I can't visualize the video I save

Python
1

Hi guys. I have this system where I do detection using yolov8 and classification using efficientnetB3. I test them on a video and I want to modify and save this video with all the bounding box indìformation, but I cannot visualize the video I save. What do I do wrong??

import torch
from tensorflow.keras.models import load_model
import cv2
from ultralytics import YOLO
import numpy as np

detector =  YOLO('/myPath/best.pt')

classifier = load_model('/myPath/efficientnet_model_unfreeze_128.h5')

video_path = '/myPath/video_test.mp4'
cap = cv2.VideoCapture(video_path)

output_path = '/myPath/output_video_a.mp4'
fourcc = cv2.VideoWriter_fourcc(*'mp4v')
fps =  cap.get(cv2.CAP_PROP_FPS)
width = int(cap.get(cv2.CAP_PROP_FRAME_WIDTH))
height = int(cap.get(cv2.CAP_PROP_FRAME_HEIGHT))
out = cv2.VideoWriter(output_path, fourcc, fps, (width, height))  

while cap.isOpened():
    ret, frame = cap.read()
    if not ret:
        break

    results = detector(frame)     
    detections = results[0].boxes.xyxy.cpu().numpy()  

    for detection in results:
        for box in detection.boxes:
            x1, y1, x2, y2, conf, cls = box.data.tolist()[0]
            
            roi = frame[int(y1):int(y2), int(x1):int(x2)]
            
            roi_resized = cv2.resize(roi, (300, 300))  
            roi_resized = roi_resized / 255.0 
            roi_resized = roi_resized.reshape(1, 300, 300, 3)
            
            pred = classifier.predict(roi_resized)
            class_id = pred.argmax(axis=1)[0]

            label = f'Class: {class_id}, Conf: {conf:.2f}'
            cv2.rectangle(frame, (int(x1), int(y1)), (int(x2), int(y2)), (0, 255, 0), 2)
            cv2.putText(frame, label, (int(x1), int(y1) - 10), cv2.FONT_HERSHEY_SIMPLEX, 0.9, (0, 255, 0), 2)
    
    out.write(frame)
    
cap.release()
out.release()
cv2.destroyAllWindows()
2

please explain, as we cannot see it …
what does happen (or not ?)

then, please dont expect much help with keras models, this would be quite off-topic here.

however, i find it weird, that you try a seperate classification on top of a YOLO result
(which already has a class id, no ?)
what do you expect of that ?