Duplicate from hf-ml4h/biomedical-language-models

.gitattributes

@@ -0,0 +1,34 @@
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ckpt filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.mlmodel filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.npy filter=lfs diff=lfs merge=lfs -text
+*.npz filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pickle filter=lfs diff=lfs merge=lfs -text
+*.pkl filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+*.safetensors filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.xz filter=lfs diff=lfs merge=lfs -text
+*.zip filter=lfs diff=lfs merge=lfs -text
+*.zst filter=lfs diff=lfs merge=lfs -text
+*tfevents* filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,13 @@
+---
+title: Explore Biomedical Language Models
+emoji: 🗺️
+colorFrom: indigo
+colorTo: green
+sdk: gradio
+sdk_version: 3.19.1
+app_file: app.py
+pinned: false
+duplicated_from: hf-ml4h/biomedical-language-models
+---
+
+Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference

app.py

@@ -0,0 +1,92 @@
+#!/usr/bin/env python
+
+from __future__ import annotations
+
+import gradio as gr
+
+from model_list import ModelList
+
+DESCRIPTION = '# Explore Biomedical Language Models'
+NOTES = '''
+- Stanford HAI Article, ["The Shaky Foundations of Foundation Models in Healthcare"](https://hai.stanford.edu/news/shaky-foundations-foundation-models-healthcare)
+'''
+FOOTER = ''''''
+
+def main():
+    model_list = ModelList()
+
+    with gr.Blocks(css='style.css') as demo:
+        gr.Markdown(DESCRIPTION)
+
+        search_box = gr.Textbox(
+            label='Search Model Name',
+            placeholder=
+            'You can search for titles with regular expressions. e.g. (?<!sur)face',
+            max_lines=1)
+        
+        case_sensitive = gr.Checkbox(label='Case Sensitive')
+        
+        filter_names = gr.CheckboxGroup(choices=[
+            'Paper',
+            'Code',
+            'Model Weights',
+        ], label='Filter')
+
+        data_type_names = [
+            'Biomedical',
+            'Clinical',
+            'Scientific',
+        ]
+
+        data_types = gr.CheckboxGroup(choices=data_type_names,
+                                      value=data_type_names,
+                                      label='Training Data Type(s)')
+        
+        search_button = gr.Button('Search')
+
+        number_of_models = gr.Textbox(label='Number of Models Found')
+        table = gr.HTML(show_label=False)
+
+        gr.Markdown(NOTES)
+        gr.Markdown(FOOTER)
+
+        demo.load(fn=model_list.render,
+                  inputs=[
+                      search_box,
+                      case_sensitive,
+                      filter_names,
+                      data_types,
+                  ],
+                  outputs=[
+                      number_of_models,
+                      table,
+                  ])
+        search_box.submit(fn=model_list.render,
+                          inputs=[
+                              search_box,
+                              case_sensitive,
+                              filter_names,
+                              data_types,
+                          ],
+                          outputs=[
+                              number_of_models,
+                              table,
+                          ])
+
+        search_button.click(fn=model_list.render,
+                            inputs=[
+                                search_box,
+                                case_sensitive,
+                                filter_names,
+                                data_types,
+                            ],
+                            outputs=[
+                                number_of_models,
+                                table,
+                            ])
+
+    demo.launch(enable_queue=True, share=False)
+
+
+if __name__ == '__main__':
+    main()

model_list.py

@@ -0,0 +1,89 @@
+from __future__ import annotations
+
+import numpy as np
+import pandas as pd
+import requests
+from huggingface_hub.hf_api import SpaceInfo
+
+url = 'https://docs.google.com/spreadsheets/d/1fANyV8spnEGUBMevjnb1FupkbESq9lTM2CGQt413sXQ/edit#gid=874079331'
+csv_url = url.replace('/edit#gid=', '/export?format=csv&gid=')
+
+class ModelList:
+    def __init__(self):
+        self.table = pd.read_csv(csv_url)
+        self._preprocess_table()
+
+        self.table_header = '''
+            <tr>
+                <td width="20%">Model Name</td>
+                <td width="10%">Data Type(s)</td>
+                <td width="10%">Year Published</td>
+                <td width="10%">Paper</td>
+                <td width="10%">Code on Github</td>
+                <td width="10%">Weights on 🤗</td>
+                <td width="10%">Other Weights</td>
+            </tr>'''
+
+    def _preprocess_table(self) -> None:
+        self.table['name_lowercase'] = self.table.name.str.lower()
+
+        rows = []
+        for row in self.table.itertuples():
+            paper = f'<a href="{row.paper}" target="_blank">Paper</a>' if isinstance(
+                row.paper, str) else ''
+            github = f'<a href="{row.github}" target="_blank">GitHub</a>' if isinstance(
+                row.github, str) else ''
+            hf_model = f'<a href="{row.hub}" target="_blank">Hub Model</a>' if isinstance(
+                row.hub, str) else ''
+            other_model = f'<a href="{row.other}" target="_blank">Other Weights</a>' if isinstance(
+                row.other, str) else ''
+            row = f'''
+                <tr>
+                    <td>{row.name}</td>
+                    <td>{row.type}</td>
+                    <td>{row.year}</td>
+                    <td>{paper}</td>
+                    <td>{github}</td>
+                    <td>{hf_model}</td>
+                    <td>{other_model}</td>
+                </tr>'''
+            rows.append(row)
+        self.table['html_table_content'] = rows
+
+    def render(self, search_query: str, 
+            case_sensitive: bool,
+            filter_names: list[str],
+            data_types: list[str]) -> tuple[int, str]:
+        df = self.table
+        if search_query:
+            if case_sensitive:
+                df = df[df.name.str.contains(search_query)]
+            else:
+                df = df[df.name_lowercase.str.contains(search_query.lower())]
+        has_paper = 'Paper' in filter_names
+        has_github = 'Code' in filter_names
+        has_model = 'Model Weights' in filter_names
+        df = self.filter_table(df, has_paper, has_github, has_model, data_types)
+        return len(df), self.to_html(df, self.table_header)
+
+    @staticmethod
+    def filter_table(df: pd.DataFrame, has_paper: bool, has_github: bool,
+                     has_model: bool, data_types: list[str]) -> pd.DataFrame:
+        if has_paper:
+            df = df[~df.paper.isna()]
+        if has_github:
+            df = df[~df.github.isna()]
+        if has_model:
+            df = df[~df.hub.isna() | ~df.other.isna()]
+        df = df[df.type.isin(set(data_types))]
+        return df
+
+    @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

requirements.txt

@@ -0,0 +1,2 @@
+gradio
+pandas

style.css

@@ -0,0 +1,20 @@
+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;
+  }
+
+