diff options
Diffstat (limited to 'modules/sd_samplers_common.py')
| -rw-r--r-- | modules/sd_samplers_common.py | 14 |
1 files changed, 6 insertions, 8 deletions
diff --git a/modules/sd_samplers_common.py b/modules/sd_samplers_common.py index ceda6a35..763829f1 100644 --- a/modules/sd_samplers_common.py +++ b/modules/sd_samplers_common.py @@ -26,22 +26,20 @@ approximation_indexes = {"Full": 0, "Approx NN": 1, "Approx cheap": 2, "TAESD": def single_sample_to_image(sample, approximation=None):
-
if approximation is None:
approximation = approximation_indexes.get(opts.show_progress_type, 0)
if approximation == 2:
- x_sample = sd_vae_approx.cheap_approximation(sample)
+ x_sample = sd_vae_approx.cheap_approximation(sample) * 0.5 + 0.5
elif approximation == 1:
- x_sample = sd_vae_approx.model()(sample.to(devices.device, devices.dtype).unsqueeze(0))[0].detach()
+ x_sample = sd_vae_approx.model()(sample.to(devices.device, devices.dtype).unsqueeze(0))[0].detach() * 0.5 + 0.5
elif approximation == 3:
- x_sample = sd_vae_taesd.model()(sample.to(devices.device, devices.dtype).unsqueeze(0))[0].detach()
- x_sample = sd_vae_taesd.TAESD.unscale_latents(x_sample) # returns value in [-2, 2]
- x_sample = x_sample * 0.5
+ x_sample = sample * 1.5
+ x_sample = sd_vae_taesd.model()(x_sample.to(devices.device, devices.dtype).unsqueeze(0))[0].detach()
else:
- x_sample = processing.decode_first_stage(shared.sd_model, sample.unsqueeze(0))[0]
+ x_sample = processing.decode_first_stage(shared.sd_model, sample.unsqueeze(0))[0] * 0.5 + 0.5
- x_sample = torch.clamp((x_sample + 1.0) / 2.0, min=0.0, max=1.0)
+ x_sample = torch.clamp(x_sample, min=0.0, max=1.0)
x_sample = 255. * np.moveaxis(x_sample.cpu().numpy(), 0, 2)
x_sample = x_sample.astype(np.uint8)
|