Update Mimic4Dataset.py

Mimic4Dataset.py

@@ -356,7 +356,7 @@ def getXY_deep(data,task,feat_cond,feat_proc,feat_out,feat_chart,feat_meds):
     charts = charts.unsqueeze(0)
     charts = torch.tensor(charts)
     charts = charts.type(torch.LongTensor)
-    chart=charts
+    
 ####################""
 
     keys=dyn.columns.levels[0]
@@ -387,7 +387,7 @@ def getXY_deep(data,task,feat_cond,feat_proc,feat_out,feat_chart,feat_meds):
             proc=dyn_df[k].squeeze(0)
         if keys[k]=='LAB':
             lab=dyn_df[k].squeeze(0)
-    
+    chart=charts
     stat=cond_df
     stat = stat.to_numpy()
     stat = torch.tensor(stat)
@@ -598,7 +598,7 @@ class Mimic4Dataset(datasets.GeneratorBasedBuilder):
         self.val_size = kwargs.pop("val_size",0.1)
         self.generate_cohort = kwargs.pop("generate_cohort",True)
 
-        if self.encoding == 'onehot':
+        if self.encoding == 'concat':
             self.concat = True
         else:
             self.concat = False
@@ -911,7 +911,7 @@ class Mimic4Dataset(datasets.GeneratorBasedBuilder):
         features = datasets.Features(
             {
                 "label": datasets.ClassLabel(num_classes=2,names=["0", "1"]),
-                "onehot" : datasets.Sequence(datasets.Value("float32")),
+                "features" : datasets.Sequence(datasets.Value("float32")),
             }
         )
         return datasets.DatasetInfo(
@@ -962,7 +962,7 @@ class Mimic4Dataset(datasets.GeneratorBasedBuilder):
             X=X.values.tolist()[0]
             yield int(i), {
                 "label": label,
-                "onehot": X,
+                "features": X,
             }
 ######################################################DEEP###############################################################
     def _info_deep(self):
@@ -1011,13 +1011,13 @@ class Mimic4Dataset(datasets.GeneratorBasedBuilder):
         self.path = self.create_cohort()
         self.size_cond, self.size_proc, self.size_meds, self.size_out, self.size_chart, self.size_lab, eth_vocab,gender_vocab,age_vocab,ins_vocab=vocab(self.config.name.replace(" ","_"),self.feat_cond,self.feat_proc,self.feat_out,self.feat_chart,self.feat_meds,False)
     
-        if self.encoding == 'onehot' :
+        if self.encoding == 'concat' :
             return self._info_encoded()
 
         elif self.encoding == 'aggreg' :
             return self._info_encoded()
         
-        elif self.encoding == 'deep' :
+        elif self.encoding == 'tensor' :
             return self._info_deep()
         
         else:
@@ -1035,13 +1035,13 @@ class Mimic4Dataset(datasets.GeneratorBasedBuilder):
 
     def _generate_examples(self, filepath):
         
-        if self.encoding == 'onehot' :
+        if self.encoding == 'concat' :
             yield from self._generate_examples_encoded(filepath)
         
         elif self.encoding == 'aggreg' :
             yield from self._generate_examples_encoded(filepath)
         
-        elif self.encoding == 'deep' :
+        elif self.encoding == 'tensor' :
             yield from self._generate_examples_deep(filepath)
         else :
             yield from self._generate_examples_raw(filepath)