Update app.py

app.py

@@ -7,7 +7,14 @@ tokenizer = AutoTokenizer.from_pretrained("emilyalsentzer/Bio_ClinicalBERT")
 # Convert your tokenized data into PyTorch tensors and create a PyTorch Dataset object
 import torch
 from torch.utils.data import Dataset
-    
+
+
+train_texts = st.text_input("Enter your clinical text data (separated by commas):")
+train_labels = st.text_input("Enter your corresponding labels (separated by commas):")
+
+train_texts = train_texts.split(",")
+train_labels = train_labels.split(",")
+
 class ClinicalDataset(Dataset):
          def __init__(self, texts, labels, tokenizer):
              self.texts = texts
@@ -22,7 +29,8 @@ class ClinicalDataset(Dataset):
              label = self.labels[idx]
              encoding = self.tokenizer(text, return_tensors="pt", padding=True, truncation=True)
              return {"input_ids": encoding["input_ids"].squeeze(), "attention_mask": encoding["attention_mask"].squeeze(), "labels": torch.tensor(label)}
-     
+
+
 dataset = ClinicalDataset(texts=train_texts, labels=train_labels, tokenizer=tokenizer)
 # Fine-tune the pre-trained model on your clinical dataset
 from transformers import Trainer, TrainingArguments