Initial commit of papers

.gitignore

@@ -0,0 +1 @@
+__pycache__

README.md

@@ -1,5 +1,5 @@
 ---
-title: ICLR2023
+title: ICLR2023 Paper Submissions
 emoji: 📉
 colorFrom: indigo
 colorTo: indigo

app.py

@@ -0,0 +1,70 @@
+#!/usr/bin/env python
+
+from __future__ import annotations
+
+import gradio as gr
+
+from paper_list import PaperList
+
+DESCRIPTION = '# ICLR 2023 Paper Submissions'
+NOTES = '''
+- [ICLR 2023](https://openreview.net/group?id=ICLR.cc/2023/Conference)
+- [List of submitted papers](https://docs.google.com/spreadsheets/d/1dQMjjetud2edTEREdLiuD4giC244lxY67ZxaL7NiMUc/edit#gid=1277917086)
+'''
+
+
+def main():
+    paper_list = PaperList()
+
+    with gr.Blocks(css='style.css') as demo:
+        gr.Markdown(DESCRIPTION)
+
+        search_box = gr.Textbox(
+            label='Search Title',
+            placeholder=
+            'You can search for titles with regular expressions. e.g. (?<!sur)face',
+            max_lines=1)
+
+        case_sensitive = gr.Checkbox(label='Case Sensitive')
+
+        search_button = gr.Button('Search')
+
+        number_of_papers = gr.Textbox(label='Number of Papers Found')
+        table = gr.HTML(show_label=False)
+
+        gr.Markdown(NOTES)
+
+        demo.load(fn=paper_list.render,
+                  inputs=[
+                      search_box,
+                      case_sensitive
+                  ],
+                  outputs=[
+                      number_of_papers,
+                      table,
+                  ])
+        search_box.submit(fn=paper_list.render,
+                          inputs=[
+                              search_box,
+                              case_sensitive
+                          ],
+                          outputs=[
+                              number_of_papers,
+                              table,
+                          ])
+
+        search_button.click(fn=paper_list.render,
+                            inputs=[
+                                search_box,
+                                case_sensitive
+                            ],
+                            outputs=[
+                                number_of_papers,
+                                table,
+                            ])
+
+    demo.launch(enable_queue=True, share=False)
+
+
+if __name__ == '__main__':
+    main()

get_submissions.py

@@ -0,0 +1,41 @@
+import json
+import requests
+import csv
+
+offset = 0
+limit = 1000
+max_count = 4944
+
+base_url = 'https://api.openreview.net'
+
+all_papers = []
+while offset < max_count:
+    limit = min(limit, max_count - offset)
+
+    print(offset, limit)
+    url = base_url + f"/notes?details=invitation%2Coriginal&offset={offset}&limit={limit}&invitation=ICLR.cc%2F2023%2FConference%2F-%2FBlind_Submission"
+
+    response = requests.get(url)
+    papers = json.loads(response.text)['notes']
+    all_papers += papers
+
+    offset += limit
+
+
+
+with open('iclr_submissions.csv', 'w', encoding='UTF8', newline='') as f:
+    header = ['title', 'url', 'pdf', 'tldr', 'abstract', 'keywords']
+    writer = csv.writer(f)
+    writer.writerow(header)
+
+    for paper in all_papers:
+        content = paper['content']
+
+        title = content['title']
+        url = f'https://openreview.net/forum?id={paper["forum"]}'
+        pdf =  f'https://openreview.net/pdf?id={paper["forum"]}'
+        tldr = content.get('TL;DR', '')
+        abstract = content['abstract']
+        keywords = ', '.join(content['keywords'])
+
+        writer.writerow([title, url, pdf, tldr, abstract, keywords])

paper_list.py

@@ -0,0 +1,79 @@
+from __future__ import annotations
+
+import pandas as pd
+import requests
+from huggingface_hub.hf_api import SpaceInfo
+
+
+class PaperList:
+    def __init__(self):
+        self.organization_name = 'ICLR2023'
+        self.table = pd.read_csv('iclr_submissions.csv')
+        self._preprocess_table()
+
+        self.table_header = '''
+            <tr>
+                <td width="25%">Title</td>
+                <td width="5%">PDF</td>
+                <td width="20%">Tldr</td>
+                <td width="50%">Abstract</td>
+            </tr>'''
+
+    @staticmethod
+    def load_space_info(author: str) -> list[SpaceInfo]:
+        path = 'https://huggingface.co/api/spaces'
+        r = requests.get(path, params={'author': author})
+        d = r.json()
+        return [SpaceInfo(**x) for x in d]
+
+    def add_spaces_to_table(self, organization_name: str,
+                            df: pd.DataFrame) -> pd.DataFrame:
+        spaces = self.load_space_info(organization_name)
+        name2space = {
+            s.id.split('/')[1].lower(): f'https://huggingface.co/spaces/{s.id}'
+            for s in spaces
+        }
+        return df
+
+    def _preprocess_table(self) -> None:
+        self.table = self.add_spaces_to_table(self.organization_name,
+                                              self.table)
+        self.table['title_lowercase'] = self.table.title.str.lower()
+
+        rows = []
+        for row in self.table.itertuples():
+            paper = f'<a href="{row.url}" target="_blank">{row.title}</a>' if isinstance(
+                row.url, str) else row.title
+            pdf = f'<a href="{row.pdf}" target="_blank">pdf</a>' if isinstance(
+                row.pdf, str) else ''
+            tldr = row.tldr if isinstance(row.tldr, str) else ''
+
+            row = f'''
+                <tr>
+                    <td>{paper}</td>
+                    <td>{pdf}</td>
+                    <td>{tldr}</td>
+                    <td>{row.abstract}</td>
+                </tr>'''
+            rows.append(row)
+        self.table['html_table_content'] = rows
+
+    def render(self, search_query: str, case_sensitive: bool) -> tuple[int, str]:
+        df = self.add_spaces_to_table(self.organization_name, self.table)
+        if search_query:
+            if case_sensitive:
+                df = df[df.title.str.contains(search_query)]
+            else:
+                df = df[df.title_lowercase.str.contains(search_query.lower())]
+
+        return len(df), self.to_html(df, self.table_header)
+
+    @staticmethod
+    def to_html(df: pd.DataFrame, table_header: str) -> str:
+        table_data = ''.join(df.html_table_content)
+        html = f'''
+        <table>
+            {table_header}
+            {table_data}
+        </table>'''
+        return html

style.css

@@ -0,0 +1,22 @@
+h1 {
+    text-align: center;
+}
+
+table a {
+    background-color: transparent;
+    color: #58a6ff;
+    text-decoration: none;
+}
+
+a:active, a:hover {
+    outline-width: 0;
+}
+
+a:hover {
+    text-decoration: underline;
+}
+
+table, th, td {
+    border: 1px solid;
+    padding: 8px;
+}