diff options
| author | AUTOMATIC1111 <16777216c@gmail.com> | 2022-12-03 09:58:08 +0300 |
|---|---|---|
| committer | GitHub <noreply@github.com> | 2022-12-03 09:58:08 +0300 |
| commit | a2feaa95fc0c4c94131eb75b5b1bc0eaa1696551 (patch) | |
| tree | a7be0e8b6849aae24f7f6c7879f7ddf43d118425 /modules/textual_inversion | |
| parent | c7af672186ec09a514f0e78aa21155264e56c130 (diff) | |
| parent | 0fddb4a1c06a6e2122add7eee3b001a6d473baee (diff) | |
Merge pull request #5194 from brkirch/autocast-and-mps-randn-fixes
Use devices.autocast() and fix MPS randn issues
Diffstat (limited to 'modules/textual_inversion')
| -rw-r--r-- | modules/textual_inversion/dataset.py | 4 | ||||
| -rw-r--r-- | modules/textual_inversion/textual_inversion.py | 2 |
2 files changed, 3 insertions, 3 deletions
diff --git a/modules/textual_inversion/dataset.py b/modules/textual_inversion/dataset.py index e5725f33..2dc64c3c 100644 --- a/modules/textual_inversion/dataset.py +++ b/modules/textual_inversion/dataset.py @@ -82,7 +82,7 @@ class PersonalizedBase(Dataset): torchdata = torch.from_numpy(npimage).permute(2, 0, 1).to(device=device, dtype=torch.float32)
latent_sample = None
- with torch.autocast("cuda"):
+ with devices.autocast():
latent_dist = model.encode_first_stage(torchdata.unsqueeze(dim=0))
if latent_sampling_method == "once" or (latent_sampling_method == "deterministic" and not isinstance(latent_dist, DiagonalGaussianDistribution)):
@@ -101,7 +101,7 @@ class PersonalizedBase(Dataset): entry.cond_text = self.create_text(filename_text)
if include_cond and not (self.tag_drop_out != 0 or self.shuffle_tags):
- with torch.autocast("cuda"):
+ with devices.autocast():
entry.cond = cond_model([entry.cond_text]).to(devices.cpu).squeeze(0)
self.dataset.append(entry)
diff --git a/modules/textual_inversion/textual_inversion.py b/modules/textual_inversion/textual_inversion.py index 4eb75cb5..daf8d1b8 100644 --- a/modules/textual_inversion/textual_inversion.py +++ b/modules/textual_inversion/textual_inversion.py @@ -316,7 +316,7 @@ def train_embedding(embedding_name, learn_rate, batch_size, gradient_step, data_ if shared.state.interrupted:
break
- with torch.autocast("cuda"):
+ with devices.autocast():
# c = stack_conds(batch.cond).to(devices.device)
# mask = torch.tensor(batch.emb_index).to(devices.device, non_blocking=pin_memory)
# print(mask)
|