Skip to content

GitLab

Unverified Commit f885d28a authored by VVsssssk's avatar VVsssssk Committed by GitHub
Browse files

[Refactor] Update configs name (#1757) 

* fix cfg name

* update cfg name

* fix cfg

* fix comments

* fix comment

* fix comments
parent ea22f8ec
Show whitespace changes
Inline Side-by-side
Showing with 43 additions and 43 deletions
configs/votenet/votenet_8x8_scannet-3d-18class.py → configs/votenet/votenet_8xb8_scannet-3d.py
View file @ f885d28a
_base_ = [
'../_base_/datasets/scannet-3d-18class.py', '../_base_/models/votenet.py',
'../_base_/schedules/schedule_3x.py', '../_base_/default_runtime.py'
'../_base_/datasets/scannet-3d.py', '../_base_/models/votenet.py',
'../_base_/schedules/schedule-3x.py', '../_base_/default_runtime.py'
]
# model settings
......
configs/votenet/votenet_iouloss_8x8_scannet-3d-18class.py → configs/votenet/votenet_head-iouloss_8xb8_scannet-3d.py
View file @ f885d28a
_base_ = ['./votenet_8x8_scannet-3d-18class.py']
_base_ = ['./votenet_8xb8_scannet-3d.py']
# model settings, add iou loss
model = dict(
......
docs/en/advanced_guides/datasets/kitti_det.md
View file @ f885d28a
......@@ -159,7 +159,7 @@ train_pipeline = [
An example to evaluate PointPillars with 8 GPUs with kitti metrics is as follows:
```shell
bash tools/dist_test.sh configs/pointpillars/hv_pointpillars_secfpn_6x8_160e_kitti-3d-3class.py work_dirs/hv_pointpillars_secfpn_6x8_160e_kitti-3d-3class/latest.pth 8 --eval bbox
bash tools/dist_test.sh configs/pointpillars/pointpillars_hv_secfpn_8xb6-160e_kitti-3d-3class.py work_dirs/hv_pointpillars_secfpn_6x8_160e_kitti-3d-3class/latest.pth 8 --eval bbox
```
## Metrics
......@@ -188,7 +188,7 @@ An example to test PointPillars on KITTI with 8 GPUs and generate a submission t
```shell
mkdir -p results/kitti-3class
./tools/dist_test.sh configs/pointpillars/hv_pointpillars_secfpn_6x8_160e_kitti-3d-3class.py work_dirs/hv_pointpillars_secfpn_6x8_160e_kitti-3d-3class/latest.pth 8 --out results/kitti-3class/results_eval.pkl --format-only --eval-options 'pklfile_prefix=results/kitti-3class/kitti_results' 'submission_prefix=results/kitti-3class/kitti_results'
./tools/dist_test.sh configs/pointpillars/pointpillars_hv_secfpn_8xb6-160e_kitti-3d-3class.py work_dirs/hv_pointpillars_secfpn_6x8_160e_kitti-3d-3class/latest.pth 8 --out results/kitti-3class/results_eval.pkl --format-only --eval-options 'pklfile_prefix=results/kitti-3class/kitti_results' 'submission_prefix=results/kitti-3class/kitti_results'
```
After generating `results/kitti-3class/kitti_results/xxxxx.txt` files, you can submit these files to KITTI benchmark. Please refer to the [KITTI official website](http://www.cvlibs.net/datasets/kitti/index.php) for more details.
docs/en/advanced_guides/datasets/lyft_det.md
View file @ f885d28a
......@@ -153,7 +153,7 @@ where the first 3 dimensions refer to point coordinates, and the last refers to
An example to evaluate PointPillars with 8 GPUs with Lyft metrics is as follows.
```shell
bash ./tools/dist_test.sh configs/pointpillars/hv_pointpillars_fpn_sbn-all_2x8_2x_lyft-3d.py checkpoints/hv_pointpillars_fpn_sbn-all_2x8_2x_lyft-3d_20210517_202818-fc6904c3.pth 8 --eval bbox
bash ./tools/dist_test.sh configs/pointpillars/pointpillars_hv_fpn_sbn-all_8xb2-2x_lyft-3d.py checkpoints/hv_pointpillars_fpn_sbn-all_2x8_2x_lyft-3d_20210517_202818-fc6904c3.pth 8 --eval bbox
```
## Metrics
......@@ -189,7 +189,7 @@ We employ this official method for evaluation on Lyft. An example of printed eva
An example to test PointPillars on Lyft with 8 GPUs and generate a submission to the leaderboard is as follows.
```shell
./tools/dist_test.sh configs/pointpillars/hv_pointpillars_fpn_sbn-all_2x8_2x_lyft-3d.py work_dirs/pp-lyft/latest.pth 8 --out work_dirs/pp-lyft/results_challenge.pkl --format-only --eval-options 'jsonfile_prefix=work_dirs/pp-lyft/results_challenge' 'csv_savepath=results/pp-lyft/results_challenge.csv'
./tools/dist_test.sh configs/pointpillars/pointpillars_hv_fpn_sbn-all_8xb2-2x_lyft-3d.py work_dirs/pp-lyft/latest.pth 8 --out work_dirs/pp-lyft/results_challenge.pkl --format-only --eval-options 'jsonfile_prefix=work_dirs/pp-lyft/results_challenge' 'csv_savepath=results/pp-lyft/results_challenge.csv'
```
After generating the `work_dirs/pp-lyft/results_challenge.csv`, you can submit it to the Kaggle evaluation server. Please refer to the [official website](https://www.kaggle.com/c/3d-object-detection-for-autonomous-vehicles) for more information.
......
docs/en/advanced_guides/datasets/nuscenes_det.md
View file @ f885d28a
......@@ -205,7 +205,7 @@ It follows the general pipeline of 2D detection while differs in some details:
An example to evaluate PointPillars with 8 GPUs with nuScenes metrics is as follows.
```shell
bash ./tools/dist_test.sh configs/pointpillars/hv_pointpillars_fpn_sbn-all_4x8_2x_nus-3d.py checkpoints/hv_pointpillars_fpn_sbn-all_4x8_2x_nus-3d_20200620_230405-2fa62f3d.pth 8 --eval bbox
bash ./tools/dist_test.sh configs/pointpillars/pointpillars_hv_fpn_sbn-all_8xb4-2x_nus-3d.py checkpoints/hv_pointpillars_fpn_sbn-all_4x8_2x_nus-3d_20200620_230405-2fa62f3d.pth 8 --eval bbox
```
## Metrics
......@@ -245,7 +245,7 @@ barrier 0.466 0.581 0.269 0.169 nan nan
An example to test PointPillars on nuScenes with 8 GPUs and generate a submission to the leaderboard is as follows.
```shell
./tools/dist_test.sh configs/pointpillars/hv_pointpillars_fpn_sbn-all_4x8_2x_nus-3d.py work_dirs/pp-nus/latest.pth 8 --out work_dirs/pp-nus/results_eval.pkl --format-only --eval-options 'jsonfile_prefix=work_dirs/pp-nus/results_eval'
./tools/dist_test.sh configs/pointpillars/pointpillars_hv_fpn_sbn-all_8xb4-2x_nus-3d.py work_dirs/pp-nus/latest.pth 8 --out work_dirs/pp-nus/results_eval.pkl --format-only --eval-options 'jsonfile_prefix=work_dirs/pp-nus/results_eval'
```
Note that the testing info should be changed to that for testing set instead of validation set [here](https://github.com/open-mmlab/mmdetection3d/blob/master/configs/_base_/datasets/nus-3d.py#L132).
......
docs/en/notes/benchmarks.md
View file @ f885d28a
......@@ -120,7 +120,7 @@ We compare the training speed (samples/s) with other codebases if they implement
- __MMDetection3D__: With release v0.1.0, run
```bash
./tools/dist_train.sh configs/votenet/votenet_16x8_sunrgbd-3d-10class.py 8 --no-validate
./tools/dist_train.sh configs/votenet/votenet_8xb16_sunrgbd-3d.py 8 --no-validate
```
- __votenet__: At commit [2f6d6d3](https://github.com/facebookresearch/votenet/tree/2f6d6d36ff98d96901182e935afe48ccee82d566), run
......
docs/en/user_guides/1_exist_data_model.md
View file @ f885d28a
......@@ -45,7 +45,7 @@ Assume that you have already downloaded the checkpoints to the directory `checkp
1. Test VoteNet on ScanNet and save the points and prediction visualization results.
```shell
python tools/test.py configs/votenet/votenet_8x8_scannet-3d-18class.py \
python tools/test.py configs/votenet/votenet_8xb8_scannet-3d.py \
checkpoints/votenet_8x8_scannet-3d-18class_20200620_230238-2cea9c3a.pth \
--show --show-dir ./data/scannet/show_results
```
......@@ -53,7 +53,7 @@ Assume that you have already downloaded the checkpoints to the directory `checkp
2. Test VoteNet on ScanNet, save the points, prediction, groundtruth visualization results, and evaluate the mAP.
```shell
python tools/test.py configs/votenet/votenet_8x8_scannet-3d-18class.py \
python tools/test.py configs/votenet/votenet_8xb8_scannet-3d.py \
checkpoints/votenet_8x8_scannet-3d-18class_20200620_230238-2cea9c3a.pth \
--eval mAP
--eval-options 'show=True' 'out_dir=./data/scannet/show_results'
......@@ -62,7 +62,7 @@ Assume that you have already downloaded the checkpoints to the directory `checkp
3. Test VoteNet on ScanNet (without saving the test results) and evaluate the mAP.
```shell
python tools/test.py configs/votenet/votenet_8x8_scannet-3d-18class.py \
python tools/test.py configs/votenet/votenet_8xb8_scannet-3d.py \
checkpoints/votenet_8x8_scannet-3d-18class_20200620_230238-2cea9c3a.pth \
--eval mAP
```
......@@ -78,7 +78,7 @@ Assume that you have already downloaded the checkpoints to the directory `checkp
5. Test PointPillars on nuScenes with 8 GPUs, and generate the json file to be submit to the official evaluation server.
```shell
./tools/slurm_test.sh ${PARTITION} ${JOB_NAME} configs/pointpillars/hv_pointpillars_fpn_sbn-all_4x8_2x_nus-3d.py \
./tools/slurm_test.sh ${PARTITION} ${JOB_NAME} configs/pointpillars/pointpillars_hv_fpn_sbn-all_8xb4-2x_nus-3d.py \
checkpoints/hv_pointpillars_fpn_sbn-all_4x8_2x_nus-3d_20200620_230405-2fa62f3d.pth \
--format-only --eval-options 'jsonfile_prefix=./pointpillars_nuscenes_results'
```
......@@ -198,7 +198,7 @@ If you run MMDetection3D on a cluster managed with [slurm](https://slurm.schedmd
Here is an example of using 16 GPUs to train Mask R-CNN on the dev partition.
```shell
GPUS=16 ./tools/slurm_train.sh dev pp_kitti_3class hv_pointpillars_secfpn_6x8_160e_kitti-3d-3class.py /nfs/xxxx/pp_kitti_3class
GPUS=16 ./tools/slurm_train.sh dev pp_kitti_3class pointpillars_hv_secfpn_8xb6-160e_kitti-3d-3class.py /nfs/xxxx/pp_kitti_3class
```
You can check [slurm_train.sh](https://github.com/open-mmlab/mmdetection/blob/master/tools/slurm_train.sh) for full arguments and environment variables.
......
docs/en/user_guides/config.md
View file @ f885d28a
......@@ -440,7 +440,7 @@ model = dict(
`FPN` and `SECONDFPN` use different keywords to construct.
```python
_base_ = '../_base_/models/hv_pointpillars_fpn_nus.py'
_base_ = '../_base_/models/pointpillars_hv_fpn_nus.py'
model = dict(
pts_neck=dict(
_delete_=True,
......
docs/en/user_guides/model_deployment.md
View file @ f885d28a
......@@ -55,7 +55,7 @@ python ./tools/deploy.py \
cd mmdeploy
python tools/deploy.py \
configs/mmdet3d/voxel-detection/voxel-detection_tensorrt_dynamic-kitti.py \
${$MMDET3D_DIR}/configs/pointpillars/hv_pointpillars_secfpn_6x8_160e_kitti-3d-3class.py \
${$MMDET3D_DIR}/configs/pointpillars/pointpillars_hv_secfpn_8xb6-160e_kitti-3d-3class.py \
${$MMDET3D_DIR}/checkpoints/hv_pointpillars_secfpn_6x8_160e_kitti-3d-3class_20200620_230421-aa0f3adb.pth \
${$MMDET3D_DIR}/demo/data/kitti/kitti_000008.bin \
--work-dir work-dir \
......@@ -102,7 +102,7 @@ python tools/test.py \
cd mmdeploy
python tools/test.py \
configs/mmdet3d/voxel-detection/voxel-detection_onnxruntime_dynamic.py \
${MMDET3D_DIR}/configs/centerpoint/centerpoint_02pillar_second_secfpn_circlenms_4x8_cyclic_20e_nus.py \
${MMDET3D_DIR}/configs/centerpoint/centerpoint_pillar02_second_secfpn_head-circlenms_8xb4-cyclic-20e_nus-3d.py \
--model work-dir/end2end.onnx \
--metrics bbox \
--device cpu
......
docs/en/user_guides/useful_tools.md
View file @ f885d28a
......@@ -107,7 +107,7 @@ python tools/misc/browse_dataset.py configs/_base_/datasets/kitti-3d-3class.py -
If you also want to show 2D images with 3D bounding boxes projected onto them, you need to find a config that supports multi-modality data loading, and then change the `--task` args to `multi_modality-det`. An example is showed below
```shell
python tools/misc/browse_dataset.py configs/mvxnet/dv_mvx-fpn_second_secfpn_adamw_2x8_80e_kitti-3d-3class.py --task multi_modality-det --output-dir ${OUTPUT_DIR} --online
python tools/misc/browse_dataset.py configs/mvxnet/mvxnet_fpn_dv_second_secfpn_8xb2-80e_kitti-3d-3class.py --task multi_modality-det --output-dir ${OUTPUT_DIR} --online
```
![](../../resources/browse_dataset_multi_modality.png)
......@@ -115,7 +115,7 @@ python tools/misc/browse_dataset.py configs/mvxnet/dv_mvx-fpn_second_secfpn_adam
You can simply browse different datasets using different configs, e.g. visualizing the ScanNet dataset in 3D semantic segmentation task
```shell
python tools/misc/browse_dataset.py configs/_base_/datasets/scannet_seg-3d-20class.py --task seg --output-dir ${OUTPUT_DIR} --online
python tools/misc/browse_dataset.py configs/_base_/datasets/scannet-seg.py --task seg --output-dir ${OUTPUT_DIR} --online
```
![](../../resources/browse_dataset_seg.png)
......
docs/zh_cn/advanced_guides/datasets/kitti_det.md
View file @ f885d28a
......@@ -159,7 +159,7 @@ train_pipeline = [
使用 8 个 GPU 以及 KITTI 指标评估的 PointPillars 的示例如下：
```shell
bash tools/dist_test.sh configs/pointpillars/hv_pointpillars_secfpn_6x8_160e_kitti-3d-3class.py work_dirs/hv_pointpillars_secfpn_6x8_160e_kitti-3d-3class/latest.pth 8 --eval bbox
bash tools/dist_test.sh configs/pointpillars/pointpillars_hv_secfpn_8xb6-160e_kitti-3d-3class.py work_dirs/hv_pointpillars_secfpn_6x8_160e_kitti-3d-3class/latest.pth 8 --eval bbox
```
## 度量指标
......@@ -188,7 +188,7 @@ aos AP:97.70, 89.11, 87.38
```shell
mkdir -p results/kitti-3class
./tools/dist_test.sh configs/pointpillars/hv_pointpillars_secfpn_6x8_160e_kitti-3d-3class.py work_dirs/hv_pointpillars_secfpn_6x8_160e_kitti-3d-3class/latest.pth 8 --out results/kitti-3class/results_eval.pkl --format-only --eval-options 'pklfile_prefix=results/kitti-3class/kitti_results' 'submission_prefix=results/kitti-3class/kitti_results'
./tools/dist_test.sh configs/pointpillars/pointpillars_hv_secfpn_8xb6-160e_kitti-3d-3class.py work_dirs/hv_pointpillars_secfpn_6x8_160e_kitti-3d-3class/latest.pth 8 --out results/kitti-3class/results_eval.pkl --format-only --eval-options 'pklfile_prefix=results/kitti-3class/kitti_results' 'submission_prefix=results/kitti-3class/kitti_results'
```
在生成 `results/kitti-3class/kitti_results/xxxxx.txt` 后，您可以提交这些文件到 KITTI 官方网站进行基准测试，请参考 [KITTI 官方网站](<(http://www.cvlibs.net/datasets/kitti/index.php)>)获取更多细节。
docs/zh_cn/advanced_guides/datasets/lyft_det.md
View file @ f885d28a
......@@ -151,7 +151,7 @@ train_pipeline = [
使用 8 个 GPU 以及 Lyft 指标评估的 PointPillars 的示例如下：
```shell
bash ./tools/dist_test.sh configs/pointpillars/hv_pointpillars_fpn_sbn-all_2x8_2x_lyft-3d.py checkpoints/hv_pointpillars_fpn_sbn-all_2x8_2x_lyft-3d_20210517_202818-fc6904c3.pth 8 --eval bbox
bash ./tools/dist_test.sh configs/pointpillars/pointpillars_hv_fpn_sbn-all_8xb2-2x_lyft-3d.py checkpoints/hv_pointpillars_fpn_sbn-all_2x8_2x_lyft-3d_20210517_202818-fc6904c3.pth 8 --eval bbox
```
## 度量指标
......@@ -186,7 +186,7 @@ Lyft 提出了一个更加严格的用以评估所预测的 3D 检测框的度
使用 8 个 GPU 在 Lyft 上测试 PointPillars 并生成对排行榜的提交的示例如下：
```shell
./tools/dist_test.sh configs/pointpillars/hv_pointpillars_fpn_sbn-all_2x8_2x_lyft-3d.py work_dirs/pp-lyft/latest.pth 8 --out work_dirs/pp-lyft/results_challenge.pkl --format-only --eval-options 'jsonfile_prefix=work_dirs/pp-lyft/results_challenge' 'csv_savepath=results/pp-lyft/results_challenge.csv'
./tools/dist_test.sh configs/pointpillars/pointpillars_hv_fpn_sbn-all_8xb2-2x_lyft-3d.py work_dirs/pp-lyft/latest.pth 8 --out work_dirs/pp-lyft/results_challenge.pkl --format-only --eval-options 'jsonfile_prefix=work_dirs/pp-lyft/results_challenge' 'csv_savepath=results/pp-lyft/results_challenge.csv'
```
在生成 `work_dirs/pp-lyft/results_challenge.csv`，您可以将生成的文件提交到 Kaggle 评估服务器，请参考[官方网址](https://www.kaggle.com/c/3d-object-detection-for-autonomous-vehicles)获取更多细节。
......
docs/zh_cn/advanced_guides/datasets/nuscenes_det.md
View file @ f885d28a
......@@ -202,7 +202,7 @@ train_pipeline = [
使用 8 个 GPU 以及 nuScenes 指标评估的 PointPillars 的示例如下
```shell
bash ./tools/dist_test.sh configs/pointpillars/hv_pointpillars_fpn_sbn-all_4x8_2x_nus-3d.py checkpoints/hv_pointpillars_fpn_sbn-all_4x8_2x_nus-3d_20200620_230405-2fa62f3d.pth 8 --eval bbox
bash ./tools/dist_test.sh configs/pointpillars/pointpillars_hv_fpn_sbn-all_8xb4-2x_nus-3d.py checkpoints/hv_pointpillars_fpn_sbn-all_4x8_2x_nus-3d_20200620_230405-2fa62f3d.pth 8 --eval bbox
```
## 指标
......@@ -242,7 +242,7 @@ barrier 0.466 0.581 0.269 0.169 nan nan
使用 8 个 GPU 在 nuScenes 上测试 PointPillars 并生成对排行榜的提交的示例如下
```shell
./tools/dist_test.sh configs/pointpillars/hv_pointpillars_fpn_sbn-all_4x8_2x_nus-3d.py work_dirs/hv_pointpillars_secfpn_6x8_160e_kitti-3d-3class/latest.pth 8 --out work_dirs/pp-nus/results_eval.pkl --format-only --eval-options 'jsonfile_prefix=work_dirs/pp-nus/results_eval'
./tools/dist_test.sh configs/pointpillars/pointpillars_hv_fpn_sbn-all_8xb4-2x_nus-3d.py work_dirs/hv_pointpillars_secfpn_6x8_160e_kitti-3d-3class/latest.pth 8 --out work_dirs/pp-nus/results_eval.pkl --format-only --eval-options 'jsonfile_prefix=work_dirs/pp-nus/results_eval'
```
请注意，在[这里](https://github.com/open-mmlab/mmdetection3d/blob/master/configs/_base_/datasets/nus-3d.py#L132)测试信息应更改为测试集而不是验证集。
......
docs/zh_cn/getting_started.md
View file @ f885d28a
......@@ -209,7 +209,7 @@ to_ply('./test.obj', './test.ply', 'obj')
```python
from mmdet3d.apis import init_model, inference_detector
config_file = 'configs/votenet/votenet_8x8_scannet-3d-18class.py'
config_file = 'configs/votenet/votenet_8xb8_scannet-3d.py'
checkpoint_file = 'checkpoints/votenet_8x8_scannet-3d-18class_20200620_230238-2cea9c3a.pth'
# 从配置文件和预训练的模型文件中构建模型
......
docs/zh_cn/notes/benchmarks.md
View file @ f885d28a
......@@ -119,7 +119,7 @@
- __MMDetection3D__：在 v0.1.0 版本下, 执行如下命令：
```bash
./tools/dist_train.sh configs/votenet/votenet_16x8_sunrgbd-3d-10class.py 8 --no-validate
./tools/dist_train.sh configs/votenet/votenet_8xb16_sunrgbd-3d.py 8 --no-validate
```
- __votenet__：在 commit [2f6d6d3](https://github.com/facebookresearch/votenet/tree/2f6d6d36ff98d96901182e935afe48ccee82d566) 版本下，执行如下命令：
......
docs/zh_cn/user_guides/1_exist_data_model.md
View file @ f885d28a
......@@ -45,7 +45,7 @@ python tools/test.py ${CONFIG_FILE} ${CHECKPOINT_FILE} [--out ${RESULT_FILE}] [-
1. 在 ScanNet 数据集上测试 VoteNet，保存模型，可视化预测结果
```shell
python tools/test.py configs/votenet/votenet_8x8_scannet-3d-18class.py \
python tools/test.py configs/votenet/votenet_8xb8_scannet-3d.py \
checkpoints/votenet_8x8_scannet-3d-18class_20200620_230238-2cea9c3a.pth \
--show --show-dir ./data/scannet/show_results
```
......@@ -53,7 +53,7 @@ python tools/test.py ${CONFIG_FILE} ${CHECKPOINT_FILE} [--out ${RESULT_FILE}] [-
2. 在 ScanNet 数据集上测试 VoteNet，保存模型，可视化预测结果，可视化真实标签，计算 mAP
```shell
python tools/test.py configs/votenet/votenet_8x8_scannet-3d-18class.py \
python tools/test.py configs/votenet/votenet_8xb8_scannet-3d.py \
checkpoints/votenet_8x8_scannet-3d-18class_20200620_230238-2cea9c3a.pth \
--eval mAP
--eval-options 'show=True' 'out_dir=./data/scannet/show_results'
......@@ -62,7 +62,7 @@ python tools/test.py ${CONFIG_FILE} ${CHECKPOINT_FILE} [--out ${RESULT_FILE}] [-
3. 在 ScanNet 数据集上测试 VoteNet（不保存测试结果），计算 mAP
```shell
python tools/test.py configs/votenet/votenet_8x8_scannet-3d-18class.py \
python tools/test.py configs/votenet/votenet_8xb8_scannet-3d.py \
checkpoints/votenet_8x8_scannet-3d-18class_20200620_230238-2cea9c3a.pth \
--eval mAP
```
......@@ -78,7 +78,7 @@ python tools/test.py ${CONFIG_FILE} ${CHECKPOINT_FILE} [--out ${RESULT_FILE}] [-
5. 使用8块显卡在 nuScenes 数据集上测试 PointPillars，生成提交给官方评测服务器的 json 文件
```shell
./tools/slurm_test.sh ${PARTITION} ${JOB_NAME} configs/pointpillars/hv_pointpillars_fpn_sbn-all_4x8_2x_nus-3d.py \
./tools/slurm_test.sh ${PARTITION} ${JOB_NAME} configs/pointpillars/pointpillars_hv_fpn_sbn-all_8xb4-2x_nus-3d.py \
checkpoints/hv_pointpillars_fpn_sbn-all_4x8_2x_nus-3d_20200620_230405-2fa62f3d.pth \
--format-only --eval-options 'jsonfile_prefix=./pointpillars_nuscenes_results'
```
......@@ -196,7 +196,7 @@ export CUDA_VISIBLE_DEVICES=-1
下面是一个使用16块显卡在 dev 分区上训练 Mask R-CNN 的示例：
```shell
GPUS=16 ./tools/slurm_train.sh dev pp_kitti_3class hv_pointpillars_secfpn_6x8_160e_kitti-3d-3class.py /nfs/xxxx/pp_kitti_3class
GPUS=16 ./tools/slurm_train.sh dev pp_kitti_3class pointpillars_hv_secfpn_8xb6-160e_kitti-3d-3class.py /nfs/xxxx/pp_kitti_3class
```
你可以查看 [slurm_train.sh](https://github.com/open-mmlab/mmdetection/blob/master/tools/slurm_train.sh) 来获取所有的参数和环境变量。
......
docs/zh_cn/user_guides/config.md
View file @ f885d28a
......@@ -441,7 +441,7 @@ model = dict(
`FPN` 和 `SECONDFPN` 使用不同的关键词来构建。
```python
_base_ = '../_base_/models/hv_pointpillars_fpn_nus.py'
_base_ = '../_base_/models/pointpillars_hv_fpn_nus.py'
model = dict(
pts_neck=dict(
_delete_=True,
......
docs/zh_cn/user_guides/demo.md
View file @ f885d28a
......@@ -27,7 +27,7 @@ python demo/pcd_demo.py demo/data/kitti/kitti_000008.bin configs/second/hv_secon
在 SUN RGB-D 数据上测试 [VoteNet](https://github.com/open-mmlab/mmdetection3d/tree/master/configs/votenet) 模型：
```shell
python demo/pcd_demo.py demo/data/sunrgbd/sunrgbd_000017.bin configs/votenet/votenet_16x8_sunrgbd-3d-10class.py checkpoints/votenet_16x8_sunrgbd-3d-10class_20200620_230238-4483c0c0.pth
python demo/pcd_demo.py demo/data/sunrgbd/sunrgbd_000017.bin configs/votenet/votenet_8xb16_sunrgbd-3d.py checkpoints/votenet_16x8_sunrgbd-3d-10class_20200620_230238-4483c0c0.pth
```
如果你正在使用的 mmdetection3d 版本 >= 0.6.0，记住转换 VoteNet 的模型权重文件，查看 [README](https://github.com/open-mmlab/mmdetection3d/blob/master/configs/votenet/README.md/) 来获取转换模型权重文件的详细说明。
......@@ -45,13 +45,13 @@ python demo/multi_modality_demo.py ${PCD_FILE} ${IMAGE_FILE} ${ANNOTATION_FILE}
在 KITTI 数据上测试 [MVX-Net](https://github.com/open-mmlab/mmdetection3d/tree/master/configs/mvxnet) 模型：
```shell
python demo/multi_modality_demo.py demo/data/kitti/kitti_000008.bin demo/data/kitti/kitti_000008.png demo/data/kitti/kitti_000008_infos.pkl configs/mvxnet/dv_mvx-fpn_second_secfpn_adamw_2x8_80e_kitti-3d-3class.py checkpoints/dv_mvx-fpn_second_secfpn_adamw_2x8_80e_kitti-3d-3class_20200621_003904-10140f2d.pth
python demo/multi_modality_demo.py demo/data/kitti/kitti_000008.bin demo/data/kitti/kitti_000008.png demo/data/kitti/kitti_000008_infos.pkl configs/mvxnet/mvxnet_fpn_dv_second_secfpn_8xb2-80e_kitti-3d-3class.py checkpoints/dv_mvx-fpn_second_secfpn_adamw_2x8_80e_kitti-3d-3class_20200621_003904-10140f2d.pth
```
在 SUN RGB-D 数据上测试 [ImVoteNet](https://github.com/open-mmlab/mmdetection3d/tree/master/configs/imvotenet) 模型：
```shell
python demo/multi_modality_demo.py demo/data/sunrgbd/sunrgbd_000017.bin demo/data/sunrgbd/sunrgbd_000017.jpg demo/data/sunrgbd/sunrgbd_000017_infos.pkl configs/imvotenet/imvotenet_stage2_16x8_sunrgbd-3d-10class.py checkpoints/imvotenet_stage2_16x8_sunrgbd-3d-10class_20210323_184021-d44dcb66.pth
python demo/multi_modality_demo.py demo/data/sunrgbd/sunrgbd_000017.bin demo/data/sunrgbd/sunrgbd_000017.jpg demo/data/sunrgbd/sunrgbd_000017_infos.pkl configs/imvotenet/imvotenet_stage2_8xb16_sunrgbd-3d.py checkpoints/imvotenet_stage2_16x8_sunrgbd-3d-10class_20210323_184021-d44dcb66.pth
```
### 单目 3D 检测
......@@ -67,7 +67,7 @@ python demo/mono_det_demo.py ${IMAGE_FILE} ${ANNOTATION_FILE} ${CONFIG_FILE} ${C
在 nuScenes 数据上测试 [FCOS3D](https://github.com/open-mmlab/mmdetection3d/tree/master/configs/fcos3d) 模型：
```shell
python demo/mono_det_demo.py demo/data/nuscenes/n015-2018-07-24-11-22-45+0800__CAM_BACK__1532402927637525.jpg demo/data/nuscenes/n015-2018-07-24-11-22-45+0800__CAM_BACK__1532402927637525_mono3d.coco.json configs/fcos3d/fcos3d_r101_caffe_fpn_gn-head_dcn_2x8_1x_nus-mono3d_finetune.py checkpoints/fcos3d_r101_caffe_fpn_gn-head_dcn_2x8_1x_nus-mono3d_finetune_20210717_095645-8d806dc2.pth
python demo/mono_det_demo.py demo/data/nuscenes/n015-2018-07-24-11-22-45+0800__CAM_BACK__1532402927637525.jpg demo/data/nuscenes/n015-2018-07-24-11-22-45+0800__CAM_BACK__1532402927637525_mono3d.coco.json configs/fcos3d/fcos3d_r101-caffe-dcn_fpn_head-gn_8xb2-1x_nus-mono3d_finetune.py checkpoints/fcos3d_r101_caffe_fpn_gn-head_dcn_2x8_1x_nus-mono3d_finetune_20210717_095645-8d806dc2.pth
```
### 3D 分割
......@@ -83,5 +83,5 @@ python demo/pc_seg_demo.py ${PCD_FILE} ${CONFIG_FILE} ${CHECKPOINT_FILE} [--devi
在 ScanNet 数据上测试 [PointNet++ (SSG)](https://github.com/open-mmlab/mmdetection3d/tree/master/configs/pointnet2) 模型：
```shell
python demo/pc_seg_demo.py demo/data/scannet/scene0000_00.bin configs/pointnet2/pointnet2_ssg_16x2_cosine_200e_scannet_seg-3d-20class.py checkpoints/pointnet2_ssg_16x2_cosine_200e_scannet_seg-3d-20class_20210514_143644-ee73704a.pth
python demo/pc_seg_demo.py demo/data/scannet/scene0000_00.bin configs/pointnet2/pointnet2_ssg_2xb16-cosine-200e_scannet-seg.py checkpoints/pointnet2_ssg_16x2_cosine_200e_scannet_seg-3d-20class_20210514_143644-ee73704a.pth
```
docs/zh_cn/user_guides/model_deployment.md
View file @ f885d28a
......@@ -55,7 +55,7 @@ python ./tools/deploy.py \
cd mmdeploy
python tools/deploy.py \
configs/mmdet3d/voxel-detection/voxel-detection_tensorrt_dynamic-kitti.py \
${$MMDET3D_DIR}/configs/pointpillars/hv_pointpillars_secfpn_6x8_160e_kitti-3d-3class.py \
${$MMDET3D_DIR}/configs/pointpillars/pointpillars_hv_secfpn_8xb6-160e_kitti-3d-3class.py \
${$MMDET3D_DIR}/checkpoints/hv_pointpillars_secfpn_6x8_160e_kitti-3d-3class_20200620_230421-aa0f3adb.pth \
${$MMDET3D_DIR}/demo/data/kitti/kitti_000008.bin \
--work-dir work-dir \
......@@ -102,7 +102,7 @@ python tools/test.py \
cd mmdeploy
python tools/test.py \
configs/mmdet3d/voxel-detection/voxel-detection_onnxruntime_dynamic.py \
${MMDET3D_DIR}/configs/centerpoint/centerpoint_02pillar_second_secfpn_circlenms_4x8_cyclic_20e_nus.py \
${MMDET3D_DIR}/configs/centerpoint/centerpoint_pillar02_second_secfpn_head-circlenms_8xb4-cyclic-20e_nus-3d.py \
--model work-dir/end2end.onnx \
--metrics bbox \
--device cpu
......
docs/zh_cn/user_guides/useful_tools.md
View file @ f885d28a
......@@ -105,7 +105,7 @@ python tools/misc/browse_dataset.py configs/_base_/datasets/kitti-3d-3class.py -
如果您还想显示 2D 图像以及投影的 3D 边界框，则需要找到支持多模态数据加载的配置文件，然后将 `--task` 参数更改为 `multi_modality-det`。一个例子如下所示
```shell
python tools/misc/browse_dataset.py configs/mvxnet/dv_mvx-fpn_second_secfpn_adamw_2x8_80e_kitti-3d-3class.py --task multi_modality-det --output-dir ${OUTPUT_DIR} --online
python tools/misc/browse_dataset.py configs/mvxnet/mvxnet_fpn_dv_second_secfpn_8xb2-80e_kitti-3d-3class.py --task multi_modality-det --output-dir ${OUTPUT_DIR} --online
```
![](../../resources/browse_dataset_multi_modality.png)
......@@ -113,7 +113,7 @@ python tools/misc/browse_dataset.py configs/mvxnet/dv_mvx-fpn_second_secfpn_adam
您可以简单的使用不同的配置文件，浏览不同的数据集，例如：在 3D 语义分割任务中可视化 ScanNet 数据集
```shell
python tools/misc/browse_dataset.py configs/_base_/datasets/scannet_seg-3d-20class.py --task seg --output-dir ${OUTPUT_DIR} --online
python tools/misc/browse_dataset.py configs/_base_/datasets/scannet-seg.py --task seg --output-dir ${OUTPUT_DIR} --online
```
![](../../resources/browse_dataset_seg.png)
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment