adelorenzo/stable
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Merge pull request #4596 from rokups/rk/hyper-strategy

Browse Source 
Support for creating auto-hyperoptable strategies.
This commit is contained in:
Matthias
2021-04-05 13:55:32 +02:00
committed by GitHub
parent 1f7eeb42a2 c51839dc3b
commit b470672dfd
17 changed files with 1140 additions and 312 deletions
Download Patch File Download Diff File Expand all files Collapse all files

363
docs/advanced-hyperopt.md
View File

@@ -4,79 +4,6 @@ This page explains some advanced Hyperopt topics that may require higher
coding skills and Python knowledge than creation of an ordinal hyperoptimization
class.
## Derived hyperopt classes
Custom hyperopt classes can be derived in the same way [it can be done for strategies](strategy-customization.md#derived-strategies).
Applying to hyperoptimization, as an example, you may override how dimensions are defined in your optimization hyperspace:
```python
class MyAwesomeHyperOpt(IHyperOpt):
...
# Uses default stoploss dimension
class MyAwesomeHyperOpt2(MyAwesomeHyperOpt):
@staticmethod
def stoploss_space() -> List[Dimension]:
# Override boundaries for stoploss
return [
Real(-0.33, -0.01, name='stoploss'),
]
```
and then quickly switch between hyperopt classes, running optimization process with hyperopt class you need in each particular case:
```
$ freqtrade hyperopt --hyperopt MyAwesomeHyperOpt --hyperopt-loss SharpeHyperOptLossDaily --strategy MyAwesomeStrategy ...
or
$ freqtrade hyperopt --hyperopt MyAwesomeHyperOpt2 --hyperopt-loss SharpeHyperOptLossDaily --strategy MyAwesomeStrategy ...
```
## Sharing methods with your strategy
Hyperopt classes provide access to the Strategy via the `strategy` class attribute.
This can be a great way to reduce code duplication if used correctly, but will also complicate usage for inexperienced users.
``` python
from pandas import DataFrame
from freqtrade.strategy.interface import IStrategy
import freqtrade.vendor.qtpylib.indicators as qtpylib
class MyAwesomeStrategy(IStrategy):
buy_params = {
'rsi-value': 30,
'adx-value': 35,
}
def populate_buy_trend(self, dataframe: DataFrame, metadata: dict) -> DataFrame:
return self.buy_strategy_generator(self.buy_params, dataframe, metadata)
@staticmethod
def buy_strategy_generator(params, dataframe: DataFrame, metadata: dict) -> DataFrame:
dataframe.loc[
(
qtpylib.crossed_above(dataframe['rsi'], params['rsi-value']) &
dataframe['adx'] > params['adx-value']) &
dataframe['volume'] > 0
)
, 'buy'] = 1
return dataframe
class MyAwesomeHyperOpt(IHyperOpt):
...
@staticmethod
def buy_strategy_generator(params: Dict[str, Any]) -> Callable:
"""
Define the buy strategy parameters to be used by Hyperopt.
"""
def populate_buy_trend(dataframe: DataFrame, metadata: dict) -> DataFrame:
# Call strategy's buy strategy generator
return self.StrategyClass.buy_strategy_generator(params, dataframe, metadata)
return populate_buy_trend
```
## Creating and using a custom loss function
To use a custom loss function class, make sure that the function `hyperopt_loss_function` is defined in your custom hyperopt loss class.
@@ -142,3 +69,293 @@ This function needs to return a floating point number (`float`). Smaller numbers
!!! Note
Please keep the arguments `*args` and `**kwargs` in the interface to allow us to extend this interface later.
## Overriding pre-defined spaces
To override a pre-defined space (`roi_space`, `generate_roi_table`, `stoploss_space`, `trailing_space`), define a nested class called Hyperopt and define the required spaces as follows:
```python
class MyAwesomeStrategy(IStrategy):
class HyperOpt:
# Define a custom stoploss space.
def stoploss_space(self):
return [Real(-0.05, -0.01, name='stoploss')]
```
---
## Legacy Hyperopt
This Section explains the configuration of an explicit Hyperopt file (separate to the strategy).
!!! Warning "Deprecated / legacy mode"
Since the 2021.4 release you no longer have to write a separate hyperopt class, but all strategies can be hyperopted.
Please read the [main hyperopt page](hyperopt.md) for more details.
### Prepare hyperopt file
Configuring an explicit hyperopt file is similar to writing your own strategy, and many tasks will be similar.
!!! Tip "About this page"
For this page, we will be using a fictional strategy called `AwesomeStrategy` - which will be optimized using the `AwesomeHyperopt` class.
#### Create a Custom Hyperopt File
The simplest way to get started is to use the following command, which will create a new hyperopt file from a template, which will be located under `user_data/hyperopts/AwesomeHyperopt.py`.
Let assume you want a hyperopt file `AwesomeHyperopt.py`:
``` bash
freqtrade new-hyperopt --hyperopt AwesomeHyperopt
```
#### Legacy Hyperopt checklist
Checklist on all tasks / possibilities in hyperopt
Depending on the space you want to optimize, only some of the below are required:
* fill `buy_strategy_generator` - for buy signal optimization
* fill `indicator_space` - for buy signal optimization
* fill `sell_strategy_generator` - for sell signal optimization
* fill `sell_indicator_space` - for sell signal optimization
!!! Note
`populate_indicators` needs to create all indicators any of thee spaces may use, otherwise hyperopt will not work.
Optional in hyperopt - can also be loaded from a strategy (recommended):
* `populate_indicators` - fallback to create indicators
* `populate_buy_trend` - fallback if not optimizing for buy space. should come from strategy
* `populate_sell_trend` - fallback if not optimizing for sell space. should come from strategy
!!! Note
You always have to provide a strategy to Hyperopt, even if your custom Hyperopt class contains all methods.
Assuming the optional methods are not in your hyperopt file, please use `--strategy AweSomeStrategy` which contains these methods so hyperopt can use these methods instead.
Rarely you may also need to override:
* `roi_space` - for custom ROI optimization (if you need the ranges for the ROI parameters in the optimization hyperspace that differ from default)
* `generate_roi_table` - for custom ROI optimization (if you need the ranges for the values in the ROI table that differ from default or the number of entries (steps) in the ROI table which differs from the default 4 steps)
* `stoploss_space` - for custom stoploss optimization (if you need the range for the stoploss parameter in the optimization hyperspace that differs from default)
* `trailing_space` - for custom trailing stop optimization (if you need the ranges for the trailing stop parameters in the optimization hyperspace that differ from default)
#### Defining a buy signal optimization
Let's say you are curious: should you use MACD crossings or lower Bollinger
Bands to trigger your buys. And you also wonder should you use RSI or ADX to
help with those buy decisions. If you decide to use RSI or ADX, which values
should I use for them? So let's use hyperparameter optimization to solve this
mystery.
We will start by defining a search space:
```python
def indicator_space() -> List[Dimension]:
"""
Define your Hyperopt space for searching strategy parameters
"""
return [
Integer(20, 40, name='adx-value'),
Integer(20, 40, name='rsi-value'),
Categorical([True, False], name='adx-enabled'),
Categorical([True, False], name='rsi-enabled'),
Categorical(['bb_lower', 'macd_cross_signal'], name='trigger')
]
```
Above definition says: I have five parameters I want you to randomly combine
to find the best combination. Two of them are integer values (`adx-value` and `rsi-value`) and I want you test in the range of values 20 to 40.
Then we have three category variables. First two are either `True` or `False`.
We use these to either enable or disable the ADX and RSI guards.
The last one we call `trigger` and use it to decide which buy trigger we want to use.
So let's write the buy strategy generator using these values:
```python
@staticmethod
def buy_strategy_generator(params: Dict[str, Any]) -> Callable:
"""
Define the buy strategy parameters to be used by Hyperopt.
"""
def populate_buy_trend(dataframe: DataFrame, metadata: dict) -> DataFrame:
conditions = []
# GUARDS AND TRENDS
if 'adx-enabled' in params and params['adx-enabled']:
conditions.append(dataframe['adx'] > params['adx-value'])
if 'rsi-enabled' in params and params['rsi-enabled']:
conditions.append(dataframe['rsi'] < params['rsi-value'])
# TRIGGERS
if 'trigger' in params:
if params['trigger'] == 'bb_lower':
conditions.append(dataframe['close'] < dataframe['bb_lowerband'])
if params['trigger'] == 'macd_cross_signal':
conditions.append(qtpylib.crossed_above(
dataframe['macd'], dataframe['macdsignal']
))
# Check that volume is not 0
conditions.append(dataframe['volume'] > 0)
if conditions:
dataframe.loc[
reduce(lambda x, y: x & y, conditions),
'buy'] = 1
return dataframe
return populate_buy_trend
```
Hyperopt will now call `populate_buy_trend()` many times (`epochs`) with different value combinations.
It will use the given historical data and make buys based on the buy signals generated with the above function.
Based on the results, hyperopt will tell you which parameter combination produced the best results (based on the configured [loss function](#loss-functions)).
!!! Note
The above setup expects to find ADX, RSI and Bollinger Bands in the populated indicators.
When you want to test an indicator that isn't used by the bot currently, remember to
add it to the `populate_indicators()` method in your strategy or hyperopt file.
#### Sell optimization
Similar to the buy-signal above, sell-signals can also be optimized.
Place the corresponding settings into the following methods
* Inside `sell_indicator_space()` - the parameters hyperopt shall be optimizing.
* Within `sell_strategy_generator()` - populate the nested method `populate_sell_trend()` to apply the parameters.
The configuration and rules are the same than for buy signals.
To avoid naming collisions in the search-space, please prefix all sell-spaces with `sell-`.
### Execute Hyperopt
Once you have updated your hyperopt configuration you can run it.
Because hyperopt tries a lot of combinations to find the best parameters it will take time to get a good result. More time usually results in better results.
We strongly recommend to use `screen` or `tmux` to prevent any connection loss.
```bash
freqtrade hyperopt --config config.json --hyperopt <hyperoptname> --hyperopt-loss <hyperoptlossname> --strategy <strategyname> -e 500 --spaces all
```
Use `<hyperoptname>` as the name of the custom hyperopt used.
The `-e` option will set how many evaluations hyperopt will do. Since hyperopt uses Bayesian search, running too many epochs at once may not produce greater results. Experience has shown that best results are usually not improving much after 500-1000 epochs.
Doing multiple runs (executions) with a few 1000 epochs and different random state will most likely produce different results.
The `--spaces all` option determines that all possible parameters should be optimized. Possibilities are listed below.
!!! Note
Hyperopt will store hyperopt results with the timestamp of the hyperopt start time.
Reading commands (`hyperopt-list`, `hyperopt-show`) can use `--hyperopt-filename <filename>` to read and display older hyperopt results.
You can find a list of filenames with `ls -l user_data/hyperopt_results/`.
#### Running Hyperopt using methods from a strategy
Hyperopt can reuse `populate_indicators`, `populate_buy_trend`, `populate_sell_trend` from your strategy, assuming these methods are **not** in your custom hyperopt file, and a strategy is provided.
```bash
freqtrade hyperopt --hyperopt AwesomeHyperopt --hyperopt-loss SharpeHyperOptLossDaily --strategy AwesomeStrategy
```
### Understand the Hyperopt Result
Once Hyperopt is completed you can use the result to create a new strategy.
Given the following result from hyperopt:
```
Best result:
44/100: 135 trades. Avg profit 0.57%. Total profit 0.03871918 BTC (0.7722Σ%). Avg duration 180.4 mins. Objective: 1.94367
Buy hyperspace params:
{ 'adx-value': 44,
'rsi-value': 29,
'adx-enabled': False,
'rsi-enabled': True,
'trigger': 'bb_lower'}
```
You should understand this result like:
* The buy trigger that worked best was `bb_lower`.
* You should not use ADX because `adx-enabled: False`)
* You should **consider** using the RSI indicator (`rsi-enabled: True` and the best value is `29.0` (`rsi-value: 29.0`)
You have to look inside your strategy file into `buy_strategy_generator()`
method, what those values match to.
So for example you had `rsi-value: 29.0` so we would look at `rsi`-block, that translates to the following code block:
```python
(dataframe['rsi'] < 29.0)
```
Translating your whole hyperopt result as the new buy-signal would then look like:
```python
def populate_buy_trend(self, dataframe: DataFrame) -> DataFrame:
dataframe.loc[
(
(dataframe['rsi'] < 29.0) & # rsi-value
dataframe['close'] < dataframe['bb_lowerband'] # trigger
),
'buy'] = 1
return dataframe
```
### Validate backtesting results
Once the optimized parameters and conditions have been implemented into your strategy, you should backtest the strategy to make sure everything is working as expected.
To achieve same results (number of trades, their durations, profit, etc.) than during Hyperopt, please use same configuration and parameters (timerange, timeframe, ...) used for hyperopt `--dmmp`/`--disable-max-market-positions` and `--eps`/`--enable-position-stacking` for Backtesting.
Should results don't match, please double-check to make sure you transferred all conditions correctly.
Pay special care to the stoploss (and trailing stoploss) parameters, as these are often set in configuration files, which override changes to the strategy.
You should also carefully review the log of your backtest to ensure that there were no parameters inadvertently set by the configuration (like `stoploss` or `trailing_stop`).
### Sharing methods with your strategy
Hyperopt classes provide access to the Strategy via the `strategy` class attribute.
This can be a great way to reduce code duplication if used correctly, but will also complicate usage for inexperienced users.
``` python
from pandas import DataFrame
from freqtrade.strategy.interface import IStrategy
import freqtrade.vendor.qtpylib.indicators as qtpylib
class MyAwesomeStrategy(IStrategy):
buy_params = {
'rsi-value': 30,
'adx-value': 35,
}
def populate_buy_trend(self, dataframe: DataFrame, metadata: dict) -> DataFrame:
return self.buy_strategy_generator(self.buy_params, dataframe, metadata)
@staticmethod
def buy_strategy_generator(params, dataframe: DataFrame, metadata: dict) -> DataFrame:
dataframe.loc[
(
qtpylib.crossed_above(dataframe['rsi'], params['rsi-value']) &
dataframe['adx'] > params['adx-value']) &
dataframe['volume'] > 0
)
, 'buy'] = 1
return dataframe
class MyAwesomeHyperOpt(IHyperOpt):
...
@staticmethod
def buy_strategy_generator(params: Dict[str, Any]) -> Callable:
"""
Define the buy strategy parameters to be used by Hyperopt.
"""
def populate_buy_trend(dataframe: DataFrame, metadata: dict) -> DataFrame:
# Call strategy's buy strategy generator
return self.StrategyClass.buy_strategy_generator(params, dataframe, metadata)
return populate_buy_trend
```

392
docs/hyperopt.md
View File

@@ -1,19 +1,22 @@
# Hyperopt
This page explains how to tune your strategy by finding the optimal
parameters, a process called hyperparameter optimization. The bot uses several
algorithms included in the `scikit-optimize` package to accomplish this. The
search will burn all your CPU cores, make your laptop sound like a fighter jet
and still take a long time.
parameters, a process called hyperparameter optimization. The bot uses algorithms included in the `scikit-optimize` package to accomplish this.
The search will burn all your CPU cores, make your laptop sound like a fighter jet and still take a long time.
In general, the search for best parameters starts with a few random combinations (see [below](#reproducible-results) for more details) and then uses Bayesian search with a ML regressor algorithm (currently ExtraTreesRegressor) to quickly find a combination of parameters in the search hyperspace that minimizes the value of the [loss function](#loss-functions).
Hyperopt requires historic data to be available, just as backtesting does.
Hyperopt requires historic data to be available, just as backtesting does (hyperopt runs backtesting many times with different parameters).
To learn how to get data for the pairs and exchange you're interested in, head over to the [Data Downloading](data-download.md) section of the documentation.
!!! Bug
Hyperopt can crash when used with only 1 CPU Core as found out in [Issue #1133](https://github.com/freqtrade/freqtrade/issues/1133)
!!! Note
Since 2021.4 release you no longer have to write a separate hyperopt class, but can configure the parameters directly in the strategy.
The legacy method is still supported, but it is no longer the recommended way of setting up hyperopt.
The legacy documentation is available at [Legacy Hyperopt](advanced-hyperopt.md#legacy-hyperopt).
## Install hyperopt dependencies
Since Hyperopt dependencies are not needed to run the bot itself, are heavy, can not be easily built on some platforms (like Raspberry PI), they are not installed by default. Before you run Hyperopt, you need to install the corresponding dependencies, as described in this section below.
@@ -34,7 +37,6 @@ pip install -r requirements-hyperopt.txt
## Hyperopt command reference
```
usage: freqtrade hyperopt [-h] [-v] [--logfile FILE] [-V] [-c PATH] [-d PATH]
[--userdir PATH] [-s NAME] [--strategy-path PATH]
@@ -136,47 +138,19 @@ Strategy arguments:
```
## Prepare Hyperopting
Before we start digging into Hyperopt, we recommend you to take a look at
the sample hyperopt file located in [user_data/hyperopts/](https://github.com/freqtrade/freqtrade/blob/develop/freqtrade/templates/sample_hyperopt.py).
Configuring hyperopt is similar to writing your own strategy, and many tasks will be similar.
!!! Tip "About this page"
For this page, we will be using a fictional strategy called `AwesomeStrategy` - which will be optimized using the `AwesomeHyperopt` class.
The simplest way to get started is to use the following, command, which will create a new hyperopt file from a template, which will be located under `user_data/hyperopts/AwesomeHyperopt.py`.
``` bash
freqtrade new-hyperopt --hyperopt AwesomeHyperopt
```
### Hyperopt checklist
Checklist on all tasks / possibilities in hyperopt
Depending on the space you want to optimize, only some of the below are required:
* fill `buy_strategy_generator` - for buy signal optimization
* fill `indicator_space` - for buy signal optimization
* fill `sell_strategy_generator` - for sell signal optimization
* fill `sell_indicator_space` - for sell signal optimization
* define parameters with `space='buy'` - for buy signal optimization
* define parameters with `space='sell'` - for sell signal optimization
!!! Note
`populate_indicators` needs to create all indicators any of the spaces may use, otherwise hyperopt will not work.
Optional in hyperopt - can also be loaded from a strategy (recommended):
* `populate_indicators` - fallback to create indicators
* `populate_buy_trend` - fallback if not optimizing for buy space. should come from strategy
* `populate_sell_trend` - fallback if not optimizing for sell space. should come from strategy
!!! Note
You always have to provide a strategy to Hyperopt, even if your custom Hyperopt class contains all methods.
Assuming the optional methods are not in your hyperopt file, please use `--strategy AweSomeStrategy` which contains these methods so hyperopt can use these methods instead.
Rarely you may also need to override:
Rarely you may also need to create a [nested class](advanced-hyperopt.md#overriding-pre-defined-spaces) named `HyperOpt` and implement
* `roi_space` - for custom ROI optimization (if you need the ranges for the ROI parameters in the optimization hyperspace that differ from default)
* `generate_roi_table` - for custom ROI optimization (if you need the ranges for the values in the ROI table that differ from default or the number of entries (steps) in the ROI table which differs from the default 4 steps)
@@ -184,31 +158,19 @@ Rarely you may also need to override:
* `trailing_space` - for custom trailing stop optimization (if you need the ranges for the trailing stop parameters in the optimization hyperspace that differ from default)
!!! Tip "Quickly optimize ROI, stoploss and trailing stoploss"
You can quickly optimize the spaces `roi`, `stoploss` and `trailing` without changing anything (i.e. without creation of a "complete" Hyperopt class with dimensions, parameters, triggers and guards, as described in this document) from the default hyperopt template by relying on your strategy to do most of the calculations.
You can quickly optimize the spaces `roi`, `stoploss` and `trailing` without changing anything in your strategy.
```python
# Have a working strategy at hand.
freqtrade new-hyperopt --hyperopt EmptyHyperopt
freqtrade hyperopt --hyperopt EmptyHyperopt --hyperopt-loss SharpeHyperOptLossDaily --spaces roi stoploss trailing --strategy MyWorkingStrategy --config config.json -e 100
freqtrade hyperopt --hyperopt-loss SharpeHyperOptLossDaily --spaces roi stoploss trailing --strategy MyWorkingStrategy --config config.json -e 100
```
### Create a Custom Hyperopt File
Let assume you want a hyperopt file `AwesomeHyperopt.py`:
``` bash
freqtrade new-hyperopt --hyperopt AwesomeHyperopt
```
This command will create a new hyperopt file from a template, allowing you to get started quickly.
### Configure your Guards and Triggers
There are two places you need to change in your hyperopt file to add a new buy hyperopt for testing:
There are two places you need to change in your strategy file to add a new buy hyperopt for testing:
* Inside `indicator_space()` - the parameters hyperopt shall be optimizing.
* Within `buy_strategy_generator()` - populate the nested `populate_buy_trend()` to apply the parameters.
* Define the parameters at the class level hyperopt shall be optimizing.
* Within `populate_buy_trend()` - use defined parameter values instead of raw constants.
There you have two different types of indicators: 1. `guards` and 2. `triggers`.
@@ -224,24 +186,46 @@ Hyper-optimization will, for each epoch round, pick one trigger and possibly
multiple guards. The constructed strategy will be something like "*buy exactly when close price touches lower Bollinger band, BUT only if
ADX > 10*".
If you have updated the buy strategy, i.e. changed the contents of `populate_buy_trend()` method, you have to update the `guards` and `triggers` your hyperopt must use correspondingly.
```python
from freqtrade.strategy import IntParameter, IStrategy
class MyAwesomeStrategy(IStrategy):
# If parameter is prefixed with `buy_` or `sell_` then specifying `space` parameter is optional
# and space is inferred from parameter name.
buy_adx_min = IntParameter(0, 100, default=10)
def populate_buy_trend(self, dataframe: 'DataFrame', metadata: dict) -> 'DataFrame':
dataframe.loc[
(
(dataframe['adx'] > self.buy_adx_min.value)
), 'buy'] = 1
return dataframe
```
#### Sell optimization
Similar to the buy-signal above, sell-signals can also be optimized.
Place the corresponding settings into the following methods
* Inside `sell_indicator_space()` - the parameters hyperopt shall be optimizing.
* Within `sell_strategy_generator()` - populate the nested method `populate_sell_trend()` to apply the parameters.
* Define the parameters at the class level hyperopt shall be optimizing.
* Within `populate_sell_trend()` - use defined parameter values instead of raw constants.
The configuration and rules are the same than for buy signals.
To avoid naming collisions in the search-space, please prefix all sell-spaces with `sell-`.
#### Using timeframe as a part of the Strategy
```python
class MyAwesomeStrategy(IStrategy):
# There is no strict parameter naming scheme. If you do not use `buy_` or `sell_` prefixes -
# please specify to which space parameter belongs using `space` parameter. Possible values:
# 'buy' or 'sell'.
adx_max = IntParameter(0, 100, default=50, space='sell')
The Strategy class exposes the timeframe value as the `self.timeframe` attribute.
The same value is available as class-attribute `HyperoptName.timeframe`.
In the case of the linked sample-value this would be `AwesomeHyperopt.timeframe`.
def populate_sell_trend(self, dataframe: DataFrame, metadata: dict) -> DataFrame:
dataframe.loc[
(
(dataframe['adx'] < self.adx_max.value)
), 'buy'] = 1
return dataframe
```
## Solving a Mystery
@@ -251,65 +235,51 @@ help with those buy decisions. If you decide to use RSI or ADX, which values
should I use for them? So let's use hyperparameter optimization to solve this
mystery.
We will start by defining a search space:
We will start by defining hyperoptable parameters:
```python
def indicator_space() -> List[Dimension]:
"""
Define your Hyperopt space for searching strategy parameters
"""
return [
Integer(20, 40, name='adx-value'),
Integer(20, 40, name='rsi-value'),
Categorical([True, False], name='adx-enabled'),
Categorical([True, False], name='rsi-enabled'),
Categorical(['bb_lower', 'macd_cross_signal'], name='trigger')
]
class MyAwesomeStrategy(IStrategy):
buy_adx = IntParameter(20, 40, default=30)
buy_rsi = IntParameter(20, 40, default=30)
buy_adx_enabled = CategoricalParameter([True, False]),
buy_rsi_enabled = CategoricalParameter([True, False]),
buy_trigger = CategoricalParameter(['bb_lower', 'macd_cross_signal']),
```
Above definition says: I have five parameters I want you to randomly combine
to find the best combination. Two of them are integer values (`adx-value`
and `rsi-value`) and I want you test in the range of values 20 to 40.
Above definition says: I have five parameters I want to randomly combine to find the best combination.
Two of them are integer values (`buy_adx` and `buy_rsi`) and I want you test in the range of values 20 to 40.
Then we have three category variables. First two are either `True` or `False`.
We use these to either enable or disable the ADX and RSI guards. The last
one we call `trigger` and use it to decide which buy trigger we want to use.
We use these to either enable or disable the ADX and RSI guards.
The last one we call `trigger` and use it to decide which buy trigger we want to use.
So let's write the buy strategy using these values:
```python
@staticmethod
def buy_strategy_generator(params: Dict[str, Any]) -> Callable:
"""
Define the buy strategy parameters to be used by Hyperopt.
"""
def populate_buy_trend(dataframe: DataFrame, metadata: dict) -> DataFrame:
conditions = []
# GUARDS AND TRENDS
if 'adx-enabled' in params and params['adx-enabled']:
conditions.append(dataframe['adx'] > params['adx-value'])
if 'rsi-enabled' in params and params['rsi-enabled']:
conditions.append(dataframe['rsi'] < params['rsi-value'])
def populate_buy_trend(self, dataframe: DataFrame, metadata: dict) -> DataFrame:
conditions = []
# GUARDS AND TRENDS
if self.buy_adx_enabled.value:
conditions.append(dataframe['adx'] > self.buy_adx.value)
if self.buy_rsi_enabled.value:
conditions.append(dataframe['rsi'] < self.buy_rsi.value)
# TRIGGERS
if 'trigger' in params:
if params['trigger'] == 'bb_lower':
conditions.append(dataframe['close'] < dataframe['bb_lowerband'])
if params['trigger'] == 'macd_cross_signal':
conditions.append(qtpylib.crossed_above(
dataframe['macd'], dataframe['macdsignal']
))
# TRIGGERS
if self.buy_trigger.value == 'bb_lower':
conditions.append(dataframe['close'] < dataframe['bb_lowerband'])
if self.buy_trigger.value == 'macd_cross_signal':
conditions.append(qtpylib.crossed_above(
dataframe['macd'], dataframe['macdsignal']
))
# Check that volume is not 0
conditions.append(dataframe['volume'] > 0)
# Check that volume is not 0
conditions.append(dataframe['volume'] > 0)
if conditions:
dataframe.loc[
reduce(lambda x, y: x & y, conditions),
'buy'] = 1
if conditions:
dataframe.loc[
reduce(lambda x, y: x & y, conditions),
'buy'] = 1
return dataframe
return populate_buy_trend
return dataframe
```
Hyperopt will now call `populate_buy_trend()` many times (`epochs`) with different value combinations.
@@ -321,6 +291,20 @@ Based on the results, hyperopt will tell you which parameter combination produce
When you want to test an indicator that isn't used by the bot currently, remember to
add it to the `populate_indicators()` method in your strategy or hyperopt file.
## Parameter types
There are four parameter types each suited for different purposes.
* `IntParameter` - defines an integral parameter with upper and lower boundaries of search space.
* `DecimalParameter` - defines a floating point parameter with a limited number of decimals (default 3). Should be preferred instead of `RealParameter` in most cases.
* `RealParameter` - defines a floating point parameter with upper and lower boundaries and no precision limit. Rarely used as it creates a space with a near infinite number of possibilities.
* `CategoricalParameter` - defines a parameter with a predetermined number of choices.
!!! Tip "Disabling parameter optimization"
Each parameter takes two boolean parameters:
* `load` - when set to `False` it will not load values configured in `buy_params` and `sell_params`.
* `optimize` - when set to `False` parameter will not be included in optimization process.
Use these parameters to quickly prototype various ideas.
## Loss-functions
Each hyperparameter tuning requires a target. This is usually defined as a loss function (sometimes also called objective function), which should decrease for more desirable results, and increase for bad results.
@@ -342,16 +326,14 @@ Creation of a custom loss function is covered in the [Advanced Hyperopt](advance
## Execute Hyperopt
Once you have updated your hyperopt configuration you can run it.
Because hyperopt tries a lot of combinations to find the best parameters it will take time to get a good result. More time usually results in better results.
Because hyperopt tries a lot of combinations to find the best parameters it will take time to get a good result.
We strongly recommend to use `screen` or `tmux` to prevent any connection loss.
```bash
freqtrade hyperopt --config config.json --hyperopt <hyperoptname> --hyperopt-loss <hyperoptlossname> --strategy <strategyname> -e 500 --spaces all
freqtrade hyperopt --config config.json --hyperopt-loss <hyperoptlossname> --strategy <strategyname> -e 500 --spaces all
```
Use `<hyperoptname>` as the name of the custom hyperopt used.
The `-e` option will set how many evaluations hyperopt will do. Since hyperopt uses Bayesian search, running too many epochs at once may not produce greater results. Experience has shown that best results are usually not improving much after 500-1000 epochs.
Doing multiple runs (executions) with a few 1000 epochs and different random state will most likely produce different results.
@@ -365,30 +347,23 @@ The `--spaces all` option determines that all possible parameters should be opti
### Execute Hyperopt with different historical data source
If you would like to hyperopt parameters using an alternate historical data set that
you have on-disk, use the `--datadir PATH` option. By default, hyperopt
uses data from directory `user_data/data`.
you have on-disk, use the `--datadir PATH` option. By default, hyperopt uses data from directory `user_data/data`.
### Running Hyperopt with a smaller test-set
Use the `--timerange` argument to change how much of the test-set you want to use.
For example, to use one month of data, pass the following parameter to the hyperopt call:
For example, to use one month of data, pass `--timerange 20210101-20210201` (from january 2021 - february 2021) to the hyperopt call.
Full command:
```bash
freqtrade hyperopt --hyperopt <hyperoptname> --strategy <strategyname> --timerange 20180401-20180501
```
### Running Hyperopt using methods from a strategy
Hyperopt can reuse `populate_indicators`, `populate_buy_trend`, `populate_sell_trend` from your strategy, assuming these methods are **not** in your custom hyperopt file, and a strategy is provided.
```bash
freqtrade hyperopt --hyperopt AwesomeHyperopt --hyperopt-loss SharpeHyperOptLossDaily --strategy AwesomeStrategy
freqtrade hyperopt --hyperopt <hyperoptname> --strategy <strategyname> --timerange 20210101-20210201
```
### Running Hyperopt with Smaller Search Space
Use the `--spaces` option to limit the search space used by hyperopt.
Letting Hyperopt optimize everything is a huuuuge search space.
Letting Hyperopt optimize everything is a huuuuge search space.
Often it might make more sense to start by just searching for initial buy algorithm.
Or maybe you just want to optimize your stoploss or roi table for that awesome new buy strategy you have.
@@ -405,40 +380,9 @@ Legal values are:
The default Hyperopt Search Space, used when no `--space` command line option is specified, does not include the `trailing` hyperspace. We recommend you to run optimization for the `trailing` hyperspace separately, when the best parameters for other hyperspaces were found, validated and pasted into your custom strategy.
### Position stacking and disabling max market positions
In some situations, you may need to run Hyperopt (and Backtesting) with the
`--eps`/`--enable-position-staking` and `--dmmp`/`--disable-max-market-positions` arguments.
By default, hyperopt emulates the behavior of the Freqtrade Live Run/Dry Run, where only one
open trade is allowed for every traded pair. The total number of trades open for all pairs
is also limited by the `max_open_trades` setting. During Hyperopt/Backtesting this may lead to
some potential trades to be hidden (or masked) by previously open trades.
The `--eps`/`--enable-position-stacking` argument allows emulation of buying the same pair multiple times,
while `--dmmp`/`--disable-max-market-positions` disables applying `max_open_trades`
during Hyperopt/Backtesting (which is equal to setting `max_open_trades` to a very high
number).
!!! Note
Dry/live runs will **NOT** use position stacking - therefore it does make sense to also validate the strategy without this as it's closer to reality.
You can also enable position stacking in the configuration file by explicitly setting
`"position_stacking"=true`.
### Reproducible results
The search for optimal parameters starts with a few (currently 30) random combinations in the hyperspace of parameters, random Hyperopt epochs. These random epochs are marked with an asterisk character (`*`) in the first column in the Hyperopt output.
The initial state for generation of these random values (random state) is controlled by the value of the `--random-state` command line option. You can set it to some arbitrary value of your choice to obtain reproducible results.
If you have not set this value explicitly in the command line options, Hyperopt seeds the random state with some random value for you. The random state value for each Hyperopt run is shown in the log, so you can copy and paste it into the `--random-state` command line option to repeat the set of the initial random epochs used.
If you have not changed anything in the command line options, configuration, timerange, Strategy and Hyperopt classes, historical data and the Loss Function -- you should obtain same hyper-optimization results with same random state value used.
## Understand the Hyperopt Result
Once Hyperopt is completed you can use the result to create a new strategy.
Once Hyperopt is completed you can use the result to update your strategy.
Given the following result from hyperopt:
```
@@ -446,49 +390,38 @@ Best result:
44/100: 135 trades. Avg profit 0.57%. Total profit 0.03871918 BTC (0.7722Σ%). Avg duration 180.4 mins. Objective: 1.94367
Buy hyperspace params:
{ 'adx-value': 44,
'rsi-value': 29,
'adx-enabled': False,
'rsi-enabled': True,
'trigger': 'bb_lower'}
# Buy hyperspace params:
buy_params = {
'buy_adx': 44,
'buy_rsi': 29,
'buy_adx_enabled': False,
'buy_rsi_enabled': True,
'buy_trigger': 'bb_lower'
}
```
You should understand this result like:
- The buy trigger that worked best was `bb_lower`.
- You should not use ADX because `adx-enabled: False`)
- You should **consider** using the RSI indicator (`rsi-enabled: True` and the best value is `29.0` (`rsi-value: 29.0`)
* The buy trigger that worked best was `bb_lower`.
* You should not use ADX because `'buy_adx_enabled': False`.
* You should **consider** using the RSI indicator (`'buy_rsi_enabled': True`) and the best value is `29.0` (`'buy_rsi': 29.0`)
You have to look inside your strategy file into `buy_strategy_generator()`
method, what those values match to.
Your strategy class can immediately take advantage of these results. Simply copy hyperopt results block and paste them at class level, replacing old parameters (if any). New parameters will automatically be loaded next time strategy is executed.
So for example you had `rsi-value: 29.0` so we would look at `rsi`-block, that translates to the following code block:
Transferring your whole hyperopt result to your strategy would then look like:
```python
(dataframe['rsi'] < 29.0)
class MyAwesomeStrategy(IStrategy):
# Buy hyperspace params:
buy_params = {
'buy_adx': 44,
'buy_rsi': 29,
'buy_adx_enabled': False,
'buy_rsi_enabled': True,
'buy_trigger': 'bb_lower'
}
```
Translating your whole hyperopt result as the new buy-signal would then look like:
```python
def populate_buy_trend(self, dataframe: DataFrame) -> DataFrame:
dataframe.loc[
(
(dataframe['rsi'] < 29.0) & # rsi-value
dataframe['close'] < dataframe['bb_lowerband'] # trigger
),
'buy'] = 1
return dataframe
```
By default, hyperopt prints colorized results -- epochs with positive profit are printed in the green color. This highlighting helps you find epochs that can be interesting for later analysis. Epochs with zero total profit or with negative profits (losses) are printed in the normal color. If you do not need colorization of results (for instance, when you are redirecting hyperopt output to a file) you can switch colorization off by specifying the `--no-color` option in the command line.
You can use the `--print-all` command line option if you would like to see all results in the hyperopt output, not only the best ones. When `--print-all` is used, current best results are also colorized by default -- they are printed in bold (bright) style. This can also be switched off with the `--no-color` command line option.
!!! Note "Windows and color output"
Windows does not support color-output natively, therefore it is automatically disabled. To have color-output for hyperopt running under windows, please consider using WSL.
### Understand Hyperopt ROI results
If you are optimizing ROI (i.e. if optimization search-space contains 'all', 'default' or 'roi'), your result will look as follows and include a ROI table:
@@ -498,11 +431,13 @@ Best result:
44/100: 135 trades. Avg profit 0.57%. Total profit 0.03871918 BTC (0.7722Σ%). Avg duration 180.4 mins. Objective: 1.94367
ROI table:
{ 0: 0.10674,
21: 0.09158,
78: 0.03634,
118: 0}
# ROI table:
minimal_roi = {
0: 0.10674,
21: 0.09158,
78: 0.03634,
118: 0
}
```
In order to use this best ROI table found by Hyperopt in backtesting and for live trades/dry-run, copy-paste it as the value of the `minimal_roi` attribute of your custom strategy:
@@ -548,13 +483,16 @@ Best result:
44/100: 135 trades. Avg profit 0.57%. Total profit 0.03871918 BTC (0.7722Σ%). Avg duration 180.4 mins. Objective: 1.94367
Buy hyperspace params:
{ 'adx-value': 44,
'rsi-value': 29,
'adx-enabled': False,
'rsi-enabled': True,
'trigger': 'bb_lower'}
Stoploss: -0.27996
# Buy hyperspace params:
buy_params = {
'buy_adx': 44,
'buy_rsi': 29,
'buy_adx_enabled': False,
'buy_rsi_enabled': True,
'buy_trigger': 'bb_lower'
}
stoploss: -0.27996
```
In order to use this best stoploss value found by Hyperopt in backtesting and for live trades/dry-run, copy-paste it as the value of the `stoploss` attribute of your custom strategy:
@@ -584,11 +522,11 @@ Best result:
45/100: 606 trades. Avg profit 1.04%. Total profit 0.31555614 BTC ( 630.48Σ%). Avg duration 150.3 mins. Objective: -1.10161
Trailing stop:
{ 'trailing_only_offset_is_reached': True,
'trailing_stop': True,
'trailing_stop_positive': 0.02001,
'trailing_stop_positive_offset': 0.06038}
# Trailing stop:
trailing_stop = True
trailing_stop_positive = 0.02001
trailing_stop_positive_offset = 0.06038
trailing_only_offset_is_reached = True
```
In order to use these best trailing stop parameters found by Hyperopt in backtesting and for live trades/dry-run, copy-paste them as the values of the corresponding attributes of your custom strategy:
@@ -610,6 +548,46 @@ If you are optimizing trailing stop values, Freqtrade creates the 'trailing' opt
Override the `trailing_space()` method and define the desired range in it if you need values of the trailing stop parameters to vary in other ranges during hyperoptimization. A sample for this method can be found in [user_data/hyperopts/sample_hyperopt_advanced.py](https://github.com/freqtrade/freqtrade/blob/develop/freqtrade/templates/sample_hyperopt_advanced.py).
### Reproducible results
The search for optimal parameters starts with a few (currently 30) random combinations in the hyperspace of parameters, random Hyperopt epochs. These random epochs are marked with an asterisk character (`*`) in the first column in the Hyperopt output.
The initial state for generation of these random values (random state) is controlled by the value of the `--random-state` command line option. You can set it to some arbitrary value of your choice to obtain reproducible results.
If you have not set this value explicitly in the command line options, Hyperopt seeds the random state with some random value for you. The random state value for each Hyperopt run is shown in the log, so you can copy and paste it into the `--random-state` command line option to repeat the set of the initial random epochs used.
If you have not changed anything in the command line options, configuration, timerange, Strategy and Hyperopt classes, historical data and the Loss Function -- you should obtain same hyper-optimization results with same random state value used.
## Output formatting
By default, hyperopt prints colorized results -- epochs with positive profit are printed in the green color. This highlighting helps you find epochs that can be interesting for later analysis. Epochs with zero total profit or with negative profits (losses) are printed in the normal color. If you do not need colorization of results (for instance, when you are redirecting hyperopt output to a file) you can switch colorization off by specifying the `--no-color` option in the command line.
You can use the `--print-all` command line option if you would like to see all results in the hyperopt output, not only the best ones. When `--print-all` is used, current best results are also colorized by default -- they are printed in bold (bright) style. This can also be switched off with the `--no-color` command line option.
!!! Note "Windows and color output"
Windows does not support color-output natively, therefore it is automatically disabled. To have color-output for hyperopt running under windows, please consider using WSL.
## Position stacking and disabling max market positions
In some situations, you may need to run Hyperopt (and Backtesting) with the
`--eps`/`--enable-position-staking` and `--dmmp`/`--disable-max-market-positions` arguments.
By default, hyperopt emulates the behavior of the Freqtrade Live Run/Dry Run, where only one
open trade is allowed for every traded pair. The total number of trades open for all pairs
is also limited by the `max_open_trades` setting. During Hyperopt/Backtesting this may lead to
some potential trades to be hidden (or masked) by previously open trades.
The `--eps`/`--enable-position-stacking` argument allows emulation of buying the same pair multiple times,
while `--dmmp`/`--disable-max-market-positions` disables applying `max_open_trades`
during Hyperopt/Backtesting (which is equal to setting `max_open_trades` to a very high
number).
!!! Note
Dry/live runs will **NOT** use position stacking - therefore it does make sense to also validate the strategy without this as it's closer to reality.
You can also enable position stacking in the configuration file by explicitly setting
`"position_stacking"=true`.
## Show details of Hyperopt results
After you run Hyperopt for the desired amount of epochs, you can later list all results for analysis, select only best or profitable once, and show the details for any of the epochs previously evaluated. This can be done with the `hyperopt-list` and `hyperopt-show` sub-commands. The usage of these sub-commands is described in the [Utils](utils.md#list-hyperopt-results) chapter.