Would appreciate getting any documentation on how weights are stored in a neon model file. Is it similar/the same as keras/pytorch/tensorflow ? I'm building a converter from neon to keras/pytorch and being able to figure out how the model weights are stored will be super helpful in doing so.
Would appreciate getting any documentation on how weights are stored in a neon model file. Is it similar/the same as keras/pytorch/tensorflow ? I'm building a converter from neon to keras/pytorch and being able to figure out how the model weights are stored will be super helpful in doing so.