modify dygraph_model.py

Author: simuler

models/multitask/metaheac/dygraph_model.py

@@ -33,8 +33,6 @@ def create_model(self, config):
 
         meta_model = net.WideAndDeepModel(max_idxs, embed_dim, mlp_dims,
                                           num_expert, num_output)
-        # model_state_dict = paddle.load('paddle.pkl')
-        # meta_model.set_dict(model_state_dict)
 
         return meta_model
 
@@ -118,6 +116,10 @@ def train_forward(self, dy_model, metric_list, batch, config):
             loss_q = criterion(query_set_y_pred, label)
             losses_q.append(loss_q)  # Save the loss on the subtask dataset
 
+            pred = paddle.unsqueeze(query_set_y_pred, 1)
+            pred = paddle.concat([1 - pred, pred], 1)
+            metric_list[0].update(preds=pred.numpy(), labels=label.numpy())
+
         loss_average = paddle.stack(losses_q).mean(0)
         print_dict = {'loss': loss_average}