2021/5/18 voc原数据集 yolov4训练

1设置 config， train

1）MODEL_TYPE = {
    "TYPE": "YOLOv4"
}  # YOLO type:YOLOv4, Mobilenet-YOLOv4 or Mobilenetv3-YOLOv4

CONV_TYPE = {"TYPE": "GENERAL"}  # conv type:DO_CONV or GENERAL

ATTENTION = {"TYPE": "NONE"}  # attention type:SEnet、CBAM or NONE

# train
TRAIN = {
    "DATA_TYPE": "VOC",  # DATA_TYPE: VOC ,COCO or Customer
    "TRAIN_IMG_SIZE": 416,
    "AUGMENT": True,
    "BATCH_SIZE": 2,
    "MULTI_SCALE_TRAIN": True,
    "IOU_THRESHOLD_LOSS": 0.5,
    "YOLO_EPOCHS": 50,
    "Mobilenet_YOLO_EPOCHS": 120,
    "NUMBER_WORKERS": 2,
    "MOMENTUM": 0.9,
    "WEIGHT_DECAY": 0.0005,
    "LR_INIT": 1e-4,
    "LR_END": 1e-6,
    "WARMUP_EPOCHS": 2,  # or None
    "showatt": False
}


# val
VAL = {
    "TEST_IMG_SIZE": 416,
    "BATCH_SIZE": 2,
    "NUMBER_WORKERS": 2,
    "CONF_THRESH": 0.005,
    "NMS_THRESH": 0.45,
    "MULTI_SCALE_VAL": False,
    "FLIP_VAL": False,
    "Visual": False,
    "showatt": False
}

2）parser.add_argument('--weight_path', type=str, default='weight/yolov4.weights', help='weight file path')#weight/darknet53_448.weights#修改权重

2结果

[2021-05-17 23:35:44,342]-[train.py line:235]:  === Epoch:[ 30/50],step:[2500/2505],img_size:[384],total_loss:17.8603|loss_ciou:3.8230|loss_conf:6.3693|loss_cls:7.6680|lr:0.0001 
[2021-05-17 23:35:45,728]-[train.py line:279]:===== Validate ===== 
[2021-05-17 23:35:45,728]-[train.py line:281]:val img size is 416 
Traceback (most recent call last):
  File "C:/GitHub/YOLOv4-pytorch/train.py", line 346, in <module>
    fp_16=opt.fp_16,
  File "C:/GitHub/YOLOv4-pytorch/train.py", line 284, in train
    self.yolov4, showatt=self.showatt
  File "C:\GitHub\YOLOv4-pytorch\eval\evaluator.py", line 44, in APs_voc
    with open(img_inds_file, "r") as f:
FileNotFoundError: [Errno 2] No such file or directory: 'C:\\GitHub\\YOLOv4-pytorch\\data\\VOCtest-2007\\VOCdevkit\\VOC2007\\ImageSets\\Main\\test.txt'

3分析

...文件路径错误 or 路径下没有路径

需要修改 evaluator.py 文件