Daniel Voigt Godoy - Deep Learning with PyTorch Step-by-Step A Beginner’s Guide-leanpub

For a complete list of available tasks, please check HuggingFace’s
pipeline [212] documentation.
Let’s run the first sentence of our training set through the sentiment analysis
pipeline:
sentence = train_dataset[0]['sentence']
print(sentence)
print(sentiment(sentence))
Output
And, so far as they knew, they were quite right.
[{'label': 'POSITIVE', 'score': 0.9998356699943542}]
Positive, indeed!
If you’re curious about which model is being used under the hood, you can check
the SUPPORTED_TASKS dictionary. For sentiment analysis, it uses the distilbertbase-uncased-finetuned-sst-2-english
model:
from transformers.pipelines import SUPPORTED_TASKS
SUPPORTED_TASKS['sentiment-analysis']
Output
{'impl': transformers.pipelines.text_classification
.TextClassificationPipeline,
'tf': None,
'pt': types.AutoModelForSequenceClassification,
'default': {'model': {'pt': 'distilbert-base-uncased-finetuned-sst-
2-english',
'tf': 'distilbert-base-uncased-finetuned-sst-2-english'}}}
"What about text generation?" Fine-Tuning with HuggingFace | 1003