Michal Busta comments

Results 112 comments of


                                            Michal Busta

Many NaN errors with the new Tensor-based version

Hi, I believe it's just wrong assert. The assert is after input assignment, so ".d" derivatives parts are still un-initialized. (for larger networks you just increase probability that there will...

Error While running demo script

#49

Error While running demo script

On 27/03/2020 13:42, duxiangcheng wrote: > > @MichalBusta Thank you! But the error > is "can't open camera by index". Can you give me some advice? > opencv stuff (not...

All loss values become nan

Hi, The horizontal and vertical text are supported. - with no additional information, I think the issue will be in data feeding. (nan values can be result of len(gt) >...

3 times call to function net.forward_ocr in method process_boxes

Hi Aniket, 3rd call is training on GT boxes. It can speed up training in early stage (since a prediction network does not produce proposals with good overlap) in short...

How did you happen use the data from SynthText

Hi Mohammed, you may find some scripts here: - https://github.com/MichalBusta/dataset_conversions Hope it helps, Michal

How did you happen use the data from SynthText

> Using some inspiration from your conversion scripts and SynthText, I managed to create a `train folder` (~35gb) that contains gt_image_name.txt and image_name.jpg, as suggested by your train readme file...

What is the structure of the dataset?

https://github.com/MichalBusta/E2E-MLT/tree/master/sample_train_data/MLT

the problem about process_boxes

Hi, 1. - see: https://github.com/MichalBusta/E2E-MLT/blob/2f0b54e31ebb414cd2daad824d7d474062ebe834/train.py#L75 function just feeds actual proposals from detector to OCR module. 2. - we do not use any validation loss - the strategy for taking model...

the problem about process_boxes

What is your batch size? My guess it too small - try something > 64