I am using this config on a translation model (Helsinki-NLP/opus-mt-zh-en), and I check the size of the model using the following function before and after running init_compression and deepspeed.initialize:
Weirdly, the size of the model increases after running init_compression and deepspeed.initialize. Even after I use redundancy_clean at the end of training and save the model to disk, the size of the model stays what had been returned by print_size_of_model after running init_compression and deepspeed.initialize.
Hello,
I am using this config on a translation model (Helsinki-NLP/opus-mt-zh-en), and I check the size of the model using the following function before and after running
init_compression
anddeepspeed.initialize
:Weirdly, the size of the model increases after running
init_compression
anddeepspeed.initialize
. Even after I useredundancy_clean
at the end of training and save the model to disk, the size of the model stays what had been returned byprint_size_of_model
after runninginit_compression
anddeepspeed.initialize
.Am I missing something? Can you please explain?
Thanks a lot