-
# Issue Metrics
| Metric | Value |
| --- | ---: |
| Average time to first response | 1 day, 1:22:36.177305 |
| Average time to close | 3 days, 18:25:34.011628 |
| Number of items that remain open | 2…
-
![image](https://github.com/FMInference/H2O/assets/13203040/39ff7799-98b8-4482-86f8-3302b5bca087)
1.full和h2o都正常得到了结果的latex(如下),但是local运行出来是空白。
![image](https://github.com/FMInference/H2O/assets/1320…
-
https://developers.perspectiveapi.com/s/?language=en_US
Looks like a google api that's free to use to measure "toxicity" in online text. Might be interesting to try on our data to see if it can predi…
-
# Issue Metrics
| Metric | Average | Median | 90th percentile |
| --- | --- | --- | ---: |
| Time to first response | 2 days, 16:47:21 | 6:43:30 | 4 days, 0:35:06 |
| Time to close | 13 days, 17:17:3…
-
I think that it could be interesting to add support for google's perspectiveapi, a machine learning service that works similarly to openai's moderation endpoint, and is also free.
[](https://perspec…
-
General Idea is that peoples can write comments under checkins.
In the Settings you can set who can comment under your checkins like: People who Follow you or People who you follow or Public all th…
-
A Custom Performance setting would be nice. So when you select Custom Radio Button, it will also show a slider for Steps and a slider for the step where to swap refiner.
I do not like the preset on…
-
Thank you for creating this wrapper.
I'm experiencing a weird problem. Even though the python package `perspective` is definitely installed on my system, when I'm running the script I get the foll…
-
看文章里引用的几篇用LLM做评测的论文,好像都是针对生成比较有优势,像这钟通用领域的安全性判别问题,也会有优势吗
-
https://www.perspectiveapi.com/