Apsara Clouder Big Data Skills Certification: Using deep learning TensorFlow framework for image recognition
Compared to Caffe, Theano, Torch, and MXNet, TensorFlow has the most forks and stars on Github, and has rich applications in graphics classification, audio processing, recommendation systems, and natural language processing. The recently popular Keras framework uses TensorFlow by default at the bottom. The famous Stanford CS231n course uses TensorFlow as the programming language for teaching and homework. A number of TensorFlow books at home and abroad have been prepared or sold. AlphaGo developer Deepmind is also planning to migrate neural network applications to TensorFlow, which confirms TensorFlow’s popularity in the industry. TensorFlow is not only open source on Github, but also available in TensorFlow: Large-scale Machine Learning on Heterogeneous Distributed Systems also introduced the design and implementation of the system framework. The training cluster with 200 nodes tested is also unmatched by other distributed deep learning frameworks.
This certification system introduces some basic knowledge of deep learning, and the working principle of Tensorflow. Through Ali Cloud machine learning PAI based on the classic CIFAR-10 data set to achieve image recognition. Through this experiment, students can have a simple and intuitive understanding of deep learning Tensorflow framework.
Certification entry: Click here