AngledLuffa
commented
1 year ago This is not my experience. I have a Mac, and I can load the Italian pipeline just fine. Of course, there's the question of what versions are being run. I did this on a MacBook Air (M2, 2022), with an Apple M2 (not used because of pytorch mps bugs), OS version 12.6.3. Python 3.9.6, Stanza 1.5.0. So, my suggestion is that I'll try updating my OS, and you try updating Stanza, and we'll see if either of us gets different results.
On Tue, Mar 14, 2023 at 9:00 AM Robert Alexander @.***> wrote:
Describe the bug When launching my project on Apple M1 hardware I get the following error:
(isagog-ai-py3.10) (base) @.*** isagog-ai % python src/isagog_api/nlp_api.py 2023-03-14 16:05:29 INFO: Loading these models for language: it (Italian):
| Processor | Package |
| tokenize | combined | | mwt | combined | | ner | fbk |
2023-03-14 16:05:29 INFO: Use device: cpu 2023-03-14 16:05:29 INFO: Loading: tokenize 2023-03-14 16:05:29 INFO: Loading: mwt Traceback (most recent call last): File "/Users/bob/Documents/work/code/isagog-ai/src/isagog_api/nlp_api.py", line 14, in
from isagog_ai.nlp_it import StanzaLanguageProcessorIt File "/Users/bob/Documents/work/code/isagog-ai/src/isagog_ai/nlp_it.py", line 29, in "entity": stanza.Pipeline( File "/Users/bob/Documents/work/code/isagog-ai/.venv/lib/python3.10/site-packages/stanza/pipeline/core.py", line 278, in init self.processors[processor_name] = NAME_TO_PROCESSOR_CLASS[processor_name](config=curr_processor_config, File "/Users/bob/Documents/work/code/isagog-ai/.venv/lib/python3.10/site-packages/stanza/pipeline/processor.py", line 173, in init self._set_up_model(config, pipeline, use_gpu) File "/Users/bob/Documents/work/code/isagog-ai/.venv/lib/python3.10/site-packages/stanza/pipeline/mwt_processor.py", line 21, in _set_up_model self._trainer = Trainer(model_file=config['model_path'], use_cuda=use_gpu) File "/Users/bob/Documents/work/code/isagog-ai/.venv/lib/python3.10/site-packages/stanza/models/mwt/trainer.py", line 36, in init self.load(model_file, use_cuda) File "/Users/bob/Documents/work/code/isagog-ai/.venv/lib/python3.10/site-packages/stanza/models/mwt/trainer.py", line 149, in load self.model.load_state_dict(checkpoint['model']) File "/Users/bob/Documents/work/code/isagog-ai/.venv/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1671, in load_state_dict raise RuntimeError('Error(s) in loading state_dict for {}:\n\t{}'.format( RuntimeError: Error(s) in loading state_dict for Seq2SeqModel: Unexpected key(s) in state_dict: "SOS_tensor". Expected behavior Service launching just like with the exact same project on Intel.
Environment (please complete the following information):
- OS: MacOS Ventura 13.2.1
- Python version: 3.10.9
- Stanza version: 1.4.2
— Reply to this email directly, view it on GitHub https://github.com/stanfordnlp/stanza/issues/1216, or unsubscribe https://github.com/notifications/unsubscribe-auth/AA2AYWNA43S72AMPETEP5Q3W4CI2TANCNFSM6AAAAAAV2URTZM . You are receiving this because you are subscribed to this thread.Message ID: @.***>
rjalexa
Describe the bug When launching my project on Apple M1 hardware I get the following error:
Expected behavior Service launching just like with the exact same project on Intel.
Environment (please complete the following information):