From 1ea2a74ba15f043ad47bfcab05f8514d9cd61669 Mon Sep 17 00:00:00 2001 From: wangguanzhong Date: Mon, 26 Aug 2019 16:50:33 +0800 Subject: [PATCH] Add full config (#3095) * Add full configs example & refine doc * refine docs --- README.md | 2 + README_cn.md | 2 + docs/CONFIG.md | 1 + docs/CONFIG_cn.md | 1 + docs/config_example/mask_rcnn_r50_fpn_1x.yml | 498 ++++++++++++++++++ ppdet/data/transform/operators.py | 2 +- ppdet/modeling/anchor_heads/rpn_head.py | 2 + .../modeling/roi_extractors/roi_extractor.py | 4 +- ppdet/modeling/roi_heads/mask_head.py | 2 +- ppdet/optimizer.py | 2 +- 10 files changed, 511 insertions(+), 5 deletions(-) create mode 100644 docs/config_example/mask_rcnn_r50_fpn_1x.yml diff --git a/README.md b/README.md index cad7dfb12..3bba80b61 100644 --- a/README.md +++ b/README.md @@ -85,6 +85,8 @@ python tools/infer.py -c configs/mask_rcnn_r50_1x.yml \ For detailed training and evaluation workflow, please refer to [GETTING_STARTED.md](docs/GETTING_STARTED.md). +For detailed configuration and parameter description, please refer to [Complete config files](docs/config_example/) + We also recommend users to take a look at the [IPython Notebook demo](demo/mask_rcnn_demo.ipynb) Further information can be found in these documentations: diff --git a/README_cn.md b/README_cn.md index d4a7f792e..5ac0e3030 100644 --- a/README_cn.md +++ b/README_cn.md @@ -72,6 +72,8 @@ python tools/infer.py -c configs/mask_rcnn_r50_1x.yml \ 更多训练及评估流程，请参考[GETTING_STARTED_cn.md](docs/GETTING_STARTED_cn.md). +详细的配置信息和参数说明，请参考[示例配置文件](docs/config_example/). + 同时推荐用户参考[IPython Notebook demo](demo/mask_rcnn_demo.ipynb) 其他更多信息可参考以下文档内容： diff --git a/docs/CONFIG.md b/docs/CONFIG.md index 0e44325d5..ea05b3978 100644 --- a/docs/CONFIG.md +++ b/docs/CONFIG.md @@ -157,6 +157,7 @@ LearningRate: steps: 500 ``` +[Complete config files](config_example/) of multiple detection architectures are given and brief description of each parameter. ## Requirements diff --git a/docs/CONFIG_cn.md b/docs/CONFIG_cn.md index a1412aad6..8b7eaa653 100644 --- a/docs/CONFIG_cn.md +++ b/docs/CONFIG_cn.md @@ -149,6 +149,7 @@ LearningRate: steps: 500 ``` +[示例配置文件](config_example/)中给出了多种检测结构的完整配置文件，以及其中各个超参的简要说明。 ## 安装依赖 diff --git a/docs/config_example/mask_rcnn_r50_fpn_1x.yml b/docs/config_example/mask_rcnn_r50_fpn_1x.yml new file mode 100644 index 000000000..50d38777a --- /dev/null +++ b/docs/config_example/mask_rcnn_r50_fpn_1x.yml @@ -0,0 +1,498 @@ +# Architecture of detection, which is also the prefix of data feed module +architecture: MaskRCNN + +# Data feed module +train_feed: MaskRCNNTrainFeed +eval_feed: MaskRCNNEvalFeed +test_feed: MaskRCNNTestFeed + +# Use GPU or CPU, true by default +use_gpu: true + +# Maximum number of iteration. +# In rcnn models, max_iters is 180000 if lr schedule is 1x and batch_size is 1. +max_iters: 180000 + +# Snapshot period. If training and test at same time, evaluate model at each snapshot_iter. 10000 by default. +snapshot_iter: 10000 + +# Smooth the log output in specified iterations, 20 by default. +log_smooth_window: 20 + +# The number of iteration interval to display in training log. +log_iter: 20 + +# The directory to save models. +save_dir: output + +# The path of oretrained wegiths. If url is provided, it will download the pretrain_weights and decompress automatically. +pretrain_weights: https://paddle-imagenet-models-name.bj.bcebos.com/ResNet50_cos_pretrained.tar + +# Evalution method, COCO and VOC are available. +metric: COCO + +# The path of final model for evaluation and test. +weights: output/mask_rcnn_r50_fpn_1x/model_final/ + +# Number of classes, 81 for COCO and 21 for VOC +num_classes: 81 + +# Mask RCNN architecture, see https://arxiv.org/abs/1703.06870 +MaskRCNN: + backbone: ResNet + fpn: FPN + roi_extractor: FPNRoIAlign + rpn_head: FPNRPNHead + bbox_assigner: BBoxAssigner + bbox_head: BBoxHead + mask_assigner: MaskAssigner + mask_head: MaskHead + rpn_only: false + +# Backbone module +ResNet: + # Index of stages using deformable conv v2, [] by default + dcn_v2_stages: [] + # ResNet depth, 50 by default + depth: 50 + # Stage index of returned feature map, [2,3,4,5] by default + feature_maps: + - 2 + - 3 + - 4 + - 5 + # Stage Index of backbone to freeze, 2 by default + freeze_at: 2 + # Whether freeze normalization layers, true by default + freeze_norm: true + # Weight decay for normalization layer weights, 0. by default + norm_decay: 0.0 + # Normalization type, bn/sync_bn/affine_channel, affine_channel by default + norm_type: affine_channel + # ResNet variant, supports 'a', 'b', 'c', 'd' currently, b by default + variant: b + +# FPN module +FPN: + # Whether has extra conv in higher levels, false by default + has_extra_convs: false + # Highest level of the backbone feature map to use, 6 by default + max_level: 6 + # Lowest level of the backbone feature map to use, 6 by default + min_level: 2 + # FPN normalization type, bn/sync_bn/affine_channel, null by default + norm_type: null + # Number of feature channels, 256 by default + num_chan: 256 + # Feature map scaling factors, [0.03125, 0.0625, 0.125, 0.25] by default + spatial_scale: + - 0.03125 + - 0.0625 + - 0.125 + - 0.25 + +# RPN module, if use non-FPN architecture, use RPNHead instead +# Extract proposals according to anchors and assign box targets and +# score targets to selected proposals to compute RPN loss. For FPN +# architecture, RPN is computed from each levels and collect proposals +# together. +FPNRPNHead: + # fluid.layers.anchor_generator + # Generate anchors for RCNN models. Each position of input produces + # N anchors. N = anchor_sizes * aspect_ratios. In FPNRPNHead, aspect_ratios + # is provided and anchor_sizes depends on FPN levels and anchor_start_size. + anchor_generator: + aspect_ratios: + - 0.5 + - 1.0 + - 2.0 + variance: + - 1.0 + - 1.0 + - 1.0 + - 1.0 + # fluid.layers.rpn_target_assign + # Assign classification and regression targets to each anchor according + # to Intersection-over-Union(IoU) overlap between anchors and ground + # truth boxes. The classification targets is binary class labels. the + # positive labels are two kinds of anchors: the anchors with the highest + # IoU overlap with a ground-truth box, or an anchor that has an IoU overlap + # higher than rpn_positive_overlap with any ground-truth box. + rpn_target_assign: + rpn_batch_size_per_im: 256 + rpn_fg_fraction: 0.5 + rpn_negative_overlap: 0.3 + rpn_positive_overlap: 0.7 + rpn_straddle_thresh: 0.0 + # fluid.layers.generate_proposals in training + # Generate RoIs according to each box with probability to be a foreground + # object. The operation performs following steps: Transposes and resizes + # scores and bbox_deltas; Calculate box locations as proposal candidates; + # Clip boxes to image; Remove predicted boxes with small area; Apply NMS to + # get final proposals as output. + train_proposal: + min_size: 0.0 + nms_thresh: 0.7 + post_nms_top_n: 2000 + pre_nms_top_n: 2000 + # fluid.layers.generate_proposals in test + test_proposal: + min_size: 0.0 + nms_thresh: 0.7 + post_nms_top_n: 1000 + pre_nms_top_n: 1000 + # Size of anchor at the first scale, 32 by default + anchor_start_size: 32 + # highest level of FPN output, 6 by default + max_level: 6 + # Lowest level of FPN output, 2 by default + min_level: 2 + # Number of FPN output channels, 256 by default + num_chan: 256 + # Number of classes in RPN output, 1 by default + num_classes: 1 + +# RoI extractor module, if use non-FPN architecture, use RoIAlign instead +# For FPN architecture, proposals are distributed to different levels and +# apply roi align at each level. Then concat the outputs. +FPNRoIAlign: + # The canconical FPN feature map level, 4 by default + canconical_level: 4 + # The canconical FPN feature map size, 224 by default + canonical_size: 224 + # The highest level of FPN layer, 5 by default + max_level: 5 + # The lowest level of FPN layer, 2 by default + min_level: 2 + # Number of sampling points, 0 by default + sampling_ratio: 2 + # Box resolution, 7 by default + box_resolution: 7 + # Mask RoI resolution, 14 by default + mask_resolution: 14 + +# Mask head module +# Generate mask output and compute loss mask. +MaskHead: + # Number of convolutions, 4 for FPN, 0 otherwise. 0 by default + num_convs: 4 + # size of the output mask, 14 by default + resolution: 28 + # Dilation rate, 1 by default + dilation: 1 + # Number of channels after first conv, 256 by default + num_chan_reduced: 256 + # Number of output classes, 81 by default + num_classes: 81 + +# fluid.layers.generate_proposal_labels +# Combine boxes and gt_boxes, and sample foreground proposals and background +# prosals.Then assign classification and regression targets to selected RoIs. +BBoxAssigner: + batch_size_per_im: 512 + bbox_reg_weights: + - 0.1 + - 0.1 + - 0.2 + - 0.2 + bg_thresh_hi: 0.5 + bg_thresh_lo: 0.0 + fg_fraction: 0.25 + fg_thresh: 0.5 + num_classes: 81 + shuffle_before_sample: true + +# fluid.layers.generate_mask_labels +# For given the RoIs and corresponding labels, sample foreground RoIs. +# Assign mask targets to selected RoIs which are encoded to K binary masks +# of resolution M x M. +MaskAssigner: + resolution: 28 + num_classes: 81 + +# BBox head module +# Faster bbox head following the RoI extractor, and apply post process, such as +# NMS and box coder.. +BBoxHead: + # Head after RoI extractor, ResNetC5/TwoFCHead + head: TwoFCHead + # fluid.layers.multiclass_nms + # Select a subset of detection bounding boxes that have high scores larger + # than score_threshold. Then prune away boxes that have high IoU overlap + # with already selected boxes by nms_threshold. + keep_top_k: 100 + nms_threshold: 0.5 + score_threshold: 0.05 + # fluid.layers.box_coder + box_coder: + axis: 1 + box_normalized: false + code_type: decode_center_size + prior_box_var: + - 0.1 + - 0.1 + - 0.2 + - 0.2 + num_classes: 81 + +# RCNN head with two Fully Connected layers +TwoFCHead: + # The number of output channels, 1024 by default + num_chan: 1024 + +# Learning rate configuration +LearningRate: + # Base learning rate, 0.01 by default + base_lr: 0.01 + # Learning rate schedulers, PiecewiseDecay and LinearWarmup by default + schedulers: + # fluid.layers.piecewise_decay + # Values has higher priority and if values is null, learning rate is multipled by gamma at each stage + - !PiecewiseDecay + gamma: 0.1 + milestones: + - 120000 + - 160000 + values: null + # fluid.layers.linear_lr_warmup + # Start learning rate equals to base_lr * start_factor + - !LinearWarmup + start_factor: 0.3333333333333333 + steps: 500 + +# Optimizer module +OptimizerBuilder: + # fluid.optimizer + optimizer: + momentum: 0.9 + type: Momentum + # fluid.regularizer + regularizer: + factor: 0.0001 + type: L2 + +# Data feed module for training +MaskRCNNTrainFeed: + # Batch size per device, 1 by default + batch_size: 1 + # Dataset module + dataset: + # Annotation file path + annotation: annotations/instances_train2017.json + # Dataset directory + dataset_dir: dataset/coco + # Directory where image files are stored + image_dir: train2017 + # List of data fields needed + fields: + - image + - im_info + - im_id + - gt_box + - gt_label + - is_crowd + - gt_mask + # list of image dims + image_shape: + - 3 + - 800 + - 1333 + # List of sample transformations to use + sample_transforms: + # Transform the image data to numpy format. + - !DecodeImage + to_rgb: true # default: true + with_mixup: false # default: false + # Flip images randomly + # Transform the x coordinates of bboxes and segmentations + - !RandomFlipImage + is_mask_flip: true # default: false + # Whether bbox is normalized + is_normalized: false # default: false + prob: 0.5 # default: 0.5 + # Normalize the image + - !NormalizeImage + # The format of image, [H, W, C]/[C, H, W], true by default + is_channel_first: false + # Whether divide by 255, true by default + is_scale: true + # default: [0.485, 0.456, 0.406] + mean: + - 0.485 + - 0.456 + - 0.406 + # default: [1, 1, 1] + std: + - 0.229 + - 0.224 + - 0.225 + # Rescale image to the specified target size, and capped at max_size + - !ResizeImage + # Resize method, cv2.INTER_LINEAR(1) by default + interp: 1 + max_size: 1333 + target_size: 800 + use_cv2: true # default: true + # Change the channel + - !Permute + # The format of image, [H, W, C]/[C, H, W], true by default + channel_first: true + to_bgr: false # default: true + # List of batch transformations to use + batch_transforms: + # Pad a batch of samples to same dimensions + - !PadBatch + pad_to_stride: 32 # default: 32 + # Drop last batch if size is uneven, false by default + drop_last: false + # Number of workers processes(or threads), 2 by default + num_workers: 2 + # Number of samples, -1 represents all samples. -1 by default + samples: -1 + # If samples should be shuffled, true by default + shuffle: true + # If update im_info after padding, false by default + use_padded_im_info: false + # If use multi-process, false by default + use_process: false + +# Data feed module for test +MaskRCNNEvalFeed: + # Batch size per device, 1 by default + batch_size: 1 + # Dataset module + dataset: + # Annotation file path + annotation: annotations/instances_val2017.json + # Dataset directory + dataset_dir: dataset/coco + # Directory where image files are stored + image_dir: val2017 + # List of data fields needed + fields: + - image + - im_info + - im_id + - im_shape + # list of image dims + image_shape: + - 3 + - 800 + - 1333 + # List of sample transformations to use + sample_transforms: + # Transform the image data to numpy format. + - !DecodeImage + to_rgb: true # default: true + with_mixup: false # default: false + # Normalize the image + - !NormalizeImage + # The format of image, [H, W, C]/[C, H, W], true by default + is_channel_first: false + # Whether divide by 255, true by default + is_scale: true + # default: [0.485, 0.456, 0.406] + mean: + - 0.485 + - 0.456 + - 0.406 + # default: [1, 1, 1] + std: + - 0.229 + - 0.224 + - 0.225 + # Rescale image to the specified target size, and capped at max_size + - !ResizeImage + # Resize method, cv2.INTER_LINEAR(1) by default + interp: 1 + max_size: 1333 + target_size: 800 + use_cv2: true # default: true + # Change the channel + - !Permute + # The format of image, [H, W, C]/[C, H, W], true by default + channel_first: true + to_bgr: false # default: true + # List of batch transformations to use + batch_transforms: + # Pad a batch of samples to same dimensions + - !PadBatch + pad_to_stride: 32 # default: 32 + # Drop last batch if size is uneven, false by default + drop_last: false + # Number of workers processes(or threads), 2 by default + num_workers: 2 + # Number of samples, -1 represents all samples. -1 by default + samples: -1 + # If samples should be shuffled, true by default + shuffle: false + # If update im_info after padding, false by default + use_padded_im_info: true + # If use multi-process, false by default + use_process: false + +# Data feed module for test +MaskRCNNTestFeed: + # Batch size per device, 1 by default + batch_size: 1 + # Dataset module + dataset: + # Annotation file path + annotation: dataset/coco/annotations/instances_val2017.json + # List of data fields needed + fields: + - image + - im_info + - im_id + - im_shape + # list of image dims + image_shape: + - 3 + - 800 + - 1333 + # List of sample transformations to use + sample_transforms: + # Transform the image data to numpy format. + - !DecodeImage + to_rgb: true # default: true + with_mixup: false # default: false + # Normalize the image + - !NormalizeImage + # The format of image, [H, W, C]/[C, H, W], true by default + is_channel_first: false + # Whether divide by 255, true by default + is_scale: true + # default: [0.485, 0.456, 0.406] + mean: + - 0.485 + - 0.456 + - 0.406 + # default: [1, 1, 1] + std: + - 0.229 + - 0.224 + - 0.225 + # Change the channel + - !Permute + # The format of image, [H, W, C]/[C, H, W], true by default + channel_first: true + to_bgr: false # default: true + # List of batch transformations to use + batch_transforms: + # Pad a batch of samples to same dimensions + - !PadBatch + pad_to_stride: 32 # default: 32 + # Drop last batch if size is uneven, false by default + drop_last: false + # Number of workers processes(or threads), 2 by default + num_workers: 2 + # Number of samples, -1 represents all samples. -1 by default + samples: -1 + # If samples should be shuffled, true by default + shuffle: false + # If update im_info after padding, false by default + use_padded_im_info: true + # If use multi-process, false by default + use_process: false + + diff --git a/ppdet/data/transform/operators.py b/ppdet/data/transform/operators.py index 6b430b165..8cb4830e5 100644 --- a/ppdet/data/transform/operators.py +++ b/ppdet/data/transform/operators.py @@ -154,7 +154,7 @@ class ResizeImage(BaseOperator): raise TypeError("{}: input type is invalid.".format(self)) def __call__(self, sample, context=None): - """ Resise the image numpy. + """ Resize the image numpy. """ im = sample['image'] if not isinstance(im, np.ndarray): diff --git a/ppdet/modeling/anchor_heads/rpn_head.py b/ppdet/modeling/anchor_heads/rpn_head.py index 527fb948d..876aafe36 100644 --- a/ppdet/modeling/anchor_heads/rpn_head.py +++ b/ppdet/modeling/anchor_heads/rpn_head.py @@ -38,6 +38,7 @@ class RPNHead(object): rpn_target_assign (object): `RPNTargetAssign` instance train_proposal (object): `GenerateProposals` instance for training test_proposal (object): `GenerateProposals` instance for testing + num_classes (int): number of classes in rpn output """ __inject__ = [ 'anchor_generator', 'rpn_target_assign', 'train_proposal', @@ -281,6 +282,7 @@ class FPNRPNHead(RPNHead): num_chan (int): number of FPN output channels min_level (int): lowest level of FPN output max_level (int): highest level of FPN output + num_classes (int): number of classes in rpn output """ __inject__ = [ diff --git a/ppdet/modeling/roi_extractors/roi_extractor.py b/ppdet/modeling/roi_extractors/roi_extractor.py index 085e7f9e8..1caf3936f 100644 --- a/ppdet/modeling/roi_extractors/roi_extractor.py +++ b/ppdet/modeling/roi_extractors/roi_extractor.py @@ -29,13 +29,13 @@ class FPNRoIAlign(object): """ RoI align pooling for FPN feature maps Args: - pooled_height (int): output height - pooled_height (int): output width sampling_ratio (int): number of sampling points min_level (int): lowest level of FPN layer max_level (int): highest level of FPN layer canconical_level (int): the canconical FPN feature map level canonical_size (int): the canconical FPN feature map size + box_resolution (int): box resolution + mask_resolution (int): mask roi resolution """ def __init__(self, diff --git a/ppdet/modeling/roi_heads/mask_head.py b/ppdet/modeling/roi_heads/mask_head.py index ad59de00e..e7b313e1f 100644 --- a/ppdet/modeling/roi_heads/mask_head.py +++ b/ppdet/modeling/roi_heads/mask_head.py @@ -31,7 +31,7 @@ class MaskHead(object): """ RCNN mask head Args: - num_convs (int): num of convolutions, 4 for FPN, 1 otherwise + num_convs (int): num of convolutions, 4 for FPN, 0 otherwise num_chan_reduced (int): num of channels after first convolution resolution (int): size of the output mask dilation (int): dilation rate diff --git a/ppdet/optimizer.py b/ppdet/optimizer.py index dff6c83dc..e695aad04 100644 --- a/ppdet/optimizer.py +++ b/ppdet/optimizer.py @@ -40,7 +40,7 @@ class PiecewiseDecay(object): milestones (list): steps at which to decay learning rate """ - def __init__(self, gamma=0.1, milestones=[6000, 8000], values=None): + def __init__(self, gamma=0.1, milestones=[60000, 80000], values=None): super(PiecewiseDecay, self).__init__() self.gamma = gamma self.milestones = milestones -- GitLab