Update src/pipeline.py
Browse files- src/pipeline.py +2 -3
src/pipeline.py
CHANGED
|
@@ -11,7 +11,6 @@ from torch import Generator
|
|
| 11 |
from torchao.quantization import quantize_, int8_weight_only
|
| 12 |
from transformers import T5EncoderModel, CLIPTextModel, logging
|
| 13 |
import torch._dynamo
|
| 14 |
-
import torch_tensorrt
|
| 15 |
# torch._dynamo.config.suppress_errors = True
|
| 16 |
|
| 17 |
Pipeline: TypeAlias = FluxPipeline
|
|
@@ -47,11 +46,11 @@ def load_pipeline() -> Pipeline:
|
|
| 47 |
local_files_only=True,
|
| 48 |
torch_dtype=torch.bfloat16,
|
| 49 |
).to("cuda")
|
| 50 |
-
|
| 51 |
# pipeline.transformer.to(memory_format=torch.channels_last)
|
| 52 |
# pipeline.vae.to(memory_format=torch.channels_last)
|
| 53 |
|
| 54 |
-
quantize_(pipeline.vae, int8_weight_only())
|
| 55 |
|
| 56 |
PROMPT = 'semiconformity, peregrination, quip, twineless, emotionless, tawa, depickle'
|
| 57 |
with torch.inference_mode():
|
|
|
|
| 11 |
from torchao.quantization import quantize_, int8_weight_only
|
| 12 |
from transformers import T5EncoderModel, CLIPTextModel, logging
|
| 13 |
import torch._dynamo
|
|
|
|
| 14 |
# torch._dynamo.config.suppress_errors = True
|
| 15 |
|
| 16 |
Pipeline: TypeAlias = FluxPipeline
|
|
|
|
| 46 |
local_files_only=True,
|
| 47 |
torch_dtype=torch.bfloat16,
|
| 48 |
).to("cuda")
|
| 49 |
+
pipeline.to(memory_format=torch.channels_last)
|
| 50 |
# pipeline.transformer.to(memory_format=torch.channels_last)
|
| 51 |
# pipeline.vae.to(memory_format=torch.channels_last)
|
| 52 |
|
| 53 |
+
# quantize_(pipeline.vae, int8_weight_only())
|
| 54 |
|
| 55 |
PROMPT = 'semiconformity, peregrination, quip, twineless, emotionless, tawa, depickle'
|
| 56 |
with torch.inference_mode():
|