----------|------------:|---------------:|---------------:|-------------------:|:----------------|---------:|-------:|
| Strategy1 | 19 | -0.76 | -14.39 | -0.01440287 | 15:48:00 | 15 | 4 |
| Strategy2 | 6 | -2.73 | -16.40 | -0.01641299 | 1 day, 14:12:00 | 3 | 3 |
```
## Next step
Great, your strategy is profitable. What if the bot can give your the

View File

@@ -1,13 +1,15 @@
# Bot usage
This page explains the difference parameters of the bot and how to run
it.
This page explains the difference parameters of the bot and how to run it.
## Table of Contents
- [Bot commands](#bot-commands)
- [Backtesting commands](#backtesting-commands)
- [Hyperopt commands](#hyperopt-commands)
## Bot commands
```
usage: freqtrade [-h] [-v] [--version] [-c PATH] [-d PATH] [-s NAME]
[--strategy-path PATH] [--dynamic-whitelist [INT]]
@@ -41,6 +43,7 @@ optional arguments:
```
### How to use a different config file?
The bot allows you to select which config file you want to use. Per
default, the bot will load the file `./config.json`
@@ -49,6 +52,7 @@ python3 ./freqtrade/main.py -c path/far/far/away/config.json
```
### How to use --strategy?
This parameter will allow you to load your custom strategy class.
Per default without `--strategy` or `-s` the bot will load the
`DefaultStrategy` included with the bot (`freqtrade/strategy/default_strategy.py`).
@@ -60,6 +64,7 @@ To load a strategy, simply pass the class name (e.g.: `CustomStrategy`) in this
**Example:**
In `user_data/strategies` you have a file `my_awesome_strategy.py` which has
a strategy class called `AwesomeStrategy` to load it:
```bash
python3 ./freqtrade/main.py --strategy AwesomeStrategy
```
@@ -70,6 +75,7 @@ message the reason (File not found, or errors in your code).
Learn more about strategy file in [optimize your bot](https://github.com/freqtrade/freqtrade/blob/develop/docs/bot-optimization.md).
### How to use --strategy-path?
This parameter allows you to add an additional strategy lookup path, which gets
checked before the default locations (The passed path must be a folder!):
```bash
@@ -77,21 +83,25 @@ python3 ./freqtrade/main.py --strategy AwesomeStrategy --strategy-path /some/fol
```
#### How to install a strategy?
This is very simple. Copy paste your strategy file into the folder
`user_data/strategies` or use `--strategy-path`. And voila, the bot is ready to use it.
### How to use --dynamic-whitelist?
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.
```bash
python3 ./freqtrade/main.py --dynamic-whitelist
```
**Customize the number of currencies to retrieve**
Get the 30 currencies based on BaseVolume.
```bash
python3 ./freqtrade/main.py --dynamic-whitelist 30
```
@@ -102,6 +112,7 @@ negative value (e.g -2), `--dynamic-whitelist` will use the default
value (20).
### How to use --db-url?
When you run the bot in Dry-run mode, per default no transactions are
stored in a database. If you want to store your bot actions in a DB
using `--db-url`. This can also be used to specify a custom database
@@ -111,14 +122,14 @@ in production mode. Example command:
python3 ./freqtrade/main.py -c config.json --db-url sqlite:///tradesv3.dry_run.sqlite
```
## Backtesting commands
Backtesting also uses the config specified via `-c/--config`.
```
usage: main.py backtesting [-h] [-i TICKER_INTERVAL] [--eps] [--dmmp]
usage: freqtrade backtesting [-h] [-i TICKER_INTERVAL] [--eps] [--dmmp]
[--timerange TIMERANGE] [-l] [-r]
[--strategy-list STRATEGY_LIST [STRATEGY_LIST ...]]
[--export EXPORT] [--export-filename PATH]
optional arguments:
@@ -139,6 +150,13 @@ optional arguments:
refresh the pairs files in tests/testdata with the
latest data from the exchange. Use it if you want to
run your backtesting with up-to-date data.
--strategy-list STRATEGY_LIST [STRATEGY_LIST ...]
Provide a commaseparated list of strategies to
backtest Please note that ticker-interval needs to be
set either in config or via command line. When using
this together with --export trades, the strategy-name
is injected into the filename (so backtest-data.json
becomes backtest-data-DefaultStrategy.json
--export EXPORT export backtest results, argument are: trades Example
--export=trades
--export-filename PATH
@@ -151,6 +169,7 @@ optional arguments:
```
### How to use --refresh-pairs-cached parameter?
The first time your run Backtesting, it will take the pairs you have
set in your config file and download data from Bittrex.
@@ -162,7 +181,6 @@ to come back to the previous version.**
To test your strategy with latest data, we recommend continuing using
the parameter `-l` or `--live`.
## Hyperopt commands
To optimize your strategy, you can use hyperopt parameter hyperoptimization
@@ -194,10 +212,11 @@ optional arguments:
```
## A parameter missing in the configuration?
All parameters for `main.py`, `backtesting`, `hyperopt` are referenced
in [misc.py](https://github.com/freqtrade/freqtrade/blob/develop/freqtrade/misc.py#L84)
## Next step
The optimal strategy of the bot will change with time depending of the
market trends. The next step is to
The optimal strategy of the bot will change with time depending of the market trends. The next step is to
[optimize your bot](https://github.com/freqtrade/freqtrade/blob/develop/docs/bot-optimization.md).

View File

@@ -267,6 +267,7 @@ Official webpage: https://mrjbq7.github.io/ta-lib/install.html
wget http://prdownloads.sourceforge.net/ta-lib/ta-lib-0.4.0-src.tar.gz
tar xvzf ta-lib-0.4.0-src.tar.gz
cd ta-lib
sed -i "s|0.00000001|0.000000000000000001 |g" src/ta_func/ta_utility.h
./configure --prefix=/usr
make
make install

View File

@@ -1,4 +1,5 @@
# Sandbox API testing
Where an exchange provides a sandbox for risk-free integration, or end-to-end, testing CCXT provides access to these.
This document is a *light overview of configuring Freqtrade and GDAX sandbox.
@@ -11,8 +12,11 @@ https://public.sandbox.gdax.com
https://api-public.sandbox.gdax.com
---
# Configure a Sandbox account on Gdax
Aim of this document section
- An sanbox account
- create 2FA (needed to create an API)
- Add test 50BTC to account
@@ -30,122 +34,108 @@ After registration and Email confimation you wil be redirected into your sanbox
> https://public.sandbox.pro.coinbase.com/
## Enable 2Fa (a prerequisite to creating sandbox API Keys)
From within sand box site select your profile, top right.
>Or as a direct link: https://public.sandbox.pro.coinbase.com/profile
From the menu panel to the left of the screen select
From the menu panel to the left of the screen select
> Security: "*View or Update*"
In the new site select "enable authenticator" as typical google Authenticator.
- open Google Authenticator on your phone
- scan barcode
- enter your generated 2fa
In the new site select "enable authenticator" as typical google Authenticator.
- open Google Authenticator on your phone
- scan barcode
- enter your generated 2fa
## Enable API Access
## Enable API Access
From within sandbox select profile>api>create api-keys
>or as a direct link: https://public.sandbox.pro.coinbase.com/profile/api
Click on "create one" and ensure **view** and **trade** are "checked" and sumbit your 2Fa
Click on "create one" and ensure **view** and **trade** are "checked" and sumbit your 2FA
- **Copy and paste the Passphase** into a notepade this will be needed later
- **Copy and paste the API Secret** popup into a notepad this will needed later
- **Copy and paste the API Key** into a notepad this will needed later
## Add 50 BTC test funds
To add funds, use the web interface deposit and withdraw buttons.
To add funds, use the web interface deposit and withdraw buttons.
To begin select 'Wallets' from the top menu.
> Or as a direct link: https://public.sandbox.pro.coinbase.com/wallets
- Deposits (bottom left of screen)
- - Deposit Funds Bitcoin
- - - Coinbase BTC Wallet
- - - - Max (50 BTC)
- - Deposit Funds Bitcoin
- - - Coinbase BTC Wallet
- - - - Max (50 BTC)
- - - - - Deposit
*This process may be repeated for other currencies, ETH as example*
---
# Configure Freqtrade to use Gax Sandbox
The aim of this document section
- Enable sandbox URLs in Freqtrade
- Configure API
- - secret
- - key
- - passphrase
- Enable sandbox URLs in Freqtrade
- Configure API
- - secret
- - key
- - passphrase
## Sandbox URLs
Freqtrade makes use of CCXT which in turn provides a list of URLs to Freqtrade.
These include `['test']` and `['api']`.
Freqtrade makes use of CCXT which in turn provides a list of URLs to Freqtrade.
These include `['test']` and `['api']`.
- `[Test]` if available will point to an Exchanges sandbox.
- `[Api]` normally used, and resolves to live API target on the exchange
To make use of sandbox / test add "sandbox": true, to your config.json
```
```json
"exchange": {
"name": "gdax",
"sandbox": true,
"key": "5wowfxemogxeowo;heiohgmd",
"secret": "/ZMH1P62rCVmwefewrgcewX8nh4gob+lywxfwfxwwfxwfNsH1ySgvWCUR/w==",
"password": "1bkjfkhfhfu6sr",
"outdated_offset": 5
"pair_whitelist": [
"BTC/USD"
```
Also insert your
- api-key (noted earlier)
- api-secret (noted earlier)
- password (the passphrase - noted earlier)
---
## You should now be ready to test your sandbox!
## You should now be ready to test your sandbox
Ensure Freqtrade logs show the sandbox URL, and trades made are shown in sandbox.
** Typically the BTC/USD has the most activity in sandbox to test against.
** Typically the BTC/USD has the most activity in sandbox to test against.
## GDAX - Old Candles problem
It is my experience that GDAX sandbox candles may be 20+- minutes out of date. This can cause trades to fail as one of Freqtrades safety checks
To disable this check, edit:
>strategy/interface.py
Look for the following section:
```
# Check if dataframe is out of date
signal_date = arrow.get(latest['date'])
interval_minutes = constants.TICKER_INTERVAL_MINUTES[interval]
if signal_date < (arrow.utcnow().shift(minutes=-(interval_minutes * 2 + 5))):
logger.warning(
'Outdated history for pair %s. Last tick is %s minutes old',
pair,
(arrow.utcnow() - signal_date).seconds // 60
)
return False, False
```
It is my experience that GDAX sandbox candles may be 20+- minutes out of date. This can cause trades to fail as one of Freqtrades safety checks.
You could Hash out the entire check as follows:
To disable this check, add / change the `"outdated_offset"` parameter in the exchange section of your configuration to adjust for this delay.
Example based on the above configuration:
```json
"exchange": {
"name": "gdax",
"sandbox": true,
"key": "5wowfxemogxeowo;heiohgmd",
"secret": "/ZMH1P62rCVmwefewrgcewX8nh4gob+lywxfwfxwwfxwfNsH1ySgvWCUR/w==",
"password": "1bkjfkhfhfu6sr",
"outdated_offset": 30
"pair_whitelist": [
"BTC/USD"
```
# # Check if dataframe is out of date
# signal_date = arrow.get(latest['date'])
# interval_minutes = constants.TICKER_INTERVAL_MINUTES[interval]
# if signal_date < (arrow.utcnow().shift(minutes=-(interval_minutes * 2 + 5))):
# logger.warning(
# 'Outdated history for pair %s. Last tick is %s minutes old',
# pair,
# (arrow.utcnow() - signal_date).seconds // 60
# )
# return False, False
```
Or inrease the timeout to offer a level of protection/alignment of this test to freqtrade in live.
As example, to allow an additional 30 minutes. "(interval_minutes * 2 + 5 + 30)"
```
# Check if dataframe is out of date
signal_date = arrow.get(latest['date'])
interval_minutes = constants.TICKER_INTERVAL_MINUTES[interval]
if signal_date < (arrow.utcnow().shift(minutes=-(interval_minutes * 2 + 5 + 30))):
logger.warning(
'Outdated history for pair %s. Last tick is %s minutes old',
pair,
(arrow.utcnow() - signal_date).seconds // 60
)
return False, False
```