ftobler/llama
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

whitespace

Browse Source
This commit is contained in:
Florin Tobler
2025-01-13 23:33:22 +01:00
parent 3218e7eb63
commit a697f49698
1 changed files with 2 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
inference.py
View File

@@ -101,7 +101,7 @@ class Inference:
if print_stdout: if print_stdout:
print(out_text) print(out_text)
return outputs, out_text return outputs, out_text
def generate_incremental_2(self, input_ids: torch.Tensor, print_stdout:bool=True) -> tuple[torch.Tensor, str]: def generate_incremental_2(self, input_ids: torch.Tensor, print_stdout:bool=True) -> tuple[torch.Tensor, str]:
generated_tokens = input_ids generated_tokens = input_ids
@@ -180,7 +180,7 @@ class Inference:
while True: while True:
# Call the model with the current tokens # Call the model with the current tokens
outputs = self.model( outputs = self.model(
input_ids=generated_tokens, input_ids=generated_tokens,
use_cache=True, use_cache=True,
num_beams = 1 num_beams = 1
# past_key_values=past_key_values # past_key_values=past_key_values