Merge branch 'develop' into agefilter-max-days-listed
This commit is contained in:
Kevin Julian
GitHub
@@ -32,6 +32,7 @@ class SuperDuperHyperOptLoss(IHyperOptLoss):
|
||||
def hyperopt_loss_function(results: DataFrame, trade_count: int,
|
||||
min_date: datetime, max_date: datetime,
|
||||
config: Dict, processed: Dict[str, DataFrame],
|
||||
backtest_stats: Dict[str, Any],
|
||||
*args, **kwargs) -> float:
|
||||
"""
|
||||
Objective function, returns smaller number for better results
|
||||
@@ -53,7 +54,7 @@ class SuperDuperHyperOptLoss(IHyperOptLoss):
|
||||
|
||||
Currently, the arguments are:
|
||||
|
||||
* `results`: DataFrame containing the result
|
||||
* `results`: DataFrame containing the resulting trades.
|
||||
The following columns are available in results (corresponds to the output-file of backtesting when used with `--export trades`):
|
||||
`pair, profit_ratio, profit_abs, open_date, open_rate, fee_open, close_date, close_rate, fee_close, amount, trade_duration, is_open, sell_reason, stake_amount, min_rate, max_rate, stop_loss_ratio, stop_loss_abs`
|
||||
* `trade_count`: Amount of trades (identical to `len(results)`)
|
||||
@@ -61,6 +62,7 @@ Currently, the arguments are:
|
||||
* `min_date`: End date of the timerange used
|
||||
* `config`: Config object used (Note: Not all strategy-related parameters will be updated here if they are part of a hyperopt space).
|
||||
* `processed`: Dict of Dataframes with the pair as keys containing the data used for backtesting.
|
||||
* `backtest_stats`: Backtesting statistics using the same format as the backtesting file "strategy" substructure. Available fields can be seen in `generate_strategy_stats()` in `optimize_reports.py`.
|
||||
|
||||
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.
|
||||
|
||||
|
||||
@@ -302,7 +302,6 @@ A backtesting result will look like that:
|
||||
| Days win/draw/lose | 12 / 82 / 25 |
|
||||
| Avg. Duration Winners | 4:23:00 |
|
||||
| Avg. Duration Loser | 6:55:00 |
|
||||
| Zero Duration Trades | 4.6% (20) |
|
||||
| Rejected Buy signals | 3089 |
|
||||
| | |
|
||||
| Min balance | 0.00945123 BTC |
|
||||
@@ -390,7 +389,6 @@ It contains some useful key metrics about performance of your strategy on backte
|
||||
| Days win/draw/lose | 12 / 82 / 25 |
|
||||
| Avg. Duration Winners | 4:23:00 |
|
||||
| Avg. Duration Loser | 6:55:00 |
|
||||
| Zero Duration Trades | 4.6% (20) |
|
||||
| Rejected Buy signals | 3089 |
|
||||
| | |
|
||||
| Min balance | 0.00945123 BTC |
|
||||
@@ -420,7 +418,6 @@ It contains some useful key metrics about performance of your strategy on backte
|
||||
- `Best day` / `Worst day`: Best and worst day based on daily profit.
|
||||
- `Days win/draw/lose`: Winning / Losing days (draws are usually days without closed trade).
|
||||
- `Avg. Duration Winners` / `Avg. Duration Loser`: Average durations for winning and losing trades.
|
||||
- `Zero Duration Trades`: A number of trades that completed within same candle as they opened and had `trailing_stop_loss` sell reason. A significant amount of such trades may indicate that strategy is exploiting trailing stoploss behavior in backtesting and produces unrealistic results.
|
||||
- `Rejected Buy signals`: Buy signals that could not be acted upon due to max_open_trades being reached.
|
||||
- `Min balance` / `Max balance`: Lowest and Highest Wallet balance during the backtest period.
|
||||
- `Drawdown`: Maximum drawdown experienced. For example, the value of 50% means that from highest to subsequent lowest point, a 50% drop was experienced).
|
||||
|
||||
@@ -51,7 +51,7 @@ usage: freqtrade hyperopt [-h] [-v] [--logfile FILE] [-V] [-c PATH] [-d PATH]
|
||||
[--spaces {all,buy,sell,roi,stoploss,trailing,default} [{all,buy,sell,roi,stoploss,trailing,default} ...]]
|
||||
[--print-all] [--no-color] [--print-json] [-j JOBS]
|
||||
[--random-state INT] [--min-trades INT]
|
||||
[--hyperopt-loss NAME]
|
||||
[--hyperopt-loss NAME] [--disable-param-export]
|
||||
|
||||
optional arguments:
|
||||
-h, --help show this help message and exit
|
||||
@@ -118,6 +118,8 @@ optional arguments:
|
||||
ShortTradeDurHyperOptLoss, OnlyProfitHyperOptLoss,
|
||||
SharpeHyperOptLoss, SharpeHyperOptLossDaily,
|
||||
SortinoHyperOptLoss, SortinoHyperOptLossDaily
|
||||
--disable-param-export
|
||||
Disable automatic hyperopt parameter export.
|
||||
|
||||
Common arguments:
|
||||
-v, --verbose Verbose mode (-vv for more, -vvv to get all messages).
|
||||
@@ -403,6 +405,9 @@ While this strategy is most likely too simple to provide consistent profit, it s
|
||||
!!! Note
|
||||
`self.buy_ema_short.range` will act differently between hyperopt and other modes. For hyperopt, the above example may generate 48 new columns, however for all other modes (backtesting, dry/live), it will only generate the column for the selected value. You should therefore avoid using the resulting column with explicit values (values other than `self.buy_ema_short.value`).
|
||||
|
||||
!!! Note
|
||||
`range` property may also be used with `DecimalParameter` and `CategoricalParameter`. `RealParameter` does not provide this property due to infinite search space.
|
||||
|
||||
??? Hint "Performance tip"
|
||||
By doing the calculation of all possible indicators in `populate_indicators()`, the calculation of the indicator happens only once for every parameter.
|
||||
While this may slow down the hyperopt startup speed, the overall performance will increase as the Hyperopt execution itself may pick the same value for multiple epochs (changing other values).
|
||||
@@ -509,7 +514,13 @@ You should understand this result like:
|
||||
* 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`)
|
||||
|
||||
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.
|
||||
### Automatic parameter application to the strategy
|
||||
|
||||
When using Hyperoptable parameters, the result of your hyperopt-run will be written to a json file next to your strategy (so for `MyAwesomeStrategy.py`, the file would be `MyAwesomeStrategy.json`).
|
||||
This file is also updated when using the `hyperopt-show` sub-command, unless `--disable-param-export` is provided to either of the 2 commands.
|
||||
|
||||
|
||||
Your strategy class can also contain these results explicitly. 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.
|
||||
|
||||
Transferring your whole hyperopt result to your strategy would then look like:
|
||||
|
||||
@@ -525,6 +536,10 @@ class MyAwesomeStrategy(IStrategy):
|
||||
}
|
||||
```
|
||||
|
||||
!!! Note
|
||||
Values in the configuration file will overwrite Parameter-file level parameters - and both will overwrite parameters within the strategy.
|
||||
The prevalence is therefore: config > parameter file > strategy
|
||||
|
||||
### 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:
|
||||
|
||||
@@ -55,7 +55,7 @@ class AwesomeStrategy(IStrategy):
|
||||
dataframe, _ = self.dp.get_analyzed_dataframe(pair, self.timeframe)
|
||||
|
||||
# Obtain last available candle. Do not use current_time to look up latest candle, because
|
||||
# current_time points to curret incomplete candle whose data is not available.
|
||||
# current_time points to current incomplete candle whose data is not available.
|
||||
last_candle = dataframe.iloc[-1].squeeze()
|
||||
# <...>
|
||||
|
||||
@@ -83,7 +83,7 @@ It is possible to define custom sell signals, indicating that specified position
|
||||
|
||||
For example you could implement a 1:2 risk-reward ROI with `custom_sell()`.
|
||||
|
||||
Using custom_sell() signals in place of stoplosses though *is not recommended*. It is a inferior method to using `custom_stoploss()` in this regard - which also allows you to keep the stoploss on exchange.
|
||||
Using custom_sell() signals in place of stoploss though *is not recommended*. It is a inferior method to using `custom_stoploss()` in this regard - which also allows you to keep the stoploss on exchange.
|
||||
|
||||
!!! Note
|
||||
Returning a `string` or `True` from this method is equal to setting sell signal on a candle at specified time. This method is not called when sell signal is set already, or if sell signals are disabled (`use_sell_signal=False` or `sell_profit_only=True` while profit is below `sell_profit_offset`). `string` max length is 64 characters. Exceeding this limit will cause the message to be truncated to 64 characters.
|
||||
@@ -243,7 +243,7 @@ class AwesomeStrategy(IStrategy):
|
||||
current_rate: float, current_profit: float, **kwargs) -> float:
|
||||
|
||||
if current_profit < 0.04:
|
||||
return -1 # return a value bigger than the inital stoploss to keep using the inital stoploss
|
||||
return -1 # return a value bigger than the initial stoploss to keep using the initial stoploss
|
||||
|
||||
# After reaching the desired offset, allow the stoploss to trail by half the profit
|
||||
desired_stoploss = current_profit / 2
|
||||
|
||||
@@ -130,6 +130,39 @@ trades = load_backtest_data(backtest_dir)
|
||||
trades.groupby("pair")["sell_reason"].value_counts()
|
||||
```
|
||||
|
||||
## Plotting daily profit / equity line
|
||||
|
||||
|
||||
```python
|
||||
# Plotting equity line (starting with 0 on day 1 and adding daily profit for each backtested day)
|
||||
|
||||
from freqtrade.configuration import Configuration
|
||||
from freqtrade.data.btanalysis import load_backtest_data, load_backtest_stats
|
||||
import plotly.express as px
|
||||
import pandas as pd
|
||||
|
||||
# strategy = 'SampleStrategy'
|
||||
# config = Configuration.from_files(["user_data/config.json"])
|
||||
# backtest_dir = config["user_data_dir"] / "backtest_results"
|
||||
|
||||
stats = load_backtest_stats(backtest_dir)
|
||||
strategy_stats = stats['strategy'][strategy]
|
||||
|
||||
equity = 0
|
||||
equity_daily = []
|
||||
for dp in strategy_stats['daily_profit']:
|
||||
equity_daily.append(equity)
|
||||
equity += float(dp)
|
||||
|
||||
dates = pd.date_range(strategy_stats['backtest_start'], strategy_stats['backtest_end'])
|
||||
|
||||
df = pd.DataFrame({'dates': dates,'equity_daily': equity_daily})
|
||||
|
||||
fig = px.line(df, x="dates", y="equity_daily")
|
||||
fig.show()
|
||||
|
||||
```
|
||||
|
||||
### Load live trading results into a pandas dataframe
|
||||
|
||||
In case you did already some trading and want to analyze your performance
|
||||
|
||||
@@ -702,7 +702,8 @@ You can show the details of any hyperoptimization epoch previously evaluated by
|
||||
usage: freqtrade hyperopt-show [-h] [-v] [--logfile FILE] [-V] [-c PATH]
|
||||
[-d PATH] [--userdir PATH] [--best]
|
||||
[--profitable] [-n INT] [--print-json]
|
||||
[--hyperopt-filename PATH] [--no-header]
|
||||
[--hyperopt-filename FILENAME] [--no-header]
|
||||
[--disable-param-export]
|
||||
|
||||
optional arguments:
|
||||
-h, --help show this help message and exit
|
||||
@@ -714,6 +715,8 @@ optional arguments:
|
||||
Hyperopt result filename.Example: `--hyperopt-
|
||||
filename=hyperopt_results_2020-09-27_16-20-48.pickle`
|
||||
--no-header Do not print epoch details header.
|
||||
--disable-param-export
|
||||
Disable automatic hyperopt parameter export.
|
||||
|
||||
Common arguments:
|
||||
-v, --verbose Verbose mode (-vv for more, -vvv to get all messages).
|
||||
|
||||
Reference in New Issue
Block a user