Batch_Parallel_LatticeLSTM icon indicating copy to clipboard operation
Batch_Parallel_LatticeLSTM copied to clipboard

Chinese NER using Lattice LSTM. Reproduction for ACL 2018 paper.

Results 19 Batch_Parallel_LatticeLSTM issues
Sort by recently updated
recently updated
newest added

if not self.skip_before_head: c_x = c_[[index_0, index_1 + 1]] h_x = h_[[index_0, index_1 + 1]] else: c_x = c_[[index_0, index_1]] h_x = h_[[index_0, index_1]] 请教一下作者,https://github.com/LeeSureman/Batch_Parallel_LatticeLSTM/blob/master/modules.py 里面的420那几行,是什么意思?[里面也是二维tensor] 看官方doc没有看到这样获取元素的例子,麻烦作者告知,或者有例子链接推荐一下,谢谢!

你好!我已经获取了ontonotes4.0原数据集,但是不知道如何处理,网上只有5.0的处理教程。还希望能分享一下4.0数据集预处理流程

谢谢,我已经解决了,主要是不熟悉那个cache部分,跑weibo的时候前面cache代码没有改。还有我问过原作者,他说他后来不用bi了,说是直接用gaz匹配,我也不清楚是个什么原理 _Originally posted by @18222278799 in https://github.com/LeeSureman/Batch_Parallel_LatticeLSTM/issues/6#issuecomment-563279258_

Hi, thank you so much for you elegant reproduction of batched version of Lattice LSTM. But I can't get the same results on Weibo Datasets, using the same emb\params\data. I...

![image](https://user-images.githubusercontent.com/40818990/102505379-df574580-40bc-11eb-92d5-45588873b7fb.png) ![image](https://user-images.githubusercontent.com/40818990/102505460-f9912380-40bc-11eb-9d41-fd47488dfe8c.png) ![image](https://user-images.githubusercontent.com/40818990/102505482-031a8b80-40bd-11eb-9816-1b1de02bf103.png) 環境: GPU : RTX 2060 CUDA : ![image](https://user-images.githubusercontent.com/40818990/102505675-3d842880-40bd-11eb-880a-e0342d825cad.png) ![image](https://user-images.githubusercontent.com/40818990/102505737-5260bc00-40bd-11eb-9a52-38ea304f02a0.png)

您好,我用的weibo数据跑,结果最大是56,没有达到您说的58,我能看下您的一些参数设置log么。谢谢 我的一些参数如下: device cuda:0 debug False norm_embed False batch 1 test_batch 1024 optim sgd lr 0.045 model lattice skip_before_head False hidden 113 momentum 0 bi True dataset weibo use_bigram True...

请问,args中的bi是什么含义,当我将bi设置为True的时候,运行main_without_fitlog.py,报了如下错误:(batch_size为10,数据集用的是resumeNER) Traceback (most recent call last): File "D:/work/NerAPIS/Batch_Parallel_LatticeLSTM-master/Batch_Parallel_LatticeLSTM-master/main_without_fitlog.py", line 205, in callbacks=callbacks) File "C:\ProgramData\Anaconda3\lib\site-packages\fastNLP\core\trainer.py", line 520, in __init__ batch_size=check_batch_size) File "C:\ProgramData\Anaconda3\lib\site-packages\fastNLP\core\trainer.py", line 920, in _check_code pred_dict = model(**refined_batch_x) File "C:\ProgramData\Anaconda3\lib\site-packages\torch\nn\modules\module.py",...

您好,想请教一下如果想用别的数据集的话,要做什么改动呢? 我在load_data.py文件中定义了一个 load_msra_ner函数,这个函数与load_weibo_ner这个函数设置的相同可以吗?