Streamlit Altair Bar chart Y axismax min

Question

According to this answer, I'm suppose to use scale = alt.Scae(domain=[0,1.4]) to scale the axis to my preference. But when I do it, the entire bar chart moves lower, hiding the entire x-axis. Why is that?

This is when I'm doing:

    chart_df = alt.Chart(agg_df).mark_bar().encode(
        y=alt.Y('rsi_value'
                ),
        x=alt.X('idx', sort='y'),
        color='type'
    )
    st.altair_chart(chart_df)

When I am doing this, I am able to control the max value to 60.

    chart_df = alt.Chart(agg_df).mark_bar().encode(
        y=alt.Y('rsi_value', scale=alt.Scale(domain=[0, 60])
                ),
        x=alt.X('ticker_idx', sort='y'),
        color='type'
    )
    st.altair_chart(chart_df)

But when I do this:

    chart_df = alt.Chart(agg_df).mark_bar().encode(
        y=alt.Y('rsi_value', scale=alt.Scale(domain=[20, 60])
                ),
        x=alt.X('ticker_idx', sort='y'),
        color='type'
    )

The entire X-axis disappears.

I am just trying to make the bar heights more differentiable...any ideas?

Answer 1

You can use clamp=True in alt.Scale:

chart_df = alt.Chart(agg_df).mark_bar().encode(
    y=alt.Y('rsi_value', scale=alt.Scale(domain=[20, 60], clamp=True)),
    x=alt.X('ticker_idx', sort='y'),
    color='type'
)

---

A minimal working example to show the difference:

import altair as alt
import pandas as pd
import streamlit as st

source = pd.DataFrame({
    'a': ['A', 'B', 'C', 'D', 'E', 'F', 'G', 'H', 'I'],
    'b': [28, 55, 43, 91, 81, 53, 19, 87, 52]
})

cols = st.columns([1, 1])

with cols[0]:
    st.header("With clamp=False")
    chart_df_0 = alt.Chart(source).mark_bar().encode(
        y=alt.Y('b', scale=alt.Scale(domain=[20, 60], clamp=False)),
        x=alt.X('a', sort='y'),
        
    )

    st.altair_chart(chart_df_0)


with cols[1]:
    st.header("With clamp=True")
    chart_df_1 = alt.Chart(source).mark_bar().encode(
        y=alt.Y('b', scale=alt.Scale(domain=[20, 60], clamp=True)),
        x=alt.X('a', sort='y'),
        
    )

    st.altair_chart(chart_df_1)