linyqh / NarratoAI

利用AI大模型,一键解说并剪辑视频; Using AI models to automatically provide commentary and edit videos with a single click.
MIT License
1.96k stars 236 forks source link

BUG反馈:分析进度会卡住 #63

Open ccvv3655 opened 1 day ago

ccvv3655 commented 1 day ago

分析进度会卡住,换个网络才会继续走进度

2024-11-21 15:05:39.123 | INFO | app.config.config:load_config:22 - load config from file: G:\NarratoAI\NarratoAI/config.toml 2024-11-21 15:05:39.128 | INFO | app.config.config::71 - NarratoAI v0.3.5 2024-11-21 15:05:47.283 Examining the path of torch.classes raised: Tried to instantiate class 'path.path', but it does not exist! Ensure that it is registered via torch::class 2024-11-21 15:06:20 | INFO | "./webui\components\script_settings.py:343": generate_script - 使用已缓存的关键帧: .\storage\temp\keyframes\ae1e8afca2204a760805b1d5a6ddc7a7 2024-11-21 15:06:20 | DEBUG | "./webui\components\script_settings.py:393": generate_script - Vision LLM 提供商: gemini 2024-11-21 15:06:20 | INFO | "./app\utils\vision_analyzer.py:63": analyze_images - 正在加载图片... 分析进度: 67%|████████████████████████████████████████████████ | 12/18 [01:38<00:51, 8.56s/it]

linyqh commented 8 hours ago

这应该不是卡住了,是当前正在请求 gemini api,连续请求 18 次,但是官方对接口有速率限制,1分钟最多请求 15 次,有可能是超速率请求了,后台检测到429报错;自动进入重试,默认重试是等待1分钟后再起请求,你可以多等一段时间试试

下面是官方对速率限制的说明: https://ai.google.dev/pricing?hl=zh-cn#1_5flash