hackaday

Results 5 issues of hackaday

启动命令 ``` ./gost-linux-amd64-3.0.0-beta.5 -L tcp://:8000 -F socks5+ws://user:pass@:2080 ``` 请求`8000`端口后出现panic: ``` {"level":"warning","msg":"load global TLS certificate files failed, use random generated certificate","time":"2022-09-22T01:17:51.709+08:00"} {"handler":"tcp","kind":"service","level":"info","listener":"tcp","msg":"listening on [::]:8000/tcp","service":"service-0","time":"2022-09-22T01:17:51.709+08:00"} {"handler":"tcp","kind":"handler","level":"info","listener":"tcp","local":"127.0.0.1:8000","msg":"127.0.0.1:30242 127.0.0.1:8000","remote":"127.0.0.1:30242","service":"service-0","time":"2022-09-22T01:18:19.597+08:00"} {"duration":168347,"handler":"tcp","kind":"handler","level":"info","listener":"tcp","local":"127.0.0.1:8000","msg":"127.0.0.1:30242 >< 127.0.0.1:8000","remote":"127.0.0.1:30242","service":"service-0","time":"2022-09-22T01:18:19.597+08:00"} panic: runtime...

### Verify steps - [X] 如果你可以自己 debug 并解决的话,提交 PR 吧 Is this something you can **debug and fix**? Send a pull request! Bug fixes and documentation fixes are welcome. -...

Is is possible to support GLM(https://github.com/THUDM/GLM) as PLM which also available in transformers of huggingface?

### 软件环境 ```Markdown - paddlepaddle:2.5.1 - paddlepaddle-gpu: 无 - paddlenlp: 2.6.0 ``` ### 重复问题 - [X] I have searched the existing issues ### 错误描述 ```Markdown token_id 12084和18005的token重复,均为美元符号`$`。load vocab的时候为map赋值操作,未检测重复token,导致token_id=12084没有对应token。 相关issue:https://github.com/PaddlePaddle/PaddleNLP/issues/6429 ```...

bug
triage

``` Fatal signal 6 (SIGABRT), code -1 (SI_QUEUE) in tid 21861 (Thread-3), pid 21791 (t.benchmarkncnn) 2025-03-27 17:53:03.993 21864-21864 DEBUG pid-21864 A Cmdline: com.tencent.benchmarkncnn 2025-03-27 17:53:03.993 21864-21864 DEBUG pid-21864 A pid:...