I tried implementing it a while back (as evident with the commented out samplers here, and adding k-diffusion as a requirement), but no dice. I get a tensor mismatch error: The size of tensor a (100) must match the size of tensor b (200) at non-singleton dimension 1
Judging from the tortoise-tts-fast repo someone here linked the other day, it seems it doesn't actually use any sampler besides the default one (not even DDIM).
I'll see about trying to remedy the error during my free time.
I tried implementing it a while back (as evident with the commented out samplers [here](https://git.ecker.tech/mrq/tortoise-tts/src/branch/main/webui.py#L673), and adding k-diffusion as a requirement), but no dice. I get a tensor mismatch error: `The size of tensor a (100) must match the size of tensor b (200) at non-singleton dimension 1`
Judging from [the tortoise-tts-fast repo](https://github.com/152334H/tortoise-tts-fast/blob/main/tortoise/api.py#L162) someone here linked the other day, it seems it doesn't actually use any sampler besides the default one (not even DDIM).
I'll see about trying to remedy the error during my free time.
DPM++2M sampler
I tried implementing it a while back (as evident with the commented out samplers here, and adding k-diffusion as a requirement), but no dice. I get a tensor mismatch error:
The size of tensor a (100) must match the size of tensor b (200) at non-singleton dimension 1Judging from the tortoise-tts-fast repo someone here linked the other day, it seems it doesn't actually use any sampler besides the default one (not even DDIM).
I'll see about trying to remedy the error during my free time.