Upload loconet

modeling_loconet.py

@@ -11,6 +11,9 @@ class loconet(PreTrainedModel):
         super().__init__(config)
 
         self.model = locoencoder(config)
+        self.lossAV = lossAV()
+        self.lossA = lossA()
+        self.lossV = lossV()
 
     def forward(self, audioFeature, visualFeature, masks, labels=None):
         b, s, t = visualFeature.shape[:3]

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6918e8391d48c40cfd90b332687508bb4b2269879ba1303dacb5a26937ecda87
-size 137464429
+oid sha256:e7aa6a388e424747d5b0b204b383ec53cc466a37b7f1c4c7e34a9fc766ce11f2
+size 137470241