Training The Model

Since we’re done with creating and instantiating our model, now it’s time to train it. We’ll use the fit method to train our model. This method takes features and targets as targets, and we can also pass the validation_data with it, it will automatically try your model on validation and note the loss metrics. We also provide the batch_size, e what this does is breaks our data into small batches and feed it to our model for training in each epoch, this is very helpful when you have large datasets because it reduces the RAM and CPU consumption on your machine. 

Now here we have only trained our model for 15 epochs because our purpose here is to get familiar with the process and not the accuracy itself, but you’ll have to increase or decrease the number of epochs on your machine.  There are optimization methods that you can use such as early stopping that will automatically stop the training when the model starts overfitting, so you can also try using these, I have provided a link at the bottom if you want to read about it.

Output:

Here we have only trained for 15 epochs but you should definitely train for more and try changing the model itself.

Deep learning has been on the rise in this decade and its applications are so wide-ranging and amazing that it’s almost hard to believe that it’s been only a few years in its advancements. And at the core of deep learning lies a basic “unit” that governs its architecture, yes, It’s neural networks.

A neural network architecture comprises a number of neurons or activation units as we call them, and this circuit of units serves their function of finding underlying relationships in data. And it’s mathematically proven that neural networks can find any kind of relation/function regardless of its complexity, provided it is deep/optimized enough, that is how much potential it has.

Now let’s learn to implement a neural network using TensorFlow