INNER CODE UNIT · Python
forward
huggingface/transformers-bloom-inference · inference_server/server.py:96
def forward():
try:
x = request.get_json()
x = ForwardRequest(**x)
if len(x.conditioning_text) != len(x.response):
raise Exception("unequal number of elements in conditioning_text and response arguments")
response, total_time_taken = run_and_log_time(partial(model.forward, request=x))
response.query_id = query_ids.forward_query_id
query_ids.forward_query_id += 1
response.total_time_taken = "{:.2f} secs".format(total_time_taken)
return response.dict(), status.HTTP_200_OK
except Exception:
response = get_exception_response(query_ids.forward_query_id, args.debug)
query_ids.forward_query_id += 1