Beginner's Guide to Deep Learning

## Steps for perceptron algorithm
### Step 1 : Initialize weights
 Initialize with random weights

### Step 2: Predict
**Calculate the perceptron output**

The boundary separating the two classes is given by:
<center>

$\hat{y} = w_1x_1 + w_2x_2 + w_3x_3 + ... + w_nx_n + b$
</center>

### Step 3: Compare
**Compute the error**
 
The error is computed by subtracting the predicted output($\hat{y}$) from the actual output($y$).
<center>

$error = y' - y$

 
</center>

> If the error is not zero 0, then go to step 4.

### Step 4: Train 

**Update weight and bias using weight update rule**


For every misclassified point $i$:

<center>

 $w$i = 
 $w$i + $\Delta w$i

 $\Delta w$i = error * $x$i

</center>

# Steps for perceptron algorithm
## Step 1 : Initialize weights
 Initialize with random weights

## Step 2: Predict
**Calculate the perceptron output**

The boundary separating the two classes is given by:
<center>

$\hat{y} = w_1x_1 + w_2x_2 + w_3x_3 + ... + w_nx_n + b$
</center>

## Step 3: Compare
**Compute the error**
 
The error is computed by subtracting the predicted output($\hat{y}$) from the actual output($y$).
<center>

$error = y' - y$

 
</center>

> If the error is not zero 0, then go to step 4.

## Step 4: Train 

**Update weight and bias using weight update rule**


For every misclassified point $i$:

<center>

 $w$i = 
 $w$i + $\Delta w$i

 $\Delta w$i = error * $x$i

</center>

Chapter 1

Perceptron Algorithm

Steps for perceptron algorithm

Step 1 : Initialize weights

Step 2: Predict