File size: 5,782 Bytes
1cc2077
 
 
 
 
 
 
 
25f445b
1cc2077
 
a2e6203
1cc2077
 
 
 
 
 
 
 
 
 
53feeb3
1cc2077
 
 
 
 
 
 
 
53feeb3
a2e6203
 
1cc2077
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
c806fef
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1cc2077
 
9063698
53feeb3
1cc2077
 
 
 
 
 
 
 
 
 
c806fef
 
 
 
 
1cc2077
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
__all__ = ['block', 'make_clickable_model', 'make_clickable_user', 'get_submissions']

import gradio as gr
import pandas as pd
import re
import pandas as pd
import os
import json
import yaml

from src.about import *
from src.bin.PROBE import run_probe

global data_component, filter_component


def get_baseline_df():
    df = pd.read_csv(CSV_RESULT_PATH)
    present_columns = ["Method"] + checkbox_group.value
    df = df[present_columns]
    return df


def add_new_eval(
    human_file,
    skempi_file,
    model_name_textbox: str,
    revision_name_textbox: str,
    benchmark_type: str,
):
    representation_name = model_name_textbox if revision_name_textbox == '' else revision_name_textbox

    run_probe(benchmark_type, representation_name, human_file, skempi_file)
        

    return None

block = gr.Blocks()

with block:
    gr.Markdown(
        LEADERBOARD_INTRODUCTION
    )
    with gr.Tabs(elem_classes="tab-buttons") as tabs:
        # table jmmmu bench
        with gr.TabItem("🏅 PROBE Benchmark", elem_id="probe-benchmark-tab-table", id=1):
            # selection for column part:
            checkbox_group = gr.CheckboxGroup(
                choices=TASK_INFO,
                label="Benchmark Type",
                interactive=True,
            ) # user can select the evaluation dimension

            baseline_value = get_baseline_df()
            baseline_header = ["Method"] + checkbox_group.value
            baseline_datatype = ['markdown'] + ['number'] * len(checkbox_group.value)

            data_component = gr.components.Dataframe(
                value=baseline_value,
                headers=baseline_header,
                type="pandas",
                datatype=baseline_datatype,
                interactive=False,
                visible=True,
                )

        # table 5
        with gr.TabItem("📝 About", elem_id="probe-benchmark-tab-table", id=2):
            with gr.Row():
                gr.Markdown(LLM_BENCHMARKS_TEXT, elem_classes="markdown-text")

        with gr.TabItem("🚀 Submit here! ", elem_id="probe-benchmark-tab-table", id=3):
            with gr.Row():
                gr.Markdown(EVALUATION_QUEUE_TEXT, elem_classes="markdown-text")

            with gr.Row():
                gr.Markdown("# ✉️✨ Submit your model's representation files here!", elem_classes="markdown-text")

            with gr.Row():
                with gr.Column():
                    model_name_textbox = gr.Textbox(
                        label="Model name",
                        )
                    revision_name_textbox = gr.Textbox(
                        label="Revision Model Name",
                    )
                    # Selection for benchmark type from (similartiy, family, function, affinity) to eval the representations (chekbox)
                    benchmark_type = gr.CheckboxGroup(
                        choices=TASK_INFO,
                        label="Benchmark Type",
                        interactive=True,
                    )
                    similarity_tasks = gr.CheckboxGroup(
                        choices=similarity_tasks_options,
                        label="Select Similarity Tasks",
                        interactive=True,
                    )
                
                    # Dropdown for function prediction aspect
                    function_prediction_aspect = gr.Radio(
                        choices=function_prediction_aspect_options,
                        label="Select Function Prediction Aspect",
                        interactive=True,
                    )
                
                    # Dropdown for function prediction dataset
                    function_prediction_dataset = gr.Radio(
                        choices=function_prediction_dataset_options,
                        label="Select Function Prediction Dataset",
                        interactive=True,
                    )
                
                    # Checkbox for family prediction dataset
                    family_prediction_dataset = gr.CheckboxGroup(
                        choices=family_prediction_dataset_options,
                        label="Select Family Prediction Dataset",
                        interactive=True,
                    )

            with gr.Column():
                human_file = gr.components.File(label="Click to Upload the representation file (csv) for Human dataset", file_count="single", type='filepath')
                skempi_file = gr.components.File(label="Click to Upload the representation file (csv) for SKEMPI dataset", file_count="single", type='filepath')
    
                submit_button = gr.Button("Submit Eval")
                submission_result = gr.Markdown()
                submit_button.click(
                    add_new_eval,
                    inputs = [
                        human_file,
                        skempi_file,
                        model_name_textbox,
                        revision_name_textbox,
                        benchmark_type,
                        similarity_tasks,
                        function_prediction_aspect,
                        function_prediction_dataset,
                        family_prediction_dataset,
                    ],
                )

    def refresh_data():
        value = get_baseline_df()

        return value

    with gr.Row():
        data_run = gr.Button("Refresh")
        data_run.click(
            refresh_data, outputs=[data_component]
        )

    with gr.Accordion("Citation", open=False):
        citation_button = gr.Textbox(
            value=CITATION_BUTTON_TEXT,
            label=CITATION_BUTTON_LABEL,
            elem_id="citation-button",
            show_copy_button=True,
        )

block.launch()