Sanda-Maria Dragoș Avram

4 Questions 4 Answers 0 Followers

Questions related from Sanda-Maria Dragoș Avram

Is this the way one computes macro-accuracy for BERT multi-classification text?

The code is: import evaluate metric = evaluate.load("accuracy") def compute_metrics(eval_pred): logits, labels = eval_pred predictions = np.argmax(logits, axis=-1) return...

29 December 2022 2,246 3 View

Why there are other results with the same seed for Transformers?

I have: from transformers import TrainingArguments, Trainer training_args = TrainingArguments(output_dir=“testTrainer”, evaluation_strategy=“epoch”, seed=42, num_train_epochs=10.0,...

18 December 2022 6,892 1 View

Is Hugging Face's evaluate-metric/accuracy the same as macro-accuracy?

I am running tests on BERT transformers and using the evaluate Python library. On their site, it says: "computed with Accuracy = (TP + TN) / (TP + TN + FP + FN) Where: TP: True positive TN: True...

11 December 2022 7,132 2 View

Does anyone know of applications of Benford's law on text?

I know that research has been done for using Benford's Law on images. I am wondering if anyone tried to do the same on text.

25 August 2022 4,554 1 View