Gedanke

### Short description of current behavior [https://docs.mindsdb.com/use-cases/predictive_analytics/expenditures-statsforecast](https://docs.mindsdb.com/use-cases/predictive_analytics/expenditures-statsforecast) In this tutorial, I create a model to forecast expenditures based on historical data using the Nixtla’s StatsForecast engine step by step. Here,...

Fix: KeyError: 'season_length'

1

## Description KeyError: 'season_length', raised at: /home/local/mindsdb/./mindsdb/integrations/libs/ml_handler_process/learn_process.py#115 Fixes #9546 ## Type of change (Please delete options that are not relevant) - [x] 🐛 Bug fix (non-breaking change which fixes an...

[BUG] worker-1 exited with code 1 and api-1 exited with code 1

7

**Describe the Bug** I tried to start Firecrawl from docker. ```sh $ docker compose up WARN[0000] docker-compose.yaml: `version` is obsolete [+] Running 5/5 ✔ Network firecrawl_backend Created 0.1s ✔ Container...

【求助】关于 GPU 利用率以及并发相关的问题

26

# 问题 设备是 4090 24 G，一次推理的文本大概几百字左右 单个 worker 时，GPU 显存占用很低，利用率也不高，大概 10 多点 在流式并发的场景下，同时发送 2 或 3 个请求，返回的流式语音不会有中断，卡顿的情况，再多点就不行了 请问，如何才能提升 GPU 利用率以及并发量呢 # 尝试 个人做了一些尝试 加了一些 async 和 await，增加了 workers 数量(从命令行中获取)，这里设置成 6...