Git Product home page Git Product logo

Comments (4)

jhj0517 avatar jhj0517 commented on September 17, 2024 1

Hi. Is this issue about the compute type fp16 not working in some gpu?
I would add compute_type in "Advanced Parameters" tab soon.

from whisper-webui.

Tom-Neverwinter avatar Tom-Neverwinter commented on September 17, 2024 1

some compute types appear to work here is what I found on the tesla m40
LargeV2, automatic, SRT [from youtube]

int8
worked as intended :)

int8_float32
worked as intended :)

int8_float16
Error: Requested int8_float16 compute type, but the target device or backend do not support efficient int8_float16 computation.

int8_bfloat16
Error: Requested int8_bfloat16 compute type, but the target device or backend do not support efficient int8_bfloat16 computation.

int16
Error: Requested int16 compute type, but the target device or backend do not support efficient int16 computation.

float16
Error: Requested float16 compute type, but the target device or backend do not support efficient float16 computation.

bfloat16
Error: Requested bfloat16 compute type, but the target device or backend do not support efficient bfloat16 computation.

float32
Error: CUDA failed with error out of memory

it should run most of these just fine, however it gets an error and ceases to go further sadly?

hmm maybe not: Regarding FP16, your GPU could support it but it does not have Tensor Cores. Currently we disable FP16 without Tensor Cores as it has worse performance than FP32 in my experience. You can override this behavior by setting the environment variable CT2_CUDA_ALLOW_FP16=1. SYSTRAN/faster-whisper#42

as an edit and I should say this first: THANK YOU! your work means a lot to people all over the world and will hopefully help many people

from whisper-webui.

jhj0517 avatar jhj0517 commented on September 17, 2024

I'm not 100% sure this would works fine, but I tried to display the available compute_type by device in #50.
Please re-open it if it does not work.

from whisper-webui.

Tom-Neverwinter avatar Tom-Neverwinter commented on September 17, 2024

hmm, I would change it back but highlight the options it thinks are supported. would give more flexibility.

I should also note I am running multiple cards so I'm not sure how it picks.

the 1080 would be different from the m40/p40

from whisper-webui.

Related Issues (20)

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.