The Complete Application

We have made some great progress with our pictionary application. We can now provide the player with a canvas to draw and show the model’s feedback on the drawing. However, most of this is possible due to the amazing widgets provided by Educative. What if we wanted to create our own application instead, like the one shown in the demo?

Our current application lacks a few elements:

• We need a frontend where the player is given a description of the game, a canvas to draw on, and feedback from Gemini.

• Since the goal is to make Gemini guess the drawing, we would need to track the time it takes the player to draw well enough for Gemini to guess. Tracking the time taken can be used as a performance measure. The lower, the better.

• The word given to the player to draw must be random. This will keep the game interesting.

Hello Streamlit

Let’s focus on the frontend, as that will be the base for our application. The AI pictionary demo application shown earlier in the course is created using Streamlit. Streamlit is an open-source Python framework to create visual and interactive applications. It is very beginner-friendly and easy to set up. Streamlit can be installed using pip with the following command: