Jiatong (Julius) Han

Thanks for your helpful answers!

> ModuleNotFoundError: No module named 'colossalai' It could be a problem of having ambiguity in the python you used for installing `colossalai`. Can you please show `which python`?

I think it is being auto-added during inference without having to manually add to configs. See: https://github.com/hpcaitech/Open-Sora/blob/ee909a7d6611bcc9c5cf1ac055a7e9cc74157e09/docs/report_03.md?plain=1#L121

What is your result printed from `print("Camera motion: ", cfg.get("camera_motion", None))`?

I believe it is already been possible, as T5 encoder loads from pretrained weights from huggingface: https://github.com/hpcaitech/Open-Sora/blob/ee909a7d6611bcc9c5cf1ac055a7e9cc74157e09/opensora/models/text_encoder/t5.py#L44.

Thanks for raising your concern. Can @zhengzangw please take a look to see if any action is needed? Thanks!

It is well likely due to the dataset. Can you probably follow our stage 1 first: https://github.com/hpcaitech/Open-Sora/blob/main/docs/report_03.md#first-stage, before you use your own dataset to continue training?

Can you check with `telnet yanjiuzongyuan 54437` to see if the port is available?

Your command might have been obsolete due to our iterations. Use this instead: `python scripts/inference.py configs/opensora-v1-1/inference/sample.py --prompt "A beautiful sunset over the city" --num-frames 32 --image-size 480 854`.

It should be okay since I can run on my machine with python version `3.10`; also, refer to https://github.com/hpcaitech/Open-Sora/blob/ee909a7d6611bcc9c5cf1ac055a7e9cc74157e09/docs/zh_CN/READMEv1.1.md?plain=1#L96.