pysp not needed

model.py

@@ -4,7 +4,6 @@
 from sklearn.preprocessing import LabelEncoder, MinMaxScaler
 import math
 import matplotlib.pyplot as plt, joblib
-import pysplitter as pysp
 
 
 # from sklearn.linear_model import LinearRegression
@@ -210,11 +209,6 @@ def train(fname, max_depth=-1):
 
     joblib.dump(model, f"model/{os.path.basename(fname)}.joblib")
 
-    pysp.split(
-        f"model/{os.path.basename(fname)}.joblib",
-        f"model",
-        5000000,
-    )
     return model
 
 

serve.py

@@ -19,7 +19,6 @@
 
 from model import features
 import streamlit_analytics
-import pysplitter as pysp
 
 # features = [
 #     "batting_team",
@@ -78,25 +77,16 @@ def save_history(fname, row, total_balls):
 
 
 def load_model(format):
-    format = format.lower()
-    pysp.unsplit(
-        f"model/{format}*.split",
-        f"model/{format}.feather.joblib",
+    return joblib.load(
+        "model/"
+        + (
+            "t20features.feather.joblib"
+            if format == "T20"
+            else "odifeatures.feather.joblib"
+            if format == "ODI"
+            else None
+        )
     )
-    return joblib.load(f"model/{format}.feather.joblib")
-
-
-# def load_model(format):
-#     return joblib.load(
-#         "model/"
-#         + (
-#             "t20features.feather.joblib"
-#             if format == "T20"
-#             else "odifeatures.feather.joblib"
-#             if format == "ODI"
-#             else None
-#         )
-#     )
 
 
 def simulator(args, format):