github/fauxpilot
1
0
Fork 0
mirror of https://github.com/fauxpilot/fauxpilot.git synced 2025-07-16 10:03:25 -07:00
Code Issues Projects Releases Packages Wiki Activity Actions

Error handling

Browse source
This commit is contained in:
Fred de Gier 2022-10-03 11:47:54 +02:00
parent 6671196ec8
commit 6f49915d2a
3 changed files with 9 additions and 4 deletions
Download patch file Download diff file Expand all files Collapse all files

2
copilot_proxy/Dockerfile
View file

@ -10,4 +10,4 @@ COPY copilot_proxy .
EXPOSE 5000
CMD [ "uvicorn", "--host", "0.0.0.0", "--port", "5000", "app:app"]
CMD ["uvicorn", "--host", "0.0.0.0", "--port", "5000", "app:app"]

2
copilot_proxy/models.py
View file

@ -4,7 +4,7 @@ from pydantic import BaseModel
class OpenAIinput(BaseModel):
model: str
model: str = "fastertransformer"
prompt: Optional[str]
suffix: Optional[str]
max_tokens: Optional[int] = 16

7
copilot_proxy/utils/codegen.py
View file

@ -6,7 +6,7 @@ import time
import numpy as np
import tritonclient.grpc as client_util
from tokenizers import Tokenizer
from tritonclient.utils import np_to_triton_dtype
from tritonclient.utils import np_to_triton_dtype, InferenceServerException
np.finfo(np.dtype("float32"))
np.finfo(np.dtype("float64"))
@ -230,7 +230,12 @@ class CodeGenProxy:
def __call__(self, data: dict):
st = time.time()
try:
completion, choices = self.generate(data)
except InferenceServerException as E:
print(E)
completion = {}
choices = []
ed = time.time()
print(f"Returned completion in {(ed - st) * 1000} ms")
if data.get('stream', False):