Sem título

Machine learning from basics

• Artificial Intelligence: A field of computer science that aims to make computers achieve intelligence like humans do have.
• Machine Learning: A set of techniques in which computers are trained to perform a particular task not by programming them.
• Neural Network: A construct in Machine Learning inspired by the network of neurons (nerve cells) in the biological brain.
• Neural networks are a fundamental part of deep learning.
• Deep Learning: A subfield of machine learning that uses multi-layered neural networks. Often, “machine learning” and “deep learning” are used interchangeably.

What is a machine learning?

• Machine learning and deep learning also have many subfields, branches, and special techniques. Most of the Ml and DL is the separation of Supervised Learning and Unsupervised Learning.
• Supervised learning you know what you want to teach the computer, while unsupervised learning is about letting the computer figure out what can be learned. Supervised learning is the most common type of machine learning.
• In machine learning input and putput is known but algorithm to convert input to output is not known.
• ML takes pairs of input and output and figure out algorithm.
• Inputs are called features and Outputs are called as labels.
• Examples or training examples are pair of input and output.
• Simplest ML model is dense model with one layer and one neuron.

Dense model

• model = tf.keras.Sequential([
• tf.keras.layers.Dense(units =1 , input_shape = [1])
• ])
• units tell number of neurons in that layer
• input shape tells the deimensions of input.
• model.compile(loss ="mean_squared_error", optimizer = tf.keras.optimizers.Adam(0.1))
• loss tells the diiference between output and excepted output.
• optimizer is used to set the internal variables to reduce the loss.
• 0.1 is a learning rate.
• Model is trained by using fit method.
• history = model.fit(celsius_q, fahernheit_a, epochs = 500, verbose = False)
• epochs is the number of iterations or times the nerual network train.
• verbose tells how to control output.