dosubot[bot]

Hi @himadri-bhattacharjee! I'm [Dosu](https://go.dosu.dev/dosubot) and I’m helping the langfuse team. Langfuse currently only supports basic Redis TLS options—CA, cert, and key—via environment variables (`REDIS_TLS_CA_PATH`, `REDIS_TLS_CERT_PATH`, `REDIS_TLS_KEY_PATH`). There is no support...

Hi @elevemonkey! I'm [Dosu](https://go.dosu.dev/dosubot) and I’m helping the langfuse team. This is a known issue in Langfuse version 3.116.1: the trace viewer may show the node type instead of your...

The warning `"min_value" action is not supported on type "integer"` is a known bug in the current version of `@valibot/to-json-schema`—the maintainer has acknowledged it and plans to fix it in...

Hi @Lyppo! I'm [Dosu](https://go.dosu.dev/dosubot) and I’m helping the Screenbox team. Screenbox does not currently support playing Blu-ray or UHD 4K ISO files. The app's supported formats do not include .iso...

This bug is caused by the way the backend handles the updated_at timestamp for workflow revisions. The WorkflowVersion model uses onupdate=func.now() for the updated_at field, so when a new revision...

I couldn't find any existing tickets or discussions about this issue. In the code, the cooling logic (sometimes called "ResampledCooling") uses adaptive ODE integration with an absolute tolerance tied to...

I couldn't find any details or discussion from ticket #1126 to use as context. Based on the code and docs, the cooling routines use adaptive ODE integration with tolerances tied...

Hi @LeeMonkey! I'm [Dosu](https://go.dosu.dev/dosubot) and I’m helping the ncnn team. 这个报错 "FATAL ERROR! pool allocator destroyed too early" 是 ncnn 在内存分配器（PoolAllocator/UnlockedPoolAllocator）析构时发现还有未释放的内存块时主动抛出的，通常是资源生命周期管理出了问题，比如分配器提前析构但还有 tensor/mat 等对象在用它分配的内存【参考源码实现和官方说明】([源码](https://github.com/Tencent/ncnn/blob/021af96e707f6f16bfb05f028915e42076a91b7d/src/allocator.cpp))。 常见原因和解决思路如下： - Python 脚本里，ncnn 的 Mat/tensor 输出对象如果还被...

你的代码整体写法是正确的，使用了 with 语句管理 ncnn.Net 和 Extractor 的生命周期，并且在 with 块内就把 ncnn.Mat 输出（out0）通过 np.array 拷贝成 numpy，再转 torch tensor，这样理论上不会有 ncnn 内存泄漏或引用遗留问题。 但 ncnn Python 绑定里，这类 “pool allocator destroyed too early” 报错，通常是因为 ncnn.Mat（比如 out0）对象还被...

The edit mode in the evaluator Playground uses the SharedEditor component, which extracts only the plain text content from the editor input—this means any JSON Markdown formatting (like code block...