Successfully reported this slideshow.
We use your LinkedIn profile and activity data to personalize ads and to show you more relevant ads. You can change your ad preferences anytime.

Learn Basics of Artificial intelligence with TensorFlow


Published on

This is a deck to understand the main concepts behind AI. It covers important questions like what is an AI model, Supervised and Unsupervised model. It also covers basic information on TensorFlow.

Published in: Education
  • Be the first to comment

  • Be the first to like this

Learn Basics of Artificial intelligence with TensorFlow

  1. 1. Artificial Intelligence
  2. 2. Gautam Gupta Architect Fin-tech $$$
  3. 3. What is Artificial Intelligence?
  4. 4. Use YP to search coffee shop or Search Google for “Coffee near me” Artificial Intelligence: Examples
  5. 5. Calling 1-800 call center IVR or Ask SIRI/Alexa a question Artificial Intelligence: Examples
  6. 6. Searching in aisles of a store or Shopping on Artificial Intelligence: Examples
  7. 7. What does a regular program do?
  8. 8. We have X. We want Y. We write a function to get Y. E.g. Income Tax calculation Regular Program
  9. 9. What does an AI program do?
  10. 10. We have X. We know some Y. We try different functions f to get Y. The approx. f is output of AI. E.g. Image Recognition AI Program
  11. 11. We have data of fruits. Weight, color, height, length, shape etc. AI Program
  12. 12. Between Apple and Banana, shape is the classifier AI Program
  13. 13. Between Apple and pine apple, size is the classifier AI Program
  14. 14. Between Apple and Orange, color is the classifier AI Program
  15. 15. What is the difference between AI and ML?
  16. 16. AI vs. ML Artificial Intelligence Machine Learning Data Engineering Computer Science Uses
  17. 17. What is an AI Model?
  18. 18. AI Model Model 1 Model 2 … Model 3
  19. 19. AI Model Output Input C Input B Input A
  20. 20. to identify text signs. to read characters of language. to translate the words. to render the translation on your phone.
  21. 21. What are the main types of AI Models?
  22. 22. Supervised Model Prediction Demand of a product based on previous sale Y = f(X)
  23. 23. Unsupervised Model Find Association Clustering Is there a Y = f(X)
  24. 24. How does an AI program work?
  25. 25. We have some data and price of houses. We want to estimate the price of more houses. AI Program: Regression Example
  26. 26. It is a regression problem. We have some data about the houses like- built up area, lot size, year built etc. We use algorithm like- Linear Regression. AI Program : How
  27. 27. AI Program: Linear Regression Price = aX + bY + c X = sq ft Y = year built
  28. 28. We have some data about animals. We want to classify more data to identify the animal type for a record. AI Program: Classification
  29. 29. It is a clustering problem. We have some data about the animals like- height, weight, shape, color, has tail etc. We use algorithm like- K nearest neighbors. AI Program: How
  30. 30. AI Program: KNN
  31. 31. How do we train a model?
  32. 32. Training a Model We start with initial data. Divide data into training and test data. Create model on training data. Test model on test data
  33. 33. Let say we have sale price, sq ft., year built, number of rooms, tax information of 100 houses. Training a Model Initial Data Training Data Test Data 25 houses75 houses100 houses
  34. 34. Create a Model on training data Training Output 75 houses Training Data
  35. 35. Test the model on test data Test Output 25 houses Test Data
  36. 36. Run the model on production data Production Output houses Production Data
  37. 37. What are the tools for Artificial Intelligence?
  38. 38. Language: Python, R, Java Data tools: SQL, Pandas etc. Data storage: BigQuery, DynamoDB etc. Data Stream: Kinesis, Kafka etc. ML libraries: Scikit learn, TensorFlow, Theano etc. Environment: Jupyter Notebook Production: Google Cloud, AWS Cloud, Azure Artificial Intelligence: Tools
  39. 39. What is Jupyter Notebook?
  40. 40. Julia Python R Open Source Web based Interactive output Share documents Big data support Jupyter Notebook
  41. 41. Jupyter Notebook Demo
  42. 42. What is Deep Learning?
  43. 43. Inspired from nervous system Uses layers for ML Learns supervised and unsupervised Learns by example Requires large amount of data E.g. Driverless cars, voice control Deep Learning
  44. 44. What is Neural Network?
  45. 45. One input Layer Multiple hidden layers One output Layer Weights Neural Network
  46. 46. TensorFlow
  47. 47. What is a Tensor?
  48. 48. Matrix 2 D 1 2 3 4 5 6 7 8 9 Tensor Tensor n D 0D 1 1D 1 2 3 2D 1 2 3 4
  49. 49. What is TensorFlow?
  50. 50. Open Source Library Machine Learning and Deep Learning Developed by Google Brain team Flexible Architecture TensorFlow
  51. 51. What are the benefits of TensorFlow?
  52. 52. TensorFlow : Benefits Written in Python Google support Faster compile time Tensorboard for visualization Deep learning
  53. 53. What is the API used in TensorFlow?
  54. 54. Keras API High Level API for TensorFlow TensorFlow API
  55. 55. What is an Epoch in TensorFlow?
  56. 56. “An epoch is a full iteration over samples. The number of epochs is how many times the algorithm is going to run.” TensorFlow Epoch
  57. 57. What are the big companies using TensorFlow?
  58. 58. TensorFlow Users
  59. 59. TensorFlow Demo
  60. 60. Want to know more?
  61. 61. Thanks!