import pandas as pd
from datasets import load_dataset
import streamlit as st

from clarin_datasets.dataset_to_show import DatasetToShow


class AspectEmoDataset(DatasetToShow):
    def __init__(self):
        self.dataset_name = "clarin-pl/aspectemo"
        self.subsets = ["train", "test"]
        self.description = """
        Description AspectEmo Corpus is an extended version of a publicly available PolEmo 2.0 
        corpus of Polish customer reviews used in many projects on the use of different methods in sentiment 
        analysis. The AspectEmo corpus consists of four subcorpora, each containing online customer reviews from the 
        following domains: school, medicine, hotels, and products. All documents are annotated at the aspect level 
        with six sentiment categories: strong negative (minus_m), weak negative (minus_s), neutral (zero), 
        weak positive (plus_s), strong positive (plus_m).
        
        Tasks (input, output and metrics)
        
        Aspect-based sentiment analysis (ABSA) is a text analysis method that 
        categorizes data by aspects and identifies the sentiment assigned to each aspect. It is the sequence tagging 
        task.
        
        Input ('tokens' column): sequence of tokens
        
        Output ('labels' column): sequence of predicted tokens’ classes ("O" + 6 possible classes: strong negative (
        a_minus_m), weak negative (a_minus_s), neutral (a_zero), weak positive (a_plus_s), strong positive (
        a_plus_m), ambiguous (a_amb) )
        
        Domain: school, medicine, hotels and products
        
        Measurements:
        
        Example: ['Dużo', 'wymaga', ',', 'ale', 'bardzo', 'uczciwy', 'i', 'przyjazny', 'studentom', '.', 'Warto', 'chodzić', 
        'na', 'konsultacje', '.', 'Docenia', 'postępy', 'i', 'zaangażowanie', '.', 'Polecam', '.'] → ['O', 'a_plus_s', 'O', 
        'O', 'O', 'a_plus_m', 'O', 'O', 'O', 'O', 'O', 'O', 'O', 'a_zero', 'O', 'a_plus_m', 'O', 'O', 'O', 'O', 'O', 'O'] 
        """

    def load_data(self):
        raw_dataset = load_dataset(self.dataset_name)
        self.data_dict = {
            subset: raw_dataset[subset].to_pandas() for subset in self.subsets
        }

    def show_dataset(self):
        header = st.container()
        description = st.container()
        dataframe_head = st.container()

        with header:
            st.title(self.dataset_name)

        with description:
            st.header("Dataset description")
            st.write(self.description)

        full_dataframe = pd.concat(self.data_dict.values(), axis="rows")
        with dataframe_head:
            df_to_show = full_dataframe.head(10)
            st.header("First 10 observations of the dataset")
            st.dataframe(df_to_show)
            st.text_area(label="Latex code", value=df_to_show.style.to_latex())