wolfe

how to use custom SUTime rules?

2

Thanks for create this package. I am using it with corenlp 3.9.2 just change the following version of jars: ``` _required_jars = { 'stanford-corenlp-3.9.2-models.jar', 'stanford-corenlp-3.9.2.jar', 'gson-2.8.5.jar', 'slf4j-simple-1.7.25.jar' } ``` But...

I previous think that the model can be pre-load. But after test (create an sutime api in django), the model can't pre-load in memory, is there any other solution?

As an online application, can i add new item in regexner mapping file without restart the server?

why label_map start from 1 ?

3

why label not start from 0 ? Is there other label use 0 ?

可否公开训练结果的词频文件？

4

可否公开训练结果的词频文件？

How to understand this line of code? while l < r and nums[l] == nums[l - 1]: l += 1 while l < r and nums[r] == nums[r + 1]:...

根据词性给query打标签构建训练集吗？具体如何训练模型？

is there any example for relation extraction (ubert)?

1

``` import argparse from fengshen import UbertPiplines total_parser = argparse.ArgumentParser("TASK NAME") total_parser = UbertPiplines.piplines_args(total_parser) args = total_parser.parse_args() args.pretrained_model_path = 'IDEA-CCNL/Erlangshen-Ubert-110M-Chinese' #预训练模型路径 test_data=[ { "task_type": "抽取任务", "subtask_type": "关系抽取", "text": "姚明妻子叶莉罕见现身!39岁气质出众端庄,姚明却发福严重", "choices":...

The correct one: ``` with tf.variable_scope(scope) as scope: w = tf.get_variable("weights", [x.shape[1], out_dim], initializer=tf.random_normal_initializer()) b = tf.get_variable("biases", [out_dim], initializer=tf.constant_initializer(0.0)) ``` The following code will raise the error: `TypeError: Failed to...