Results 204 comments of Ayaka

I found the cause of the problem > Unpickling of an array [uses jax.device_put](https://github.com/google/jax/blob/3243e23aa528db390fdece3fa32517c28a50318b/jax/_src/device_array.py#L326) with no device argument, so I believe the default default device context manager should do the...

可以將「方程式」條目加入轉換表中

這個比較麻煩,因為 [OpenCC 標準以「匯」為正](https://github.com/BYVoid/OpenCC/blob/455072cdc66a36d72f2361934b78dbd11d8dbf92/data/scheme/st_multi.txt#L15),也就是認為「滙」是異體字,但是根據「名從主人」的原則又要用「滙」。這種情況可以在兩個地方修改: 1. s2t 不變,在 t2tw 和 t2hk 中修改 2. 在 s2t 中修改,t2tw 和 t2hk 不變(目前做法) 我認為 2 較合理,因為 OpenCC 標準也應遵循名從主人原則。但遵從名從主人原則的條目似乎應該獨立成一個檔案(類似 [TWPhrasesName.txt 中的「純喫茶」](https://github.com/BYVoid/OpenCC/blob/455072cdc66a36d72f2361934b78dbd11d8dbf92/data/dictionary/TWPhrasesName.txt#L66)) 所以我暫時不知道應該如何處理

It's not too strict. My new projects are using Python `"~=3.8"`, and I am planning to use `"~=3.9"` for `list[str]` instead of `List[str]`, and even further `"~=3.10"` for the better...

`"~= 3.7"` means `">= 3.7, == 3.*"`

Perhaps calling `HTMLConvertHandler.convert();` again may work

> Just curious, what is your usage case? If you don't need differentiation, then we can use scipy implementation for it. Just for learning, so I can use the scipy...

好似喺工作組 group chat 入面討論過