Merge pull request #6707 from koradiyakaushal/develop

Ref: timeseries friendly merge_ordered in merge_informative_pair func…

freqtrade/strategy/strategy_helper.py

@@ -56,12 +56,18 @@ def merge_informative_pair(dataframe: pd.DataFrame, informative: pd.DataFrame,
 
     # Combine the 2 dataframes
     # all indicators on the informative sample MUST be calculated before this point
-    dataframe = pd.merge(dataframe, informative, left_on='date',
-                         right_on=date_merge, how='left')
+    if ffill:
+        # https://pandas.pydata.org/docs/user_guide/merging.html#timeseries-friendly-merging
+        # merge_ordered - ffill method is 2.5x faster than seperate ffill()
+        dataframe = pd.merge_ordered(dataframe, informative, fill_method="ffill", left_on='date',
+                                     right_on=date_merge, how='left')
+    else:
+        dataframe = pd.merge(dataframe, informative, left_on='date',
+                             right_on=date_merge, how='left')
     dataframe = dataframe.drop(date_merge, axis=1)
 
-    if ffill:
-        dataframe = dataframe.ffill()
+    # if ffill:
+    #     dataframe = dataframe.ffill()
 
     return dataframe
 

tests/strategy/test_strategy_helpers.py

@@ -68,6 +68,21 @@ def test_merge_informative_pair():
     assert result.iloc[7]['date_1h'] == result.iloc[4]['date']
     assert result.iloc[8]['date_1h'] == result.iloc[4]['date']
 
+    informative = generate_test_data('1h', 40)
+    result = merge_informative_pair(data, informative, '15m', '1h', ffill=False)
+    # First 3 rows are empty
+    assert result.iloc[0]['date_1h'] is pd.NaT
+    assert result.iloc[1]['date_1h'] is pd.NaT
+    assert result.iloc[2]['date_1h'] is pd.NaT
+    # Next 4 rows contain the starting date (0:00)
+    assert result.iloc[3]['date_1h'] == result.iloc[0]['date']
+    assert result.iloc[4]['date_1h'] is pd.NaT
+    assert result.iloc[5]['date_1h'] is pd.NaT
+    assert result.iloc[6]['date_1h'] is pd.NaT
+    # Next 4 rows contain the next Hourly date original date row 4
+    assert result.iloc[7]['date_1h'] == result.iloc[4]['date']
+    assert result.iloc[8]['date_1h'] is pd.NaT
+
 
 def test_merge_informative_pair_same():
     data = generate_test_data('15m', 40)