## Yolov8: cfg
config:
resume = True
### (Do not use resume = True)

Unexpected Events occured:
1. Configuration parameters passed from model.train is not used
2. Creates new directory and trains with 90 epoch

Preview of Engine/Trainer loaded config:

{
  "engine/trainer": {
    "task": "classify",
    "mode": "train",
    "model": "/home/shreeram/workspace/ambl/YOLOv8_custom/BaseModels/Classify/yolov8n-cls.pt",
    "data": "/home/shreeram/workspace/ambl/YOLOv8_custom/12to14_data/018_lagenda_FairFace_agender_nomask_combined_train_aug/age_dataset",
    "epochs": 90,
    "patience": 50,
    "batch": 256,
    "imgsz": 224,
    "save": true,
    "save_period": -1,
    "cache": false,
    "device": 0,
    "workers": 16,
    "project": "YOLOv8",
    "name": "yolov8n-cls-nonorm3",
    "exist_ok": false,
    "pretrained": false,
    "optimizer": "SGD",
    "verbose": true,
    "seed": 0,
    "deterministic": true,
    "single_cls": false,
    "rect": false,
    "cos_lr": false,
    "close_mosaic": 10,
    "resume": false,
    "amp": true,
    "fraction": 1.0,
    "profile": false,
    "freeze": null,
    "overlap_mask": true,
    "mask_ratio": 4,
    "dropout": 0.0,
    "val": true,
    "split": "val",
    "save_json": false,
    "save_hybrid": false,
    "conf": null,
    "iou": 0.7,
    "max_det": 300,
    "half": false,
    "dnn": false,
    "plots": true,
    "source": null,
    "vid_stride": 1,
    "stream_buffer": false,
    "visualize": false,
    "augment": false,
    "agnostic_nms": false,
    "classes": null,
    "retina_masks": false,
    "show": false,
    "save_frames": false,
    "save_txt": false,
    "save_conf": false,
    "save_crop": false,
    "show_labels": true,
    "show_conf": true,
    "show_boxes": true,
    "line_width": null,
    "format": "torchscript",
    "keras": false,
    "optimize": false,
    "int8": false,
    "dynamic": false,
    "simplify": false,
    "opset": null,
    "workspace": 4,
    "nms": false,
    "lr0": 0.1,
    "lrf": 0.01,
    "momentum": 0.937,
    "weight_decay": 0.0001,
    "warmup_epochs": 0,
    "warmup_momentum": 0.8,
    "warmup_bias_lr": 0.1,
    "box": 7.5,
    "cls": 0.5,
    "dfl": 1.5,
    "pose": 12.0,
    "kobj": 1.0,
    "label_smoothing": 0.1,
    "nbs": 256,
    "hsv_h": 0.015,
    "hsv_s": 0.7,
    "hsv_v": 0.4,
    "degrees": 0.0,
    "translate": 0.1,
    "scale": 0.5,
    "shear": 0.0,
    "perspective": 0.0,
    "flipud": 0.0,
    "fliplr": 0.5,
    "mosaic": 1.0,
    "mixup": 0.0,
    "copy_paste": 0.0,
    "cfg": null,
    "tracker": "botsort.yaml",
    "image_weights": false,
    "hide_labels": false,
    "hide_conf": false,
    "line_thickness": 3,
    "boxes": true,
    "fl_gamma": 0.0,
    "v5loader": false,
    "save_dir": "YOLOv8/yolov8n-cls-nonorm3"
  }
}

<!-- -->
### How to resume train?
(High probability the training may not be continued because there is no change in the val loss or other pre-defined scenario from yolov8)

load the previously trained model as base model .i.e last.pt
change atleast one parameter value i.e. patience = XX

run train file

Still delets old folder and retrains with new one

I found a link from issues in Ultralytics:
I am not sure either to implement it now so, as of now I will just put it here for reference.
#### Related issues -->
https://github.com/ultralytics/ultralytics/issues/6655

#### Purposed solution -->
https://github.com/ultralytics/ultralytics/blob/9c6989a1b0d0b94c51eacf7cb123c028145b224e/ultralytics/engine/trainer.py#L124

Note:
In classification network dropout is not used as its default value is set to 0.0
https://docs.ultralytics.com/modes/train/#arguments