kohya-ss / kohya-ss/sd-scripts

Dataset maker sd1.5 - wd-tagger module stopped working on colab

Open
#1,645 0 comments 0 reactions 0 assignees View on GitHub
Dominant language
Python
Stars
7.2k
Forks
1.2k
Avg merge
11m
Merged PRs (30d)
2

Description

After installing all dependencies i get that result. And not tagging at all
It goes like this for me from like 2 days.

```
Launching program...

env: PYTHONPATH=/content/kohya-trainer
2024-09-27 06:51:14.875562: E external/local_xla/xla/stream_executor/cuda/cuda_dnn.cc:9261] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered
2024-09-27 06:51:14.875618: E external/local_xla/xla/stream_executor/cuda/cuda_fft.cc:607] Unable to register cuFFT factory: Attempting to register factory for plugin cuFFT when one has already been registered
2024-09-27 06:51:14.877620: E external/local_xla/xla/stream_executor/cuda/cuda_blas.cc:1515] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered
```

Edit:
I have tried to change gpu into TPU z2-8 and getting other error
```
🚶‍♂️ Launching program...

env: PYTHONPATH=/content/kohya-trainer
[libprotobuf FATAL external/com_google_protobuf/src/google/protobuf/message.cc:258] File is already registered: xla/service/cpu/backend_config.proto
terminate called after throwing an instance of 'google::protobuf::FatalException'
what(): File is already registered: xla/service/cpu/backend_config.proto
https://symbolize.stripped_domain/r/?trace=7b566a4bb9fc,7b566a46751f&map=
*** SIGABRT received by PID 4856 (TID 4856) on cpu 24 from PID 4856; stack trace: ***
PC: @ 0x7b566a4bb9fc (unknown) pthread_kill
@ 0x7b55a26214f9 928 (unknown)
@ 0x7b566a467520 (unknown) (unknown)
https://symbolize.stripped_domain/r/?trace=7b566a4bb9fc,7b55a26214f8,7b566a46751f&map=5edeb7d86db111100e979a74159a3982:7b5592a00000-7b55a2840ba0
E0927 20:57:21.295803 4856 coredump_hook.cc:447] RAW: Remote crash data gathering hook invoked.
E0927 20:57:21.295822 4856 client.cc:272] RAW: Coroner client retries enabled (b/136286901), will retry for up to 30 sec.
E0927 20:57:21.295828 4856 coredump_hook.cc:542] RAW: Sending fingerprint to remote end.
E0927 20:57:21.295847 4856 coredump_hook.cc:551] RAW: Cannot send fingerprint to Coroner: [NOT_FOUND] stat failed on crash reporting socket /var/google/services/logmanagerd/remote_coredump.socket (Is the listener running?): No such file or directory
E0927 20:57:21.295854 4856 coredump_hook.cc:603] RAW: Dumping core locally.
E0927 20:57:24.065734 4856 process_state.cc:808] RAW: Raising signal 6 with default behavior
```

I dont know if that helps. Just trying to run it, and i cant.

Contributor guide

No contributing guide indexed for this repository

Research direction

Start with the Colab setup and the dataset maker's wd-tagger launch path; the report names no repository files or tests. Reproduce the CUDA plugin-registration failure and the alternate protobuf crash, then verify that the wd-tagger completes image tagging in the reported environment.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
machine-learning
Issue type
Bug
Difficulty
4/5
Estimated time
3-5 days
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
25/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.