Reputation: 107
I have a dataframe. When trying to update cells, all updated cell values are zero. This is my code:
for column in data:
if column != "id" and column != "diagnosis":
# change the dtype to 'float64'
data[column] = data[column].astype("float")
columnArray = data[column].values
column_max = max(columnArray)
column_min = min(columnArray)
print(column_max, " ", column_min,column)
for index in range(columnArray.shape[0]):
cell_value = columnArray[index]
new_value = (cell_value-column_min)/(column_max-column_min)
# print(new_value)
data.at[index,column] = new_value
I also should mention that I am a little bit new to pandas and NumPy and that there might be a built-in function that normalizes my features without any pain.
Upvotes: 0
Views: 141
Reputation: 36765
There is no need to do any of the for loops:
columns = ~data.columns.isin(['id', 'diagnosis'])
data.loc[:, columns] = (data.loc[:, columns] - data.loc[:, columns].min()) / (data.loc[:, columns].max() - data.loc[:, columns].min())
Upvotes: 1