Timothyxxx commited on
Commit
1033534
1 Parent(s): 7913a66
Files changed (2) hide show
  1. app.py +1 -1
  2. resources/demo_description.md +2 -2
app.py CHANGED
@@ -28,7 +28,7 @@ st.set_page_config(
28
  ROOT_DIR = os.path.join(os.path.dirname(__file__), "./")
29
  # todo: Add more binder questions, need careful cherry-picks
30
  EXAMPLE_TABLES = {
31
- "Estonia men's national volleyball team": (558, "what is the total number of players from france?"),
32
  # 'how old is kert toobal'
33
  "Highest mountain peaks of California": (5, "which is the lowest mountain?"),
34
  # 'which mountain is in the most north place?'
 
28
  ROOT_DIR = os.path.join(os.path.dirname(__file__), "./")
29
  # todo: Add more binder questions, need careful cherry-picks
30
  EXAMPLE_TABLES = {
31
+ "Estonia men's national volleyball team": (558, "who is the youngest player from europe?"),
32
  # 'how old is kert toobal'
33
  "Highest mountain peaks of California": (5, "which is the lowest mountain?"),
34
  # 'which mountain is in the most north place?'
resources/demo_description.md CHANGED
@@ -2,9 +2,9 @@ This is an interactive demo of [Binder](https://lm-code-binder.github.io/) based
2
  You can input a question(may not be answered by pure SQL/Python),
3
  The demo will generate a Binder program (e.g., SQL/Python bound with API calls to GPT3 Codex),
4
  and then execute the Binder program (API calls by prompting Codex and SQL/Python by using SQL/Python interpreters) to drive the final answer.
5
- For more details, check out the project [website](https://lm-code-binder.github.io/).
6
 
 
7
 
8
- *Note:*
9
  - The demo might be slow when (high) concurrent requests occur because of the query limit of OpenAI GPT3 Codex.
10
  - To speed up the demonstration, we only employed a simper version of Binder (e.g., no majority vote), please check the [paper](https://arxiv.org/abs/2210.02875) and [code](https://github.com/HKUNLP/Binder) for more detail.
 
2
  You can input a question(may not be answered by pure SQL/Python),
3
  The demo will generate a Binder program (e.g., SQL/Python bound with API calls to GPT3 Codex),
4
  and then execute the Binder program (API calls by prompting Codex and SQL/Python by using SQL/Python interpreters) to drive the final answer.
 
5
 
6
+ *For more details, check out the [project website](https://lm-code-binder.github.io/).*
7
 
8
+ **Note:**
9
  - The demo might be slow when (high) concurrent requests occur because of the query limit of OpenAI GPT3 Codex.
10
  - To speed up the demonstration, we only employed a simper version of Binder (e.g., no majority vote), please check the [paper](https://arxiv.org/abs/2210.02875) and [code](https://github.com/HKUNLP/Binder) for more detail.