Jun Wang comments

Results 21 comments of


                                            Jun Wang

Scripts to batch download and convert the dataset

> Yeah sure, but creating an own project (id) costs money, or not? Similar question. Does that cost money to create a new project in GCP? I would like to...

Train Error

Same problem, did you try to crop smaller subimages and get it worked on image (256, 256) @jiangshashen

how to get the datasets(most of the urls you offered has been invalid

Hello, did you successfully get the data? Thanks in advance!

ValueError: Expected 2D array, got 1D array instead:

Same problem, anyone solve it yet?

ValueError:X has 22032 features per sample;expecting 38916

You may resize the image to keep the length of feature vector consistent before you extract the HOG features. e.g. add a line 'im = cv2.imresize(im, (150, 300))' after #38...

LinearSVC fit error

Can you elaborate a little bit more? I still got the error 'ValueError: X has 324 features per sample; expecting 2430'

Do you have any plan that release the code for dataset preparation?

@StevenyzZhang follow up on this. Specifically, I'm interested in the code that merges the OCR texts from PaddleOCR to paragraphs based on the geometric relationships mentioned in Section3 of paper....

The ``device`` args doesn't take effect

same here.

Batch inference for LLaVA One Vision

@dshatwell23 Thanks for your template. Do we need to resize the images to be the same dimension inside each batch?

Do batch inference to speed up the process

Did you figure out a way to do that? @piperino11 @tink2123