You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
labels = target_text_encoding["input_ids"]
labels[
labels == 0
] = -100 # to make sure we have correct labels for T5 text generation
return dict(
source_text_input_ids=source_text_encoding["input_ids"].flatten(),
source_text_attention_mask=source_text_encoding["attention_mask"].flatten(),
labels=labels.flatten(),
labels_attention_mask=target_text_encoding["attention_mask"].flatten(),
)
as i know, the decoder_input_ids is default to be got by shifting labels, but at these codes, the decoder_attention_mask is matched to labels. so i think the decoder_input_ids prepared by models will not be matched to decoder_attention_mask.
is it a bug or my understanding is wrong?
The text was updated successfully, but these errors were encountered:
target_text_encoding = self.tokenizer(
data_row["target_text"],
max_length=self.target_max_token_len,
padding="max_length",
truncation=True,
return_attention_mask=True,
add_special_tokens=True,
return_tensors="pt",
)
as i know, the decoder_input_ids is default to be got by shifting labels, but at these codes, the decoder_attention_mask is matched to labels. so i think the decoder_input_ids prepared by models will not be matched to decoder_attention_mask.
is it a bug or my understanding is wrong?
The text was updated successfully, but these errors were encountered: