fairseq/models/bart/hub_interface.py

Survived

Mutant 2953

--- fairseq/models/bart/hub_interface.py
+++ fairseq/models/bart/hub_interface.py
@@ -17,7 +17,7 @@
 from fairseq.data import encoders
 
 
-logger = logging.getLogger(__name__)
+logger = None
 
 
 class BARTHubInterface(nn.Module):

Mutant 2954

--- fairseq/models/bart/hub_interface.py
+++ fairseq/models/bart/hub_interface.py
@@ -42,7 +42,6 @@
         # this is useful for determining the device
         self.register_buffer('_float_tensor', torch.tensor([0], dtype=torch.float))
 
-    @property
     def device(self):
         return self._float_tensor.device
 

Mutant 2955

--- fairseq/models/bart/hub_interface.py
+++ fairseq/models/bart/hub_interface.py
@@ -46,7 +46,7 @@
     def device(self):
         return self._float_tensor.device
 
-    def encode(self, sentence: str, *addl_sentences, no_separator=True) -> torch.LongTensor:
+    def encode(self, sentence: str, *addl_sentences, no_separator=False) -> torch.LongTensor:
         """
         BPE-encode a sentence (or multiple sentences).
 

Mutant 2956

--- fairseq/models/bart/hub_interface.py
+++ fairseq/models/bart/hub_interface.py
@@ -102,7 +102,7 @@
         )
         return sample
 
-    def sample(self, sentences: List[str], beam: int = 1, verbose: bool = False, **kwargs) -> str:
+    def sample(self, sentences: List[str], beam: int = 2, verbose: bool = False, **kwargs) -> str:
         input = [self.encode(sentence) for sentence in sentences]
         hypos = self.generate(input, beam, verbose, **kwargs)
         return [self.decode(x['tokens']) for x in hypos]

Mutant 2957

--- fairseq/models/bart/hub_interface.py
+++ fairseq/models/bart/hub_interface.py
@@ -102,7 +102,7 @@
         )
         return sample
 
-    def sample(self, sentences: List[str], beam: int = 1, verbose: bool = False, **kwargs) -> str:
+    def sample(self, sentences: List[str], beam: int = 1, verbose: bool = True, **kwargs) -> str:
         input = [self.encode(sentence) for sentence in sentences]
         hypos = self.generate(input, beam, verbose, **kwargs)
         return [self.decode(x['tokens']) for x in hypos]

Mutant 2958

--- fairseq/models/bart/hub_interface.py
+++ fairseq/models/bart/hub_interface.py
@@ -107,7 +107,7 @@
         hypos = self.generate(input, beam, verbose, **kwargs)
         return [self.decode(x['tokens']) for x in hypos]
 
-    def generate(self, tokens: List[torch.LongTensor], beam: int = 5, verbose: bool = False, **kwargs) -> torch.LongTensor:
+    def generate(self, tokens: List[torch.LongTensor], beam: int = 6, verbose: bool = False, **kwargs) -> torch.LongTensor:
         sample = self._build_sample(tokens)
 
         # build generator using current args as well as any kwargs

Mutant 2959

--- fairseq/models/bart/hub_interface.py
+++ fairseq/models/bart/hub_interface.py
@@ -107,7 +107,7 @@
         hypos = self.generate(input, beam, verbose, **kwargs)
         return [self.decode(x['tokens']) for x in hypos]
 
-    def generate(self, tokens: List[torch.LongTensor], beam: int = 5, verbose: bool = False, **kwargs) -> torch.LongTensor:
+    def generate(self, tokens: List[torch.LongTensor], beam: int = 5, verbose: bool = True, **kwargs) -> torch.LongTensor:
         sample = self._build_sample(tokens)
 
         # build generator using current args as well as any kwargs

Mutant 2960

--- fairseq/models/bart/hub_interface.py
+++ fairseq/models/bart/hub_interface.py
@@ -135,7 +135,7 @@
         hypos = [v for _, v in sorted(zip(sample['id'].tolist(), hypos))]
         return hypos
 
-    def extract_features(self, tokens: torch.LongTensor, return_all_hiddens: bool = False) -> torch.Tensor:
+    def extract_features(self, tokens: torch.LongTensor, return_all_hiddens: bool = True) -> torch.Tensor:
         if tokens.dim() == 1:
             tokens = tokens.unsqueeze(0)
         if tokens.size(-1) > min(self.model.max_positions()):

Mutant 2961

--- fairseq/models/bart/hub_interface.py
+++ fairseq/models/bart/hub_interface.py
@@ -172,7 +172,7 @@
             name, num_classes=num_classes, embedding_size=embedding_size, **kwargs
         )
 
-    def predict(self, head: str, tokens: torch.LongTensor, return_logits: bool = False):
+    def predict(self, head: str, tokens: torch.LongTensor, return_logits: bool = True):
         if tokens.dim() == 1:
             tokens = tokens.unsqueeze(0)
         features = self.extract_features(tokens.to(device=self.device))