alihan/MoniGoMani-freqtrade-trading
1
0
Fork 0
mirror of https://github.com/Rikj000/MoniGoMani.git synced 2022-03-06 00:08:05 +03:00
Code Issues Projects Releases Wiki Activity

Updated Documentation + created mgm_tools folder + Moved calculator to mgm_tools

Browse Source
This commit is contained in:
Rikj000
2021-04-18 13:14:40 +02:00
parent c5ab66b54f
commit 26cccdf70d
4 changed files with 42 additions and 35 deletions
Download Patch File Download Diff File Expand all files Collapse all files

32
README.md
View File

@@ -36,18 +36,12 @@ Each signal has it's own weight allocated to it & a total buy/sell signal needed
MGM (MoniGoMani) will loop through all signals, if they trigger it will add up the weight and eventually it will check if it's bigger then what's needed in total, if it is it will buy/sell.
The beauty lies in using MGM in combination with hyperopting (= a form of machine learning where you backtest a timeframe a lot of times to find the most ideal values), since all weighted signals have been made hyperoptable it can be used to find the most "ideal" weight divisions.
Also will it teach us what works where & what doesn't since MoniGoMani first detects Downwards/Sideways/Upwards trends and then does all of the above individually for each kind of trend (Creating basically 3 individual strategies, 1 for each kind of trend).
Further it will do various hyperoptable checks upon the open trades to see if there are "bad" ones to unclog.
## Feature List:
- [Auto-HyperOptable Strategy](https://github.com/freqtrade/freqtrade/pull/4596)! \**(No more need for legacy MoniGoMani, legacy MoniGoManiHyperOpt and MoniGoManiHyperOpted strategy classes)*
- All Configurable & HyperOptable settings are **easily copy/paste-able from the HyperOpt Results
- Configurable & HyperOptable Buy/Sell Signal Weight Influence Tables for Downwards/Sideways/Upwards trends
- Configurable & HyperOptable Total Buy/Sell Signal Percentages for Downwards/Sideways/Upwards trends
- Turn On/Off Trading on Downwards/Sideways/Upwards trends for Buys/Sells *(HyperOptable)*
- Settings to Enable/Disable HyperOpting for individual `buy_params` & `sell_params` through [HyperOpt Setting Overrides](https://github.com/Rikj000/MoniGoMani/blob/main/VERYQUICKSTART.md#hyperopt-setting-overrides)
- [TimeFrame-Zoom](MoniGoMani/blob/main/VERYQUICKSTART.md#timeframe-zoom) during backtesting/hyperopting to prevent profit exploitation! *(Read: [Backtesting-Traps](https://brookmiles.github.io/freqtrade-stuff/2021/04/12/backtesting-traps/))*
- [Precision Setting](https://github.com/Rikj000/MoniGoMani/blob/main/VERYQUICKSTART.md#precision-setting) to alter the step-size used during HyperOpts
- Turn On/Off **All** Individual Weighted Signal DataFrame entries for easy debugging in an IDE or better speed while dry/live running or hyperopting
- Each Table has 9 Buy & 9 Sell signals implemented each Configurable & HyperOptable:
- [Auto-HyperOptable Strategy](https://github.com/freqtrade/freqtrade/pull/4596)! \*No more need for legacy MoniGoMani, legacy MoniGoManiHyperOpt and MoniGoManiHyperOpted strategy classes!
- All HyperOptable settings are \*\*easily copy/paste-able from the HyperOpt Results
- Configurable Buy/Sell Signal Weight Influence Tables for Downwards/Sideways/Upwards trends, each table **currently** has 9 Buy & 9 Sell signals implemented ***(HyperOptable!)***:
- ADX + Strong Up/Strong Down
- RSI
- MACD
@@ -57,12 +51,18 @@ Also will it teach us what works where & what doesn't since MoniGoMani first det
- EMA Long Death/Golden Cross
- Bollinger Band Re-Entrance
- VWAP Cross
- Configurable & Hyperoptable [Open Trade Unclogger](https://github.com/Rikj000/MoniGoMani/blob/main/VERYQUICKSTART.md#open-trade-unclogger), if enabled it attempts to unclog the bot when it's stuck with losing trades & unable to trade more new trades :rocket:
- Configurable Total Buy/Sell Signal Percentages for Downwards/Sideways/Upwards trends ***(HyperOptable!)***
- Configurable Trading on Downwards/Sideways/Upwards trends for Buys/Sells ***(HyperOptable!)***
- Settings to Enable/Disable HyperOpting for individual `buy_params` & `sell_params` and setting them to a static value through [HyperOpt Setting Overrides](https://github.com/Rikj000/MoniGoMani/blob/main/VERYQUICKSTART.md#hyperopt-setting-overrides)
- Configurable [Open Trade Unclogger](https://github.com/Rikj000/MoniGoMani/blob/main/VERYQUICKSTART.md#open-trade-unclogger), if enabled it attempts to unclog the bot when it's stuck with losing trades & unable to trade more new trades ***(HyperOptable!)*** :rocket:
- [TimeFrame-Zoom](MoniGoMani/blob/main/VERYQUICKSTART.md#timeframe-zoom) during backtesting/hyperopting to prevent profit exploitation! *(Read: [Backtesting-Traps](https://brookmiles.github.io/freqtrade-stuff/2021/04/12/backtesting-traps/))*
- [Precision Setting](https://github.com/Rikj000/MoniGoMani/blob/main/VERYQUICKSTART.md#precision-setting) to alter the step-size used during HyperOpting
- [Total Overall Signal Importance Calculator](https://github.com/Rikj000/MoniGoMani/blob/main/VERYQUICKSTART.md#total-overall-signal-importance-calculator) for Total Average Signal Importance Calculation upon the HyperOpt Results (With some really handy subcommands)
- Main/Sub Plot Configurations for all indicators used *(Handy in FreqUI)*
- Pre-Configured Main/Sub Plot Configurations for visualisation of all indicators used in FreqUI
- Turn On/Off **All** Individual Weighted Signal DataFrame entries for easy debugging in an IDE or better speed while dry/live running or hyperopting
*\*Support/Updates for Legacy versions stopped since Auto-HyperOptable Strategies are merged into the official Freqtrade Development Branch! **Please switch to the new MoniGoManiHyperStrategy!***
*\*\*Overrides will exclude some from hyperopts results check*
*\*Support/Updates for Legacy versions stopped since Auto-HyperOptable Strategies are merged into the official Freqtrade Development Branch! Please switch to the new MoniGoManiHyperStrategy!*
*\*\*If you set up overrides then currently these will be missing from hyperopts results! Please add these back in manually to prevent unexpected behaviour!*
## Go-To Commands:
For Hyper Opting *(the new [MoniGoManiHyperStrategy.py](https://github.com/Rikj000/MoniGoMani/blob/main/user_data/strategies/MoniGoManiHyperStrategy.py))*:
@@ -73,9 +73,9 @@ For Back Testing *(the new [MoniGoManiHyperStrategy.py](https://github.com/Rikj0
```properties
freqtrade backtesting -s MoniGoManiHyperStrategy -c ./user_data/config-btc.json -c ./user_data/config-private.json --timerange 20210101-20210316
```
For Total Average Signal Importance Calculation *(with the [Total-Overall-Signal-Importance-Calculator.py](https://github.com/Rikj000/MoniGoMani/blob/main/user_data/Total-Overall-Signal-Importance-Calculator.py))*:
For Total Average Signal Importance Calculation *(with the [Total-Overall-Signal-Importance-Calculator.py](https://github.com/Rikj000/MoniGoMani/blob/main/user_data/mgm_tools/Total-Overall-Signal-Importance-Calculator.py))*:
```properties
python ./user_data/Total-Overall-Signal-Importance-Calculator.py -sc BTC
python ./user_data/mgm_tools/Total-Overall-Signal-Importance-Calculator.py -sc BTC
```
For Hyper Opting *(the legacy [MoniGoMani.py](https://github.com/Rikj000/MoniGoMani/blob/main/Legacy%20MoniGoMani/user_data/strategies/MoniGoMani.py) + legacy [MoniGoManiHyperOpt.py](https://github.com/Rikj000/MoniGoMani/blob/main/Legacy%20MoniGoMani/user_data/hyperopts/MoniGoManiHyperOpt.py))*:
```properties

44
VERYQUICKSTART.md
View File

@@ -2,7 +2,8 @@
**MoniGoManiHyperStrategy found in releases already has a decent hyperopt applied to it for BTC pairs!**
**When changing anything in one of the `config.json`'s please [re-optimize](https://github.com/Rikj000/MoniGoMani/blob/main/VERYQUICKSTART.md#how-to-optimize-monigomani)!**
# Very Quick Start (With Docker)
# Very Quick Start (With Docker):
1) [Download](https://github.com/Rikj000/MoniGoMani/releases) the latest `MoniGoMani` release and unzip it somewhere. Or clone the `main` branch through git.
2) Install [Docker Desktop](https://www.docker.com/get-started)
3) Open and edit `MoniGoMani/user_data/config-private.json` & `MoniGoMani/user_data/config.json`
@@ -22,7 +23,8 @@
That's it you successfully set up Freqtrade, connected to Telegram, with FreqUI! Congratulations :partying_face:
*Need a more detailed guide? Checkout the [Official Freqtrade Website](https://www.freqtrade.io/en/stable/docker_quickstart/)!*
# Very Quick Start (From Source Code)
# Very Quick Start (From Source Code):
1) Install [Git](https://git-scm.com/downloads)
2) Open a terminal window and navigate to where you want to put `Freqtrade`
3) Type `git clone https://github.com/freqtrade/freqtrade.git` to clone the Freqtrade repo
@@ -37,21 +39,23 @@ That's it you successfully set up Freqtrade, connected to Telegram, with FreqUI!
That's it you successfully set up Freqtrade natively, you can now use `MoniGoManiHyperStrategy` for hyperopting/backtesting/dry/live-running! Congratulations :partying_face:
Check the [Go-To Commands](https://github.com/Rikj000/MoniGoMani/blob/main/VERYQUICKSTART.md#go-to-commands) for how to use it.
# How to Optimize MoniGoMani
*(These are just my ideas/theories, if you have other ideas, please test them & report your results to [#moni-go-mani-testing](https://discord.gg/xFZ9bB6vEz) so we can learn and improve the workflow!)*
# How to Optimize MoniGoMani:
*(These are just my ideas/theories, if you have other ideas, please test them & report your results to [#moni-go-mani-testing](https://discord.gg/xFZ9bB6vEz) so we can learn and improve the workflow! Also yes the current process is lengthy but we hope to automate this where possible in further versions)*
0) When you change anything in your `config.json`, `config-private.json` (besides personal info etc) or in `MoniGoManiHyperStrategy` itself you should always re-hyperopt to find the new ideal weights for your setup. This should also be done when the market changes in it's long term direction!
1) Do some Technical Analysis on how your stake currency has been behaving in the last months/weeks & pick a logical timeframe to do your hyperopt upon (The timeframe in the go-to commands for example resembles some bullish rise/correction cycles & I believe 2021 will be a bullish year thus I think it's a good timeframe to test upon).
2) Disable HyperOpting for the 6 `buy/sell___trades_when_downwards/sideways/upwards` settings and override them manually with some logical thinking (Noticed HyperOpt often doesn't place these locically on it's own, sometimes creating a hodlr bot), instructions for how to do this are under [HyperOpt Setting Overrides](https://github.com/Rikj000/MoniGoMani/blob/main/VERYQUICKSTART.md#hyperopt-setting-overrides)
2) Disable HyperOpting for the 6 `buy/sell___trades_when_downwards/sideways/upwards` settings and override them manually with some logical thinking (Noticed HyperOpt often doesn't place these locically on it's own, sometimes creating a hodlr bot + it narrows down the search spaces so hyperopt will look more where you want it too), instructions for how to do this are under [HyperOpt Setting Overrides](https://github.com/Rikj000/MoniGoMani/blob/main/VERYQUICKSTART.md#hyperopt-setting-overrides)
3) HyperOpt MoniGoManiHyperStrategy for a 1st run with the command provided in the [Go-To Commands](https://github.com/Rikj000/MoniGoMani/blob/main/VERYQUICKSTART.md#go-to-commands) (Free to alter the command if you have a good idea that you want to test)
4) Apply the HyperOpt results from your 1st run into the HyperOpt Results Copy/Paste Section of `MoniGoManiHyperStrategy.py` (Manually add the 6 `buy/sell___trades_when_downwards/sideways/upwards` settings that where overridden in step 2 since they will be excluded from HyperOpts results)
5) Further Disable HyperOpting for `buy/sell_downwards/sideways/upwards_trend_signal_weight` settings that scored lower then `10%` in the 1st HyperOpt run and override them manually to `0%` (Indication of weak signal / signal not working well during these trends with your current weight allocation setup. Also `10%` was just an idea, feel free to change this) instructions for how to do this are under [HyperOpt Setting Overrides](https://github.com/Rikj000/MoniGoMani/blob/main/VERYQUICKSTART.md#hyperopt-setting-overrides)
6) HyperOpt MoniGoManiHyperStrategy for a 2nd run with the command provided in the [Go-To Commands](https://github.com/Rikj000/MoniGoMani/blob/main/VERYQUICKSTART.md#go-to-commands). This is needed since fully disabling the weak signals in step 4 altered the setup, thus "ideal" weights that take this into consideration will have to be re-calculated again with a 2nd hyperopt run.
6) HyperOpt MoniGoManiHyperStrategy for a 2nd run with the command provided in the [Go-To Commands](https://github.com/Rikj000/MoniGoMani/blob/main/VERYQUICKSTART.md#go-to-commands). This is needed since fully disabling the weak signals in step 4 altered the setup, thus "ideal" weights that take this into consideration will have to be re-calculated again with a 2nd hyperopt run. This should not cause overfitting since it's a complete fresh hyperopt run.
7) Copy/Paste your results into the `Total-Overall-Signal-Importance-Calculator.py` & run it's [Go-To Command](https://github.com/Rikj000/MoniGoMani/blob/main/VERYQUICKSTART.md#go-to-commands) but include `-fm` or `--fix-missing` at the end of your command. This is needed to make them whole again since overridden weighted buy/sell_params will be missing from your HyperOpt results. (This command will re-include them with 0 as their value & re-print the fixed copy/paste-able weighted buy/sell_params)
8) Apply the weighted `buy/sell_params` that the calculator fixed & the other result values from the 2nd hyperopt run into the HyperOpt Results Copy/Paste Section of `MoniGoManiHyperStrategy.py` (Manually add the 6 `buy/sell___trades_when_downwards/sideways/upwards` settings that where overridden in step 2 since they will be excluded from HyperOpt & Calculator results)
9) Turn off all Override settings again & you are done :smile:
# HyperOpt Setting Overrides
# HyperOpt Setting Overrides:
When the Parameters in the HyperOpt Space Parameters sections are altered as following examples then they can be used as overrides while hyperopting / backtesting / dry/live-running
(only truly useful when hyperopting though!) Meaning you can use this to set individual buy_params/sell_params to a fixed value when hyperopting!
*(MoniGoManiHyperStrategy v0.8.1 or above Required!)*
@@ -76,8 +80,8 @@ sell_downwards_trend_macd_weight = \
| **optimize**=False | Exclude from hyperopting (Make static) |
| **load**=False | Don't load from the HyperOpt Results Copy/Paste Section |
# Open Trade Unclogger:
# Open Trade Unclogger:
When the Open Trade Unclogger is enabled it attempts to unclog the bot when it's stuck with losing trades & unable to trade more new trades.
This `custom_stoploss` function should be able to work in tandem with `Trailing stoploss`.
@@ -92,10 +96,10 @@ It will only unclog a losing trade when all of following checks have been full-f
Please configurable/hyperoptable in the sell_params dictionary under the hyperopt results copy/paste section.
Only used when `use_custom_stoploss` & `sell_params['sell___unclogger_enabled']` are both set to `True`!
# Total Overall Signal Importance Calculator
# Total Overall Signal Importance Calculator:
Paste the `buy_params` & `sell_params` results from your HyperOpt over in the `/user_data/Total-Overall-Signal-Importance-Calculator.py` file.
Then execute: `python ./user_data/Total-Overall-Signal-Importance-Calculator.py -sc BTC` from your favorite terminal / CLI to calculate the overall importance of the signals being used.
Then execute: `python ./user_data/mgm_tools/Total-Overall-Signal-Importance-Calculator.py -sc BTC` from your favorite terminal / CLI to calculate the overall importance of the signals being used.
The higher the score of a signal the better! And now it will also export to a `importance.log` file in the same folder for easy sharing!
Share these results in [#moni-go-mani-testing](https://discord.gg/xFZ9bB6vEz) so we can improve the signals!
@@ -105,8 +109,8 @@ Share these results in [#moni-go-mani-testing](https://discord.gg/xFZ9bB6vEz) so
- Optional fill in `-fm` or `--fix-missing` to re-include missing weighted buy/sell_params with 0 as their value & re-print them as copy/paste-able results. Also keeps the tool from crashing when there are missing weighted values (Mostly useful after a hyperopt with overridden values)
- Optional fill in `-pu` or `--precision-used` to re-calculate the weights to what would be expected after running hyperopt with precision enabled. Always use this after running hyperopt with precision different from 1!
# Precision Setting
# Precision Setting:
The `precision` setting can be used to control the precision / step size used during hyperopting.
A value **smaller than 1** will limit the search space, but may skip over good values.
While a value **larger than 1** increases the search space, but will increase the duration of hyperopting.
@@ -121,13 +125,17 @@ To disable `precision` / for old the work mode **just** use **1**.
| **1/5** or **0.2** | **5** (0, 5, 10 ...) |
| **5** | **1/5** or **0.2** (0, 0.2, 0.4, 0.8, ...) |
# TimeFrame-Zoom
**<span style="color:darkorange">WARNING:</span> When backtesting/hyperopting without TimeFrame-Zoom the profit shown will be unrealisticly high!**
# TimeFrame-Zoom:
To prevent profit exploitation during backtesting/hyperopting we backtest/hyperopt MoniGoMani which would normally use a `timeframe` (1h candles) using a smaller `backtest_timeframe` (5m candles) instead. This happens while still using an `informative_timeframe` (original 1h candles) to generate the buy/sell signals.
To prevent profit exploitation during hyperopting the strategy will zoom in upon the timeframe during backtesting/hyperopting.
This should result to way more reliable/realistic results after hyperopting/backtesting.
For more information on why this is needed please read [Backtesting-Traps](https://brookmiles.github.io/freqtrade-stuff/2021/04/12/backtesting-traps/)!
With this more realistic results should be found during backtesting/hyperopting. Since the buy/sell signals will operate on the same `timeframe` that live would use (1h candles), while at the same time `backtest_timeframe` (5m or 1m candles) will simulate price movement during that `timeframe` (1h candle), providing more realistic trailing stoploss and ROI behaviour during backtesting/hyperopting.
For more information on why this is needed please read [Backtesting-Traps](https://brookmiles.github.io/freqtrade-stuff/2021/04/12/backtesting-traps/)!
**<span style="color:darkorange">WARNING:</span> Candle data for both `timeframe` as `backtest_timeframe` will have to be downloaded before you will be able to backtest/hyperopt! (Since both will be used)**
**<span style="color:darkorange">WARNING:</span> This will be slower than backtesting at 1h and 1m is a CPU killer. But if you plan on using trailing stoploss or ROI, you probably want to know that your backtest results are not complete lies.**
**<span style="color:darkorange">WARNING:</span> To disable TimeFrame-Zoom just use the same candles for `timeframe` & `backtest_timeframe`**
### TimeFrame-Zoom Examples:
| Parameter | Meaning |
@@ -145,9 +153,9 @@ For Back Testing *(the new [MoniGoManiHyperStrategy.py](https://github.com/Rikj0
```properties
freqtrade backtesting -s MoniGoManiHyperStrategy -c ./user_data/config-btc.json -c ./user_data/config-private.json --timerange 20210101-20210316
```
For Total Average Signal Importance Calculation *(with the [Total-Overall-Signal-Importance-Calculator.py](https://github.com/Rikj000/MoniGoMani/blob/main/user_data/Total-Overall-Signal-Importance-Calculator.py))*:
For Total Average Signal Importance Calculation *(with the [Total-Overall-Signal-Importance-Calculator.py](https://github.com/Rikj000/MoniGoMani/blob/main/user_data/mgm_tools/Total-Overall-Signal-Importance-Calculator.py))*:
```properties
python ./user_data/Total-Overall-Signal-Importance-Calculator.py -sc BTC
python ./user_data/mgm_tools/Total-Overall-Signal-Importance-Calculator.py -sc BTC
```
For Hyper Opting *(the legacy [MoniGoMani.py](https://github.com/Rikj000/MoniGoMani/blob/main/Legacy%20MoniGoMani/user_data/strategies/MoniGoMani.py) + legacy [MoniGoManiHyperOpt.py](https://github.com/Rikj000/MoniGoMani/blob/main/Legacy%20MoniGoMani/user_data/hyperopts/MoniGoManiHyperOpt.py))*:
```properties

1
user_data/logs/freqtrade.log
View File

@@ -1 +0,0 @@

0
user_data/Total-Overall-Signal-Importance-Calculator.py → user_data/mgm_tools/Total-Overall-Signal-Importance-Calculator.py
View File