I tried to run a train on the personal short video (in reallity only 4 frames and 1 mask -- video of me moving a speaker behind the laptop and return) for the model from this paper omnimotion.
I run it with set limit to 500 steps:
python train.py --config configs/default.txt --data_dir "Users\XXX" --num_iters 500
but the train does not stop on the 500 steps and then get such error:
OMP: Error #15: Initializing libiomp5md.dll,
but found libiomp5md.dll already initialized.
OMP: Hint This means that multiple copies of the OpenMP runtime
have been linked into the program.
That is dangerous, since it can degrade performance or
cause incorrect results.
The best thing to do is to ensure that only a single OpenMP runtime
is linked into the process, e.g. by avoiding static linking of the OpenMP runtime
in any library.
As an unsafe, unsupported, undocumented workaround you can set the environment variable
KMP_DUPLICATE_LIB_OK=TRUE to allow the program to continue to execute,
but that may cause crashes or silently produce incorrect results.
For more information, please see http://www.intel.com/software/products/support/
Is it safe to try to remove such limitation?
I have 3070 laptop version with 8GB
I want to try to run this train localy on my laptop. I did all the preprocess parts from the githab and didn't change much the code. I hope that when I manually put the number of iteration and other information in calls in terminal I should get exact hyperparameters as I want.