app.py

import gradio as gr
from pyvis.network import Network

import numpy as np
import pandas as pd
import os
from datasets import load_dataset
from datasets import Features
from datasets import Value
from datasets import Dataset
import matplotlib.pyplot as plt



Secret_token = os.getenv('HF_Token')

dataset = load_dataset('FDSRashid/hadith_info',data_files = 'Basic_Edge_Information.csv', token = Secret_token, split = 'train')
dataset2 = load_dataset('FDSRashid/hadith_info',data_files = 'Taraf_Info.csv', token = Secret_token, split = 'train')



features = Features({'Rawi ID': Value('int32'), 'Famous Name': Value('string'), 'Narrator Rank': Value('string'), 'Number of Narrations': Value('string'), 'Official Name':Value('string'), 'Title Name':Value('string'), 'Generation': Value('string')} )
narrator_bios = load_dataset("FDSRashid/hadith_info", data_files = 'Teacher_Bios.csv', token = Secret_token,features=features )
narrator_bios = narrator_bios['train'].to_pandas()
narrator_bios.loc[49845, 'Narrator Rank'] = 'رسول الله'
narrator_bios.loc[49845, 'Number of Narrations'] = 0
narrator_bios['Number of Narrations'] = narrator_bios['Number of Narrations'].astype(int)
narrator_bios.loc[49845, 'Number of Narrations'] = 327512 
narrator_bios['Generation'] = narrator_bios['Generation'].replace([None], [-1])
narrator_bios['Generation'] = narrator_bios['Generation'].astype(int)



edge_info = dataset.to_pandas()
taraf_info = dataset2.to_pandas()
min_year = int(taraf_info['Year'].min())
max_year = int(taraf_info['Year'].max())
cmap = plt.colormaps['cool']
def value_to_hex(value):
    rgba_color = cmap(value)
    return "#{:02X}{:02X}{:02X}".format(int(rgba_color[0] * 255), int(rgba_color[1] * 255), int(rgba_color[2] * 255))

def subsetEdges(fstyear, lstyear):
    info = taraf_info[(taraf_info['Year'] >= fstyear)& (taraf_info['Year'] <= lstyear)]
    narrators = edge_info[edge_info['Edge_ID'].isin(info['ID'].unique())]
    return narrators
def splitIsnad(dataframe):
    teacher_student =dataframe['Edge_Name'].str.split(' TO ')
    dataframe['Teacher'] = teacher_student.apply(lambda x: x[0])
    dataframe['Student'] = teacher_student.apply(lambda x: x[1])
    return dataframe


def network_narrator(narrator_id, fst_year, lst_year, yaxis):
    edges = subsetEdges(fst_year, lst_year)
    edges_single = edges[(edges['Teacher_ID']==narrator_id) | (edges['Student_ID']==narrator_id)]
    edges_prepped = splitIsnad(edges_single)
    net = Network(directed =True)
    for _, row in edges_prepped.iterrows():
        source = row['Teacher']
        target = row['Student']
        attribute_value = row[yaxis]
        edge_color = value_to_hex(attribute_value)
        teacher_info = narrator_bios[narrator_bios['Rawi ID'] == row['Teacher_ID']]
        student_info = narrator_bios[narrator_bios['Rawi ID'] == row['Student_ID']]
        teacher_narrations = teacher_info['Number of Narrations'].to_list()[0]
        student_narrations = student_info['Number of Narrations'].to_list()[0]
        net.add_node(source, color=value_to_hex(teacher_narrations), font = {'size':30, 'color': 'orange'}, label = f"{source}\n{teacher_narrations}")
        net.add_node(target, color=value_to_hex(student_narrations), font = {'size': 20, 'color': 'red'}, label = f"{target}\n{student_narrations}")
        net.add_edge(source, target, color=edge_color, value=attribute_value, label = f"{yaxis}:{attribute_value}")


    net.barnes_hut(gravity=-3000, central_gravity=0.3, spring_length=200)
    html = net.generate_html()
    html = html.replace("'", "\"")

    return f"""<iframe style="width: 100%; height: 600px;margin:0 auto" name="result" allow="midi; geolocation; microphone; camera; 
    display-capture; encrypted-media;" sandbox="allow-modals allow-forms 
    allow-scripts allow-same-origin allow-popups 
    allow-top-navigation-by-user-activation allow-downloads" allowfullscreen="" 
    allowpaymentrequest="" frameborder="0" srcdoc='{html}'></iframe>""", edges_prepped[['Teacher', 'Student', 'Tarafs', 'Hadiths', 'Isnads', 'Books']]

def narrator_retriever(name):
        return narrator_bios[(narrator_bios['Official Name'].str.contains(name)) | (narrator_bios['Famous Name'].str.contains(name)) | (narrator_bios['Rawi ID'].astype(str) == name)][['Rawi ID', 'Title Name', 'Official Name', 'Famous Name', 'Number of Narrations', 'Narrator Rank', 'Generation' ]]
    


with gr.Blocks() as demo:
    gr.Markdown("Search Narrators using this tool or Visualize Network of a Narrator")
    with gr.Tab("Search Narrator"):
        text_input = gr.Textbox()
        text_output = gr.DataFrame()
        text_button = gr.Button("Search")
        text_button.click(narrator_retriever, inputs=text_input, outputs=text_output)
    with gr.Tab("Visualize Network"):
        with gr.Row():
            image_input = gr.Number()
            FirstYear = gr.Slider(min_year, max_year, value = -11, label = 'Begining', info = 'Choose the first year to display Narrators')
            Last_Year = gr.Slider(min_year, max_year, value = 9, label = 'End', info = 'Choose the  last year to display Narrators')
            Yaxis = gr.Dropdown(choices = ['Tarafs', 'Hadiths', 'Isnads', 'Books'], value = 'Tarafs', label = 'Variable to Display', info = 'Choose the variable to visualize.')
        image_output = gr.HTML()
        image_button = gr.Button("Visualize!")
        image_button.click(network_narrator, inputs=[image_input, FirstYear, Last_Year, Yaxis], outputs=[image_output, gr.DataFrame()])

    
    

demo.launch()