Huggingface trainer is not showing any steps and training loss logs when training (Jupyter Notebook)

Question

I am trying to do a multilabel classification using the model emilyalsentzer/Bio_ClinicalBERT. I have medical text data that I am using. I have attached a snippet of it where there are quite a few labels that each have some sort of text. The text itself are paragraphs relating to the label. snippet of my label and text columns

When I am training the model, I just get this: error that im facing

I will provide my code below (my dataframe with 2 columns (label and text) is called data_c). Where am I going wrong and how can I fix this?

I tried to change the training arguments but wasnt successful.

My code:

###################################################################

from sklearn.preprocessing import LabelEncoder
from sklearn.model_selection import train_test_split
from transformers import AutoTokenizer, AutoModelForSequenceClassification, TrainingArguments, Trainer
from datasets import Dataset

###################################################################

tokenizer = AutoTokenizer.from_pretrained('emilyalsentzer/Bio_ClinicalBERT')

def preprocess_function(examples):
    return tokenizer(examples["text"], 
                     truncation=True,
                     padding="max_length",
                     max_length=512)

###################################################################

doc_train, doc_test = train_test_split(data_c, test_size = 0.3)

dataset_train = Dataset.from_pandas(doc_train)
dataset_test = Dataset.from_pandas(doc_test)

tokenized_dataset_train = dataset_train.map(preprocess_function, batched=True)
tokenized_dataset_test = dataset_test.map(preprocess_function, batched=True)

from transformers import AutoModelForSequenceClassification, TrainingArguments, Trainer
from transformers import DataCollatorWithPadding

###################################################################

num_labels = len(data_c.label.unique())

disease = AutoModelForSequenceClassification.from_pretrained('emilyalsentzer/Bio_ClinicalBERT', num_labels=num_labels)

data_collator = DataCollatorWithPadding(tokenizer=tokenizer)

###################################################################

training_args = TrainingArguments(
    output_dir="./results",
    learning_rate=5e-6,
    per_device_train_batch_size=4,
    per_device_eval_batch_size=4,
    num_train_epochs=2,
    weight_decay=0.01,
    report_to = "none"
)

from sklearn.metrics import precision_recall_fscore_support, accuracy_score

def compute_metrics(pred):
    labels = pred.label_ids
    preds = pred.predictions.argmax(-1)
    precision, recall, f1, _ = precision_recall_fscore_support(labels, preds, average='micro')
    acc = accuracy_score(labels, preds)
    loss = pred.loss
    
    return {
        'accuracy': acc,
        'f1': f1,
        'precision': precision,
        'recall': recall
    }

trainer = Trainer(
    model=disease,
    args=training_args,
    train_dataset=tokenized_dataset_train,
    eval_dataset=tokenized_dataset_test,
    tokenizer=tokenizer,
    data_collator=data_collator,
    compute_metrics = compute_metrics,
)

###################################################################

trainer.train()