Closed AyaMohamedS closed 5 years ago
The important message is here:
ResourceExhaustedError: OOM when allocating tensor with shape[3,3,256,256]
This message indicates that you GPU memory is not big enough. You should check the batch size you use.
Really thanks you made my day ^_^ i made batch size =1 and it works i 'm little confused about this batch size and the number of training steps 1/ is that true "each step will take 1 training image"? 2/ shall i set the number of steps = no. of training images to make 1 epoch over the dataset?
can you help me in understanding the training steps w.r.t the training dataset?
I'm glad that I can make you day haha ^_^.
num_images
can obtained from the image list.
num_steps_of_epoch = num_images / batch_size
i have selected some training images from ade20k dataset with their color codded _seg.png mask images i set certain parameters at first as the train.py
but when i tried training code on them they give me the following error.
how can i get rid of this error? and how can the train be done on the colored masks of ade20k dataset not gray masks? Thanks