Time of instruction:
This course will begin on April 1. The duration of the course is approximately 14 weeks.
Subject:
People who are interested in deep learning AI, who want to learn about deep learning practices.
learners need a little bit of the basics of Python development and deep learning, neural network fundamentals
Course Environment:
Windows10 + Anaconda
Harvest Expectations:
Master the basic use of TensorFlow and Tensorboard, can skillfully use tensorflow to do image recognition and verification code identification, in-depth understanding and mastering the use of TensorFlow in NLP
Instructor Profile:
Ben, a technology enthusiast with years of experience in software algorithm development, is currently a first-line engineer in deep learning, with extensive in-depth learning experience and first-line practical experience!
Course Outline:
First Lesson TensorFlow introduction, Anaconda installation, TensorFlow CPU version installation.
The second lesson TensorFlow the basics of use, including some interpretation and manipulation of graphs (graphs), sessions (session), tensor (tensor), variables (Variable).
lesson three TensorFlow linear regression and simple use of classifications.
The fourth lesson Softmax, cross-entropy (cross-entropy), dropout, and the introduction of various optimizations in TensorFlow.
Fifth Lesson, CNN, and CNN to solve the problem of mnist classification.
The sixth lesson uses Tensorboard to visualize the structure and visualize the process of the network operation.
The seventh lesson is the explanation of recurrent neural network lstm and the use of LSTM network.
The eighth lesson is to save and load the model, using Google's image recognition network inception-v3 for image recognition.
Nineth Lesson tensorflow GPU version installation. Design your own network model and train your own network model for image recognition.
The tenth lesson uses the TensorFlow to identify the verification code.
The 11th lesson TensorFlow use in NLP (i).
The 12th lesson TensorFlow use in NLP (ii).
in recent years, deep learning technology has been widely applied and disseminated in academia and industry. The spread of deep learning is not only due to the progress of the algorithm, but also because the deep learning technology has achieved very good application effect in all walks of life.
deep Learning, as a combination of theory and practice, has emerged in the new algorithm theory, and various deep learning frameworks have been appearing in people's Field of vision. Like Torch,mxnet,theano,caffe and so on. Google announced on November 9, 2015 that its own second-generation machine learning system, TensorFlow, was open. Deep learning is a key part of future new products and technologies. Research in this area is global and fast-growing, but lacks a standardized tool. Google wants to make TensorFlow the standard for the deep learning industry.
TensorFlow supports Python and C + + languages, supports algorithms such as CNN, RNN, and lstm, and can be used in many deep learning areas such as speech recognition or image processing. It can run on one or more CPUs or GPUs. It can be run in embedded systems (such as mobile phones, tablets), in PCs, and in distributed systems. It is currently the world's hottest deep learning platform (no one).
The course content is basically code-based programming, there will be a small amount of deep learning theoretical content. The course starts with some of the most basic knowledge from TensorFlow's most basic diagrams (graphs), sessions (session), tensor (tensor), variables (Variable), and gradually talks about the basics of TensorFlow, And the use of CNN and LSTM in TensorFlow. After the course, we will take you to do a few practical projects, such as training your own model to do image recognition, using TensorFlow to identify the verification code, and TensorFlow in the use of NLP.
: Baidu Network disk download
Original address: http://linyunbbs.com/thread-2275-1-1.html
The study and application of into gold deep learning tensorflow framework in smelting number video tutorial
