Messing with configuration.

app.py

@@ -3,8 +3,7 @@ import torch
 import gradio
 import json
 import onnxruntime
-from optimum.onnxruntime import ORTModelForSequenceClassification
-from transformers import AutoTokenizer
+import time
 from transformers import pipeline
 from fastapi import FastAPI
 from fastapi.middleware.cors import CORSMiddleware
@@ -31,16 +30,13 @@ print(f"CUDA device: {torch.cuda.get_device_name(torch.cuda.current_device())}")
 # "xenova/nli-deberta-v3-small" "cross-encoder/nli-deberta-v3-small" Was using this for a good while and it was...okay
 
 model_name = "MoritzLaurer/deberta-v3-base-zeroshot-v2.0"
-# file_name = "onnx/model.onnx"
 tokenizer_name = "MoritzLaurer/deberta-v3-base-zeroshot-v2.0"
 
-# model = ORTModelForSequenceClassification.from_pretrained(model_name, export=True, provider="CUDAExecutionProvider")
-# tokenizer = AutoTokenizer.from_pretrained(tokenizer_name, model_max_length=512)
-
 classifier = pipeline(task="zero-shot-classification", model=model_name, tokenizer=tokenizer_name, device="cuda:0")
 # classifier = pipeline(task="zero-shot-classification", model=model_name, tokenizer=tokenizer_name)
 
 def classify(data_string, request: gradio.Request):
+    print(f"classify")
     if request:
         if request.headers["origin"] not in ["https://statosphere-3704059fdd7e.c5v4v4jx6pq5.win", "https://crunchatize-77a78ffcc6a6.c5v4v4jx6pq5.win", "https://crunchatize-2-2b4f5b1479a6.c5v4v4jx6pq5.win", "https://tamabotchi-2dba63df3bf1.c5v4v4jx6pq5.win", "https://ravenok-statosphere-backend.hf.space", "https://lord-raven.github.io"]:
             return "{}"
@@ -52,14 +48,18 @@ def classify(data_string, request: gradio.Request):
     # if 'task' in data and data['task'] == 'few_shot_classification':
     #     return few_shot_classification(data)
     # else:
-    return json.dumps(zero_shot_classification(data))
+    start_time = time.time()
+    result = zero_shot_classification(data)
+    end_time = time.time()
+    elapsed_time = end_time - start_time
+    print(f"classification took {elapsed_time}.")
+    return json.dumps(result)
 
 @spaces.GPU(duration=3)
 def zero_shot_classification(data):
     return classifier(data['sequence'], candidate_labels=data['candidate_labels'], hypothesis_template=data['hypothesis_template'], multi_label=data['multi_label'])
 
 def create_sequences(data):
-    # return ['###Given:\n' + data['sequence'] + '\n###End Given\n###Hypothesis:\n' + data['hypothesis_template'].format(label) + "\n###End Hypothesis" for label in data['candidate_labels']]
     return [data['sequence'] + '\n' + data['hypothesis_template'].format(label) for label in data['candidate_labels']]
 
 # def few_shot_classification(data):

requirements.txt

@@ -3,5 +3,6 @@ fastapi==0.88.0
 huggingface_hub==0.26.0
 json5==0.9.25
 numpy
+time
 optimum[onnxruntime-gpu]==1.24.0
 transformers==4.36