Dice coefficient Vs Soft Dice coefficient - machine-learning

What is the difference between Dice coefficient and Soft Dice coefficient?
Background : semantic segmentation

Related

Selecting an IoU and confidence threshold for evaluation of model performance

mAP is commonly used to evaluate the performance of object detection models. However, there are two variables that need to be set when calculating mAP:
confidence threshold
IoU threshold
Just to clarify, confidence threshold is the minimum score that the model will consider the prediction to be a true prediction (otherwise it will ignore this prediction entirely). IoU threshold is the minimum overlap between ground truth and prediction boxes for the prediction to be considered a true positive.
Setting both of these thresholds to be low would result in a greater mAP. However, the low thresholds would most likely be inconsistent with the mAP scores from other studies. How does one select, and justify, these threshold values?
In Yolov5, we do NMS on the outputs of network, then calculate mAP. So, there is a conf_thres and an iou_thres in NMS to filter some boxes, these are set to 0.001 and 0.6, see: https://github.com/ultralytics/yolov5/blob/2373d5470e386a0c63c6ab77fbee6d699665e27b/val.py#L103.
When calculating mAP, we set iou threshold to 0.5 for mAP#0.5, or 0.5 to 0.95 with step 0.05 for mAP#0.5:0.95.
I guess the way of calculating mAP in Yolov5 is aligned to other framework. If I'm wrong, please correct me.

Improving SNR (signal to Noise) - sine wave over white noise

I'm reading a signal which is a constant frequency sine wave (F = 1KHz), constant phase and changing amplitude.
For example, if the read-back signal is "high" the amplitude will be (say) 1V.
If the read-back signal is "low" the amplitude will be (say) 0.01 V.
The signal is deep in white noise (full spectra).
How can I improve SNR to remove the noise but remain with the signal ?
Of course I tried steep filters BPF at the known frequency.
Any other ideas to remove the noise and get a better SNR ?

What machine learning algorithm for predicting inputs to get a certain output

I have a simple golf game where the only inputs are angle and power. Submitting these values shoots the ball towards the hole. After the ball stops, a score is created based on the distance between the hole and the ball.
I want to use a machine learning algorithm to predict angle and power values that will give me the perfect score (sinking the ball in the hole).
I understand that I can use a Linear Regression to predict a score based on a chosen angle and power, but I'm not sure how to do it the other way around (get a suggested angle and power from a given score).
Learning of a linear regression model in your case is to solve (a,b,c) in:
score = a * angle + b * power + c
After the training you have values of (a,b,c). When you want to suggest angle and power to obrain particular score you have
The angle is in range of <0,90), so you can calculate the power needed to use for particular angle to obtain a perfect score.

Gamma normalization in HOG (Histogram of Gradients)

What is the meaning of gamma normalization in HOG(Histogram of Gradients)? Is it the same as gamma correction? Because I came across many journals saying that gamma normalization in HOG is the square root of the image intensity. But it is different compared to gamma correction's formula.
Gamma Normalization in HOG is actually Power Law Transformation
s = cr^γ
where s is output pixel, r is input pixel, c is constant and γ is exponent. Different devices used for image capture, display and printing use this power law transformation to correct image intensity values and this process is known as gamma correction. In short, gamma normalization in HOG is same as gamma correction.

Finding the function approximated by a neural network

If I have a feed-forward multilayer perceptron with sigmoid activation function, which is trained and has known weights, how can I find the equation of the curve that is approximated by the network (the curve that separates between 2 types of data)?
In general, there is no closed form solution for the input points where your NN output is 0.5 (or 0, in case of -1/1 instead of 0/1).
What is usually done for visualization in low-dimensional input space is gridding up the input space and computing the contours of the NN output. (The contours are smooth estimate of what the NN response surface looks like.)
In MATLAB, one would do
[X,Y] = meshgrid(linspace(-1,1), linspace(-1,1));
contour(f(X,Y))
where f is your trained NN, and assuming [-1,1] x [-1,1] space.

Resources