MCPcopy Create free account
hub / github.com/NVIDIA/TensorRT / get_sample_input

Method get_sample_input

demo/Diffusion/models.py:373–375  ·  view source on GitHub ↗
(self, batch_size, image_height, image_width)

Source from the content-addressed store, hash-verified

371 }
372
373 def get_sample_input(self, batch_size, image_height, image_width):
374 latent_height, latent_width = self.check_dims(batch_size, image_height, image_width)
375 return torch.randn(batch_size, 4, latent_height, latent_width, dtype=torch.float32, device=self.device)
376
377def make_VAE(version, hf_token, device, verbose, max_batch_size, inpaint=False):
378 return VAE(hf_token=hf_token, device=device, verbose=verbose, path=get_path(version, inpaint=inpaint),

Callers 1

loadEnginesMethod · 0.45

Calls 1

check_dimsMethod · 0.80

Tested by

no test coverage detected