Class 21 - INTRODUCTION TO TENSORFLOW Notes from the AI Basic Course by Irfan Malik & Dr Sheraz Naseer (Xeven Solutions)

Class 21 - INTRODUCTION TO TENSORFLOW

Notes from the AI Basic Course by Irfan Malik & Dr Sheraz Naseer (Xeven Solutions)

Conventional ML works on features. It has limitations, cannot work on raw data.

They can process features, but cann't process unstructred data.

Scikit learn is the famous library for conventional ML.

Now, we are going to discuss DEEP LEARNING.

Branch of ML, in which we uses neural network for different tasks called deep learning.

What is neuron?

Unit of cell in our brain.

Brain is composed of billion of neurons.

Neural network inspiration is from Biology.

When you have complex data, linear models will not work.

Patterns are like you daily routine habbit, same goes for data.

From now onwards, It's Practical World.

Welcome to Deep Learning:

It has Input, Feature Extraction + Classification & Output.

If you know what you want (output), neural network is here for your help.

Google Slides:

https://docs.google.com/presentation/d/1vRYqKt-13fgL1_uxubC4jc_RcnlS8ucN/edit#slide=id.g25fee922ceb_0_5

Simple Neural Network:

It like a math's function.

Tensor Flow:

TensorFlow is an open-source machine learning framework developed by Google.

It enables building, training, and deploying machine learning models.

Primarily used for deep learning tasks, it's versatile for various other types of machine learning.

Tensor Flow Playground:

https://playground.tensorflow.org/#activation=tanh&batchSize=10&dataset=circle&regDataset=reg-plane&learningRate=0.03&regularizationRate=0&noise=0&networkShape=4,2&seed=0.21712&showTestData=false&discretize=false&percTrainData=50&x=true&y=true&xTimesY=false&xSquared=false&ySquared=false&cosX=false&sinX=false&cosY=false&sinY=false&collectStats=false&problem=classification&initZero=false&hideText=false

Framework:

Like a toolbox.

Ready to use library

Built-in functions

Hidden Layer means Feature Extraction + Classification

Loss function:

Distance b/w actual & predicted layer

Loss near to 0 means, model is performing well.

If Epoch is 1, it means all data has been seen.

Single good feature can solve alot of your Problems in Neural Networks.

#AI #artificialintelligence #datascience #irfanmalik #drsheraz #xevensolutions #hamzanadeem