Andy Hsu
Andy Hsu
I'm running pytorch 0.2, and the code `dqn_learn.py` fail to work.. the error as follow ``` Traceback (most recent call last): File "ram.py", line 57, in main(env) File "ram.py", line...
if `window.h` included before, will have redefined errors. it can be solved by using `winsock` instead of `winsock2` http://stackoverflow.com/questions/39849684/header-correct-but-compile-erroridentifier-not-found/39867561#39867561
See below example, load_dataset works normally but after `to_sharegpt`, the dataset become empty. ```py from pprint import pprint from datasets import load_dataset # fizzbuzz123.csv ''' Query,Answer "123?","456" "1234?","456" "12345?","456" "123456?","456"...
When dataset have same prompt, it will throw error when calling `standardize_sharegpt`, see the following example, the only difference in `good.csv` and `bad.csv` is name column, which `good.csv` I added...