When we are using an image transformer, why do we need a feature extractor (TrOCR processor is Feature Extractor + Roberta Tokenizer)? And I saw the output image given by the processor, it’s the same as the original image, just the shape is changed, it resized smaller. @nielsr is the processor do…

Yes feature extractors also have a from_pretrained method, to just load the same configuration as the one of a particular checkpoint on the hub. e.g. if you do ViTFeatureExtractor.from_pretrained("google/vit-base-patch16-224"), it will make sure the size attribute of the feature extractor is set to…

Why TrOCR processor has a feature extractor?

Beginners

Kforcode November 25, 2021, 4:59am 8

yes you got it right, np.repeat along the last dimension, should do the job

Topic		Replies	Views
Extract visual and contextual features from images Models	5	4463	August 27, 2021
Get original image from trocr processor Intermediate	1	682	October 10, 2022
Processor while fine-tuning TrOCR on IAM 🤗Transformers	0	217	November 28, 2023
Finetuning TrOCR on the IAM dataset 🤗Transformers	1	1139	August 11, 2022
Error finding processor's image class. Loading based on pattern matching with feature extractor 🤗Transformers	11	12750	October 27, 2023

Why TrOCR processor has a feature extractor?

Related topics