Coding error！！

[AheadSnail]

I want to complete the distillation of text similarity using the following script。

python run_glue_no_trainer_distillation.py
--max_seq_length 128 --model_name_or_path ./student_model
--teacher_model_name_or_path BAAI/bge-small-zh-v1.5 --do_distillation
--per_device_train_batch_size 16 --learning_rate 1e-5 --num_train_epochs 1
--output_dir ./output_dir --loss_weights 0 1
--train_file ./train_data/dda-train.json
--temperature 2 --seed 5143

The student model here is local, and then I want to load the local data set, so the corresponding task_name is not set.
The following is the wrong information

Traceback (most recent call last):
File "/Users/jasonth/MyWork/AI/neural-compressor/examples/pytorch/nlp/huggingface_models/text-classification/distillation/eager/run_glue_no_trainer_distillation.py", line 708, in
main()
File "/Users/jasonth/MyWork/AI/neural-compressor/examples/pytorch/nlp/huggingface_models/text-classification/distillation/eager/run_glue_no_trainer_distillation.py", line 326, in main
is_regression = datasets["train"].features["label"].dtype in ["float32", "float64"]
TypeError: 'module' object is not subscriptable

is_regression = datasets["train"].features["label"].dtype in ["float32", "float64"]
Should the datasets in this code be changed to raw_datasets?