sgnlp.models.emotion_entailment.utils.convert_df_to_dataset

convert_df_to_dataset(df: pandas.core.frame.DataFrame, max_seq_length: int, tokenizer: sgnlp.models.emotion_entailment.tokenization.RecconEmotionEntailmentTokenizer) → torch.utils.data.dataset.TensorDataset

Convert pandas dataframe to TensorDataset

Parameters

• df (pd.DataFrame) – DataFrame containing ‘text’ and ‘labels’ columns

• max_seq_length (int) – max sequence length

• tokenizer (RecconEmotionEntailmentTokenizer) – RecconEmotionEntailmentTokenizer from sgnlp

Returns

A tensordataset object containing all examples

Return type

TensorDataset