Skip to content

Latest commit

 

History

History
282 lines (199 loc) · 8.18 KB

GETTING_STARTED.md

File metadata and controls

282 lines (199 loc) · 8.18 KB

Getting Started

This page provides basic tutorials about the usage of mmdetection. For installation instructions, please see INSTALL.md.

Prepare DOTA dataset.

It is recommended to symlink the dataset root to AerialDetection/data.

Here, we give an example for single scale data preparation of DOTA-v1.0.

First, make sure your initial data are in the following structure.

data/dota
├── train
│   ├──images
│   └── labelTxt
├── val
│   ├── images
│   └── labelTxt
└── test
    └── images

Split the original images and create COCO format json.

python DOTA_devkit/prepare_dota1.py --srcpath path_to_dota --dstpath path_to_split_1024

Then you will get data in the following structure

dota1_1024
├── test1024
│   ├── DOTA_test1024.json
│   └── images
└── trainval1024
     ├── DOTA_trainval1024.json
     └── images

For data preparation with data augmentation, refer to "DOTA_devkit/prepare_dota1_aug.py"

For data preparation of dota1.5, refer to "DOTA_devkit/prepare_dota1_5.py" and "DOTA_devkit/prepare_dota1_5_aug.py"

Inference with pretrained models

Test a dataset

  • single GPU testing
  • multiple GPU testing

You can use the following commands to test a dataset.

# single-gpu testing
python tools/test.py ${CONFIG_FILE} ${CHECKPOINT_FILE} [--out ${RESULT_FILE}]

# multi-gpu testing
./tools/dist_test.sh ${CONFIG_FILE} ${CHECKPOINT_FILE} ${GPU_NUM} [--out ${RESULT_FILE}]

Optional arguments:

  • RESULT_FILE: Filename of the output results in pickle format. If not specified, the results will not be saved to a file.

Examples:

Assume that you have already downloaded the checkpoints to work_dirs/.

  1. Test Faster R-CNN.
python tools/test.py configs/DOTA/faster_rcnn_RoITrans_r50_fpn_1x_dota.py \
    work_dirs/faster_rcnn_RoITrans_r50_fpn_1x_dota/epoch_12.pth \ 
    --out work_dirs/faster_rcnn_RoITrans_r50_fpn_1x_dota/results.pkl
  1. Test Mask R-CNN with 4 GPUs.
./tools/dist_test.sh configs/DOTA/mask_rcnn_r50_fpn_1x_dota.py \
    work_dirs/mask_rcnn_r50_fpn_1x_dota/epoch_12.pth \
    4 --out work_dirs/mask_rcnn_r50_fpn_1x_dota/results.pkl 
  1. Parse the results.pkl to the format needed for DOTA evaluation

For methods with only OBB Head, set the type OBB.

python tools/parse_results.py --config configs/DOTA/faster_rcnn_RoITrans_r50_fpn_1x_dota.py --type OBB

For methods with both OBB and HBB Head, set the type HBBOBB.

python tools/parse_results.py --config configs/DOTA/faster_rcnn_h-obb_r50_fpn_1x_dota.py --type OBB

For methods with HBB and Mask Head, set the type Mask

python tools/parse_results.py --config configs/DOTA/mask_rcnn_r50_fpn_1x_dota.py --type Mask

For methods with only HBB Head, se the type HBB

python tools/parse_results.py --config configs/DOTA/faster_rcnn_r50_fpn_1x_dota.py --type HBB

Demo of inference in a large size image.

python demo_large_image.py

Train a model

mmdetection implements distributed training and non-distributed training, which uses MMDistributedDataParallel and MMDataParallel respectively.

All outputs (log files and checkpoints) will be saved to the working directory, which is specified by work_dir in the config file.

*Important*: The default learning rate in config files is for 8 GPUs. If you use less or more than 8 GPUs, you need to set the learning rate proportional to the GPU num, e.g., 0.01 for 4 GPUs and 0.04 for 16 GPUs.

Train with a single GPU

python tools/train.py ${CONFIG_FILE}

If you want to specify the working directory in the command, you can add an argument --work_dir ${YOUR_WORK_DIR}.

Train with multiple GPUs

./tools/dist_train.sh ${CONFIG_FILE} ${GPU_NUM} [optional arguments]

Optional arguments are:

  • --validate (recommended): Perform evaluation at every k (default=1) epochs during the training.
  • --work_dir ${WORK_DIR}: Override the working directory specified in the config file.
  • --resume_from ${CHECKPOINT_FILE}: Resume from a previous checkpoint file.

Train with multiple machines

If you run mmdetection on a cluster managed with slurm, you can just use the script slurm_train.sh.

./tools/slurm_train.sh ${PARTITION} ${JOB_NAME} ${CONFIG_FILE} ${WORK_DIR} [${GPUS}]

Here is an example of using 16 GPUs to train Mask R-CNN on the dev partition.

./tools/slurm_train.sh dev mask_r50_1x configs/mask_rcnn_r50_fpn_1x.py /nfs/xxxx/mask_rcnn_r50_fpn_1x 16

You can check slurm_train.sh for full arguments and environment variables.

If you have just multiple machines connected with ethernet, you can refer to pytorch launch utility. Usually it is slow if you do not have high speed networking like infiniband.

How-to

Use my own datasets

The simplest way is to convert your dataset to existing dataset formats (COCO or PASCAL VOC).

Here we show an example of adding a custom dataset of 5 classes, assuming it is also in COCO format.

In mmdet/datasets/my_dataset.py:

from .coco import CocoDataset


class MyDataset(CocoDataset):

    CLASSES = ('a', 'b', 'c', 'd', 'e')

In mmdet/datasets/__init__.py:

from .my_dataset import MyDataset

Then you can use MyDataset in config files, with the same API as CocoDataset.

It is also fine if you do not want to convert the annotation format to COCO or PASCAL format. Actually, we define a simple annotation format and all existing datasets are processed to be compatible with it, either online or offline.

The annotation of a dataset is a list of dict, each dict corresponds to an image. There are 3 field filename (relative path), width, height for testing, and an additional field ann for training. ann is also a dict containing at least 2 fields: bboxes and labels, both of which are numpy arrays. Some datasets may provide annotations like crowd/difficult/ignored bboxes, we use bboxes_ignore and labels_ignore to cover them.

Here is an example.

[
    {
        'filename': 'a.jpg',
        'width': 1280,
        'height': 720,
        'ann': {
            'bboxes': <np.ndarray, float32> (n, 4),
            'labels': <np.ndarray, float32> (n, ),
            'bboxes_ignore': <np.ndarray, float32> (k, 4),
            'labels_ignore': <np.ndarray, float32> (k, ) (optional field)
        }
    },
    ...
]

There are two ways to work with custom datasets.

  • online conversion

    You can write a new Dataset class inherited from CustomDataset, and overwrite two methods load_annotations(self, ann_file) and get_ann_info(self, idx), like CocoDataset and VOCDataset.

  • offline conversion

    You can convert the annotation format to the expected format above and save it to a pickle or json file, like pascal_voc.py. Then you can simply use CustomDataset.

Develop new components

We basically categorize model components into 4 types.

  • backbone: usually a FCN network to extract feature maps, e.g., ResNet, MobileNet.
  • neck: the component between backbones and heads, e.g., FPN, PAFPN.
  • head: the component for specific tasks, e.g., bbox prediction and mask prediction.
  • roi extractor: the part for extracting RoI features from feature maps, e.g., RoI Align.

Here we show how to develop new components with an example of MobileNet.

  1. Create a new file mmdet/models/backbones/mobilenet.py.
import torch.nn as nn

from ..registry import BACKBONES


@BACKBONES.register
class MobileNet(nn.Module):

    def __init__(self, arg1, arg2):
        pass

    def forward(x):  # should return a tuple
        pass
  1. Import the module in mmdet/models/backbones/__init__.py.
from .mobilenet import MobileNet
  1. Use it in your config file.
model = dict(
    ...
    backbone=dict(
        type='MobileNet',
        arg1=xxx,
        arg2=xxx),
    ...

For more information on how it works, you can refer to TECHNICAL_DETAILS.md (TODO).