f863a6804ba1f46c3f216c6ba75b642b9b61ad17,lore/pipelines/holdout.py,Base,encode_x,#Base#Any#,140

Before Change

        encoded = OrderedDict()
        for encoder in self.encoders:
            transformed = encoder.transform(self.read_column(data, encoder.source_column))
            if hasattr(encoder, "sequence_length"):
                for i in range(encoder.sequence_length):
                    encoded[encoder.sequence_name(i)] = transformed.iloc[:, i]
            else:
                encoded[encoder.name] = transformed
        
        for column in self.index:
            encoded[column] = self.read_column(data, column)
        
        // Using a DataFrame as a container temporarily requires double the memory,

After Change

        encoded = OrderedDict()
        if self.multiprocessing:
            pool = multiprocessing.Pool(self.workers)
            results = []
            for encoder in self.encoders:
                results.append((encoder, pool.apply_async(self.transform, (encoder, data))))

            for encoder, result in results:
                self.merged_transformed(encoded, encoder, result.get())
    
        else:
            for encoder in self.encoders:
                self.merged_transformed(encoded, encoder, self.transform(encoder, data))

Instances