I want to train flux lora on small text crops. And usually the size of these crops are small. So my multidatabackened.json looks like this
[
{
"id": "pseudo-camera-10k-flux",
"type": "local",
"crop": true,
"crop_aspect": "square",
"crop_style": "center",
"resolution": 512,
"minimum_image_size": 32,
"maximum_image_size": 32,
"target_downsample_size": 512,
"resolution_type": "pixel_area",
"cache_dir_vae": "cache/vae/flux/pseudo-camera-10k",
"instance_data_dir": "datasets/pseudo-camera-10k",
"disabled": false,
"skip_file_discovery": "",
"caption_strategy": "filename",
"metadata_backend": "discovery",
"repeats": 0
}
But I get this error
Could not load library libnvrtc.so.12. Error: libnvrtc.so.12: cannot open shared object file: No such file or directory
File "SimpleTuner/.venv/lib/python3.10/site-packages/transformers/models/clip/modeling_clip.py", line 491, in forward
attn_output = torch.nn.functional.scaled_dot_product_attention(
RuntimeError: cuDNN Frontend error: [cudnn_frontend] Error: No execution plans support the graph.
bghira
commented
2 days ago
I want to train flux lora on small text crops. And usually the size of these crops are small. So my multidatabackened.json looks like this [ { "id": "pseudo-camera-10k-flux", "type": "local", "crop": true, "crop_aspect": "square", "crop_style": "center", "resolution": 512, "minimum_image_size": 32, "maximum_image_size": 32, "target_downsample_size": 512, "resolution_type": "pixel_area", "cache_dir_vae": "cache/vae/flux/pseudo-camera-10k", "instance_data_dir": "datasets/pseudo-camera-10k", "disabled": false, "skip_file_discovery": "", "caption_strategy": "filename", "metadata_backend": "discovery", "repeats": 0 }
But I get this error
Could not load library libnvrtc.so.12. Error: libnvrtc.so.12: cannot open shared object file: No such file or directory File "SimpleTuner/.venv/lib/python3.10/site-packages/transformers/models/clip/modeling_clip.py", line 491, in forward attn_output = torch.nn.functional.scaled_dot_product_attention( RuntimeError: cuDNN Frontend error: [cudnn_frontend] Error: No execution plans support the graph.
What can I do to solve this issue?