--- a
+++ b/chexbert/src/datasets/impressions_dataset.py
@@ -0,0 +1,47 @@
+import torch
+import pandas as pd
+import numpy as np
+from bert_tokenizer import load_list
+from torch.utils.data import Dataset, DataLoader
+
+class ImpressionsDataset(Dataset):
+        """The dataset to contain report impressions and their labels."""
+        
+        def __init__(self, csv_path, list_path):
+                """ Initialize the dataset object
+                @param csv_path (string): path to the csv file containing labels
+                @param list_path (string): path to the list of encoded impressions
+                """
+                self.df = pd.read_csv(csv_path)
+                self.df =  self.df[['Enlarged Cardiomediastinum', 'Cardiomegaly', 'Lung Opacity',
+                                    'Lung Lesion', 'Edema', 'Consolidation', 'Pneumonia', 'Atelectasis',
+                                    'Pneumothorax', 'Pleural Effusion', 'Pleural Other', 'Fracture',
+                                    'Support Devices', 'No Finding']]
+                self.df.replace(0, 2, inplace=True) #negative label is 2
+                self.df.replace(-1, 3, inplace=True) #uncertain label is 3
+                self.df.fillna(0, inplace=True) #blank label is 0
+                self.encoded_imp = load_list(path=list_path)
+
+        def __len__(self):
+                """Compute the length of the dataset
+
+                @return (int): size of the dataframe
+                """
+                return self.df.shape[0]
+
+        def __getitem__(self, idx):
+                """ Functionality to index into the dataset
+                @param idx (int): Integer index into the dataset
+
+                @return (dictionary): Has keys 'imp', 'label' and 'len'. The value of 'imp' is
+                                      a LongTensor of an encoded impression. The value of 'label'
+                                      is a LongTensor containing the labels and 'the value of
+                                      'len' is an integer representing the length of imp's value
+                """
+                if torch.is_tensor(idx):
+                        idx = idx.tolist()
+                label = self.df.iloc[idx].to_numpy()
+                label = torch.LongTensor(label)
+                imp = self.encoded_imp[idx]
+                imp = torch.LongTensor(imp)
+                return {"imp": imp, "label": label, "len": imp.shape[0]}