Is this latest version ok with centos 7.6 / 2023.1 as well?
@talosh I’m getting an error that I’ve never seen before:
Executing command rm -f
[Errno 39] Directory not empty:
and then I get the same for path to outgoing
Any ideas? I’ve tried chmod 777 -R the whole directory where it’s writing incoming/outgoing
1 Like
Trying to make it work on my M1 pro. Got this in terminal. What should i do? It look like its rendering but nothing happen.
in/bash: /Users/seb/Documents/flameTimewarpML/miniconda3/bin/conda: No such file or directory
/bin/bash: conda: command not found
Traceback (most recent call last):
File “/Users/seb/Documents/flameTimewarpML/bundle/inference_sequence.py”, line 3, in
import cv2
ModuleNotFoundError: No module named ‘cv2’
Saving session…
…copying shared history…
…saving history…truncating history files…
…completed.
I have tried running it as well on my M1 Max and actually it does work for a while. Getting slower and slower and then at one point it says:
rosetta eror: thread_suspend failed
Thats when I’m trying to render a timewarp. I successfully rendered a fluidmorph once , then it started to fail as well. Would be really nice to have a stable silicon version
1 Like
Can’t wait to have this working on my M1!!
I updated the PyTorch today following the Link posted here earlier.
if you missed it, its this one here
Since I did that, it kind of works, still slow but I get at least the amazing results. But also I would love to have this natively integrated. Would be a smart step from Autodesk in my mind
1 Like
amazing !! thanks.
We have talked to adsk about this at the detroit user group, its definetely on their radar 
2 Likes
I may have to uninstall re instal TimewarpML . how should i do this ?
Updated . but still not working. i may have miss something along the road:
Here my shell when i start a render; Any idea?
/bin/bash: /Users/seb/Documents/flameTimewarpML/miniconda3/bin/conda: No such file or directory
CommandNotFoundError: Your shell has not been properly configured to use ‘conda activate’.
To initialize your shell, run
$ conda init <SHELL_NAME>
Currently supported shells are:
- bash
- fish
- tcsh
- xonsh
- zsh
- powershell
See ‘conda init --help’ for more information and options.
IMPORTANT: You may need to close and restart your shell after running ‘conda init’.
Traceback (most recent call last):
File “/Users/seb/Documents/flameTimewarpML/bundle/inference_sequence.py”, line 3, in
import cv2
ModuleNotFoundError: No module named ‘cv2’
Saving session…
…saving history…truncating history files…
…completed.
[Process completed]
Hey Andy,
Did you manage to fix this, I’m suddenly seeing this as well too…
Cheers, Ton
Hey Ton,
Apologies…I forgot to post the solution. I had thought I had copied all the necessary files to my new machine, but I had missed something. A fresh install solved the problem.
My coworker writes: “When using the ML Timewarp, be sure to have the media hub set to File Sequence. Otherwise a bunch of stills will be imported.” So keep an eye out for that one!
4 Likes
Woah! Thank your co-worker for me. I could not for the life of me figure out why I had to keep loading and joining all the imported stills out of nowhere! It all makes sense now!
Genius!
2 Likes
Hi Seb,
I’m getting this on my (non-M1) Mac now. Did you ever get it working?
I’ve deleted it from //shared/python/ and from the docs folder, so the minicoda3 error has gone away, but still no ‘cv2’ file found.
Thanks,
Rich
Have you tried enabling the “CPU Proc” in the export dialogue? That normally fixes this error for me.
Fixed this on Mac by uninstalling all components and loading up the latest Mac specific version from github (from 2021) - I had assumed the linux would be compatible but either it wasn’t or I had screwed up the install…
Sorry Rich,
its build upon pytorch that is a facebook open source machine learinig engine and it is fairly different on mac and linux.
Linux is using NVIDIA and CUDA whilst Mac is CPU-only on Intel and there are some sort of acceleration on for Apple silicon but I didn’t had a chance to test it yet
2 Likes