LM: handle empty context

lm_eval/base.py

@@ -15,7 +15,8 @@ class LM(abc.ABC):
         :param requests: list
             A list of pairs (context, continuation)
             context: str
-                Context string
+                Context string. Implementations of LM must be able to handle an 
+                empty context string.
             continuation: str
                 The continuation over which log likelihood will be calculated. If 
                 there is a word boundary, the space should be in the continuation. 

lm_eval/models/gpt2.py

@@ -24,7 +24,13 @@ class GPT2LM(LM):
         # TODO: vectorize properly
         for context, continuation in tqdm(requests):
             # when too long to fit in context, truncate from the left
+            
+            if context == "":
+                # end of text as context
+                context_enc = [50256]
+            else:
                 context_enc = self.tokenizer.encode(context)
+            
             continuation_enc = self.tokenizer.encode(continuation)
             inp = torch.tensor([(context_enc + continuation_enc)[-1024:]], dtype=torch.long).to(self.device)
             ctxlen = len(context_enc) - max(0, len(context_enc) + len(continuation_enc) - 1024)

lm_eval/models/gpt3.py

@@ -72,7 +72,12 @@ class GPT3LM(LM):
             inps = []
             ctxlens = []
             for context, continuation in chunk:
+                if context == "":
+                    # end of text as context
+                    context_enc = [50256]
+                else:
                     context_enc = self.tokenizer.encode(context)
+                    
                 continuation_enc = self.tokenizer.encode(continuation)
                 inp = (context_enc + continuation_enc)[-self.MAX_LENGTH:]
                 ctxlen = len(context_enc) - max(0, len(context_enc) + len(continuation_enc) - self.MAX_LENGTH)

tests/test_models.py

@@ -11,3 +11,5 @@ def test_gpt2():
     assert ll_dog > ll_cat
     assert not ig_cat
 
+    # test empty context
+    gpt2.loglikelihood([('', 'test')])
\ No newline at end of file