SSD-Tensorflow icon indicating copy to clipboard operation
SSD-Tensorflow copied to clipboard

Training For Custom Images and Classes

Open SrikanthVelpuri opened this issue 6 years ago • 4 comments

Can you please write a good readme for implemetation of SSD for custom objects and classes. or Someone Please help me train on my own images...I have all images and thier bounding boxes...All I need is the process to implement it in SSD

SrikanthVelpuri avatar Mar 21 '18 11:03 SrikanthVelpuri

I have the same question, do you have any solution?

anamissblonde avatar Jun 21 '18 15:06 anamissblonde

I have the same question,too

Salight avatar Jun 28 '18 05:06 Salight

I have problem restoring the checkpoint when training custom classes:

NotFoundError (see above for traceback): Restoring from checkpoint failed. This is most likely due to a Variable name or other graph key that is missing from the checkpoint. Please ensure that you have not altered the graph expected based on the checkpoint. Original error:

Key beta1_power not found in checkpoint
	 [[node save/RestoreV2 (defined at train_ssd_network.py:372)  = RestoreV2[dtypes=[DT_FLOAT, DT_FLOAT, DT_INT64, DT_FLOAT, DT_FLOAT, ..., DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT], _device="/job:localhost/replica:0/task:0/device:CPU:0"](_arg_save/Const_0_0, save/RestoreV2/tensor_names, save/RestoreV2/shape_and_slices)]]

Anyone knows how to solve it? Thanks a lot!!!

xiyuanzh avatar Feb 27 '19 06:02 xiyuanzh

@xiyuanzh The checkpoint from the provided model?

satyajithj avatar May 16 '19 17:05 satyajithj