Yujian Liu
Yujian Liu
I met the same MemoryError when fitting parameters of a large Bayesian Network. Is there any recommended solution?
@ankurankan Thanks! The data file is large so it might not be convenient to share with you. But particularly, the model has a CPD for one node that has about...
@ankurankan Thank you so much! BIFReader didn't give me error traceback, but it took more than 10 hours and still couldn't load the .bif file, and I just terminated it.
@khalibartan Thanks! Yeah that might be the problem if BIFReader tries to read the whole file into memory at once (the .bif file is 25 GB). So is there any...
@ankurankan Oh great! Thanks! I haven't try the new reader. I successfully did parameter learning in pgmpy and stored the model through pickle. I'll try the reader. BTW, I have...
@delima87 Thanks! This solves the error. However, I find that if I run the code on GPU, the model will detect nothing. The same code works fine if I run...
I see. The [config.json](https://huggingface.co/decapoda-research/llama-7b-hf/blob/main/config.json) and [generation_config.json](https://huggingface.co/decapoda-research/llama-7b-hf/blob/main/generation_config.json) both set it to 1. So I will change it to 2 for now.
@Hzfinfdu Thanks for the great work! I have a follow up question. When you say per-word NLL, do you mean to calculate $\mathcal{L}_{vlb}$ in Eq. 3 for each token? Do...
@Hzfinfdu Thanks for the reply! I have another low-level question. When you calculate NLL on test set, do you sum for all T diffusion steps, or do you sample a...