Skip to content

You can download the code of this step here or all the steps here.

For Notebooks

The "Getting Started" Notebook is available here.

Step 2: Interactive GUI

Now, the page has several visual elements:

  • A slider that is connected to the Python variable n_week ;

  • A chart and a table controls that represent the DataFrame content.

Taipy GUI manages everything. To go further into Taipy GUI, let's consider the concept of state.

Multi-client - state

Try to open a few clients with the same URL. You will see that every client is independent from each other; you can change n_week on a client, and n_week will not change in other clients. This is due to the concept of state.

The state holds the value of all the variables that are used in the user interface, for one specific connection.

For example, at the beginning, state.n_week = 10. When n_week is modified by the slider (through a given graphical client), this is, in fact, state.n_week that is modified, not n_week (the global Python variable). Therefore, if you open 2 different clients, n_week will have 2 state values (state.n_week), one for each client.

In the code below, this concept will be used to connect a variable (n_week) to other variables:

  • We will create a chart that will only display one week of data corresponding to the selected week of the slider.

  • A connection has to be made between the slider's value (state.n_week) and the chart data (state.dataset_week).

How to connect two variables - the on_change function

In Taipy, the on_change() function is a "special" function. Taipy will check if you created a function with this name and will use it. Whenever the state of a variable is modified, the callback function is called with three parameters:

  • state (the state object containing all the variables)

  • The name of the modified variable

  • Its value.

Here, on_change() will be called whenever the slider's value (state.n_week) changes. Each time this happens, state.dataset_week will be updated according to the new value of the selected week. Then, Taipy will propagate this change automatically to the associated chart.

# Select the week based on the slider value
dataset_week = dataset[dataset["Date"].dt.isocalendar().week == n_week]

page = """
# Getting started with Taipy

Select week: *<|{n_week}|>*



# on_change is the function that is called when any variable is changed
def on_change(state, var_name: str, var_value):
    if var_name == "n_week":
        # Update the dataset when the slider is moved
        state.dataset_week = dataset[dataset["Date"].dt.isocalendar().week == var_value]


Interactive GUI