Fixed pandas FutureWarning (#1073)

* Fixed pandas FutureWarning `FutureWarning: Passing a set as an indexer is deprecated and will raise in a future version. Use a list instead.` * fixed another pandas FutureWarning ``` scripts/data_collector/index.py:228: FutureWarning: The frame.append method is deprecated and will be removed from pandas in a future version. Use pandas.concat instead. new_df = new_df.append(_tmp_df, sort=False) ``` * fixed more pandas futurewarnings
microsoft · Apr 27, 2022 · 84ff662 · 84ff662
1 parent 00e40e7
commit 84ff662
Show file tree

Hide file tree

Showing 5 changed files with 7 additions and 7 deletions.
diff --git a/qlib/contrib/report/analysis_position/parse_position.py b/qlib/contrib/report/analysis_position/parse_position.py
@@ -68,9 +68,9 @@ def parse_position(position: dict = None) -> pd.DataFrame:
             if not _trading_day_sell_df.empty:
                 _trading_day_sell_df["status"] = -1
                 _trading_day_sell_df["date"] = _trading_date
-                _trading_day_df = _trading_day_df.append(_trading_day_sell_df, sort=False)
+                _trading_day_df = pd.concat([_trading_day_df, _trading_day_sell_df], sort=False)
 
-        result_df = result_df.append(_trading_day_df, sort=True)
+        result_df = pd.concat([result_df, _trading_day_df], sort=True)
 
         previous_data = dict(
             date=_trading_date,

diff --git a/qlib/contrib/report/analysis_position/risk_analysis.py b/qlib/contrib/report/analysis_position/risk_analysis.py
@@ -85,7 +85,7 @@ def _get_monthly_risk_analysis_with_report(report_normal_df: pd.DataFrame) -> pd
             # _m_report_long_short,
             pd.Timestamp(year=gp_m[0], month=gp_m[1], day=month_days),
         )
-        _monthly_df = _monthly_df.append(_temp_df, sort=False)
+        _monthly_df = pd.concat([_monthly_df, _temp_df], sort=False)
 
     return _monthly_df
 

diff --git a/scripts/data_collector/base.py b/scripts/data_collector/base.py
@@ -170,7 +170,7 @@ def save_instrument(self, symbol, df: pd.DataFrame):
         df["symbol"] = symbol
         if instrument_path.exists():
             _old_df = pd.read_csv(instrument_path)
-            df = _old_df.append(df, sort=False)
+            df = pd.concat([_old_df, df], sort=False)
         df.to_csv(instrument_path, index=False)
 
     def cache_small_data(self, symbol, df):

diff --git a/scripts/data_collector/index.py b/scripts/data_collector/index.py
@@ -225,7 +225,7 @@ def parse_instruments(self):
                 ] = _row.date
             else:
                 _tmp_df = pd.DataFrame([[_row.symbol, self.bench_start_date, _row.date]], columns=instruments_columns)
-                new_df = new_df.append(_tmp_df, sort=False)
+                new_df = pd.concat([new_df, _tmp_df], sort=False)
 
         inst_df = new_df.loc[:, instruments_columns]
         _inst_prefix = self.INST_PREFIX.strip()

diff --git a/scripts/data_collector/yahoo/collector.py b/scripts/data_collector/yahoo/collector.py
@@ -245,7 +245,7 @@ def download_index_data(self):
             _path = self.save_dir.joinpath(f"sh{_index_code}.csv")
             if _path.exists():
                 _old_df = pd.read_csv(_path)
-                df = _old_df.append(df, sort=False)
+                df = pd.concat([_old_df, df], sort=False)
             df.to_csv(_path, index=False)
             time.sleep(5)
 
@@ -404,7 +404,7 @@ def normalize_yahoo(
                 .index
             )
         df.sort_index(inplace=True)
-        df.loc[(df["volume"] <= 0) | np.isnan(df["volume"]), set(df.columns) - {symbol_field_name}] = np.nan
+        df.loc[(df["volume"] <= 0) | np.isnan(df["volume"]), list(set(df.columns) - {symbol_field_name})] = np.nan
 
         change_series = YahooNormalize.calc_change(df, last_close)
         # NOTE: The data obtained by Yahoo finance sometimes has exceptions