What Colab Offers You#

As a programmer, you can perform the following using Google Colab:

• Write and execute code in Python
• Document your code with support for mathematical equations
• Create, upload, and share notebooks
• Import and save notebooks from/to Google Drive
• Import and publish notebooks from GitHub
• Import external datasets (e.g., from Kaggle)
• Integrate PyTorch, TensorFlow, Keras, OpenCV
• Access free cloud service with free GPU

Step 1: Access Colab#

Open the following URL in your browser: https://colab.research.google.com

You’ll see a welcome screen with various options for creating or opening notebooks.

Step 2: Create a New Notebook#

Click on the NEW PYTHON 3 NOTEBOOK link at the bottom of the screen. A new notebook will open up with a code window where you can enter your Python code.

Setting the Notebook Name#

By default, the notebook uses the naming convention UntitledXX.ipynb. To rename it, click on the name and type your desired name in the edit box. For example, you might name it “MyFirstColabNotebook.”

Entering and Executing Code#

Let’s try a simple example. Enter the following Python statements in the code window:

import time
print(time.ctime())

To execute the code, click on the arrow (play button) on the left side of the code cell. After a moment, you’ll see the output underneath the code window showing the current date and time.

Adding More Code Cells#

To add more code to your notebook, you can either:

• Select Insert / Code Cell from the menu
• Hover your mouse at the bottom center of a code cell until CODE and TEXT buttons appear, then click CODE

Try adding a new cell with this code:

time.sleep(5)
print(time.ctime())

Running All Cells#

To run all the code in your notebook without interruption, select:

Runtime / Reset and run all…

This will execute all cells sequentially, and you’ll notice the time difference between outputs is exactly 5 seconds.

Managing Cells#

Changing Cell Order: Use the UP CELL or DOWN CELL buttons to move cells up or down in your notebook.

Deleting Cells: Click on the vertical-dotted icon at the top right corner of a code cell and select “Delete cell.”

Markdown Basics#

In a Text cell, you can use Markdown syntax for formatting:

This is **bold**.
This is *italic*.
This is ~strikethrough~.

Mathematical Equations#

Colab supports LaTeX for mathematical representations. For inline equations, use single dollar signs:

$\sqrt{3x-1}+(1+x)^2$

For block equations, use double dollar signs:

$$e^x=\sum_{i=0}^\infty \frac{1}{i!}x^i$$

Here’s an example of more complex mathematical notation:

$$f(x_1, x_2) = 20 + e - 20exp(-0.2 \sqrt{\frac{1}{n}(x_1^2 + x_2^2)}) - exp(\frac{1}{n}(cos(2\pi x_1) + cos(2\pi x_2)))$$

You can even create matrices:

$$A_{m,n} = \begin{pmatrix}
a_{1,1} & a_{1,2} & \cdots & a_{1,n} \\
a_{2,1} & a_{2,2} & \cdots & a_{2,n} \\
\vdots & \vdots & \ddots & \vdots \\
a_{m,1} & a_{m,2} & \cdots & a_{m,n}
\end{pmatrix}$$

Saving to Google Drive#

Select: File / Save a copy in Drive…

This creates a copy of your notebook in your Google Drive, which you can rename later.

Saving to GitHub#

You can also save directly to your GitHub repository:

Select: File / Save a copy in GitHub…

You’ll be prompted to log in to GitHub. If you don’t have a repository, you can create one and save your project there.

Simple Commands#

You can use system commands by prefixing them with an exclamation mark:

message = 'A Great Tutorial on Colab by Tutorialspoint!'
greeting = !echo -e "$message\n$message"
greeting

Getting Remote Data#

You can download files from remote servers:

!wget https://wordpress.org/latest.zip

Cloning Git Repositories#

You can clone entire GitHub repositories into Colab:

!git clone https://github.com/wxs/keras-mnist-tutorial.git

After cloning, you can right-click on any .ipynb file in the repository and select “Open With / Colaboratory” to open it in Colab.

System Aliases#

To get a list of available system commands, execute:

!ls /bin

This will show you all the available system utilities you can use in your notebook.

Mounting Your Drive#

First, mount your Google Drive:

Select: Tools / Command palette, then search for and select “Mount Drive”

This will insert the following code:

from google.colab import drive
drive.mount('/content/drive')

When you run this, you’ll be asked to authenticate and grant permissions. Once completed, your drive will be mounted at /content/drive.

Listing Drive Contents#

You can now list files in your drive:

!ls "/content/drive/My Drive/Colab Notebooks"

Running Python Code#

To execute a Python file from your drive:

!python3 "/content/drive/My Drive/Colab Notebooks/hello.py"

Function Lists#

When you type a module name followed by a dot, Colab can show you available functions. Simply hover over function names to get more information.

Function Documentation#

To see documentation for any function or class, type the function name with an open parenthesis and hover over it:

Tensor = torch.cos()

Hovering over cos will display its documentation, including parameters, return values, and examples.

Line Magics#

Line magics are prepended with a single % character:

%ldir  # Lists directory contents
%history  # Shows command history

Cell Magics#

Cell magics are prepended with %% and affect the entire cell:

%%html
<marquee style='width: 50%; color: Green;'>
  <b>Welcome to Tutorialspoint!</b>
</marquee>

This renders HTML directly in your notebook.

You can also create SVG graphics:

%%html
<svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 600 400" width="400" height="400">
  <rect x="10" y="0" width="300" height="100" style="fill:orange; stroke:black;" />
  <rect x="10" y="100" width="300" height="100" style="fill:white; stroke:black;" />
  <rect x="10" y="200" width="300" height="100" style="fill:green; stroke:black;" />
</svg>

Getting a List of Magics#

To see all available magics:

%lsmagic

This displays both line and cell magics available in your environment.

Creating a Form#

Click on the Options menu (vertical dots) in a code cell and select “Add a form.” This adds a form to your cell with a default title.

Adding Form Fields#

To add input fields:

1. Click the Options menu again
2. Select “Form” to reveal submenus
3. Choose “Add a form field”
4. Configure the field (variable name, type, default value)

Example of an integer input:

#@title Form
sleeptime = 0 #@param {type:"integer"}

Form Field Types#

Text Input:

name = 'Tutorialspoint' #@param {type:"string"}
print(name)

Dropdown List:

color = 'green' #@param ["red", "green", "blue"]
print(color)

Date Input:

#@title Date fields
date_input = '2019-06-03' #@param {type:"date"}
print(date_input)

Using Form Values#

Once you’ve created form fields, you can use the variables in your code:

import time
print(time.ctime())
time.sleep(sleeptime)
print(time.ctime())

Keras#

!pip install -q keras

PyTorch#

!pip3 install torch torchvision

MxNet#

!apt install libnvrtc8.0
!pip install mxnet-cu80

OpenCV#

!apt-get -qq install -y libsm6 libxext6 && pip install -q -U opencv-python

XGBoost#

!pip install -q xgboost==0.4a30

GraphViz#

!apt-get -qq install -y graphviz && pip install -q pydot

Enabling GPU#

Select: Runtime / Change runtime type

In the dialog that appears, set “Hardware accelerator” to “GPU” and save.

Testing for GPU#

To verify GPU is enabled:

import tensorflow as tf
tf.test.gpu_device_name()

If GPU is enabled, this returns: '/device:GPU:0'

Listing Available Devices#

To see all devices used during execution:

from tensorflow.python.client import device_lib
device_lib.list_local_devices()

This shows details about CPUs, GPUs, and other accelerators available to your notebook.

Checking RAM#

To see available memory:

!cat /proc/meminfo

This displays detailed information about your runtime’s memory resources.