training google colab error

[unoroberto12]

System Info

when I started training Laura in google colab, I got this error. Please help me if you can😫😫
Checking dataset...
📁MyDrive/Loras/Kamixxx_style_lora/dataset
📈 Found 11 images with 10 repeats, equaling 110 steps.
📉 Divide 110 steps by 2 batch size to get 55.0 steps per epoch.
🔮 There will be 10 epochs, for around 550 total training steps.

✅ Dependencies already installed.

🔄 Model already downloaded.

📄 Config saved to /content/drive/MyDrive/Loras/Kamixxx_style_lora/training_config.toml
📄 Dataset config saved to /content/drive/MyDrive/Loras/Kamixxx_style_lora/dataset_config.toml

⭐ Starting trainer...

ipex flag is deprecated, will be removed in Accelerate v1.10. From 2.7.0, PyTorch has all needed optimizations for Intel CPU and XPU.
WARNING: All log messages before absl::InitializeLog() is called are written to STDERR
E0000 00:00:1752956937.419400 2914 cuda_dnn.cc:8310] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered
E0000 00:00:1752956937.429199 2914 cuda_blas.cc:1418] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered
Traceback (most recent call last):
File "/content/kohya-trainer/train_network.py", line 17, in
import library.train_util as train_util
File "/content/kohya-trainer/library/train_util.py", line 1768, in
def replace_unet_modules(unet: diffusers.models.unet_2d_condition.UNet2DConditionModel, mem_eff_attn, xformers):
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.11/dist-packages/diffusers/utils/import_utils.py", line 876, in getattr
raise AttributeError(f"module {self.name} has no attribute {name}")
AttributeError: module diffusers.models has no attribute unet_2d_condition. Did you mean: 'unets.unet_2d_condition'?
Traceback (most recent call last):
File "/usr/local/bin/accelerate", line 10, in
sys.exit(main())
^^^^^^
File "/usr/local/lib/python3.11/dist-packages/accelerate/commands/accelerate_cli.py", line 50, in main
args.func(args)
File "/usr/local/lib/python3.11/dist-packages/accelerate/commands/launch.py", line 1199, in launch_command
simple_launcher(args)
File "/usr/local/lib/python3.11/dist-packages/accelerate/commands/launch.py", line 785, in simple_launcher
raise subprocess.CalledProcessError(returncode=process.returncode, cmd=cmd)
subprocess.CalledProcessError: Command '['/usr/bin/python3', 'train_network.py', '--dataset_config=/content/drive/MyDrive/Loras/Kamixxx_style_lora/dataset_config.toml', '--config_file=/content/drive/MyDrive/Loras/Kamixxx_style_lora/training_config.toml']' returned non-zero exit status 1.

Who can help?

No response

Information

• The official example scripts
• My own modified scripts

Tasks

• An officially supported task in the examples folder (such as GLUE/SQuAD, ...)
• My own task or dataset (give details below)

Reproduction

.

Expected behavior

.

[ArkVex]

System Info

when I started training Laura in google colab, I got this error. Please help me if you can😫😫 Checking dataset... 📁MyDrive/Loras/Kamixxx_style_lora/dataset 📈 Found 11 images with 10 repeats, equaling 110 steps. 📉 Divide 110 steps by 2 batch size to get 55.0 steps per epoch. 🔮 There will be 10 epochs, for around 550 total training steps.

✅ Dependencies already installed.

🔄 Model already downloaded.

📄 Config saved to /content/drive/MyDrive/Loras/Kamixxx_style_lora/training_config.toml 📄 Dataset config saved to /content/drive/MyDrive/Loras/Kamixxx_style_lora/dataset_config.toml

⭐ Starting trainer...

ipex flag is deprecated, will be removed in Accelerate v1.10. From 2.7.0, PyTorch has all needed optimizations for Intel CPU and XPU. WARNING: All log messages before absl::InitializeLog() is called are written to STDERR E0000 00:00:1752956937.419400 2914 cuda_dnn.cc:8310] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered E0000 00:00:1752956937.429199 2914 cuda_blas.cc:1418] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered Traceback (most recent call last): File "/content/kohya-trainer/train_network.py", line 17, in import library.train_util as train_util File "/content/kohya-trainer/library/train_util.py", line 1768, in def replace_unet_modules(unet: diffusers.models.unet_2d_condition.UNet2DConditionModel, mem_eff_attn, xformers): ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/usr/local/lib/python3.11/dist-packages/diffusers/utils/import_utils.py", line 876, in getattr raise AttributeError(f"module {self.name} has no attribute {name}") AttributeError: module diffusers.models has no attribute unet_2d_condition. Did you mean: 'unets.unet_2d_condition'? Traceback (most recent call last): File "/usr/local/bin/accelerate", line 10, in sys.exit(main()) ^^^^^^ File "/usr/local/lib/python3.11/dist-packages/accelerate/commands/accelerate_cli.py", line 50, in main args.func(args) File "/usr/local/lib/python3.11/dist-packages/accelerate/commands/launch.py", line 1199, in launch_command simple_launcher(args) File "/usr/local/lib/python3.11/dist-packages/accelerate/commands/launch.py", line 785, in simple_launcher raise subprocess.CalledProcessError(returncode=process.returncode, cmd=cmd) subprocess.CalledProcessError: Command '['/usr/bin/python3', 'train_network.py', '--dataset_config=/content/drive/MyDrive/Loras/Kamixxx_style_lora/dataset_config.toml', '--config_file=/content/drive/MyDrive/Loras/Kamixxx_style_lora/training_config.toml']' returned non-zero exit status 1.

Who can help?

No response

Information

• The official example scripts[ ] My own modified scripts

Tasks

• An officially supported task in the examples folder (such as GLUE/SQuAD, ...)[ ] My own task or dataset (give details below)

Reproduction

.

Expected behavior

.

Hey! The error is happening because the diffusers library has changed its internal structure; net_2d_condition is no longer directly under diffusers.models.

Fix: Downgrade diffusers to version 0.20.2, which works with the trainer script.

Run this:

pip install diffusers==0.20.2

Then rerun your training. This should solve the "AttributeError: module diffusers.models has no attribute unet_2d_condition" issue.

Let me know if it works!

[unoroberto12]

@ArkVex sorry for asking a stupid question, but how do I run this command?😅😅 Like just write it in the command line or something..

[ArkVex]

@ArkVex sorry for asking a stupid question, but how do I run this command?😅😅 Like just write it in the command line or something..

No worries, not a stupid question at all! 😄
Yes just open a new code cell in your Google Colab notebook, paste this line, and run it:

!pip install diffusers==0.20.2

The exclamation mark at the beginning tells Colab to run it as a terminal command.
Once it finishes installing, re-run your training cell. Let me know if anything breaks!

[unoroberto12]

@ArkVex Unfortunately, it didn't work

[ArkVex]

@ArkVex Unfortunately, it didn't work

Can you show me the logs

[unoroberto12]

@ArkVex I don't have any logs. Well, I have this folder in gd, but there's nothing in it

[yaswanth19]

@unoroberto12 You have to sometimes restart the kernel for the installed package to load otherwise it uses the previous package version.

First install using !pip install diffusers==0.20.2 and then restart the kernel.
Later verify using below code whether the correct package version is installed or not.

import diffusers
diffusers.__version__

[unoroberto12]

@yaswanth19 That didn't work either.