Ruoxi comments

Results 107 comments of


                                            Ruoxi

Create wandb run with run name that was previously deleted

I also met this problem today. Deleted things should behave as if they never existed, or they are _NOT_ deleted.

Metacache: find entry error: get /xxx : leveldb: closed

Met this problem today again

Internal Server Error and EIO on fuse mount for remote EC shard read failure.

> It is tricky to set up credentials on remote server for google drive, so I made a s3 share instead. > > endpoint: https://s3-haosu.nrp-nautilus.io bucket: seaweed > > Could...

@tf.function works the first time running the code but fails afterwards on Spyder IDE.

Sorry for my delay in reply, and.. no, since we only have the modules generated after tensorflow is run, and when umr is initialized there is no such tmp module...

would you release the training code in the future?

Though not official, you can try the finetune code from https://github.com/TencentARC/InstantMesh I cannot release the training code for non-technical reasons..

CUDA out of memory. Tried to allocate error

Nowadays we do have more techniques to reduce inference-time memory, including [model offloading](https://huggingface.co/docs/diffusers/en/optimization/memory#model-offloading), [autotune compile](https://huggingface.co/docs/diffusers/en/optimization/torch2.0#torchcompile), quantization and maybe others. We do not have the code for these ready now but...

Depth map size ratios for depth ControlNet

The size should be the resolution of your render output and can be fixed to a same value. I am not sure why you have different-sized depths.

Can I get train code?

Though not official, you can try the finetune code from https://github.com/TencentARC/InstantMesh

Ruoxi