generating prediction-stat:"P" ,"R" , "mAP" per image in test.py

[tjbe2021]

❔Question

Additional context

Hi there, I need some help to generate prediction-stats per image in the "Val" folder when run the test.py script, I couldn't find a way to do it by going through the code. Appreciate the help.

[github-actions]

👋 Hello @tjbe2021, thank you for your interest in 🚀 YOLOv5! Please visit our ⭐️ Tutorials to get started, where you can find quickstart guides for simple tasks like Custom Data Training all the way to advanced concepts like Hyperparameter Evolution.

If this is a 🐛 Bug Report, please provide screenshots and minimum viable code to reproduce your issue, otherwise we can not help you.

If this is a custom training ❓ Question, please provide as much information as possible, including dataset images, training logs, screenshots, and a public link to online W&B logging if available.

For business inquiries or professional support requests please visit https://www.ultralytics.com or email Glenn Jocher at [email protected].

Requirements

Python 3.8 or later with all requirements.txt dependencies installed, including torch>=1.7. To install run:

$ pip install -r requirements.txt

Environments

YOLOv5 may be run in any of the following up-to-date verified environments (with all dependencies including CUDA/CUDNN, Python and PyTorch preinstalled):

• Google Colab and Kaggle notebooks with free GPU:
• Google Cloud Deep Learning VM. See GCP Quickstart Guide
• Amazon Deep Learning AMI. See AWS Quickstart Guide
• Docker Image. See Docker Quickstart Guide

Status

If this badge is green, all YOLOv5 GitHub Actions Continuous Integration (CI) tests are currently passing. CI tests verify correct operation of YOLOv5 training (train.py), testing (test.py), inference (detect.py) and export (export.py) on MacOS, Windows, and Ubuntu every 24 hours and on every commit.

[maheshmechengg]

When i give image size other than model (416) to test.py my anchors misplace on test labels & test predict:

[glenn-jocher]

@tjbe2021 test.py runs by default on your data.yaml test: directory. You can point it to the data.yaml val: directory by:
python test.py --task val

@maheshmechengg please raise a new issue for new topics. Your labels are incorrect.

[tjbe2021]

@glenn-jocher Thanks for the prompt reply, I've changed the default to val, tI get the values for each class, and for all.

However , I'm trying to generate the P, R, mAP for the each image inside the val directory. So that I could perform some sandy checks, is that possible with this ? I tried tweaking the code, but doesn't seems it's working.

[glenn-jocher]

@tjbe2021 AP is not computed per image, it is computed per class over all images, and then averaged as mAP. If you want to obtain mAP on one image, then your dataset would have to be 1 image.

[tjbe2021]

@glenn-jocher great thanks for the answer, that make sense! By any chance under metrics.py are you producing any results on FP,TP, values?

[glenn-jocher]

@tjbe2021 YOLOv5 TP and FP vectors are computed here:

yolov5/utils/general.py

Lines 250 to 319 in c8c5ef3

	def ap_per_class(tp, conf, pred_cls, target_cls, plot=False, fname='precision-recall_curve.png'):
	""" Compute the average precision, given the recall and precision curves.
	Source: https://github.com/rafaelpadilla/Object-Detection-Metrics.
	# Arguments
	tp: True positives (nparray, nx1 or nx10).
	conf: Objectness value from 0-1 (nparray).
	pred_cls: Predicted object classes (nparray).
	target_cls: True object classes (nparray).
	plot: Plot precision-recall curve at [email protected]
	fname: Plot filename
	# Returns
	The average precision as computed in py-faster-rcnn.
	"""
	# Sort by objectness
	i = np.argsort(-conf)
	tp, conf, pred_cls = tp[i], conf[i], pred_cls[i]
	# Find unique classes
	unique_classes = np.unique(target_cls)
	# Create Precision-Recall curve and compute AP for each class
	px, py = np.linspace(0, 1, 1000), [] # for plotting
	pr_score = 0.1 # score to evaluate P and R https://github.com/ultralytics/yolov3/issues/898
	s = [unique_classes.shape[0], tp.shape[1]] # number class, number iou thresholds (i.e. 10 for mAP0.5...0.95)
	ap, p, r = np.zeros(s), np.zeros(s), np.zeros(s)
	for ci, c in enumerate(unique_classes):
	i = pred_cls == c
	n_gt = (target_cls == c).sum() # Number of ground truth objects
	n_p = i.sum() # Number of predicted objects
	if n_p == 0 or n_gt == 0:
	continue
	else:
	# Accumulate FPs and TPs
	fpc = (1 - tp[i]).cumsum(0)
	tpc = tp[i].cumsum(0)
	# Recall
	recall = tpc / (n_gt + 1e-16) # recall curve
	r[ci] = np.interp(-pr_score, -conf[i], recall[:, 0]) # r at pr_score, negative x, xp because xp decreases
	# Precision
	precision = tpc / (tpc + fpc) # precision curve
	p[ci] = np.interp(-pr_score, -conf[i], precision[:, 0]) # p at pr_score
	# AP from recall-precision curve
	for j in range(tp.shape[1]):
	ap[ci, j], mpre, mrec = compute_ap(recall[:, j], precision[:, j])
	if j == 0:
	py.append(np.interp(px, mrec, mpre)) # precision at [email protected]
	# Compute F1 score (harmonic mean of precision and recall)
	f1 = 2 * p * r / (p + r + 1e-16)
	if plot:
	py = np.stack(py, axis=1)
	fig, ax = plt.subplots(1, 1, figsize=(5, 5))
	ax.plot(px, py, linewidth=0.5, color='grey') # plot(recall, precision)
	ax.plot(px, py.mean(1), linewidth=2, color='blue', label='all classes %.3f [email protected]' % ap[:, 0].mean())
	ax.set_xlabel('Recall')
	ax.set_ylabel('Precision')
	ax.set_xlim(0, 1)
	ax.set_ylim(0, 1)
	plt.legend()
	fig.tight_layout()
	fig.savefig(fname, dpi=200)
	return p, r, ap, f1, unique_classes.astype('int32')

They don't print out by default, you'd have to introduce some custom code to see them.

fpc and tpc are the FP and TP arrays of shape (n,10), for the 10 iou thresholds of 0.5:0.95. If you look at the last row, this is the FP and TP count per iou threshold:

tpc.shape
Out[3]: (3444, 10)
fpc.shape
Out[4]: (3444, 10)
tpc[-1]
Out[5]: array([138, 124, 105,  91,  80,  66,  54,  38,  22,   9])
fpc[-1]
Out[6]: array([3306, 3320, 3339, 3353, 3364, 3378, 3390, 3406, 3422, 3435])

So at 0.5 iou and 0.001 confidence threshold, for class 0, dataset inference results in 138 TPs and 3306 FPs.

[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.

[glenn-jocher]

@tjbe2021 good news 😃! Your original issue may now be fixed ✅ in PR #5727. This PR explicitly computes TP and FP from the existing Labels, P, and R metrics:

TP = Recall * Labels
FP = TP / Precision - TP

These TP and FP per-class vectors are left in val.py for users to access if they want:

yolov5/val.py

Line 240 in 36d12a5

tp, fp, p, r, f1, ap, ap_class = ap_per_class(*stats, plot=plots, save_dir=save_dir, names=names)

To receive this update:

• Git – git pull from within your yolov5/ directory or git clone https://github.com/ultralytics/yolov5 again
• PyTorch Hub – Force-reload model = torch.hub.load('ultralytics/yolov5', 'yolov5s', force_reload=True)
• Notebooks – View updated notebooks
• Docker – sudo docker pull ultralytics/yolov5:latest to update your image

Thank you for spotting this issue and informing us of the problem. Please let us know if this update resolves the issue for you, and feel free to inform us of any other issues you discover or feature requests that come to mind. Happy trainings with YOLOv5 🚀!