the feature scaling code:
scaler= MinMaxScaler(feature_range=(-1,1))
finalData=scaler.fit_transform(features)
is converting the features of positive samples to -1 too. All the values of 146 features in the csv is -1.
In the training dataset, when i scaled the features, all the features were scaled properly, like this: 0.158492 -0.961181 -0.524880 -0.338608. My training dataset have 400 samples 200 positive and 200 negative and each sample has 144 features.
but in the real time application after deployment of my model using tkinter, this feature scaling code is just giving 1 and -1. It may be due to entries as user may analyze one sample and more. i.e., number of samples may vary in real time.
Samavi Nasir It seems that you are facing issues with the MinMax Scaler while implementing it in your real-time application using tkinter. The MinMax Scaler should be adjusting the values to a range between -1 and 1, but you are observing that all the features are scaled to -1, which is not the expected outcome.
Here's a step-by-step guide to help you troubleshoot the issue:
By following these steps, you can identify the root cause of the issue and address any discrepancies in the feature scaling process, ensuring that the MinMax Scaler functions as expected in your real-time application.
- Badges
- Science topic
- Similar topics
- Psychology
- Emotion