Releases: Rikj000/MoniGoMani
v0.12.0 Weighted Signal Interface, Master Framework/Settings Separation, Configurable HyperOptable Stoploss & More!
⚠️ Disclaimer
- This strategy is under development. It is not recommended running it live at this moment.
- Always test this strategy before using it!
- I am in no way responsible for your live results! This strategy is still experimental and under heavy development!
- MoniGoMani should always be re-optimized after doing manual changes!
- You need to optimize your own copy of MoniGoMani while thinking logically, don't follow your computer blindly!
Please update your Freqtrade to it's official development
branch!
This release was brought to you in collaboration with @bgouvea & @raftersvk! 🎉
Changelog
- ✨ Added Weighted Signal Interface to easily implement new indicators and weighted signals used by MGM! 🚀
- ✨ Added Master Framework file. Now all settings & code are pretty much separated!
- ✨ Added Configurable HyperOptable Stoploss to fine-tune where HyperOpt should look
- ⚡ Updated Optimization Process & Go-To Commands
- ✨ Added separate command for applying the results after a 2nd HyperOpt Run
- ⚡ Updated commands so StaticPairLists will automatically be applied after downloading
- ⚡ Updated mgm-config.json
- ✨ Added
Freqtrade Protections
optimized for MGM - 📦 Grouped, moved and added settings
- ✨ Added
- ⚡ Updated so weighted signals can only trigger once in their respective lookback windows
- ⚡ Updated debuggable dataframe to separate weighted signals into individual trends
- ⚡ Updated docker-compose.yml and Dockerfile.MoniGoMani for better Docker support
- ♻️ Refactored nearly all code duplications away and improved upon code readability!
- ♻️ Refactored the Total Overall Signal Importance Calculator:
- ✨ Added Total Signals Needed Section
- ⚡ Automatically load
mgm-config-hyperopt.json
andmgm-config.json
- ⚡ Improved upon styling
- 🔥 Removed unneeded sub commands and results copy-paste section
- 🔥 Removed temporary Docker fix
- ♻️ Refactored Binance-Retrieve-All-Tradable-StaticPairList.py
- 🪲 BugFixed TimeFrame kept Zooming at all times
- 📝 Updated README.md, COMPLETELY NEW DOCUMENTATION in MGM_DOCUMENTATION.md, updated VERYQUICKSTART_FREQTRADE.md!
- 📝 Added Some Test Results
v0.11.0 Partially Automated Optimisation Process, Load Settings/Results from configs, BugFixes (TimeFrame-Zoom, Precision and startup_candle_count) & More!
WARNING: I am in no way responsible for your live results! This strategy is still experimental and under development!
WARNING: MoniGoMani should always be re-optimized after doing manual changes!
You need to optimize your own copy of MoniGoMani while thinking logically, don't follow your computer blindly!
Please install Freqtrade Rikj000/Freqtrade/Rikj000/Freqtrade/moni-go-mani/v0.11.0
at commit bd44deea
(latest)!
Changelog
- Freqtrade itself:
- No more missing signals after HyperOpting with Overrides and exporting to json files! (Pull-Request Pending)
- Partially Automated MoniGoMani Optimization Process
- Moved
HyperOpt Results Copy/Paste Section
to amgm-config-hyperopt.json
file, which can easily be applied/removed using commands - Updated mgm-config.json and mgm-config-private.json
- Added
monigomani_settings
section and moved nearly all MoniGoMani settings to it - Added settings to easily setup minimum/maximum search space values and thresholds for most of the MoniGoMani parameters
- Moved StaticPairList to
mgm-config.json
- Switched to
stake_currency: "USDT"
as the default - Switched to
sell_profit_only: true
as the default - Applied more Filters to the VolumePairList
- Added
- Updated docker-compose.yml
- Added Dockerfile.MoniGoMani with jq dependency
- Applied more Filters to Binance-Retrieve-Top-Volume-StaticPairList
- BugFixed TimeFrame-Zoom not working correctly with rolling buy/sell checks and startup_candle_count conversion
- BugFixed Precision Setting not being applied correctly everywhere
- Updated README.md, COMPLETELY NEW DOCUMENTATION in MGM_DOCUMENTATION.md, updated VERYQUICKSTART_FREQTRADE.md!
- Added Some Test Results
v0.10.0 No more Missing Signals, custom_sell() Unclogger, Long Continuous ROI-Table, Rolling Weighted-Buy/Sell Signals, Improved Optimisation Process & More!
WARNING: I am in no way responsible for your live results! This strategy is still experimental and under development!
WARNING: MoniGoMani should always be re-optimized unless you really know what you are doing when manually allocating parameters!
I strongly recommended to re-optimize your own copy of MoniGoMani while thinking logically, don't follow your computer blindly!
Please update your Freqtrade development
too! (Tested on freqtrade develop-37b71b8c
)
Changelog
- Freqtrade itself:
- No more missing signals after HyperOpting with Overrides! 🥳
- HyperOpt and BackTest Results align with eachother again (Expect lower more realistic results)
- Rewrote the Open Trade Unclogger:
- BugFixed (Trailing)StopLoss not being returned correcly
- Now it functions as a
custom_sell()
function instead of acustom_stoploss()
, which is a less hacky/more natural place for it to live. - Dropped
custom_info['trend_indicator']
storage. - Refined the default search spaces for the unclogger
- Added Custom Long Continuous ROI-Table generation with configurable
roi_table_step_size
setting. - Added HyperOptable
buy/sell__downwards/sideways/upwards_trend_total_signal_needed_candles_lookback_window
parameters, which are used in the buy/sell signal calculation to take weighted signals that triggered in the last few candles into consideration. - Added
number_of_weighted_signals
setting to reach more equally divided total weighted signal scores - Added bull/bear pairs to blacklist
- Updated default ADX threshold to 22
- BugFixed UncloggedWinRatioAndProfitRatioLoss's Divided By 0 Exceptions
- BugFixed Automatic RunMode detection
- Improved the MoniGoMani Optimization Process
- Updated README.md, renamed, updated and added new documentation sections in MGM_DOCUMENTATION.md, added separate VERYQUICKSTART_FREQTRADE.md
- Added
Some Test Results
v0.9.1 Open Trade Unclogger Update, WinRatioAndProfitRatio HyperOptLoss, Improved Optimisation Process & More!
WARNING: I am in no way responsible for your live results! This strategy is still experimental and under development!
WARNING: MoniGoMani should always be re-optimized unless you really know what you are doing when manually allocating parameters!
I strongly recommended to re-optimize your own copy of MoniGoMani while thinking logically, don't follow your computer blindly!
Brought to you in collaboration with @PoCk3T, a new addition to the development team of MoniGoMani! 🚀
Changelog
- Updated the Open Trade Unclogger:
- Added HyperOptable
sell___unclogger_trend_lookback_candles_window_percentage_needed
parameter which is used in the unclogger's last check - Renamed the
sell___unclogger_enabled_when_downwards/sideways/upwards
parameters to a more clear naming format:sell___unclogger_trend_lookback_window_uses_downwards/sideways/upwards_candles
, representing more to what their actual purpose is - Refined the default search spaces for the unclogger
Added 2 Custom HyperLoss Functions: - WinRatioAndProfitRatioLoss: Attempts to optimise for the best profit and stability (Returns smaller number for better results)
- UncloggedWinRatioAndProfitRatioLoss: Same as WinRatioAndProfitRatioLoss but has a configurable Percentage of loss to ignore while HyperOpting (Small losses are a by-product of the Unclogger)
- Added HyperOptable
- Improved the MoniGoMani Optimization Process
- Updated the Total Overall Signal Importance Calculator:
- Added
-lf
or--load-file
Optional (Unused by default): Path to.json
file to load HyperOpt Results from which will be used in the Calculator.
.json
's should be extracted withfreqtrade hyperopt-show --best --no-header --print-json > ./user_data/config-mgm-hyperopt.json
- Added
- Added Binance-Retrieve-Top-Volume-StaticPairList.json: To retrieving all tradable pairs on Binance and create your own
pairs-btc.json
file forfreqtrade data-download
(Thanks to @nightshift2k) - Improved MoniGoMani Logging by switching to
f-strings
- Updated README.md, updated and added new documentation sections in VERYQUICKSTART.md
- Added
Some Test Results
- Added
MoniGoMani Logo
v0.9.0 Open Trade Unclogger, TimeFrame-Zoom, Precision Setting & More!
WARNING: I am in no way responsible for your live results! This strategy is still experimental and under development!
WARNING: MoniGoMani should always be re-optimized unless you really know what you are doing when manually allocating parameters!
I strongly recommended to re-optimize your own copy of MoniGoMani while thinking logically, don't follow your computer blindly!
One of the biggest updates MoniGoMani has seen so far!
Brought to you in collaboration with @DisesiD, @FreshLettuce, 2 amazing additions to the development team of MoniGoMani! 🚀
Changelog
- Added Open Trade Unclogger: When enabled it attempts to unclog the bot when it's stuck with losing trades & unable to trade more new trades. (HyperOptable)
- Added TimeFrame-Zoom: When enabled a zoomed in TimeFrame can be used during backtesting/hyperopting to prevent profit exploitation
- Added Precision Setting: When different from 1 it can be used to control the precision / step size used during hyperopting.
- Updated the Total Overall Signal Importance Calculator:
- 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! - Added Open Trade Unclogger Results to the Report
- BugFix: Integers being printed as Doubles when using
-fm
or-pu
- Optional fill in
- Added Binance-Retrieve-Pair-List: To retrieving all tradable pairs on Binance and create your own
pairs-btc.json
file forfreqtrade data-download
(Thanks to @PoCk3T) - Added MoniGoMani Logging
- Added Class Level RunTime Detection
- Reworked
config.json
(Renamed toconfig-btc.json
timeframe
line removed!) - Reworked
docker-compose.yml
- Updated README.md, updated and added new documentation sections in VERYQUICKSTART.md
- Added
Some Test Results
&Cluster Results
(Thanks to @nightshift2k)
v0.8.1 HyperOpt Setting Overrides & Calculator Update!
WARNING: Old MoniGoMani became EOL (End Of Life) @ v0.8.0, please use MoniGoManiHyperStrategy from now on!
WARNING: I am in no way responsible for your live results! This strategy is still experimental and under development!
WARNING: MoniGoMani should always be re-optimized unless you really know what you are doing when manually allocating parameters!
I strongly recommended to re-optimize your own copy of MoniGoMani while thinking logically, don't follow your computer blindly!
In this update tools to help squize out more total profit / optimize better are included! 🚀
Changelog
- Settings to Enable/Disable HyperOpting for individual
buy_params
&sell_params
through HyperOpt Setting Overrides - Reworked the MoniGoMani Freqtrade
docker-compose.yml
file (be sure todocker-compose pull
&docker-compose up --build --force-recreate
) - Update to the
Total-Overall-Signal-Importance-Calculator.py
:- 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
- Reworked Documentation in README.md, updated and added new sections to VERYQUICKSTART.md:
- Moved
MoniGoMani.py
,MoniGoManiHyperOpt.py
andMoniGoManiHyperOpted.py
to/Legacy MoniGoMani/user_data/
Edit: Use the included updated MoniGoMani.v0.8.1.zip
or clone from main
but the tagged GitHub zips will still contain a minor merge error.
v0.8.0 Auto-HyperOptable Strategies & Other Improvements!
WARNING: Auto-HyperOptable Strategies still need to be merged into Freqtrade, see VeryQuickStart for how to already set it up now though!
WARNING: I am in no way responsible for your live results! This strategy is still experimental and under development!
WARNING: MoniGoMani should always be re-optimized unless you really know what you are doing when manually allocating parameters!
I strongly recommended to re-optimize your own copy of MoniGoMani while thinking logically, don't follow your computer blindly!
The biggest update so far! 🤖
- Even bigger refactor then the last one has been done to reach compatibility with Auto-HyperOptable Strategies
- Added MoniGoManiHyperStrategy the Auto-HyperOptable version of MoniGoMani! 🥳
- Reworked Documentation in all MoniGoMani files & Markdown files Made copy/paste sections more visible
- Split
config.json
intoconfig.json
config-private.json
& updated Go-To Commands to decrease risk in users leaking sensitive personal information - Reworked the MoniGoMani Freqtrade
docker-compose.yml
file - Big update to the
Total-Overall-Signal-Importance-Calculator.py
(All Thanks to @DisesiD!):- Calculate & Include
Total Overall Downwards/Sideways/Upwards Signal Importance
results to improve upon trading in certain trends - Calculate & Include
Total Buy/Sell Signal Percentages
for comparison towards the individual buy/sell signals - Auto export results to a
importance.log
file in the same folder for easy sharing (Might write up another file that calculates a grand total average of all.log
files provided in a folder which should create a really stable result) - Now you must fill in
-sc
or--staking-currency
with the one you use inconfig.json
asstake_currency
since it really matters - Optional fill in
-f
or--file
to submit a custom file name for the log file to be exported - Optional fill in
-nf
or--no-file
if you don't want a log file to be exported - BugFixed Importance calculator not averaging buy signals (Was overstating their importance) Added a variable for trend_amount
- Calculate & Include
v0.7.2 Huge code refactor (& minor performance improvements)
WARNING: I am in no way responsible for your live results! This strategy is still experimental and under development!
WARNING: MoniGoMani should always be re-optimized unless you really know what you are doing when manually allocating parameters!
I strongly recommended to re-optimize your own copy of MoniGoMani while thinking logically, don't follow your computer blindly!
Huge code refactor that changed the original trend arrays to buy_params
& sell_params
arrays throughout all MoniGoMani code.
Also removed the unneeded debuggable_weighted_signal_dataframe
from MoniGoManiHyperOpted which should cause a some minor decrease in CPU usage.
It will also help in keeping people from accidentally turning it on for their runs.
- v0.7.2 (02-04-2021 - Huge code refactor - Changed original trend array to buy_params & sell_params - Removed unneeded debuggable_weighted_signal_dataframe from MoniGoManiHyperOpted)
v0.7.1 Total Overall Signal Importance Calculator
WARNING: I am in no way responsible for your live results! This strategy is still experimental and under development!
WARNING: MoniGoMani should always be re-optimized unless you really know what you are doing when manually allocating parameters!
I strongly recommended to re-optimize your own copy of MoniGoMani while thinking logically, don't follow your computer blindly!
Paste the buy_params
& sell_params
results from your HyperOpt in /user_data/Total-Overall-Signal-Importance-Calculator.py
Then execute: python ./user_data/Total-Overall-Signal-Importance-Calculator.py
from your favorite
terminal / CLI to calculate the overall importance of the signals being used.
The higher the score of a signal the better, please share your results on the Discord server so we can improve! 🚀
(MoniGoMani has been designed so signals can easily be inserted / swapped out)
Changelog:
- v0.7.1 (01-04-2021 - Added Total Overall Signal Importance Calculator)
v0.7.0 Making Hyperopt Results Copy/Paste-able
WARNING: I am in no way responsible for your live results! This strategy is still experimental and under development!
WARNING: MoniGoMani should always be re-optimized unless you really know what you are doing when manually allocating parameters!
I strongly recommended to re-optimize your own copy of MoniGoMani while thinking logically, don't follow your computer blindly!
This release allows you to easily copy/paste the results from a hyperopt in MoniGoManiHyperOpted.
Thanks to @C9H13NO3-dev for this addition!
Changelog:
- v0.7.0 (31-03-2021 - Making Hyperopt Results Copy/Paste-able)