Two cloud computing platform funeral – Sohu Technology jodie foster

The funeral of the two cloud computing platform – Sohu science and technology Lei Feng, according to: Qu Xiaofeng, Hong Kong Polytech University, research center for human biometrics. Lei Feng network (search for "Lei Feng" public concern number) exclusive articles, please contact authorized reprint. Recently, I have used two cloud computing platform and getdatajoy, is about to die. The former has been acquired, has been shut down for individual users; the latter is about to close in January 2, 2017. At the outbreak of the artificial intelligence today, two is the core of intelligent computing cloud computing platform, not only the wind and the sky dragon, homeopathy in the new era, but fell on the threshold, have to cause people to ponder. | – on-demand allocation of computing power on the, each project can use different computing capabilities of the platform, run separately. A single project can use multiple virtual CPU and large or small memory space. is a data scientist can dynamically allocate computing power cloud computing platform, in short, is the data scientist GitHub". On Sense, data scientists can collaborate with others and generate data reports. March 22, 2016, Sense announced the acquisition of big data software company Cloudera. Sense founder Tristan Zajonc and Anand on the Sense blog released a notice of the acquisition was announced, while announced free and personal services closed in April 31, 2016, on. In Sense, users can directly use Python, R, Julia code, algorithm experiment, construction model, and then according to the algorithm needs and cost into consideration, choose suitable operation ability of cloud computing platform (CPU, virtual memory) operation, then the output and save the results. Run the program, can be a one-time function, you can also use a similar Jupyter interactive execution environment to single step, respectively, to see the results of the output. The output of the results can be data files, such as CSV, can be PNG, JPG and other formats of the picture, or can be displayed with Java Dynamic chart. Finally, you can generate markdown, PDF report. I used to do some data analysis. Its biggest advantage is the ability to configure the characteristics of computing. In the initial stage of the experiment, the use of a small amount of computing power, with a single CPU check data, debugging algorithm, test assumptions. When the experimental process is relatively clear, the code runs through, you can change the use of large computing power, with 16, 32, 64 core and large memory load all data operations, as soon as possible to obtain experimental results. Especially in the same direction of the experiment, you can simply copy the project, modify parameters, add functions or adjust.相关的主题文章: