fixed minor type conversion from onnx

When checking the type of output was causing some noisy output in the logs. Corrected the method for checking the type

doctr/models/recognition/predictor/pytorch.py

@@ -82,7 +82,7 @@ def forward(
                 batch = batch.to(self.device)
                 # batch = batch.half()
             char_logits = self.model(batch)
-            if type(char_logits) != torch.Tensor():
+            if not torch.is_tensor(char_logits):
                 char_logits = torch.tensor(char_logits)
             raw += [self.postprocessor(char_logits)]