def main():
# track web app activity
streamlit_analytics.start_tracking()
# Web Page Headers
st.write("""
# Wine Quality Prediction Web-App
This app predicts the ** Quality of your Red Wine ** using **wine features** input via the **side panel**
""")
#read in wine image and render with streamlit
image = Image.open('resources/wine_image.jpg')
st.image(image, caption='The Wine Grader Company', use_column_width=True)
st.sidebar.header(
'User Input Parameters'
) #user input parameter collection with streamlit side bar
# Load ML models
model_knn = joblib.load(open("model/autoML-model_knn.joblib", "rb"))
model_decis_tree = joblib.load(
open("model/autoML-model_Decision-Tree.joblib", "rb"))
model_log_reg = joblib.load(
open("model/autoML-model_Log-Regress.joblib", "rb"))
# user input variables
user_input_df = get_user_input()
processed_user_input = data_preprocessor(user_input_df)
st.subheader('User Input parameters')
st.write(user_input_df)
# machine learning techniques results variables
prediction_knn = model_knn.predict(processed_user_input)
prediction__decis_tree = model_decis_tree.predict(processed_user_input)
prediction_log_reg = model_log_reg.predict(processed_user_input)
# output machine learning results
st.write("Ask our A.I. wine connaisseur for their opinion")
connaisseur_choice = st.selectbox("Connaisseur opinion: ", [
'', 'The decision tree connaisseur', 'The KNN connaisseur',
'The logistic regression connaisseur'
])
# A.I opinion displayed
if (connaisseur_choice == "The decision tree connaisseur"):
st.write("The decision tree connaisseur verdict: ")
if (prediction__decis_tree == 1):
st.write(
"Good quality wine! This wine is at least a 7 in my expert opinion. You got great test!"
)
choose_image = randrange(1)
if (choose_image == 1):
image = Image.open('resources/wineconnaisseur_1.png')
st.image(image, use_column_width=True)
else:
image = Image.open('resources/wineconnaisseur_3.png')
st.image(image, use_column_width=True)
else:
st.write(
"bad quality wine! Do yourself a favor and throw away this thing you call wine...."
)
image = Image.open('resources/wineconnaisseur_2.png')
st.image(image, use_column_width=True)
if (connaisseur_choice == "The KNN connaisseur"):
st.write("The KNN connaisseur verdict: ")
if (prediction_knn == 1):
st.write(
"Good quality wine! This wine is at least a 7 in my expert opinion. You got great test!"
)
choose_image = randrange(1)
if (choose_image == 0):
image = Image.open('resources/wineconnaisseur_1.png')
st.image(image, use_column_width=True)
else:
image = Image.open('resources/wineconnaisseur_3.png')
st.image(image, use_column_width=True)
else:
st.write(
"bad quality wine! Do yourself a favor and throw away this thing you call wine...."
)
image = Image.open('resources/wineconnaisseur_2.png')
st.image(image, use_column_width=True)
if (connaisseur_choice == "The logistic regression connaisseur"):
st.write("The logistic regression connaisseur verdict: ")
if (prediction_log_reg == 1):
st.write(
"Good quality wine! This wine is at least a 7 in my expert opinion. You got great test!"
)
choose_image = randrange(1)
if (choose_image == 1):
image = Image.open('resources/wineconnaisseur_1.png')
st.image(image, use_column_width=True)
else:
image = Image.open('resources/wineconnaisseur_3.png')
st.image(image, use_column_width=True)
else:
st.write(
"bad quality wine! Do yourself a favor and throw away this thing you call wine...."
)
image = Image.open('resources/wineconnaisseur_2.png')
st.image(image, use_column_width=True)
# end tracking code block
streamlit_analytics.stop_tracking()
# Filtro genero
if genero == 'Mulher':
filtro = df['genero']=="Feminino"
df = df[filtro]
elif genero == 'Homem':
filtro = df['genero']=="Masculino"
df = df[filtro]
# Nivel_materia
filtro = df[dict_niveis[nivel]+"_"+dict_materias[materia]]==1
df = df[filtro]
return df
streamlit_analytics.start_tracking()
niveis = ['','Ensino fundamental','Ensino Médio e Pré-vestibular','Concurso']
materias = ['','Matemática','Física','Química','Inglês','Redação']
dict_materias = {
'Matemática':'mat',
'Física':'fis',
'Química':'quim',
'Inglês':'ing',
'Redação':'red'
}
dict_niveis = {
'Ensino fundamental':'ef',
'Ensino Médio e Pré-vestibular':'em',