From d07387b0a36f5030d2a33568a0a57985161bf86e Mon Sep 17 00:00:00 2001 From: Shilong Zhang <61961338+jshilong@users.noreply.github.com> Date: Fri, 10 Dec 2021 13:40:19 +0800 Subject: [PATCH] [Doc]Add doc for detect_anomalous_params (#6697) * add doc for detect_anomalous_params * add min verision requirment * fix version and key * fix version --- docs/faq.md | 3 ++- docs_zh-CN/faq.md | 3 ++- 2 files changed, 4 insertions(+), 2 deletions(-) diff --git a/docs/faq.md b/docs/faq.md index caf0d7193f9..1779d184242 100644 --- a/docs/faq.md +++ b/docs/faq.md @@ -81,7 +81,8 @@ We list some common troubles faced by many users and their corresponding solutio - "RuntimeError: Expected to have finished reduction in the prior iteration before starting a new one" 1. This error indicates that your module has parameters that were not used in producing loss. This phenomenon may be caused by running different branches in your code in DDP mode. - 2. You can set ` find_unused_parameters = True` in the config to solve the above problems or find those unused parameters manually. + 2. You can set ` find_unused_parameters = True` in the config to solve the above problems(but this will slow down the training speed. + 3. If the version of your MMCV >= 1.4.1, you can get the name of those unused parameters with `detect_anomalous_params=True` in `optimizer_config` of config. ## Evaluation diff --git a/docs_zh-CN/faq.md b/docs_zh-CN/faq.md index 9247a1c612c..ab577554368 100644 --- a/docs_zh-CN/faq.md +++ b/docs_zh-CN/faq.md @@ -82,7 +82,8 @@ 3. 使用 `config/fp16` 中的示例尝试混合精度训练。`loss_scale` 可能需要针对不同模型进行调整。 - "RuntimeError: Expected to have finished reduction in the prior iteration before starting a new one" 1. 这个错误出现在存在参数没有在 forward 中使用，容易在 DDP 中运行不同分支时发生。 - 2. 你可以在 config 设置 `find_unused_parameters = True`，或者手动查找哪些参数没有用到。 + 2. 你可以在 config 设置 `find_unused_parameters = True` 进行训练 (会降低训练速度)。 + 3. 你也可以通过在 config 中的 `optimizer_config` 里设置 `detect_anomalous_params=True` 查找哪些参数没有用到，但是需要 MMCV 的版本 >= 1.4.1。 ## Evaluation 相关