Update main.py

main.py

@@ -1,3 +1,7 @@
+import time
+total_time = time.time()
+start_time = time.time()
+
 from ultralytics import YOLO
 from base64 import b64encode
 from speech_recognition import AudioFile, Recognizer
@@ -12,13 +16,18 @@ from fastapi.responses import StreamingResponse
 from utils import tts, read_image_file, pil_to_base64, get_hist
 from typing import Optional
 from huggingface_hub import hf_hub_download
-import time
+
 
 from io import BytesIO
 import zipfile
+print("Import time", time.time() - start_time)
+start_time = time.time()
 
 model_path = hf_hub_download(repo_id="ultralyticsplus/yolov8s", filename='yolov8s.pt')
 model = YOLO(model_path)
+
+print("Load model time", time.time() - start_time)
+
 CLASS = model.model.names
 defaul_bot_voice = "おはいようございます"
 area_thres = 0.3
@@ -35,7 +44,9 @@ async def predict_api(
         file: UploadFile = File(...),
         last_seen: Union[UploadFile, None] = File(None)
 ):
+    start_time = time.time()
     image = read_image_file(await file.read())
+    print("Read image", time.time() - start_time)
     results = model.predict(image, show=False)[0]
     image = read_image_as_pil(image)
     masks, boxes = results.masks, results.boxes
@@ -43,7 +54,7 @@ async def predict_api(
     most_close = 0
     out_img = None
     diff_value = 0.5
-    total_time = time.time()
+    # total_time = time.time()
     start_time = time.time()
     if boxes is not None:
         for xyxy, conf, cls in zip(boxes.xyxy, boxes.conf, boxes.cls):