suragnair / seqGAN

A simplified PyTorch implementation of "SeqGAN: Sequence Generative Adversarial Nets with Policy Gradient." (Yu, Lantao, et al.)
642 stars 149 forks source link

How to see the orcale data #16

Closed FreeRotate closed 3 years ago

FreeRotate commented 3 years ago

Hello sir, I'm confused about the oracle_samples.trc and oracle_EMBDIM32_HIDDENDIM32_VOCAB5000_MAXSEQLEN20.trc. How to check the data in the two file?

FreeRotate commented 3 years ago

oracle_samples.trc is real data with encode as numbers, and type is tensor. oracle_EMBDIM32_HIDDENDIM32_VOCAB5000_MAXSEQLEN20.trc is a Generator model. Is that right?

suragnair commented 3 years ago

https://github.com/suragnair/seqGAN/blob/ae8ffcd54977bd9ee177994c751f86d34f5f7aa3/main.py#L30-L33

You can use torch.load to inspect them