automatically detect maximum required data based on user fed indicators (to avoid NaNs in dataset for rolling indicators), add new config parameter for backtesting to let users increase their startup_candles to accommodate high timeframe indicators, add docs to explain all. Add new feature for automatic indicator duplication according to user defined intervals (exhibited in example strat and configs now).

config_examples/config_freqai_futures.example.json

@@ -59,6 +59,7 @@
         }
     ],
     "freqai": {
+        "startup_candles": 10000,
         "timeframes": [
             "3m",
             "15m",
@@ -79,7 +80,9 @@
             "weight_factor": 0.9,
             "principal_component_analysis": false,
             "use_SVM_to_remove_outliers": true,
-            "stratify": 0
+            "stratify": 0,
+            "indicator_max_period": 20,
+            "indicator_interval": 10
         },
         "data_split_parameters": {
             "test_size": 0.33,

config_examples/config_freqai_spot.example.json

@@ -7,7 +7,7 @@
     "dry_run": true,
     "timeframe": "5m",
     "dry_run_wallet": 4000,
-    "dataformat_ohlcv": "hdf5",
+    "dataformat_ohlcv": "json",
     "cancel_open_orders_on_exit": true,
     "unfilledtimeout": {
         "entry": 10,
@@ -51,6 +51,7 @@
         }
     ],
     "freqai": {
+        "startup_candles": 10000,
         "timeframes": [
             "5m",
             "15m",
@@ -74,7 +75,9 @@
             "weight_factor": 0.9,
             "principal_component_analysis": false,
             "use_SVM_to_remove_outliers": false,
-            "stratify": 0
+            "stratify": 0,
+            "indicator_max_period": 50,
+            "indicator_interval": 10
         },
         "data_split_parameters": {
             "test_size": 0.33,