Translate 'get_start.md' and 'analysis.md' into Chinese #786
Conversation
l412198735
changed the title
|
@linyq17, Please review this PR. |
docs/zh_cn/get_started.md
Outdated
| - Linux | macOS | Windows | ||
| - Python 3.6+ | ||
| - PyTorch 1.6+ | ||
| - CUDA 9.2+ (If you build PyTorch from source, CUDA 9.0 is also compatible) |
There was a problem hiding this comment.
(If you build PyTorch from source, CUDA 9.0 is also compatible) >（如果基于 PyTorch 源码安装，也能够支持 CUDA 9.0）
There was a problem hiding this comment.
The above issue has been resolved and is ready for review again.
docs/zh_cn/get_started.md
Outdated
| - [MMEngine](https://mmengine.readthedocs.io/en/latest/get_started/installation.html) | ||
| - [MMDetection](https://mmdetection.readthedocs.io/en/latest/get_started.html#installation) | ||
|
|
||
| MMTracking, MMEngine, MMCV 和 MMDetection的兼容版本如下。请安装正确的版本以避免安装出现问题。 |
There was a problem hiding this comment.
Please note that the one blank space between English & numbers and Chinese. Check all the docs.
docs/zh_cn/get_started.md
Outdated
| conda activate open-mmlab | ||
| ``` | ||
|
|
||
| 2. 基于[PyTorch官方说明](https://pytorch.org/) 安装PyTorch和 torchvision。这里我们使用PyTorch 1.10.0和CUDA 11.1。 |
There was a problem hiding this comment.
Please check all the docs of blank space issues.
docs/zh_cn/get_started.md
Outdated
| pip install 'mmdet>=3.0.0rc0' | ||
| ``` | ||
|
|
||
| 如果你想修改代码，你也可以选择从源代码构建 MMDetection: |
docs/zh_cn/get_started.md
Outdated
| cd .. | ||
| ``` | ||
|
|
||
| **重要提示**: 如果安装了 mmcv，则需要首先运行 pip 卸载 mmcv-lite。因为如果 mmcv-lite 和 mmcv 都已安装，会出现 ModuleNotFoundError。 |
There was a problem hiding this comment.
ModuleNotFoundError > ModuleNotFoundError
docs/zh_cn/get_started.md
Outdated
|
|
||
| ## 验证 | ||
|
|
||
| 验证MMTracking和所需环境是否正确安装, 我们可以运行MOT, VIS, VID and SOT [演示脚本](https://github.com/open-mmlab/mmtracking/blob/1.x/demo/) 中的**一个** 。 |
There was a problem hiding this comment.
验证 MMTracking > 为了验证 MMTracking
| ## 多目标跟踪错误可视化 | ||
|
|
||
| `tools/analysis_tools/mot/mot_error_visualize.py` 能可视化多目标跟踪的误差。 | ||
| 这个脚本需要推理的结果。默认情况下，**红色**边界框表示假阳性，**黄色**边界框表示假阴性，**蓝色**边界框表示ID切换。 |
| python tools/analysis_tools/analyze_logs.py cal_train_time log.json [--include-outliers] | ||
| ``` | ||
|
|
||
| The output is expected to be like the following. |
| 这里，`eval_show_video_indices`用于索引`numpy.ndarray`。 | ||
| 它可以是`int`(正或负)。正数`k`表示排顶部的k个结果，而负数表示排底部的k个结果。 | ||
|
|
||
| ## Save SOT evaluation results and plot them |
| - `SHOW_INTERVAL`: 显示间隔（秒）。 | ||
| - `--not-show`: 是否不实时显示图片。 | ||
|
|
||
| ## 在视频水平上展示单目标跟踪评估结果。 |
|
l412198735 seems not to be a GitHub user. You need a GitHub account to be able to sign the CLA. If you have already a GitHub account, please add the email address used for this commit to your account. You have signed the CLA already but the status is still pending? Let us recheck it. |
|
The above issue has been resolved and is ready for review again. |
#780 Translate 'get_start.md' and 'analysis.md' into Chinese