Yu Zhang

@MBAnslow Hi, this issue has been solved by the latest commits: ```py >>> parser.predict(['This', 'is', '(not)', 'parsable', '!'], verbose=False)[0].pretty_print()\ TOP | S _____________|______________________ | VP | | ________|________ | NP...

不好意思，我似乎没有找到你说的`remove`在哪里

Hi, thanks for your suggestions. Intuitively crfpar may benefit from PLMs like XLNet. But I didn't conduct the experiments on them. I will let you know if the experiments are...

@unikcc Hi, the code and pretrained models of the two papers can be found in this [repo](https://github.com/yzhangcs/parser). I'm sorry I didn't take the time to reorganize the code here.

@Jokeren Hi, we @sustcsonglin found when using Triton 2.2 & H100 (A100 works), running [this check](https://github.com/sustcsonglin/flash-linear-attention/blob/main/tests/test_fused_chunk.py) will still give very strange results. ```py >>> python tests/test_fused_chunk.py DTYPE STORE IFCOND DIFF...

@lhoestq Hello, I'm wondering if there are any solutions to work with shuffle now. I've noticed the caveats in docs, > examples from shuffle buffers are lost when resuming and...

@lhoestq thank you for your quick response! I'll try it :}

@lhoestq Hi, just revise the `BufferShuffledExamplesIterable` and it works ```py class BufferShuffledExamplesIterable(datasets.iterable_dataset.BufferShuffledExamplesIterable): def __init__(self, *args, **kwargs): super().__init__(*args, **kwargs) def _init_state_dict(self) -> dict: self._state_dict = self.ex_iterable._init_state_dict() self._state_dict['mem_buffer'] = ([],) self._state_dict['gloabl_example_index'] =...

@Hairmore Hello，抱歉很晚回复你的问题，.conllx请尽量使用utf8编码，.txt文件有特殊用途，表示纯文本文件

recommend to use conllu format files with .conllu/.conllx extension on Linux, which is my practice.