Add warning about cache problems
This commit is contained in:
parent
062d00e8f2
commit
2bcfc0c90c
@ -462,6 +462,10 @@ The output will show a table containing the realized absolute Profit (in stake c
|
||||
|
||||
To save time, by default backtest will reuse a cached result when backtested strategy and config match that of previous backtest. To force a new backtest despite existing result for identical run specify `--no-cache` parameter.
|
||||
|
||||
!!! Warning
|
||||
Caching is automatically disabled for open-ended timeranges (`--timerange 20210101-`), as freqtrade cannot ensure reliably that the underlying data didn't change. It can also use cached results where it shouldn't if the original backtest had missing data at the end, which was fixed by downloading more data.
|
||||
In this instance, please use `--no-cache` once to get a fresh backtest.
|
||||
|
||||
### Further backtest-result analysis
|
||||
|
||||
To further analyze your backtest results, you can [export the trades](#exporting-trades-to-file).
|
||||
|
@ -753,6 +753,7 @@ class Backtesting:
|
||||
}
|
||||
|
||||
# Load previous result that will be updated incrementally.
|
||||
# This can be circumvented in certain instances in combination with downloading more data
|
||||
if self.timerange.stopts == 0 or datetime.fromtimestamp(
|
||||
self.timerange.stopts, tz=timezone.utc) > datetime.now(tz=timezone.utc):
|
||||
self.config['no_backtest_cache'] = True
|
||||
|
Loading…
Reference in New Issue
Block a user