Quantifying the Carbon Emissions of Machine Learning
Paper
•
1910.09700
•
Published
•
24
This model has been created with Argilla, trained with Transformers.
This is a sample model finetuned from prajjwal1/bert-tiny.
Training the model using the ArgillaTrainer:
# Load the dataset:
dataset = FeedbackDataset.from_huggingface("argilla/emotion")
# Create the training task:
task = TrainingTask.for_text_classification(text=dataset.field_by_name("text"), label=dataset.question_by_name("label"))
# Create the ArgillaTrainer:
trainer = ArgillaTrainer(
dataset=dataset,
task=task,
framework="transformers",
model="prajjwal1/bert-tiny",
)
trainer.update_config({
"logging_steps": 1,
"num_train_epochs": 1,
"output_dir": "tmp"
})
trainer.train(output_dir="None")
You can test the type of predictions of this model like so:
trainer.predict("This is awesome!")
Model trained with ArgillaTrainer for demo purposes