Chronos: fix type deprecations in numpy 1.24

python/chronos/src/bigdl/chronos/data/utils/roll_dataset.py

@@ -31,7 +31,7 @@ def get_roll_start_idx(df, id_col, window_size):
         id_start_idxes = df.index[df[id_col] != df[id_col].shift(1)].tolist() + [len(df.index)]
     roll_start_idx_iter = ((range(id_start_idxes[i], id_start_idxes[i+1] - window_size + 1))
                            for i in range(len(id_start_idxes) - 1))
-    roll_start_idxes = np.fromiter(itertools.chain.from_iterable(roll_start_idx_iter), np.int)
+    roll_start_idxes = np.fromiter(itertools.chain.from_iterable(roll_start_idx_iter), np.int32)
     return roll_start_idxes
 
 

python/chronos/src/bigdl/chronos/model/tcmf/time.py

@@ -59,45 +59,45 @@ def __init__(self, start_date, num_ts=100, freq="H", normalized=True):
         self.dti = pd.date_range(self.start_date, periods=self.num_ts, freq=self.freq)
 
     def _minute_of_hour(self):
-        minutes = np.array(self.dti.minute, dtype=np.float)
+        minutes = np.array(self.dti.minute, dtype=np.float32)
         if self.normalized:
             minutes = minutes / 59.0 - 0.5
         return minutes
 
     def _hour_of_day(self):
-        hours = np.array(self.dti.hour, dtype=np.float)
+        hours = np.array(self.dti.hour, dtype=np.float32)
         if self.normalized:
             hours = hours / 23.0 - 0.5
         return hours
 
     def _day_of_week(self):
-        dayWeek = np.array(self.dti.dayofweek, dtype=np.float)
+        dayWeek = np.array(self.dti.dayofweek, dtype=np.float32)
         if self.normalized:
             dayWeek = dayWeek / 6.0 - 0.5
         return dayWeek
 
     def _day_of_month(self):
-        dayMonth = np.array(self.dti.day, dtype=np.float)
+        dayMonth = np.array(self.dti.day, dtype=np.float32)
         if self.normalized:
             dayMonth = dayMonth / 30.0 - 0.5
         return dayMonth
 
     def _day_of_year(self):
-        dayYear = np.array(self.dti.dayofyear, dtype=np.float)
+        dayYear = np.array(self.dti.dayofyear, dtype=np.float32)
         if self.normalized:
             dayYear = dayYear / 364.0 - 0.5
         return dayYear
 
     def _month_of_year(self):
-        monthYear = np.array(self.dti.month, dtype=np.float)
+        monthYear = np.array(self.dti.month, dtype=np.float32)
         if self.normalized:
             monthYear = monthYear / 11.0 - 0.5
         return monthYear
 
     def _week_of_year(self):
-        weekYear = np.array(pd.Int64Index(self.dti.isocalendar().week), dtype=np.float) if\
+        weekYear = np.array(pd.Int64Index(self.dti.isocalendar().week), dtype=np.float32) if\
             version.parse(pd.__version__) >= version.parse("1.1.0") else\
-            np.array(self.dti.weekofyear, dtype=np.float)
+            np.array(self.dti.weekofyear, dtype=np.float32)
         if self.normalized:
             weekYear = weekYear / 51.0 - 0.5
         return weekYear

python/chronos/src/bigdl/chronos/model/tf1/VanillaLSTM_keras.py

@@ -71,7 +71,7 @@ def _check_config(self, **config):
                                   f"{lstm_name} should be int or an list/tuple of ints. "
                                   f"Got {config[lstm_name]}")
         if dropout_name in config:
-            if not check_iter_type(config[dropout_name], (float, np.float)):
+            if not check_iter_type(config[dropout_name], (float, np.float32)):
                 invalidInputError(False,
                                   f"{dropout_name} should be float or a list/tuple of floats. "
                                   f"Got {config[dropout_name]}")

python/chronos/src/setup.py

@@ -62,7 +62,7 @@ def setup_package():
         url='https://github.com/intel-analytics/BigDL',
         packages=get_bigdl_packages(),
         install_requires=['pandas>=1.0.5, <=1.3.5', 'scikit-learn>=0.22.0, <=1.0.2',
-                          'bigdl-nano==' + VERSION, 'numpy<=1.23.5'],
+                          'bigdl-nano==' + VERSION],
         extras_require={'pytorch': ['bigdl-nano[pytorch]==' + VERSION],
                         'tensorflow': ['bigdl-nano[tensorflow_27]=='+VERSION],
                         'automl': ['optuna<=2.10.1', 'configspace<=0.5.0', 'SQLAlchemy<=1.4.27'],

python/chronos/test/bigdl/chronos/data/experimental/test_xshardstsdataset.py

@@ -40,7 +40,7 @@ def generate_spark_df():
     spark = SparkSession(sc)
     rdd = sc.range(0, 100)
     from pyspark.ml.linalg import DenseVector
-    df = rdd.map(lambda x: (DenseVector(np.random.randn(1, ).astype(np.float)),
+    df = rdd.map(lambda x: (DenseVector(np.random.randn(1, ).astype(np.float32)),
                             int(np.random.randint(0, 2, size=())),
                             int(x))).toDF(["feature", "id", "date"])
     return df