hwaking

i got the prediction like '[["1", "B_t_h"], ["9", "M_t_h"], ["4", "M_t_h"], ["0", "M_t_h"]]', the tag of number 0 should be 'E_t_h', is there any explation for this problem? and how...

一句话中有多对实体 关系的情况您怎么考虑呢

4

一句话中有多对实体 关系的情况您怎么考虑呢（如ABC三个实体中，存在A_B,A_C两对实体关系的情况），拆成两行？ 另外，您当前模型的精度怎么样呢 还有就是您模型中前后窗口长度设置的是多少呢 可能问题比较naive，还请您耐心赐教！

当前MAX_BOXES与MAX_TURNS不生效导致单次对话不终止显存不断增加最后显存溢出问题，修改最大对话轮数和最大历史对话数量使其生效，逻辑为历史最大对话框记录轮数达到MAX_BOXES时截断历史对话为最近MAX_TURNS数。

# Row-wise cosine similarity between two equal-width matrices and return the max top_k score and index, the score all greater than threshold_score. @vowelparrot @dev2049 @hwchase17 it's useful when we want...

Add cosine_simirarity_v1 which is more faster than the original cosine_similarity method.

3

add cosine_simirarity_v1 method which is 5 times faster than the orginal cosine_similarity method. # add cosine_simirarity_v1 method which is 5 times faster than the orgianl cosine_similarity method. @vowelparrot update math_utils.py...

add "imported function" to power the generated result of tabAutocomplete

1

### Validations - [X] I believe this is a way to improve. I'll try to join the [Continue Discord](https://discord.gg/NWtdYexhMs) for questions - [x] I'm not able to find an [open...

How can I do continue pretraining?

1

can you give the script about contnue pretraining?

can you share the continue pretraining scripts and sample data?