A tool to count OPs and paramters of MXNet model.
It only works for gluon yet.
Reference: THOP: PyTorch-OpCounter
pip install --index-url https://pypi.org/simple/ mxop
- Github (latest)
pip install --upgrade git+https://github.com/hey-yahei/OpSummary.MXNet.git
- Count OPs
from mxop.gluon import count_ops op_counter = count_ops(net) # net is the gluon model you want to count OPs
- Count parameters
from mxop.gluon import count_params op_counter = count_params(net, input_size) # net is the gluon model you want to count OPs # input_size is the shape of your input
- Print summary
from mxop.gluon import op_summary op_counter = op_summary(net, input_size) # net is the gluon model you want to count OPs # input_size is the shape of your input
- Count OPs and parameters for each layer.
- Support Symbol model for MXNet.
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.