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

Merge pull request #2024 from freqtrade/custom_hyperopt_loss

Browse Source 
Custom hyperopt loss function (and sharpe-ratio)
This commit is contained in:
Matthias
2019-07-20 12:48:26 +02:00
committed by GitHub
parent 4d0cf9ec8e fa8904978b
commit 790838d897
14 changed files with 541 additions and 101 deletions
Download Patch File Download Diff File Expand all files Collapse all files

40
docs/bot-usage.md
View File

@@ -199,19 +199,22 @@ to find optimal parameter values for your stategy.
```
usage: freqtrade hyperopt [-h] [-i TICKER_INTERVAL] [--timerange TIMERANGE]
[--max_open_trades MAX_OPEN_TRADES]
[--stake_amount STAKE_AMOUNT] [-r]
[--customhyperopt NAME] [--eps] [--dmmp] [-e INT]
[-s {all,buy,sell,roi,stoploss} [{all,buy,sell,roi,stoploss} ...]]
[--print-all] [-j JOBS]
[--max_open_trades INT]
[--stake_amount STAKE_AMOUNT] [-r]
[--customhyperopt NAME] [--eps] [-e INT]
[-s {all,buy,sell,roi,stoploss} [{all,buy,sell,roi,stoploss} ...]]
[--dmmp] [--print-all] [-j JOBS]
[--random-state INT] [--min-trades INT] [--continue]
[--hyperopt-loss NAME]
optional arguments:
-h, --help show this help message and exit
-i TICKER_INTERVAL, --ticker-interval TICKER_INTERVAL
Specify ticker interval (1m, 5m, 30m, 1h, 1d).
Specify ticker interval (`1m`, `5m`, `30m`, `1h`,
`1d`).
--timerange TIMERANGE
Specify what timerange of data to use.
--max_open_trades MAX_OPEN_TRADES
--max_open_trades INT
Specify max_open_trades to use.
--stake_amount STAKE_AMOUNT
Specify stake_amount.
@@ -221,18 +224,18 @@ optional arguments:
run your optimization commands with up-to-date data.
--customhyperopt NAME
Specify hyperopt class name (default:
DefaultHyperOpts).
`DefaultHyperOpts`).
--eps, --enable-position-stacking
Allow buying the same pair multiple times (position
stacking).
-e INT, --epochs INT Specify number of epochs (default: 100).
-s {all,buy,sell,roi,stoploss} [{all,buy,sell,roi,stoploss} ...], --spaces {all,buy,sell,roi,stoploss} [{all,buy,sell,roi,stoploss} ...]
Specify which parameters to hyperopt. Space-separated
list. Default: `all`.
--dmmp, --disable-max-market-positions
Disable applying `max_open_trades` during backtest
(same as setting `max_open_trades` to a very high
number).
-e INT, --epochs INT Specify number of epochs (default: 100).
-s {all,buy,sell,roi,stoploss} [{all,buy,sell,roi,stoploss} ...], --spaces {all,buy,sell,roi,stoploss} [{all,buy,sell,roi,stoploss} ...]
Specify which parameters to hyperopt. Space separate
list. Default: all.
--print-all Print all results, not only the best ones.
-j JOBS, --job-workers JOBS
The number of concurrently running jobs for
@@ -240,6 +243,19 @@ optional arguments:
(default), all CPUs are used, for -2, all CPUs but one
are used, etc. If 1 is given, no parallel computing
code is used at all.
--random-state INT Set random state to some positive integer for
reproducible hyperopt results.
--min-trades INT Set minimal desired number of trades for evaluations
in the hyperopt optimization path (default: 1).
--continue Continue hyperopt from previous runs. By default,
temporary files will be removed and hyperopt will
start from scratch.
--hyperopt-loss NAME
Specify the class name of the hyperopt loss function
class (IHyperOptLoss). Different functions can
generate completely different results, since the
target for optimization is different. (default:
`DefaultHyperOptLoss`).
```
## Edge commands

101
docs/hyperopt.md
View File

@@ -144,16 +144,85 @@ it will end with telling you which paramter combination produced the best profit
The search for best parameters starts with a few random combinations and then uses a
regressor algorithm (currently ExtraTreesRegressor) to quickly find a parameter combination
that minimizes the value of the objective function `calculate_loss` in `hyperopt.py`.
that minimizes the value of the [loss function](#loss-functions).
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 `hyperopt.py`.
## 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.
By default, FreqTrade uses a loss function, which has been with freqtrade since the beginning and optimizes mostly for short trade duration and avoiding losses.
A different version this can be used by using the `--hyperopt-loss <Class-name>` argument.
This class should be in it's own file within the `user_data/hyperopts/` directory.
Currently, the following loss functions are builtin: `SharpeHyperOptLoss` and `DefaultHyperOptLoss`.
### 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.
For the sample below, you then need to add the command line parameter `--hyperopt-loss SuperDuperHyperOptLoss` to your hyperopt call so this fuction is being used.
A sample of this can be found below, which is identical to the Default Hyperopt loss implementation. A full sample can be found [user_data/hyperopts/](https://github.com/freqtrade/freqtrade/blob/develop/user_data/hyperopts/sample_hyperopt_loss.py)
``` python
from freqtrade.optimize.hyperopt import IHyperOptLoss
TARGET_TRADES = 600
EXPECTED_MAX_PROFIT = 3.0
MAX_ACCEPTED_TRADE_DURATION = 300
class SuperDuperHyperOptLoss(IHyperOptLoss):
"""
Defines the default loss function for hyperopt
"""
@staticmethod
def hyperopt_loss_function(results: DataFrame, trade_count: int,
min_date: datetime, max_date: datetime,
*args, **kwargs) -> float:
"""
Objective function, returns smaller number for better results
This is the legacy algorithm (used until now in freqtrade).
Weights are distributed as follows:
* 0.4 to trade duration
* 0.25: Avoiding trade loss
* 1.0 to total profit, compared to the expected value (`EXPECTED_MAX_PROFIT`) defined above
"""
total_profit = results.profit_percent.sum()
trade_duration = results.trade_duration.mean()
trade_loss = 1 - 0.25 * exp(-(trade_count - TARGET_TRADES) ** 2 / 10 ** 5.8)
profit_loss = max(0, 1 - total_profit / EXPECTED_MAX_PROFIT)
duration_loss = 0.4 * min(trade_duration / MAX_ACCEPTED_TRADE_DURATION, 1)
result = trade_loss + profit_loss + duration_loss
return result
```
Currently, the arguments are:
* `results`: DataFrame containing the result
The following columns are available in results (corresponds to the output-file of backtesting when used with `--export trades`):
`pair, profit_percent, profit_abs, open_time, close_time, open_index, close_index, trade_duration, open_at_end, open_rate, close_rate, sell_reason`
* `trade_count`: Amount of trades (identical to `len(results)`)
* `min_date`: Start date of the hyperopting TimeFrame
* `min_date`: End date of the hyperopting TimeFrame
This function needs to return a floating point number (`float`). Smaller numbers will be interpreted as better results. The parameters and balancing for this is up to you.
!!! Note
This function is called once per iteration - so please make sure to have this as optimized as possible to not slow hyperopt down unnecessarily.
!!! Note
Please keep the arguments `*args` and `**kwargs` in the interface to allow us to extend this interface later.
## 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 you will have the result (more than 30 mins).
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.
@@ -168,8 +237,11 @@ running at least several thousand evaluations.
The `--spaces all` flag determines that all possible parameters should be optimized. Possibilities are listed below.
!!! Note
By default, hyperopt will erase previous results and start from scratch. Continuation can be archived by using `--continue`.
!!! Warning
When switching parameters or changing configuration options, the file `user_data/hyperopt_results.pickle` should be removed. It's used to be able to continue interrupted calculations, but does not detect changes to settings or the hyperopt file.
When switching parameters or changing configuration options, make sure to not use the argument `--continue` so temporary results can be removed.
### Execute Hyperopt with Different Ticker-Data Source
@@ -179,12 +251,11 @@ use data from directory `user_data/data`.
### Running Hyperopt with Smaller Testset
Use the `--timerange` argument to change how much of the testset
you want to use. The last N ticks/timeframes will be used.
Example:
Use the `--timerange` argument to change how much of the testset you want to use.
For example, to use one month of data, pass the following parameter to the hyperopt call:
```bash
freqtrade hyperopt --timerange -200
freqtrade hyperopt --timerange 20180401-20180501
```
### Running Hyperopt with Smaller Search Space
@@ -197,14 +268,14 @@ new buy strategy you have.
Legal values are:
- `all`: optimize everything
- `buy`: just search for a new buy strategy
- `sell`: just search for a new sell strategy
- `roi`: just optimize the minimal profit table for your strategy
- `stoploss`: search for the best stoploss value
- space-separated list of any of the above values for example `--spaces roi stoploss`
* `all`: optimize everything
* `buy`: just search for a new buy strategy
* `sell`: just search for a new sell strategy
* `roi`: just optimize the minimal profit table for your strategy
* `stoploss`: search for the best stoploss value
* space-separated list of any of the above values for example `--spaces roi stoploss`
### Position stacking and disabling max market positions.
### 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.
@@ -252,7 +323,7 @@ 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)
```