Hi
Can someone please explain to this newb why I get a “No model was supplied…” message during the Python import call? That is surprising to me since it should only be importing namespaces and not actually calling anything in those imported libs. I’m also a newb to Python although not to coding in general This code is running on Colab.
Thank you!
print("trace 0")
from transformers import pipeline, AutoModelForTokenClassification, AutoTokenizer
#from transformers import pipeline, AutoModel, AutoTokenizer
# Sentiment analysis pipeline
print("trace 1")
analyzer = pipeline("sentiment-analysis")
print("trace 2")
# Question answering pipeline, specifying the checkpoint identifier
print("trace 3")
oracle = pipeline(
"question-answering", model="distilbert/distilbert-base-cased-distilled-squad", tokenizer="google-bert/bert-base-cased"
)
print("trace 4")
# Named entity recognition pipeline, passing in a specific model and tokenizer
print("trace 5")
model = AutoModelForTokenClassification.from_pretrained("dbmdz/bert-large-cased-finetuned-conll03-english")
#model = AutoModel.from_pretrained("dbmdz/bert-large-cased-finetuned-conll03-english")
print("trace 6")
tokenizer = AutoTokenizer.from_pretrained("google-bert/bert-base-cased")
print("trace 7")
recognizer = pipeline("ner", model=model, tokenizer=tokenizer)
print("trace 8")
Output:
trace 0
No model was supplied, defaulted to distilbert-base-uncased-finetuned-sst-2-english and revision af0f99b (distilbert/distilbert-base-uncased-finetuned-sst-2-english · Hugging Face).
Using a pipeline without specifying a model name and revision in production is not recommended.trace 1
/usr/local/lib/python3.10/dist-packages/huggingface_hub/utils/_token.py:88: UserWarning:
[snip]