confidence thresh?

❓ Questions and Help

In the code train_net.py, after the model trained, the code run into the inference part. However, in the code inference.py, I couldnot find the object confidence thresh setting. I am confused the model output based on which value of object confidence? where can we use object confidence thresh as a parameter and pass into the model prediction function.

    logger.info("Start evaluation on {} dataset({} images).".format(dataset_name, len(dataset)))
    start_time = time.time()
    predictions = compute_on_dataset(model, data_loader, device)
    # wait for all processes to complete before measuring the time
    synchronize()
    total_time = time.time() - start_time

[chengyangfu]

I’m closing this issue, please let me know if you have any further question.

[Pari-singh]

@chengyangfu could you explain why we are applying confidence threshold at the ROI head specifically and not at any other layer?