Category：Python
-
Time：2020-8-14
After training the target detection model, it is necessary to evaluate its performance. The accuracy under different thresholds and whether there is missing detection is calculated based on IOU (intersection over union). Hope to provide some ideas, if you feel useful, please praise me and praise me~ The value of IOU can be understood as […]
-
Time：2020-8-14
1 data source The pictures that appear in the answers to all the questions under the topic “beauty” 2 grab tool Python 3, and uses the third-party library requests, lxml, aipface, the code is 100 + lines 3 necessary environment MAC / Linux / windows. Before windows, there were many abnormal reactions. Later, it […]
-
Time：2020-8-13
Python does not support true multithreading. Python provides multithreaded packages, but if you want to speed up your code through multithreading, it’s not a good idea to use multithreaded packages. Python has a thing called global interpreter lock (GIL), which ensures that only one of your multiple threads is executed at any time. The execution […]
-
Time：2020-8-13
Many of the official documents, theses and networks built by the great gods all calculate the accuracy rate, and there is very little miscalculation rate, Here’s how to calculate the accuracy andMisjudgment rate and recall rateEtc 1. Calculation accuracy Get the number of predicted correct numbers of each batch train_correct = (pred == batch_y.squeeze(1)).sum() The […]
-
Time：2020-8-12
Background There is a requirement in the development: when the program runs to a certain place, it needs to be confirmed by the user, but it can’t wait all the time. The later program can’t be blocked all the time. It needs to have a timeout limit, that is, if the program has not received […]
-
Time：2020-8-12
dominant sea power preface On December 7, 2018, the last blockbuster of the year, sea king, was released on schedule. At present, the cat’s eye score has reached 9.5 points. Relying on the production cost of US $150 million, the box office is close to 900 million. This article crawls through the 3W + […]
-
Time：2020-8-11
After training the network in Python, we need to test the learning results. The methods used in the routine on the official website are all correct torch.eq () this function. But in order to evaluate the results more precisely, we need to calculate other indicators. After looking through the official website API, we found that […]
-
Time：2020-8-11
File operation Import OS ා module for operating files and foldersImport shutil ා advanced module of file operation 1. Create an empty folderfile=open(“youliang.txt”,”w”,encoding=”utf-8″)file.close() 2. Rename the file******os.rename(“youliang.txt”,”yl.txt”) 3. Delete file operationos.remove(“yl.txt”) 4. Create a folder named AAAos.mkdir(“AAA”) Create the file in the specified directoryfile=open(“AAA/1.txt”,”w”,encoding=”utf-8″)file.close()file=open(“AAA/2.txt”,”w”,encoding=”utf-8″)file.close() 6. Get all the file names in the specified directory and […]
-
Time：2020-8-10
The average accuracy of classifier is calculated correct = torch.zeros(1).squeeze().cuda() total = torch.zeros(1).squeeze().cuda() for i, (images, labels) in enumerate(train_loader): images = Variable(images.cuda()) labels = Variable(labels.cuda()) output = model(images) prediction = torch.argmax(output, 1) correct += (prediction == labels).sum().float() total += len(labels) acc_str = ‘Accuracy: %f’%((correct/total).cpu().detach().data.numpy()) The accuracy of each sub class is calculated correct = list(0. […]
-
Time：2020-8-10
def isprime(n): for i in range(2, n): if n % i == 0: return 0 return 1 N = int (input (“please enter a data”)) value_set = set() for i in range(2, n): if isprime(i) == 1: value_set.add(i) print(value_set)
-
Time：2020-8-9
What is worth learning: 1. Select the method of legal index 2. Array to image display import numpy as np from PIL import Image #input : shape(N, 4) # (x, y, z, intensity) def pointcloud2image(point_cloud): x_size = 640 y_size = 640 x_range = 60.0 y_range = 60.0 grid_size = np.array([2 * x_range / x_size, 2 […]
-
Time：2020-8-9
Although there are a lot of words in this question, it is actually a paper tiger. A list derived answer will come out. value = sum([2 ** i for i in range(64)]) print(value) Operation results:
