这些参数在yoloV4模型中意味着什么 [英] What all these parameters means in yoloV4 model

问题描述

训练YOLOv4的所有这些参数是什么意思?

What do all of these parameters from training YOLOv4 mean?

(在1300次迭代中进行下一次mAP计算)

(next mAP calculation at 1300 iterations)

最高准确度mAP@0.5 = 63.16％，最佳= 68.55％

Last accuracy mAP@0.5 = 63.16 %, best = 68.55 %

1249:26.351213、24.018257平均损失，0.001000速率，2.998398秒，39968张图像，还剩10.505599小时载入时间:0.000068秒

1249: 26.351213, 24.018257 avg loss, 0.001000 rate, 2.983998 seconds, 39968 images, 10.505599 hours left Loaded: 0.000068 seconds

(在1300次迭代中进行下一个mAP计算)最后一次准确度mAP@0.5 = 63.16％，最佳= 68.55％

(next mAP calculation at 1300 iterations) Last accuracy mAP@0.5 = 63.16 %, best = 68.55 %

1250:13.904115、23.006844平均损失，0.001000速率，4.093653秒，40000张图像，还剩10.456502小时调整大小，random_coef = 1.40

1250: 13.904115, 23.006844 avg loss, 0.001000 rate, 4.093653 seconds, 40000 images, 10.456502 hours left Resizing, random_coef = 1.40

推荐答案

这是参数的含义.

对于您给出的示例:

(next mAP calculation at 1300 iterations) Last accuracy mAP@0.5 = 63.16 %, best = 68.55 %

1250: 13.904115, 23.006844 avg loss, 0.001000 rate, 4.093653 seconds, 40000 images, 10.456502 hours left Resizing, random_coef = 1.40

• 1250 ->迭代

  精度为mAP@0.5 ->IoU阈值50％时的最后平均平均精度(mAP).每100次迭代计算一次mAP.因此，在此示例中，它是迭代= 1200的mAP

  Last accuracy mAP@0.5 --> Last mean average precision (mAP) at 50% IoU threshold. mAP is calculated every 100th iteration. So, in the example, it's the mAP from iteration = 1200

  最佳->迄今为止最高的mAP

  best --> highest mAP so far

  13.904115 ->总损失

  23.006844平均损失->平均损失，这是您应避免的训练不足的事情

  23.006844 avg loss--> average loss, this is the thing you should care about for being low in training

  0.001000费率->学习率

  4.093653秒->处理批次所花费的总时间

  4.093653 seconds --> total time spent to process the batch

  40000张图片->到目前为止，训练期间使用的图片总数(迭代*批量= 1250 * 32)

  40000 images --> total amount of images used during training so far (iteration*batch = 1250 * 32)

  还剩10.456502小时->完成配置文件中的 max_batches 所需的估计时间

  10.456502 hours left --> estimated time remaining for finishing up to the max_batches in your config file

  调整大小，random_coef = 1.40 ->确认您的数据集每10次迭代从1/1.4随机调整为1.4(在此迭代中为1.40)

  Resizing, random_coef = 1.40 --> Confirming that your dataset is being randomly resized every 10 iterations from 1/1.4 to 1.4 (in this iteration, it's 1.40)

  参考: https://github.com/AlexeyAB/darknet/blob/master/src/detector.c https://github.com/AlexeyAB/darknet/wiki/CFG-不同层中的参数

  这篇关于这些参数在yoloV4模型中意味着什么的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！