Faster_RCNN_for_Open_Images_Dataset_Keras icon indicating copy to clipboard operation
Faster_RCNN_for_Open_Images_Dataset_Keras copied to clipboard

X, Y, image_data, debug_img, debug_num_pos = next(data_gen_train)I got Infinite loop?

Open kshitiz-ds opened this issue 5 years ago • 12 comments

i am using my own dataset related to fashion and i am using just 570 images in total and even after that it is running for more than 9 hours, Can you help solving it.

kshitiz-ds avatar Sep 19 '19 11:09 kshitiz-ds

@RockyXu66 help me ..

kshitiz-ds avatar Sep 19 '19 11:09 kshitiz-ds

mine crashes too ! even with the default data - I appreciate any hint.

naarkhoo avatar Sep 29 '19 19:09 naarkhoo

It crashes for me too!

eslammofreh avatar Dec 14 '19 05:12 eslammofreh

I also used my own dataset and I was suffering of this issue. I was miss re-configure some parts of the code to match my own dataset. You have to check the code. I hope this helps you.

eslammofreh avatar Dec 16 '19 05:12 eslammofreh

@eslammofreh can you elaborate on the solution ?what do you mean by "re-configure" exactly.

naarkhoo avatar Dec 16 '19 08:12 naarkhoo

I have same problem even though I used the default dataset.

TaeheeJeong avatar Jan 15 '20 22:01 TaeheeJeong

Same problem here... On my own dataset, and I don't know what exactly I need to change in the code..

garelli avatar Feb 04 '20 16:02 garelli

did you guy's fix it? i am facing the same problem.

samrawitergrtr avatar May 25 '20 21:05 samrawitergrtr

I had the same problem , it might have something to do with the calc_rpn function calling random.sample for the number of positive and negative regions. It maybe be clashing with the np.random method, causing some errors.

To solve this issue, I imported random as rnd, and explicitly called the rnd.sample method for the calc_rpn function.

ashtanzk avatar Aug 11 '20 06:08 ashtanzk

I also used my own dataset and I was suffering of this issue. I was miss re-configure some parts of the code to match my own dataset. You have to check the code. I hope this helps you.

Can u please elaborate more on it. I am also facing the same issue for my custom dataset.

Priyanshi-Sharma-142 avatar Mar 12 '22 14:03 Priyanshi-Sharma-142

I had the same problem , it might have something to do with the calc_rpn function calling random.sample for the number of positive and negative regions. It maybe be clashing with the np.random method, causing some errors.

To solve this issue, I imported random as rnd, and explicitly called the rnd.sample method for the calc_rpn function.

I tried this but it still doesn't work. Can u please tell how to solve it

Priyanshi-Sharma-142 avatar Mar 12 '22 15:03 Priyanshi-Sharma-142

in get_anchor_gt, pls delete the try-except for calc_rpn and rerun the program. You will see the exception that made the infinite loop. Then carefully trace it.

vitusdama avatar Jun 30 '22 01:06 vitusdama