Access to preprocessed information in the inference image

Hi, Since different methods of preprocessing the data can be used, in the prediction pipeline the same methods have to be used (e.g. same variance and mean if we normalize the images). How exactly can we access those values. Thanks, Anamaria

Created by Anamaria Vizitiu Anamaria
You can add these two values to the trained model data that you retrieve at the end of the training submission. Thanks!
Hi Thomas, Let's say we standardize the data across all training data. The global mean and variance computed on the training dataset should be then used to standardize in the same manner every testing sample. Thus, those 2 values should be available during prediction. Thanks, Anamaria
Hi Anamaria, I don't understand your question. Can you clarify it? Thanks! Thomas

Your web browser must have JavaScript enabled in order for this application to display correctly.
If you are an automated web crawler from a search engine, follow this AJAX application crawl link

Hi, Since different methods of preprocessing the data can be used, in the prediction pipeline the same methods have to be used (e.g. same variance and mean if we normalize the images). How exactly can we access those values. Thanks, Anamaria

Drop files to upload

Access to preprocessed information in the inference image page is loading…