masahi opened a new pull request, #14602: URL: https://github.com/apache/tvm/pull/14602
I got an out-of-memory error on my 16G linux system when trying to trace VAE in web-stable-diffusion https://github.com/mlc-ai/web-stable-diffusion/blob/main/build.py#L82 (this is not an issue on Mac). This simple change let me complete the build successfully. Using the `time -v` command to measure the peak RAM usage of a script that does VAE tracing, I got Before: `Maximum resident set size (kbytes): 12288884` After (disable grad): `Maximum resident set size (kbytes): 7179364` So this is a good reduction of 5GB. @MasterJH5574 @spectrometerHBH @junrushao -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
