DOG VISION(A DEEP LEARNING PROJECT)

Ok so today starting my Project on “Transfer Learning”. I had just completed my course and it’s time for an end to end project. I am quite anxious as I am new to this field and not have much of knowledge about it.

As I am mentoring by my mentor Andrie on this project. There were so much of technical terms that I don’t know about, these terms just made my heart pump faster.

Let’s talk about some technical stuff, I think my experience bored you a little bit.

Transfer Learning

Transfer learning is a machine learning technique where a model trained on one task is re-used on a second related task.

What is a model?

A model is like an engine of the project, the model is trained to recognize certain types of patterns present in the dataset.

Dataset

Dataset is the images, text, csv or any raw form of data in which we’ll have to find the pattern. A dataset consists of series of input and output for which we’ll train our model and after training our model. The model can predict the output if the input as in correct manner is given to it.

Google Colab

Google colab is a free cloud service and its supports free GPU specially developed for improving skill on deep learning applications using libraries such as Keras, TensorFlow, PyTorch, and OpenCV.

You might be thinking what the hell these things are! It’s Ok we’ll gonna figure it out !

What is GPU?

GPU is designed to quickly render high-resolution images and video. Ok in simple terms GPU is faster than CPU.

And when we need faster computational power we’ll use GPU. In deep learning we’ll have high number of dataset so training of model on these dataset required high computational power.

So alright we’ve gone through the basic terminology of deep leaning. You might be thinking “Why we are doing all of these stuff?”

Ok I am releasing the truth now, First of all the topic of the project is DOG VISION. The purpose of this project is to build a model or engine that if we give the image of a dog to this model. The model can predict the name of the dog.

Isn,t it cool that just by clicking the photo of the dog you can know the dog’s breed.