What is evolve param?

[chokyungjin]

I'm sooooo happy to use your YOLOv5.
I'm training my custom data using YOLOv5 And I applied evolve parameter,
but it is not over for seven hours.
When I turned this param off, it was over in three hours.
And Model is not saving pt.

What is evolve param? Thanks again!!!!

[glenn-jocher]

@chokyungjin evolving hyperparameters is seperate from training, and takes much longer (can take weeks or months on a single GPU). See evolution tutorial: https://docs.ultralytics.com/yolov5

[chokyungjin]

@glenn-jocher Thanks again.
But I have a another issue. My Train curve is weird in results.jpg.
I don't know why I'm getting the results like this, except I changed my custom data.

my opt is batch 80 img 416 epochs 300 weights '' augmentation default.

[glenn-jocher]

@chokyungjin your val data may not have been found correctly. Make sure your paths appear, and make sure your test_batch0_gt.jpg appears correct.

[chokyungjin]

@glenn-jocher My test_batch0_gt.jpg appears correct.
So what am I supposed to do now?
I only revise augmentation meta code and iOU threshold value.

Thanks.

[github-actions]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.

[maloyan]

@chokyungjin Did you manage to resolve that problem?

[glenn-jocher]

@maloyan see YOLOv5 Hyperparameter Evolution tutorial:

YOLOv5 Tutorials

• Train Custom Data  🚀 RECOMMENDED
• Tips for Best Training Results  ☘️ RECOMMENDED
• Weights & Biases Logging  🌟 NEW
• Supervisely Ecosystem  🌟 NEW
• Multi-GPU Training
• PyTorch Hub  ⭐ NEW
• TFLite, ONNX, CoreML, TensorRT Export 🚀
• Test-Time Augmentation (TTA)
• Model Ensembling
• Model Pruning/Sparsity
• Hyperparameter Evolution
• Transfer Learning with Frozen Layers  ⭐ NEW
• TensorRT Deployment

Good luck 🍀 and let us know if you have any other questions!

[MLDavies]

Is it possible to evolve on a subset of hyperparameters? If so, how would this been done?

[glenn-jocher]

@MLDavies update meta dict here by setting mutation scale to zero for any hyperparams you want to fix in place:

yolov5/train.py

Lines 569 to 600 in 014acde

	# Hyperparameter evolution metadata (mutation scale 0-1, lower_limit, upper_limit)
	meta = {
	'lr0': (1, 1e-5, 1e-1), # initial learning rate (SGD=1E-2, Adam=1E-3)
	'lrf': (1, 0.01, 1.0), # final OneCycleLR learning rate (lr0 * lrf)
	'momentum': (0.3, 0.6, 0.98), # SGD momentum/Adam beta1
	'weight_decay': (1, 0.0, 0.001), # optimizer weight decay
	'warmup_epochs': (1, 0.0, 5.0), # warmup epochs (fractions ok)
	'warmup_momentum': (1, 0.0, 0.95), # warmup initial momentum
	'warmup_bias_lr': (1, 0.0, 0.2), # warmup initial bias lr
	'box': (1, 0.02, 0.2), # box loss gain
	'cls': (1, 0.2, 4.0), # cls loss gain
	'cls_pw': (1, 0.5, 2.0), # cls BCELoss positive_weight
	'obj': (1, 0.2, 4.0), # obj loss gain (scale with pixels)
	'obj_pw': (1, 0.5, 2.0), # obj BCELoss positive_weight
	'iou_t': (0, 0.1, 0.7), # IoU training threshold
	'anchor_t': (1, 2.0, 8.0), # anchor-multiple threshold
	'anchors': (2, 2.0, 10.0), # anchors per output grid (0 to ignore)
	'fl_gamma': (0, 0.0, 2.0), # focal loss gamma (efficientDet default gamma=1.5)
	'hsv_h': (1, 0.0, 0.1), # image HSV-Hue augmentation (fraction)
	'hsv_s': (1, 0.0, 0.9), # image HSV-Saturation augmentation (fraction)
	'hsv_v': (1, 0.0, 0.9), # image HSV-Value augmentation (fraction)
	'degrees': (1, 0.0, 45.0), # image rotation (+/- deg)
	'translate': (1, 0.0, 0.9), # image translation (+/- fraction)
	'scale': (1, 0.0, 0.9), # image scale (+/- gain)
	'shear': (1, 0.0, 10.0), # image shear (+/- deg)
	'perspective': (0, 0.0, 0.001), # image perspective (+/- fraction), range 0-0.001
	'flipud': (1, 0.0, 1.0), # image flip up-down (probability)
	'fliplr': (0, 0.0, 1.0), # image flip left-right (probability)
	'mosaic': (1, 0.0, 1.0), # image mixup (probability)
	'mixup': (1, 0.0, 1.0), # image mixup (probability)
	'copy_paste': (1, 0.0, 1.0)} # segment copy-paste (probability)

[MLDavies]

Ah. Perfect. Thank you.