曹国旭 comments

Results 9 comments of


                                            曹国旭

Is this possible to train this project to detect my own dataset's keypoints?

Thanks and I already have the annotation file, but is there any way to train with my own data with this project? I didn't find it in README

Use own images set to create voxelized shape

same issue

Use own images set to create voxelized shape

Still no solution

ValueError: total size of new array must be unchanged when loading model

Same issue. Any solution?

ValueError: total size of new array must be unchanged when loading model

Okay, I saw the solution in #7 issue.

How to write statements like "select count(*) from table_name where col_name>0"?

@meirdev @ThirVondukr just out of curiosity, how you guys found this solution? It's not in the official document

OOM issue when finetune with V100

@bingjie3216 Out of curiosity, how much RAM you use when train dolly on 8 V100? I am currently try to reproduce this on 8 V100 too but I just have...

Finished training but the inference performance doesn't look good

I'm facing the same issue with opt-1.3b model.

Finished training but the inference performance doesn't look good

@mrwyattii Could you provide the detailed training configuration for the opt-1.3b conversation results proposed in the README? I attempted to reproduce the results on 8 * V100, but unfortunately, the...