Tensorflow In Machine Learning
Download as PPTX, PDF0 likes50 views
The document discusses machine learning, its types, and applications, particularly focusing on TensorFlow, a library developed by Google for machine learning tasks. It covers the history of TensorFlow, usage scenarios, as well as the pros and cons of the library. Additionally, it touches on the differences between machine learning and deep learning, including various algorithms and their applications in real-world scenarios.
Tensorflow In Machine Learning
- 1. MOHAMED HEGAZY R&D SOFTWARE ENGINEER 1
- 2. About me R&D Engineer at eFinance Loves to learn and talk about new teach and theories Computer Science Graduate from the British University in Egypt 2
- 3. Agenda Introduction about Machine Learning Life without Machine Learning Types of Learning in machine Learning Tensorflow .. What is it Tensorflow .. Use cases Tensorflow .. Creating Simple app in python Tensorflow .. Pros and cons Summary Word of advice 3
- 4. The Era of Machine Learning Started in 1959. Study of Pattern recognition and computational Learning Theory Mainly focuses on Predictions Does it sound Familiar ? 4
- 5. Types Of learning in Machine Learning Supervised Learning Semi-Supervised Learning Active Learning (reinforcement learning) Unsupervised Learning 5
- 6. Artificial Intelligence Vs Machine Learning 6
- 7. Deep Learning Vs Machine Learning 7
- 8. Machine learning vs Deep learning Machine Learning Deep Learning Training dataset Small Large Choose your own Features Yes No Number of Classifiers available Many Few Training Time Short Long 8
- 9. Tensor flow – History Created in 2011 by Google Brain team, Firstly named as “DISTBELIF”. To identify photos The system used positive reinforcement Used for all of their AI/ML functionality Made to tackle problem of teamwork between researchers, data scientists and developers In 2015 Google Released Tensorflow for Open-Source use. 9
- 12. Tensorflow – GRAPHS and nodes 12
- 13. Tensorflow – GRAPHS and nodes 13
- 14. Tensor flow – the tool : Real World Applications Speech recognition Image recognition Object tagging videos Self-driving cars Sentiment analysis Detection of flaws Text summarization Mobile image and video processing Air, land, and sea drones 14
- 15. TENSORFLOW – USAGE : image classification 15
- 16. 16
- 17. TENSORFLOW – USAGE : voice analysis Start recording the Conversation when hearing a specific word P.S. it never heard your voice before 17
- 18. TENSORFLOW – USAGE : Data Classification 18
- 19. Tensorflow – Real World Applications Google products that use TensorFlow include •Google search •Google's voice recognition app •Google Translate. 19
- 20. Tensorflow – application perquisites Most suitable language : python Platform : Best for Linux Library download ◦ Pip install Tensorflow ◦ Pip install tensorflow-gpu (available for NVIDIA drivers) ◦ CUDA drivers from NVIDIA ◦ cuDNN from NVIDIA Setting up training 20
- 21. TenSorflow -- Sample 21 >>> import tensorflow as tf >>> hello = tf.constant('Hello, TensorFlow!') >>> sess = tf.Session() >>> print(sess.run(hello)) mport tensorflow as tf # Set up a linear classifier. classifier = tf.estimator.LinearClassifier(feature_columns) # Train the model on some sample data. classifier.train(input_fn=train_input_function, steps=2000) # Use it to predict. predictions = classifier.predict(input_fn=predict_input_function)
- 22. Pros of Tensorflow 1. Graphs 1. Faster and better than some other libraries 2. Library Management 1. Backed By Google for updates and performance 3. Debugging 4. Scalability 5. Pipelining 6. Keras 7. Distributed 22
- 23. Cons of Tensorflow 1. Slower than other algorithms and deep learning libraries 23
- 24. Extra materials Google Machine Learning Crash Course https://developers.google.com/machine-learning/crash-course/ 24
- 26. Thank you 26
Editor's Notes
- #7: Ai is like the if statement machine learning is like dog identification with little data deep learning