|
|
|
import glob, os, sys; |
|
sys.path.append('../utils') |
|
|
|
|
|
import seaborn as sns |
|
import matplotlib.pyplot as plt |
|
import numpy as np |
|
import pandas as pd |
|
import streamlit as st |
|
from st_aggrid import AgGrid |
|
from utils.tapp_classifier import load_tappClassifier, tapp_classification |
|
import logging |
|
logger = logging.getLogger(__name__) |
|
from utils.config import get_classifier_params |
|
from io import BytesIO |
|
import xlsxwriter |
|
import plotly.express as px |
|
from pandas.api.types import ( |
|
is_categorical_dtype, |
|
is_datetime64_any_dtype, |
|
is_numeric_dtype, |
|
is_object_dtype, |
|
is_list_like) |
|
|
|
|
|
tapp_classifier_identifier = 'tapp' |
|
param1 = get_classifier_params(tapp_classifier_identifier) |
|
|
|
def app(): |
|
|
|
with st.container(): |
|
if 'key0' in st.session_state: |
|
df = st.session_state.key0 |
|
|
|
|
|
classifier = load_tappClassifier(classifier_name=param1['model_name']) |
|
st.session_state['{}_classifier'.format(tapp_classifier_identifier)] = classifier |
|
|
|
if len(df) > 100: |
|
warning_msg = ": This might take sometime, please sit back and relax." |
|
else: |
|
warning_msg = "" |
|
|
|
df = tapp_classification(haystack_doc=df, |
|
threshold= param1['threshold']) |
|
|
|
st.session_state.key1 = df |