v1.14: module 'torch.library' has no attribute 'register_fake'

[thistleknot]

Describe the bug

running latest build results in torch error

python server.py --api --listen --n-gpu-layers 32 --threads 8 --numa --tensorcores --trust-remote-code

...

RuntimeError: Failed to import transformers.models.auto.processing_auto because of the following error
(look up to see its traceback):
module 'torch.library' has no attribute 'register_fake'

this url says this error is a result of torch vision? https://github.com/lllyasviel/IC-Light/issues/77

Is there an existing issue for this?

• [X] I have searched the existing issues

Reproduction

pip install -r requirements.txt

python server.py --api --listen --n-gpu-layers 32 --threads 8 --numa --tensorcores --trust-remote-code

Screenshot

No response

Logs

(textgen) [root@pve-m7330 text-generation-webui]# !903
python server.py --api --listen --n-gpu-layers 32 --threads 8 --numa --tensorcores --trust-remote-code
╭──────────────────────────────── Traceback (most recent call last) ─────────────────────────────────╮
│ /home/user/miniconda3/envs/textgen/lib/python3.10/site-packages/transformers/utils/import_utils.py │
│ :1603 in _get_module                                                                               │
│                                                                                                    │
│   1602         try:                                                                                │
│ ❱ 1603             return importlib.import_module("." + module_name, self.__name__)                │
│   1604         except Exception as e:                                                              │
│                                                                                                    │
│ /home/user/miniconda3/envs/textgen/lib/python3.10/importlib/__init__.py:126 in import_module       │
│                                                                                                    │
│   125             level += 1                                                                       │
│ ❱ 126     return _bootstrap._gcd_import(name[level:], package, level)                              │
│   127                                                                                              │
│ in _gcd_import:1050                                                                                │
│ in _find_and_load:1027                                                                             │
│ in _find_and_load_unlocked:1006                                                                    │
│                                                                                                    │
│                                      ... 4 frames hidden ...                                       │
│                                                                                                    │
│ /home/user/miniconda3/envs/textgen/lib/python3.10/site-packages/transformers/image_processing_util │
│ s.py:21 in <module>                                                                                │
│                                                                                                    │
│    20 from .image_processing_base import BatchFeature, ImageProcessingMixin                        │
│ ❱  21 from .image_transforms import center_crop, normalize, rescale                                │
│    22 from .image_utils import ChannelDimension                                                    │
│                                                                                                    │
│ /home/user/miniconda3/envs/textgen/lib/python3.10/site-packages/transformers/image_transforms.py:2 │
│ 2 in <module>                                                                                      │
│                                                                                                    │
│    21                                                                                              │
│ ❱  22 from .image_utils import (                                                                   │
│    23     ChannelDimension,                                                                        │
│                                                                                                    │
│ /home/user/miniconda3/envs/textgen/lib/python3.10/site-packages/transformers/image_utils.py:58 in  │
│ <module>                                                                                           │
│                                                                                                    │
│    57     if is_torchvision_available():                                                           │
│ ❱  58         from torchvision.transforms import InterpolationMode                                 │
│    59                                                                                              │
│                                                                                                    │
│ /home/user/miniconda3/envs/textgen/lib/python3.10/site-packages/torchvision/__init__.py:10 in      │
│ <module>                                                                                           │
│                                                                                                    │
│     9 from .extension import _HAS_OPS  # usort:skip                                                │
│ ❱  10 from torchvision import _meta_registrations, datasets, io, models, ops, transforms, util     │
│    11                                                                                              │
│                                                                                                    │
│ /home/user/miniconda3/envs/textgen/lib/python3.10/site-packages/torchvision/_meta_registrations.py │
│ :163 in <module>                                                                                   │
│                                                                                                    │
│   162                                                                                              │
│ ❱ 163 @torch.library.register_fake("torchvision::nms")                                             │
│   164 def meta_nms(dets, scores, iou_threshold):                                                   │
╰────────────────────────────────────────────────────────────────────────────────────────────────────╯
AttributeError: module 'torch.library' has no attribute 'register_fake'

The above exception was the direct cause of the following exception:

╭──────────────────────────────── Traceback (most recent call last) ─────────────────────────────────╮
│ /home/user/text-generation-webui/server.py:40 in <module>                                          │
│                                                                                                    │
│    39 import modules.extensions as extensions_module                                               │
│ ❱  40 from modules import (                                                                        │
│    41     chat,                                                                                    │
│                                                                                                    │
│ /home/user/text-generation-webui/modules/training.py:21 in <module>                                │
│                                                                                                    │
│    20 from datasets import Dataset, load_dataset                                                   │
│ ❱  21 from peft import (                                                                           │
│    22     LoraConfig,                                                                              │
│                                                                                                    │
│ /home/user/miniconda3/envs/textgen/lib/python3.10/site-packages/peft/__init__.py:22 in <module>    │
│                                                                                                    │
│    21                                                                                              │
│ ❱  22 from .auto import (                                                                          │
│    23     AutoPeftModel,                                                                           │
│                                                                                                    │
│ /home/user/miniconda3/envs/textgen/lib/python3.10/site-packages/peft/auto.py:32 in <module>        │
│                                                                                                    │
│    31 from .config import PeftConfig                                                               │
│ ❱  32 from .mapping import MODEL_TYPE_TO_PEFT_MODEL_MAPPING                                        │
│    33 from .peft_model import (                                                                    │
│                                                                                                    │
│ /home/user/miniconda3/envs/textgen/lib/python3.10/site-packages/peft/mapping.py:22 in <module>     │
│                                                                                                    │
│    21                                                                                              │
│ ❱  22 from peft.tuners.xlora.model import XLoraModel                                               │
│    23                                                                                              │
│                                                                                                    │
│                                      ... 7 frames hidden ...                                       │
│                                                                                                    │
│ /home/user/miniconda3/envs/textgen/lib/python3.10/site-packages/awq/models/base.py:35 in <module>  │
│                                                                                                    │
│    34 from awq.utils.utils import get_best_device, qbits_available                                 │
│ ❱  35 from transformers import (                                                                   │
│    36     AutoConfig,                                                                              │
│ in _handle_fromlist:1075                                                                           │
│                                                                                                    │
│ /home/user/miniconda3/envs/textgen/lib/python3.10/site-packages/transformers/utils/import_utils.py │
│ :1594 in __getattr__                                                                               │
│                                                                                                    │
│   1593             module = self._get_module(self._class_to_module[name])                          │
│ ❱ 1594             value = getattr(module, name)                                                   │
│   1595         else:                                                                               │
│                                                                                                    │
│ /home/user/miniconda3/envs/textgen/lib/python3.10/site-packages/transformers/utils/import_utils.py │
│ :1593 in __getattr__                                                                               │
│                                                                                                    │
│   1592         elif name in self._class_to_module.keys():                                          │
│ ❱ 1593             module = self._get_module(self._class_to_module[name])                          │
│   1594             value = getattr(module, name)                                                   │
│                                                                                                    │
│ /home/user/miniconda3/envs/textgen/lib/python3.10/site-packages/transformers/utils/import_utils.py │
│ :1605 in _get_module                                                                               │
│                                                                                                    │
│   1604         except Exception as e:                                                              │
│ ❱ 1605             raise RuntimeError(                                                             │
│   1606                 f"Failed to import {self.__name__}.{module_name} because of the followin    │
╰────────────────────────────────────────────────────────────────────────────────────────────────────╯
RuntimeError: Failed to import transformers.models.auto.processing_auto because of the following error
(look up to see its traceback):
module 'torch.library' has no attribute 'register_fake'

### System Info

```shell
python 3.10
rocky linux 9
p5200 (compute 60)

confirm v1.13 works

[thistleknot]

turns out to be a package issue

I installed the reqs from v1.13 then switched back to v1.14

installed torch for 12.1 and ran just fine