add softmax

paddlepalm/task_paradigm/cls.py

@@ -75,8 +75,9 @@ class TaskParadigm(task_paradigm):
                 name=scope_name+"cls_out_b", initializer=fluid.initializer.Constant(0.)))
 
         if self._is_training:
+            inputs = fluid.layers.softmax(logits)
             loss = fluid.layers.cross_entropy(
-                input=logits, label=label_ids)
+                input=inputs, label=label_ids)
             loss = layers.mean(loss)
             return {"loss": loss}
         else:

paddlepalm/task_paradigm/match.py

@@ -79,8 +79,9 @@ class TaskParadigm(task_paradigm):
                 initializer=fluid.initializer.Constant(0.)))
 
         if self._is_training:
+            inputs = fluid.layers.softmax(logits)
             ce_loss = fluid.layers.cross_entropy(
-                input=logits, label=labels)
+                input=inputs, label=labels)
             loss = fluid.layers.mean(x=ce_loss)
             return {'loss': loss}
         else:

paddlepalm/task_paradigm/mlm.py

@@ -100,8 +100,9 @@ class TaskParadigm(task_paradigm):
             is_bias=True)
 
         if self._is_training:
+            inputs = fluid.layers.softmax(fc_out)
             mask_lm_loss = fluid.layers.cross_entropy(
-                input=fc_out, label=mask_label)
+                input=inputs, label=mask_label)
             loss = fluid.layers.mean(mask_lm_loss)
             return {'loss': loss}
         else:

paddlepalm/task_paradigm/mrc.py

@@ -102,8 +102,9 @@ class TaskParadigm(task_paradigm):
         def _compute_single_loss(logits, positions):
             """Compute start/en
             d loss for mrc model"""
+            inputs = fluid.layers.softmax(logits)
             loss = fluid.layers.cross_entropy(
-                input=logits, label=positions)
+                input=inputs, label=positions)
             loss = fluid.layers.mean(x=loss)
             return loss