API do Streamlit

A classe StreamlitRenderer é responsável por renderizar os gráficos do Graphic Walker.

from pygwalker.api.streamlit import StreamlitRenderer
 
renderer = StreamlitRenderer(df, spec="./gw_config.json")

Parâmetros

dataset: DataFrame ou Connector a ser utilizado.
gid: ID para o div do container do GraphicWalker.
field_specs: Especificações dos campos.
theme_key: Tipo de tema para o GraphicWalker.
dark: Configuração do tema.
spec: Dados de configuração do gráfico.
spec_io_mode: Modo de E/S das configurações.
kernel_computation: Se verdadeiro, utiliza cálculo de kernel para os dados.
kanaries_api_key: Chave da API dos kanaries.
default_tab: Aba padrão a ser exibida.

A classe StreamlitRenderer contém os métodos:

StreamlitRenderer.explorer

renderer.explorer()

StreamlitRenderer.chart

renderer.chart(0)

StreamlitRenderer.viewer

renderer.viewer()

Exemplo de uso com kernel_computation ativado (Recomendado)

Neste exemplo, é demonstrado como utilizar o Pygwalker no Streamlit com o cálculo de kernel ativado.

from pygwalker.api.streamlit import StreamlitRenderer
import pandas as pd
import streamlit as st
 
# Adjust the width of the Streamlit page
st.set_page_config(
    page_title="Use Pygwalker In Streamlit",
    layout="wide"
)
 
# Add Title
st.title("Use Pygwalker In Streamlit")
 
# You should cache your pygwalker renderer, if you don't want your memory to explode
@st.cache_resource
def get_pyg_renderer() -> "StreamlitRenderer":
    df = pd.read_csv("data.csv")
    # If you want to use feature of saving chart config, set `spec_io_mode="rw"`
    return StreamlitRenderer(df, spec="./gw_config.json")
 
 
renderer = get_pyg_renderer()
 
st.subheader("Display Explore UI")
 
tab1, tab2, tab3, tab4 = st.tabs(
    ["graphic walker", "data profiling", "graphic renderer", "pure chart"]
)
 
with tab1:
    renderer.explorer()
 
with tab2:
    renderer.explorer(default_tab="data")
 
with tab3:
    renderer.viewer()
 
with tab4:
    st.markdown("### registered per weekday")
    renderer.chart(0)
    st.markdown("### registered per day")
    renderer.chart(1)

Exemplo de uso com kernel_computation desativado

Neste exemplo, é demonstrado como utilizar o Pygwalker no Streamlit com o cálculo de kernel desativado.

import pygwalker as pyg
import pandas as pd
import streamlit.components.v1 as components
import streamlit as st
 
# Adjust the width of the Streamlit page
st.set_page_config(
    page_title="Use Pygwalker In Streamlit",
    layout="wide"
)
 
# Add Title
st.title("Use Pygwalker In Streamlit")
 
# Import your data
df = pd.read_csv("/bike_sharing_dc.csv")
# Paste the copied Pygwalker chart code here
vis_spec = """[{"visId":"gw_rZy5","name":"Chart 1","encodings":{"dimensions":[{"dragId":"gw_BUE2","fid":"ZGF0ZV8x","name":"date","semanticType":"nominal","analyticType":"dimension"},{"dragId":"gw_x1ug","fid":"bW9udGhfMg==","name":"month","semanticType":"ordinal","analyticType":"dimension"},{"dragId":"gw_zRAa","fid":"c2Vhc29uXzM=","name":"season","semanticType":"nominal","analyticType":"dimension"},{"dragId":"gw_ZeVh","fid":"eWVhcl81","name":"year","semanticType":"nominal","analyticType":"dimension"},{"dragId":"gw_JqXv","fid":"aG9saWRheV82","name":"holiday","semanticType":"nominal","analyticType":"dimension"},{"dragId":"gw_OD2F","fid":"d29yayB5ZXMgb3Igbm90XzE0","name":"work yes or not","semanticType":"nominal","analyticType":"dimension"},{"dragId":"gw_KgQu","fid":"YW0gb3IgcG1fMTU=","name":"am or pm","semanticType":"nominal","analyticType":"dimension"},{"dragId":"gw_PqvI","fid":"RGF5IG9mIHRoZSB3ZWVrXzE2","name":"Day of the week","semanticType":"ordinal","analyticType":"dimension"}],"measures":[{"dragId":"gw_7JNg","fid":"aW5kZXhfMA==","name":"index","analyticType":"measure","semanticType":"quantitative","aggName":"sum"},{"dragId":"gw_IYM_","fid":"aG91cl80","name":"hour","analyticType":"measure","semanticType":"quantitative","aggName":"sum"},{"dragId":"gw_ofd8","fid":"dGVtcGVyYXR1cmVfNw==","name":"temperature","analyticType":"measure","semanticType":"quantitative","aggName":"sum"},{"dragId":"gw_sGlX","fid":"ZmVlbGluZ190ZW1wXzg=","name":"feeling_temp","analyticType":"measure","semanticType":"quantitative","aggName":"sum"},{"dragId":"gw_674M","fid":"aHVtaWRpdHlfOQ==","name":"humidity","analyticType":"measure","semanticType":"quantitative","aggName":"sum"},{"dragId":"gw_AxQD","fid":"d2luc3BlZWRfMTA=","name":"winspeed","analyticType":"measure","semanticType":"quantitative","aggName":"sum"},{"dragId":"gw_iy94","fid":"Y2FzdWFsXzEx","name":"casual","analyticType":"measure","semanticType":"quantitative","aggName":"sum"},{"dragId":"gw_9J2u","fid":"cmVnaXN0ZXJlZF8xMg==","name":"registered","analyticType":"measure","semanticType":"quantitative","aggName":"sum"},{"dragId":"gw_WzEF","fid":"Y291bnRfMTM=","name":"count","analyticType":"measure","semanticType":"quantitative","aggName":"sum"},{"dragId":"gw_count_fid","fid":"gw_count_fid","name":"Row count","analyticType":"measure","semanticType":"quantitative","aggName":"sum","computed":true,"expression":{"op":"one","params":[],"as":"gw_count_fid"}}],"rows":[{"dragId":"gw_gJqj","fid":"cmVnaXN0ZXJlZF8xMg==","name":"registered","analyticType":"measure","semanticType":"quantitative","aggName":"sum"}],"columns":[{"dragId":"gw_uZ9C","fid":"RGF5IG9mIHRoZSB3ZWVrXzE2","name":"Day of the week","semanticType":"ordinal","analyticType":"dimension"}],"color":[{"dragId":"gw_04s5","fid":"c2Vhc29uXzM=","name":"season","semanticType":"nominal","analyticType":"dimension"}],"opacity":[],"size":[],"shape":[],"radius":[],"theta":[],"details":[],"filters":[],"text":[]},"config":{"defaultAggregated":true,"geoms":["auto"],"stack":"stack","showActions":false,"interactiveScale":false,"sorted":"none","zeroScale":true,"size":{"mode":"auto","width":320,"height":200},"format":{}}}]"""
 
# Generate the HTML using Pygwalker
pyg_html = pyg.to_html(df, spec=vis_spec)
 
# Embed the HTML into the Streamlit app
components.html(pyg_html, height=1000, scrolling=True)

Perguntas e Respostas Relacionadas

Como construir uma alternativa de código aberto online ao Tableau/PowerBI que possa lidar com grandes quantidades de dados?

Resposta: Você pode usar o pygwalker + Streamlit + Snowflake para construir uma alternativa online de código aberto ao Tableau/PowerBI que possa lidar com grandes quantidades de dados.

Jupyter Walker APIs Html APIs