What is Google Colab ?
The main question that most of you probably have, would be…… What is Google Colab ??
Google Colab is actually the short form for Google Colaboratory. To begin with what is Google Colab; …. Its simply a Python execution environment. It can be used to execute any sort of Python code online. This execution environment for Python is similar to the interface that we get in Jupyter Notebook.
For those who don’t know about Jupyter Notebook, it is a notebook as the name suggests but for running Python codes. In a Jupyter Notebook you can execute Python codes in different cells where each cell can be linked to the others. This is very useful when we want to try different codes and repeatedly execute various different codes, thus in this case, instead of erasing the code from the same file again and again , we can execute these codes in different cells in the Jupyter Notebook. These notebooks are one of the many features that are provided by Anaconda Navigator.
Why use Colab ?
Many of you must be wondering; “If we can use Jupyter then why even use Google Colab” , I had a similar question in the beginning, but there are various reasons for switching to Colab.
Web Based:
Google Colab is actually web based that is, you don’t need to download any extra softwares or set any environment paths in order to get started. All you need is a proper web browser and an internet connection.
How it changed my coding practices ?
To give a short overview, the Machine Learning and Deep Learning community actually develops many different models in order to solve various problems. These models are essential to make the computers able to function without any explicit progamming. But the main issue with these models is, the time that they take to train on huge datasets.
In case of training these ML models, GPUs (Graphics Processing Unit) are always better and faster than CPUs. But the main problem is, GPUs are kind of costly and also setting them up for model training in a proper Jupyter Environment is not that simple. Here is the point which I like the mos about colab:
Google Colab lets you use GPUs for training and that too free of cost. This usage of GPUs is really the best feature according to me, that is provided by Colab. Also the key features of Google Colab include:
- Free GPU usage (Use runtime option in Colab)
- Browser based ( Hence you can access Colab anywhere on any computer)
- Preinstalled libraries: Colab provides you all the common libraries that are used to execute various tasks; thus there is no need to install them separately.
- User friendly interface: Colab has an absolute great interface, you can really trust me on this.
5. Storage: Google Colab lets you store all the files that you create, directly to your own Google Drive. Isn’t this awesome !!! So basically you can access your work anywhere by just logging into your Google Drive !!
“I strongly suggest you to try Google Colab though you can actually be fine with the normal Python compilers for general programming. But for all those who are currently studying Neural Networks, Machine Learning and who are into creation of any models using TensorFlow and Keras are really the most benefited by using Google Colab. Happy Programming”
