意外的非懒惰

问题描述

这周我写了这段代码 challenge 来产生丑陋的数字。

sub factors( $n ) {
  if $n > 1 {
     $_,|factors $n div $_
      given ( grep $n %% *,2..* ).first } }

.say for ( 1,|grep *.&factors.all ∈ ( 2,3,5 ),2..* )[^150];

这是有效的，从某种意义上说，它产生了正确的输出，但它的行为并不懒惰：输出不会立即开始，而是在开始 30 秒后开始。

但是，当我删除索引并遍历裸序列时

.say for 1,2..*;

我按预期立即得到输出。

这是一个错误，是吗？

解决方法

这不是错误（尽管可能记录得更好）。使用 import cv2 import face_recognition import math import csv from os import path # calculate Eye Aspect Ratio (EAR) def cal_EAR(arr): p2_p6 = math.sqrt(pow((arr[1][0] - arr[5][0]),2) + pow((arr[1][1] - arr[5][1]),2)) p3_p5 = math.sqrt(pow((arr[2][0] - arr[4][0]),2) + pow((arr[2][1] - arr[4][1]),2)) p1_p4 = math.sqrt(pow((arr[0][0] - arr[3][0]),2) + pow((arr[0][1] - arr[3][1]),2)) return (p2_p6 + p3_p5) / (2 * p1_p4) # calculate Mouth Aspect Ratio (MAR) def cal_MAR(arr_top,arr_bottom): top_bottom = math.sqrt(pow(arr_top[-3][0] - arr_bottom[-3][0],2) + pow(arr_top[-3][1] - arr_bottom[-3][1],2)) left_right = math.sqrt(pow(arr_top[0][0] - arr_bottom[0][0],2) + pow(arr_top[0][1] - arr_bottom[0][1],2)) return top_bottom / left_right # calculate pupil circularity (PUC) def cal_PUC(arr): distance_1_2 = math.sqrt(pow(arr[0][0] - arr[1][0],2) + pow(arr[0][1] - arr[1][1],2)) distance_2_3 = math.sqrt(pow(arr[1][0] - arr[2][0],2) + pow(arr[1][1] - arr[2][1],2)) distance_3_4 = math.sqrt(pow(arr[2][0] - arr[3][0],2) + pow(arr[2][1] - arr[3][1],2)) distance_4_5 = math.sqrt(pow(arr[3][0] - arr[4][0],2) + pow(arr[3][1] - arr[4][1],2)) distance_5_6 = math.sqrt(pow(arr[4][0] - arr[5][0],2) + pow(arr[4][1] - arr[5][1],2)) distance_6_1 = math.sqrt(pow(arr[5][0] - arr[0][0],2) + pow(arr[5][1] - arr[0][1],2)) distance_2_5 = math.sqrt(pow(arr[2][0] - arr[5][0],2) + pow(arr[2][1] - arr[5][1],2)) perimeter = distance_1_2 + distance_2_3 + distance_3_4 + distance_4_5 + distance_5_6 + distance_6_1 area = pow(distance_2_5 / 2,2) * math.pi circularity = (4 * math.pi * area) / pow(perimeter,2) return circularity # calculate eyebrow angle (EBA) def cal_EBA(arr): hypotenuse = math.sqrt(pow(arr[2][0] - arr[3][0],2)) opposite = abs(arr[2][1] - arr[3][1]) angle = opposite / hypotenuse return angle # calculate chin aspect ratio (CAR) def cal_CAR(arr): p0_p16 = math.sqrt(pow(arr[0][0] - arr[16][0],2) + pow(arr[0][1] - arr[16][1],2)) p0_p8 = math.sqrt(pow(arr[0][0] - arr[8][0],2) + pow(arr[0][1] - arr[8][1],2)) CAR = p0_p8 / p0_p16 return CAR def modify_csv(file_name): # ask user to enter the value of the class CLASS = input("Please enter the value of the class: ") # open the csv file with open(file_name,'w',newline='') as file: writer = csv.writer(file) writer.writerow(["L_EAR","R_EAR","MAR","PUC","MOE","ABA","CAR","CLASS"]) rows = 0 is_existed = True while is_existed: # input the name of the csv file file_name = input("Please enter the name of the csv file you want to create: ") file_name = "../csv_files/" + file_name + ".csv" # check whether the file is already existed is_existed = path.exists(file_name) # warn user if the file is already existed if is_existed: print("The file is already existed") # ask user whether he/she wants to delete it or create another file choice = input("Do you want to rewrite file? (N/Y): ") if choice.upper() == 'N': continue elif choice.upper() == 'Y': is_existed = False modify_csv(file_name) else: is_existed = False modify_csv(file_name) while True: # turn on the webcam and check the status capture = cv2.VideoCapture(0) if capture.isOpened() is False: print("Camera Error,please check your camera @_@") exit() # change the BGR frame to gray frame ret,frame = capture.read() gray = cv2.cvtColor(frame,cv2.COLOR_BGR2GRAY) # use face_recognition library to locate the landmarks face_marks = face_recognition.face_landmarks(gray,None,"large") print(face_marks) print("chin points: ",face_marks[0]["chin"]) print("left_eyebrow points: ",face_marks[0]["left_eyebrow"]) print("right_eyebrow points: ",face_marks[0]["right_eyebrow"]) print("nose_bridge points: ",face_marks[0]["nose_bridge"]) print("nose_tip points: ",face_marks[0]["nose_tip"]) print("left_eye points: ",face_marks[0]["left_eye"]) print("right_eye points: ",face_marks[0]["right_eye"]) print("top_lip points: ",face_marks[0]["top_lip"]) print("bottom_lip points: ",face_marks[0]["bottom_lip"]) # calculate EAR,MAR,PUC,and MOE if face_marks.__len__() != 0: L_EAR = cal_EAR(face_marks[0]["left_eye"]) R_EAR = cal_EAR(face_marks[0]["right_eye"]) MAR = cal_MAR(face_marks[0]["top_lip"],face_marks[0]["bottom_lip"]) PUC = cal_PUC(face_marks[0]["left_eye"]) EBA = cal_EBA(face_marks[0]["right_eyebrow"]) CAR = cal_CAR(face_marks[0]["chin"]) MOE = MAR / L_EAR # write these values to csv file with open(file_name,'a',newline='') as file: writer = csv.writer(file) writer.writerow([L_EAR,R_EAR,MOE,EBA,CAR,CLASS]) rows += 1 print("Number of records: ",rows) # mark these landmarks on the frame for landmarks_dict in face_marks: print(landmarks_dict) point_order = 0 for landmarks_key in landmarks_dict.keys(): # keys: chin,left_eyebrow,right_eyebrow,nose_bridge,nose_tip,left_eye,right_eye,top_lip,bottom_lip print(landmarks_key) for point in landmarks_dict[landmarks_key]: point_position = (point[0],point[1]) cv2.circle(frame,point,3,(255,0),-1) cv2.putText(frame,str(point_order),point_position,fontFace=cv2.FONT_HERSHEY_SCRIPT_SIMPLEX,fontScale=0.3,color=(0,255,0)) print("Point ",point_order,"is",point_position) point_order = point_order + 1 # rectangle the face face_point = face_recognition.face_locations(gray) print(face_point) for pts in face_point: cv2.rectangle(frame,(pts[3],pts[0]),(pts[1],pts[2]),2) # press q to exit the loop if rows >= 200: print("Done") break # display the frame cv2.imshow("Face_landmarks_fr",frame) # release the memory capture.release() cv2.destroyAllWindows() 进行索引将其索引的元素具体化，因此 [] 计算前 150 个元素的（非惰性）列表。 （列表的其余部分保持惰性，但那些初始元素不会）。

如果你想懒惰地迭代，你可以使用 [^150] 代替，它会给你以下最后一行：

&head