Unsorted animation frame slider

📊 Plotly Python
1

Hello,

I’m trying to make a 3D scatter plot with plotly which has an animation frame for column “predicted-time-point”, and shows data for “old-time-point” in each frame by their x,y,z values. However, the animation frame slider is not in the numerical order. It goes like 18, 15, 8, 4 etc. When

    fig = px.scatter_3d(df_high_res, x='x', y='y', z='z', color='old-time-point', 
                        animation_frame='predicted-time-point',
                        # color_discrete_map=color_map,
                        # color_discrete_sequence=custom_color_scale,  # Use custom color scale
                        title='3D Scatter Plot with Animation Frames',
                        labels={'old-time-point': 'Old Time Point', 'predicted-time-point': 'Predicted Time Point',
                                'new-time-point': 'New Time Point', 'x': 'X', 'y': 'Y', 'z': 'Z'},
                        hover_data=['old-time-point', 'predicted-time-point', 'new-time-point', 'x', 'y', 'z'],
                        )

    fig.update_layout(scene=dict(aspectmode='data'))
    fig.update_traces(marker_size = 4)

    fig.update_layout(
        scene=dict(
            xaxis=dict(title='X'),
            yaxis=dict(title='Y'),
            zaxis=dict(title='Z'),
        ),
        title=f'3D Scatter Plot for Cluster {cl_info[-1]}',
        width=900,  
        height=900, 
        showlegend=True, 
        legend=dict(title='Time-Point') 
    )    

    fig.show()

I sort the dataframe before plotting with df_high_res = df_high_res.sort_values(by=['predicted-time-point', 'old-time-point']), then I loose some data points in the figures. Instead of three old-time-point` categories, I only see two. A simple print(df_high_res[‘old-time-point’].unique()) shows that the category is still there, but it doesn’t appear in the plots. How can I solve it and have them plotted with discrete colors scheme?

2

Hi, check the dtype of your dataframe.

Can you reproduce this with a dummy dataframe?