app.py

import gradio as gr
import pandas as pd
import scipy
from utils import *

def spm_fn(SPM_path_to_csv, SPM_identifier_column, SPM_sequence_column, SPM_sortby, SPM_sliding_window_min, SPM_sliding_window_max, SPM_min_gap, SPM_max_gap, SPM_S_support_thresh, SPM_I_support_thresh, SPM_dataset_format):
    spm_params = {
        "path_to_csv": SPM_path_to_csv.name,
        "identifier_column": SPM_identifier_column,
        "sequence_column": SPM_sequence_column,
        "sortby": SPM_sortby,
        "sliding_window_min": SPM_sliding_window_min,
        "sliding_window_max": SPM_sliding_window_max,
        "min_gap": SPM_min_gap,
        "max_gap": SPM_max_gap,
        "S_support_thresh": SPM_S_support_thresh,
        "I_support_thresh": SPM_I_support_thresh,
        "dataset_format": SPM_dataset_format
    }
    
    spm_result, occurrence_matrix = SPM(spm_params)
    spm_result=spm_result.reset_index()
    spm_result.rename(columns={'index': 'Pattern'}, inplace=True)
    spm_result.to_csv("spm_result.csv")
    return "spm_result.csv",spm_result

def dsm_fn(DSM_path_to_csv_left, DSM_path_to_csv_right, DSM_identifier_column, DSM_sequence_column, DSM_sortby, DSM_sliding_window_min, DSM_sliding_window_max, DSM_min_gap, DSM_max_gap, DSM_S_support_thresh, DSM_I_support_thresh, DSM_threshold_pvalue, DSM_dataset_format, DSM_test_type):
    dsm_params = {
        "path_to_csv_left": DSM_path_to_csv_left.name,
        "path_to_csv_right": DSM_path_to_csv_right.name,
        "identifier_column": DSM_identifier_column,
        "sequence_column": DSM_sequence_column,
        "sortby": DSM_sortby,
        "sliding_window_min": DSM_sliding_window_min,
        "sliding_window_max": DSM_sliding_window_max,
        "min_gap": DSM_min_gap,
        "max_gap": DSM_max_gap,
        "S_support_thresh": DSM_S_support_thresh,
        "I_support_thresh": DSM_I_support_thresh,
        "threshold_pvalue": DSM_threshold_pvalue,
        "dataset_format": DSM_dataset_format,
        "test_type": DSM_test_type
    }
    
    ptrn_left, ptrn_right, ptrn_both_left, ptrn_both_right, dsm_result = DSM(dsm_params)
    dsm_result.to_csv("dsm_result.csv")
    return "dsm_result.csv",dsm_result

theme = gr.themes.Base(
    primary_hue="gray",
    secondary_hue="orange",
    neutral_hue="neutral",
    font=[gr.themes.GoogleFont('Source Sans Pro'), 'ui-sans-serif', gr.themes.GoogleFont('system-ui'), 'sans-serif'],
).set(
    body_background_fill='*primary_200',
    body_background_fill_dark='*primary_950',
    body_text_color_dark='*primary_100',
    body_text_weight='500',
    embed_radius='*radius_xxl',
    background_fill_primary='*primary_50',
    button_shadow='*input_shadow',
    background_fill_primary_dark='*secondary_800',
    background_fill_secondary_dark='*primary_200',
    border_color_accent='*secondary_800',
    border_color_accent_dark='*secondary_950',
    border_color_primary='*primary_950',
    color_accent_soft_dark='*neutral_500',
    link_text_color='*neutral_900',
    link_text_color_dark='*primary_200',
    code_background_fill='*secondary_200',
    block_background_fill='*body_background_fill',
    button_border_width='*block_label_border_width',
    button_primary_text_color_dark='*primary_300',
    button_secondary_text_color_dark='*table_odd_background_fill',
    button_secondary_text_color_hover='*button_secondary_background_fill'
)



def demo():
    with gr.Blocks(theme=theme) as demo:
        
        gr.Markdown(
            """<center><h2>SPM and DSM Implementation in Python</center></h2>
            <h3>Sequential Pattern Mining (SPM) and Differential Sequential Mining (DSM)</h3>
            """
        )
        
        with gr.Tab("SPM"):
            with gr.Row():
                spm_file = gr.File(label="Upload CSV file")
            with gr.Accordion("Advanced options", open=False):
                with gr.Row():
                    identifier_column = gr.Textbox(label="Identifier Column", value="Identifier")
                with gr.Row():
                    sequence_column = gr.Textbox(label="Sequence Column", value="Sequence")
                with gr.Row():
                    sortby = gr.Dropdown(label="Sort By", choices=["S-Support", "I-Support"], value="S-Support")
                with gr.Row():
                    sliding_window_min = gr.Number(label="Sliding Window Min", value=1)
                with gr.Row():
                    sliding_window_max = gr.Number(label="Sliding Window Max", value=4)
                with gr.Row():
                    min_gap = gr.Number(label="Min Gap", value=1)
                with gr.Row():
                    max_gap = gr.Number(label="Max Gap", value=12)
                with gr.Row():
                    S_support_thresh = gr.Number(label="S Support Threshold", value=0.4)
                with gr.Row():
                    I_support_thresh = gr.Number(label="I Support Threshold", value=0)
                with gr.Row():
                    dataset_format = gr.Number(label="Dataset Format", value=0)
            with gr.Row():
                spm_result=gr.File(label="SPM result")
            with gr.Row():
                spm_dataframe=gr.Dataframe()
            spm_submit = gr.Button("Generate SPM Result...")
                
        
        with gr.Tab("DSM"):
            with gr.Row():
                dsm_left_file = gr.File(label="Upload Left Dataset CSV file")
                dsm_right_file = gr.File(label="Upload Right Dataset CSV file")
            with gr.Accordion("Advanced options", open=False):
                with gr.Row():
                    identifier_column_dsm = gr.Textbox(label="Identifier Column", value="Identifier")
                with gr.Row():
                    sequence_column_dsm = gr.Textbox(label="Sequence Column", value="Sequence")
                with gr.Row():
                    sortby_dsm = gr.Dropdown(label="Sort By", choices=["S-Support", "I-Support"], value="S-Support")
                with gr.Row():
                    sliding_window_min_dsm = gr.Number(label="Sliding Window Min", value=1)
                with gr.Row():
                    sliding_window_max_dsm = gr.Number(label="Sliding Window Max", value=1)
                with gr.Row():
                    min_gap_dsm = gr.Number(label="Min Gap", value=1)
                with gr.Row():
                    max_gap_dsm = gr.Number(label="Max Gap", value=12)
                with gr.Row():
                    S_support_thresh_dsm = gr.Number(label="S Support Threshold", value=0.4)
                with gr.Row():
                    I_support_thresh_dsm = gr.Number(label="I Support Threshold", value=0)
                with gr.Row():
                    threshold_pvalue = gr.Number(label="Threshold P-value", value=0.1)
                with gr.Row():
                    dataset_format_dsm = gr.Number(label="Dataset Format", value=0)
                with gr.Row():
                    test_type = gr.Dropdown(label="Test Type", choices=["poisson_means_test", "ttest_ind", "mannwhitneyu", "bws_test", "ranksums", "brunnermunzel", "mood", "ansari", "cramervonmises_2samp", "epps_singleton_2samp", "ks_2samp", "kstest"], value="ttest_ind")
            with gr.Row():
                dsm_result=gr.File(label="DSM result")
            with gr.Row():
                dsm_dataframe=gr.Dataframe()
            dsm_submit = gr.Button("Generate DSM Result...")
        spm_submit.click(spm_fn, \
                inputs=[spm_file, identifier_column, sequence_column, sortby, sliding_window_min, sliding_window_max, min_gap, max_gap, S_support_thresh, I_support_thresh, dataset_format], \
                outputs=[spm_result,spm_dataframe])
        dsm_submit.click(dsm_fn, \
                inputs=[dsm_left_file, dsm_right_file, identifier_column_dsm, sequence_column_dsm, sortby_dsm, sliding_window_min_dsm, sliding_window_max_dsm, min_gap_dsm, max_gap_dsm, S_support_thresh_dsm, I_support_thresh_dsm, threshold_pvalue, dataset_format_dsm, test_type], \
                outputs=[dsm_result,dsm_dataframe])
        demo.queue().launch(debug=True,share=True)
#gr.themes.builder()
if __name__ == "__main__":
    demo()