cure depression and regression

py_scripts/scout.py

@@ -1,12 +1,40 @@
 # Regression/Prediction (Totally gonna do later trust bro)
+from sklearn.linear_model import LinearRegression
 
 from sklearn.impute import SimpleImputer
 import numpy as np
 
 
-# Column2 is the column you're grouping by using the given values pls
+def cure_depression(dataset):
-def cure_depression(dataset, column, column2, values):
+    # this is pog
-    # Allegedly for loop here )
+    numeric = dataset.select_dtypes(include=np.number)
-    dataset[column].fillna(dataset[column].mean(), inplace=True)
+    numeric_columns = numeric.columns
-    #    print(dataset)
+    dataset[numeric_columns] = dataset[numeric_columns].interpolate(
-    return dataset[column].to_list()
+        method="linear", limit_direct="forward"
+    )
+    # fuck around and find out with other methods maybe idk
+    return dataset
+
+
+def regression_expression(dataset, column, missing_value):
+    lr = LinearRegression()
+    numeric = dataset.select_dtypes(include=np.number)
+    # Migrate this to digger
+    # the fookin nulls
+    testdf = numeric[numeric[column].isnull() == False]
+    testdf = testdf[testdf[column] != 0]
+    # the non nulls and non 0s
+    traindf = numeric[numeric[column].isnull() == False]
+    traindf = traindf[traindf[column] != 0]
+    # print(traindf.head(20))
+    # end of migration
+
+    y = traindf[column]
+    traindf.drop(column, axis=1, inplace=True)
+    lr.fit(traindf, y)
+    testdf.drop(column, axis=1, inplace=True)
+    pred = lr.predict(testdf)
+    # can't put this in data set directly because length no match
+    # join testdf and traindf to form dataset perhaps??
+    testdf[column] = pred
+    print(testdf.head(30))