Junxiong Wang

movie_info_idx is missing

6

I try the least version and use imdbpy2sql.py to generate the IMDB csv files. However, all the csv files are generated except the movie_info_idx table. Do you need how to...

We are trying to reproduce HuBERT large. Could you please share some pretraining information, for example loss, perplexity, correct_m, correct_u? We are not sure where does our current model stand....

Evaluation metric bits-per-byte

1

Hi there, Megabyte paper uses bits-per-byte in Table 2 as their evaluation metric. It seems it has difference compared with byte level perplexity, since their number in arXiv and Code...