whoisalan/notes

notes

notes's Issues

Bias–variance tradeoff

pre-concept

The bias is an error from erroneous assumptions in the learning algorithm. High bias can cause an algorithm to miss the relevant relations between features and target outputs (underfitting).

Bias is the difference between the average prediction of our model and the correct value which we are trying to predict. Model with high bias pays very little attention to the training data and oversimplifies the model. It always leads to high error on training and test data.

The variance is an error from sensitivity to small fluctuations in the training set. High variance can cause an algorithm to model the random noise in the training data, rather than the intended outputs (overfitting).

Variance is the variability of model prediction for a given data point or a value which tells us spread of our data. Model with high variance pays a lot of attention to training data and does not generalize on the data which it hasn’t seen before. As a result, such models perform very well on training data but has high error rates on test data.

Different between Least squares and ridge regression：

Least squares wants to minimize the sum of the squared residuals

RidgeR want to minimize the sum of the squared residuals + lambda * slope^2 (penalty, lambda≥0)

strength

when we have more dims than the size of data size, Ridge can better settle this question than other regression

Least Squares :

Ridge regression:

conclusion

opencv_python3

shi...上次写的没有保存，重新开始吧...

cv2.line(img,(0,0),(150,150),(255,255,255),15)
为图片上画一条线
普通计算机都是RGB，但opencv是BGR，因此上面三个对应的分别是BGR
对于计算机来说，数字越大，颜色越亮，所以255，255，255是白色
15 means width

矩形：

cv2.rectangle(img,(15,25),(200,150),(0,0,255),15)

圆形：

cv2.circle(img,(100,63), 55, (0,255,0), -1)
-1 for thickness. This means the object will actually be filled in

多边形：

pts = np.array([[10,5],[20,30],[70,20],[50,10]], np.int32)
# OpenCV documentation had this code, which reshapes the array to a 1 x 2. I did not 
# find this necessary, but you may:
#pts = pts.reshape((-1,1,2))
cv2.polylines(img, [pts], True, (0,255,255), 3)
为啥需要[]呢？？？ 

字体：

font = cv2.FONT_HERSHEY_SIMPLEX
cv2.putText(img,'OpenCV Tuts!',(0,130), font, 1, (200,255,155), 2, cv2.LINE_AA)

basic operation

change the color of single pixel

px = img[55,55]
#Next, we could actually change a pixel:
img[55,55] = [255,255,255]
Then re-reference:
px = img[55,55]

ROI=region of image

img[100:150,100:150] = [255,255,255]

watch_face = img[37:111,107:194]
img[0:74,0:87] = watch_face
cv2.imshow('image',img)

Bagging and bagging in sk-learn

pre-concept

ensemble methods use multiple learning algorithms to obtain better predictive performance than could be obtained from any of the constituent learning algorithms alone

Bootstrap aggregating, also called bagging, is a machine learning ensemble meta-algorithm designed to improve the stability and accuracy of machine learning algorithms used in statistical classification and regression.

The idea of bagging is to reduce overfitting of one model(according to what I learned before: to adding bias and reducing variance)