File size: 4,389 Bytes
e51e8f8
58113ed
e51e8f8
 
 
 
 
 
 
 
 
 
 
58113ed
 
758ca0a
58113ed
 
 
 
 
 
 
 
 
8410cc1
58113ed
 
e51e8f8
77df2c6
968af70
 
 
e51e8f8
968af70
 
 
 
 
 
 
 
 
 
 
e51e8f8
968af70
e51e8f8
 
968af70
e51e8f8
 
 
 
 
968af70
 
 
 
e51e8f8
968af70
e51e8f8
 
968af70
 
 
 
 
 
 
 
 
 
e51e8f8
 
 
968af70
e51e8f8
 
 
968af70
e51e8f8
 
968af70
 
 
58113ed
8410cc1
 
 
 
 
58113ed
968af70
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
e51e8f8
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
import datetime
import operator
import pathlib

import pandas as pd
import tqdm.auto
import yaml
from huggingface_hub import HfApi

repo_dir = pathlib.Path(__file__).parent


class DemoList:
    COLUMN_INFO = [
        ['status', 'markdown'],
        ['hardware', 'markdown'],
        ['title', 'markdown'],
        ['arxiv', 'markdown'],
        ['github', 'markdown'],
        ['likes', 'number'],
        ['tags', 'str'],
        ['last_modified', 'str'],
        ['created', 'str'],
        ['sdk', 'markdown'],
        ['sdk_version', 'str'],
        ['suggested_hardware', 'markdown'],
    ]

    def __init__(self):
        self.api = HfApi()
        self._raw_data = self.load_data()
        self.df_raw = pd.DataFrame(self._raw_data)
        self.df = self.prettify_df()

    @property
    def column_names(self):
        return list(map(operator.itemgetter(0), self.COLUMN_INFO))

    @property
    def column_datatype(self):
        return list(map(operator.itemgetter(1), self.COLUMN_INFO))

    @staticmethod
    def get_space_id(url: str) -> str:
        return '/'.join(url.split('/')[-2:])

    def load_data(self) -> list[dict]:
        with open(repo_dir / 'list.yaml') as f:
            data = yaml.safe_load(f)
        res = []
        for url in tqdm.auto.tqdm(list(data)):
            space_id = self.get_space_id(url)
            space_info = self.api.space_info(repo_id=space_id)
            card = space_info.cardData
            info = data[url]
            for tag in ['arxiv', 'github', 'tags']:
                if tag not in info:
                    info[tag] = []
            info['url'] = url
            info['title'] = card['title']
            info['sdk'] = card['sdk']
            info['sdk_version'] = card.get('sdk_version', '')
            info['likes'] = space_info.likes
            info['last_modified'] = space_info.lastModified
            info['status'] = space_info.runtime['stage']
            info['suggested_hardware'] = card.get('suggested_hardware', '')
            info['hardware'] = space_info.runtime['hardware']['current']
            if info['hardware'] is None:
                info['hardware'] = space_info.runtime['hardware']['requested']
            res.append(info)
        return res

    def get_arxiv_link(self, links: list[str]) -> str:
        links = [self.create_link(link.split('/')[-1], link) for link in links]
        return '\n'.join(links)

    def get_github_link(self, links: list[str]) -> str:
        links = [self.create_link('github', link) for link in links]
        return '\n'.join(links)

    def get_tag_list(self, tags: list[str]) -> str:
        return ', '.join(tags)

    @staticmethod
    def create_link(text: str, url: str) -> str:
        return f'<a href={url} target="_blank">{text}</a>'

    def to_div(self, text: str | None, category_name: str) -> str:
        if text is None:
            text = ''
        class_name = f'{category_name}-{text.lower()}'
        return f'<div class="{class_name}">{text}</div>'

    @staticmethod
    def format_timestamp(timestamp: str) -> str:
        s = datetime.datetime.strptime(timestamp, '%Y-%m-%dT%H:%M:%S.000Z')
        return s.strftime('%Y/%m/%d %H:%M:%S')

    def prettify_df(self) -> pd.DataFrame:
        new_rows = []
        for _, row in self.df_raw.copy().iterrows():
            new_row = {
                'status':
                self.to_div(row.status, 'status'),
                'hardware':
                self.to_div(row.hardware, 'hardware'),
                'suggested_hardware':
                self.to_div(row.suggested_hardware, 'hardware'),
                'title':
                self.create_link(row.title, row.url),
                'arxiv':
                self.get_arxiv_link(row.arxiv),
                'github':
                self.get_github_link(row.github),
                'likes':
                row.likes,
                'tags':
                self.get_tag_list(row.tags),
                'last_modified':
                self.format_timestamp(row.last_modified),
                'created':
                self.format_timestamp(row.created),
                'sdk':
                self.to_div(row.sdk, 'sdk'),
                'sdk_version':
                row.sdk_version,
            }
            new_rows.append(new_row)
        df = pd.DataFrame(new_rows).loc[:, self.column_names]
        return df