File size: 2,121 Bytes
e58ac13
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
05/25/2023 17:40:41 - INFO - __main__ - ***** Running training *****
05/25/2023 17:40:41 - INFO - __main__ -   Num examples = 833
05/25/2023 17:40:41 - INFO - __main__ -   Num Epochs = 72
05/25/2023 17:40:41 - INFO - __main__ -   Instantaneous batch size per device = 1
05/25/2023 17:40:41 - INFO - __main__ -   Total train batch size (w. parallel, distributed & accumulation) = 4
05/25/2023 17:40:41 - INFO - __main__ -   Gradient Accumulation steps = 4
05/25/2023 17:40:41 - INFO - __main__ -   Total optimization steps = 15000
Steps:   0%|                                                                                    | 0/15000 [00:00<?, ?it/s]Traceback (most recent call last):
  File "lora_test_1.py", line 908, in <module>
    main()
  File "lora_test_1.py", line 728, in main
    latents = vae.encode(batch["pixel_values"].to(dtype=weight_dtype)).latent_dist.sample()
  File "/home/long.qul/miniconda3/envs/triton/lib/python3.8/site-packages/diffusers/utils/accelerate_utils.py", line 46, in wrapper
    return method(self, *args, **kwargs)
  File "/home/long.qul/miniconda3/envs/triton/lib/python3.8/site-packages/diffusers/models/autoencoder_kl.py", line 164, in encode
    h = self.encoder(x)
  File "/home/long.qul/miniconda3/envs/triton/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1501, in _call_impl
    return forward_call(*args, **kwargs)
  File "/home/long.qul/miniconda3/envs/triton/lib/python3.8/site-packages/diffusers/models/vae.py", line 109, in forward
    sample = self.conv_in(sample)
  File "/home/long.qul/miniconda3/envs/triton/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1501, in _call_impl
    return forward_call(*args, **kwargs)
  File "/home/long.qul/miniconda3/envs/triton/lib/python3.8/site-packages/torch/nn/modules/conv.py", line 463, in forward
    return self._conv_forward(input, self.weight, self.bias)
  File "/home/long.qul/miniconda3/envs/triton/lib/python3.8/site-packages/torch/nn/modules/conv.py", line 459, in _conv_forward
    return F.conv2d(input, weight, bias, self.stride,
RuntimeError: cuDNN error: CUDNN_STATUS_NOT_INITIALIZED