forked from traveller59/second.pytorch
-
Notifications
You must be signed in to change notification settings - Fork 0
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
1. fix pp config: wrong ped/cyc anchor center
2. add a fast-training nuscenes config
- Loading branch information
1 parent
c2f9674
commit 4040cf7
Showing
8 changed files
with
357 additions
and
12 deletions.
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
|
@@ -44,3 +44,55 @@ Use ```set_train_step``` in utils.config_tool.train if you don't want to calcula | |
|
||
* use code such as code in script_server.py instead of use commands in terminal. | ||
|
||
|
||
## Reference Performance | ||
|
||
* all.pp.lowa.config: 30 epoch, 1/2 dataset, train speed: 12 sample/s | ||
|
||
``` | ||
car Nusc dist [email protected], 1.0, 2.0, 4.0 | ||
58.85, 76.12, 80.65, 82.49 | ||
bicycle Nusc dist [email protected], 1.0, 2.0, 4.0 | ||
0.00, 0.00, 0.00, 0.00 | ||
bus Nusc dist [email protected], 1.0, 2.0, 4.0 | ||
2.55, 15.42, 27.19, 32.03 | ||
construction_vehicle Nusc dist [email protected], 1.0, 2.0, 4.0 | ||
0.00, 0.00, 0.02, 0.31 | ||
motorcycle Nusc dist [email protected], 1.0, 2.0, 4.0 | ||
8.61, 14.30, 15.00, 15.53 | ||
pedestrian Nusc dist [email protected], 1.0, 2.0, 4.0 | ||
39.14, 49.29, 53.50, 57.03 | ||
traffic_cone Nusc dist [email protected], 1.0, 2.0, 4.0 | ||
12.58, 18.92, 22.79, 27.99 | ||
trailer Nusc dist [email protected], 1.0, 2.0, 4.0 | ||
0.00, 1.10, 7.42, 20.91 | ||
truck Nusc dist [email protected], 1.0, 2.0, 4.0 | ||
5.44, 15.78, 22.77, 27.05 | ||
barrier Nusc dist [email protected], 1.0, 2.0, 4.0 | ||
7.54, 34.54, 44.52, 49.80 | ||
``` | ||
|
||
* all.pp.config: 50 epoch, 1/8 dataset, train speed: 4 sample/s | ||
|
||
``` | ||
car Nusc dist [email protected], 1.0, 2.0, 4.0 | ||
58.85, 76.12, 80.65, 82.49 | ||
bicycle Nusc dist [email protected], 1.0, 2.0, 4.0 | ||
0.00, 0.00, 0.00, 0.00 | ||
bus Nusc dist [email protected], 1.0, 2.0, 4.0 | ||
2.55, 15.42, 27.19, 32.03 | ||
construction_vehicle Nusc dist [email protected], 1.0, 2.0, 4.0 | ||
0.00, 0.00, 0.02, 0.31 | ||
motorcycle Nusc dist [email protected], 1.0, 2.0, 4.0 | ||
8.61, 14.30, 15.00, 15.53 | ||
pedestrian Nusc dist [email protected], 1.0, 2.0, 4.0 | ||
39.14, 49.29, 53.50, 57.03 | ||
traffic_cone Nusc dist [email protected], 1.0, 2.0, 4.0 | ||
12.58, 18.92, 22.79, 27.99 | ||
trailer Nusc dist [email protected], 1.0, 2.0, 4.0 | ||
0.00, 1.10, 7.42, 20.91 | ||
truck Nusc dist [email protected], 1.0, 2.0, 4.0 | ||
5.44, 15.78, 22.77, 27.05 | ||
barrier Nusc dist [email protected], 1.0, 2.0, 4.0 | ||
7.54, 34.54, 44.52, 49.80 | ||
``` |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,293 @@ | ||
model: { | ||
second: { | ||
voxel_generator { | ||
point_cloud_range : [-50, -50, -5, 50, 50, 3] | ||
voxel_size : [0.25, 0.25, 8] | ||
max_number_of_points_per_voxel : 60 | ||
} | ||
voxel_feature_extractor: { | ||
module_class_name: "PillarFeatureNet" | ||
num_filters: [64] | ||
with_distance: false | ||
num_input_features: 4 | ||
} | ||
middle_feature_extractor: { | ||
module_class_name: "PointPillarsScatter" | ||
downsample_factor: 1 | ||
num_input_features: 64 | ||
} | ||
rpn: { | ||
module_class_name: "RPNV2" | ||
layer_nums: [3, 5, 5] | ||
layer_strides: [2, 2, 2] | ||
num_filters: [64, 128, 256] | ||
upsample_strides: [0.25, 0.5, 1] | ||
num_upsample_filters: [128, 128, 128] | ||
use_groupnorm: false | ||
num_groups: 32 | ||
num_input_features: 64 | ||
} | ||
loss: { | ||
classification_loss: { | ||
weighted_sigmoid_focal: { | ||
alpha: 0.25 | ||
gamma: 2.0 | ||
anchorwise_output: true | ||
} | ||
} | ||
localization_loss: { | ||
weighted_smooth_l1: { | ||
sigma: 3.0 | ||
code_weight: [1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0] | ||
} | ||
} | ||
classification_weight: 1.0 | ||
localization_weight: 2.0 | ||
} | ||
num_point_features: 4 # model's num point feature should be independent of dataset | ||
# Outputs | ||
use_sigmoid_score: true | ||
encode_background_as_zeros: true | ||
encode_rad_error_by_sin: true | ||
|
||
use_direction_classifier: true | ||
direction_loss_weight: 0.2 | ||
|
||
# Loss | ||
pos_class_weight: 1.0 | ||
neg_class_weight: 1.0 | ||
|
||
loss_norm_type: NormByNumPositives | ||
# Postprocess | ||
post_center_limit_range: [-59.6, -59.6, -10, 59.6, 59.6, 10] | ||
use_rotate_nms: false | ||
use_multi_class_nms: false | ||
nms_pre_max_size: 1000 | ||
nms_post_max_size: 300 | ||
nms_score_threshold: 0.05 | ||
nms_iou_threshold: 0.5 | ||
|
||
box_coder: { | ||
ground_box3d_coder: { | ||
linear_dim: false | ||
encode_angle_vector: false | ||
} | ||
} | ||
target_assigner: { | ||
anchor_generators: { | ||
anchor_generator_range: { | ||
sizes: [1.95017717, 4.60718145, 1.72270761] # wlh | ||
anchor_ranges: [-50, -50, -0.93897414, 50, 50, -0.93897414] | ||
rotations: [0, 1.57] # DON'T modify this unless you are very familiar with my code. | ||
matched_threshold : 0.25 | ||
unmatched_threshold : 0.2 | ||
class_name: "car" | ||
} | ||
region_similarity_calculator: { | ||
nearest_iou_similarity: {} | ||
} | ||
} | ||
anchor_generators: { | ||
anchor_generator_range: { | ||
sizes: [0.60058911, 1.68452161, 1.27192197] # wlh | ||
anchor_ranges: [-50, -50, -1.03743013, 50, 50, -1.03743013] | ||
rotations: [0, 1.57] # DON'T modify this unless you are very familiar with my code. | ||
matched_threshold : 0.2 | ||
unmatched_threshold : 0.15 | ||
class_name: "bicycle" | ||
} | ||
region_similarity_calculator: { | ||
nearest_iou_similarity: {} | ||
} | ||
} | ||
anchor_generators: { | ||
anchor_generator_range: { | ||
sizes: [2.94046906, 11.1885991, 3.47030982] # wlh | ||
anchor_ranges: [-50, -50, -0.0715754, 50, 50, -0.0715754] | ||
rotations: [0, 1.57] # DON'T modify this unless you are very familiar with my code. | ||
matched_threshold : 0.5 | ||
unmatched_threshold : 0.35 | ||
class_name: "bus" | ||
} | ||
region_similarity_calculator: { | ||
nearest_iou_similarity: {} | ||
} | ||
} | ||
anchor_generators: { | ||
anchor_generator_range: { | ||
sizes: [2.73050468, 6.38352896, 3.13312415] # wlh | ||
anchor_ranges: [-50, -50, -0.08168083, 50, 50, -0.08168083] | ||
rotations: [0, 1.57] # DON'T modify this unless you are very familiar with my code. | ||
matched_threshold : 0.4 | ||
unmatched_threshold : 0.3 | ||
class_name: "construction_vehicle" | ||
} | ||
region_similarity_calculator: { | ||
nearest_iou_similarity: {} | ||
} | ||
} | ||
anchor_generators: { | ||
anchor_generator_range: { | ||
sizes: [0.76279481, 2.09973778, 1.44403034] # wlh | ||
anchor_ranges: [-50, -50, -0.99194854, 50, 50, -0.99194854] | ||
rotations: [0, 1.57] # DON'T modify this unless you are very familiar with my code. | ||
matched_threshold : 0.2 | ||
unmatched_threshold : 0.15 | ||
class_name: "motorcycle" | ||
} | ||
region_similarity_calculator: { | ||
nearest_iou_similarity: {} | ||
} | ||
} | ||
anchor_generators: { | ||
anchor_generator_range: { | ||
sizes: [0.66344886, 0.7256437, 1.75748069] # wlh | ||
anchor_ranges: [-50, -50, -0.73911038, 50, 50, -0.73911038] | ||
rotations: [0] # DON'T modify this unless you are very familiar with my code. | ||
matched_threshold : 0.5 | ||
unmatched_threshold : 0.35 | ||
class_name: "pedestrian" | ||
} | ||
region_similarity_calculator: { | ||
distance_similarity: { | ||
distance_norm: 1.414 # match range | ||
with_rotation: false | ||
rotation_alpha: 0.0 # rot error contribution | ||
} | ||
} | ||
} | ||
anchor_generators: { | ||
anchor_generator_range: { | ||
sizes: [0.39694519, 0.40359262, 1.06232151] # wlh | ||
anchor_ranges: [-50, -50, -1.27868911, 50, 50, -1.27868911] | ||
rotations: [0] # DON'T modify this unless you are very familiar with my code. | ||
matched_threshold : 0.5 | ||
unmatched_threshold : 0.35 | ||
class_name: "traffic_cone" | ||
} | ||
region_similarity_calculator: { | ||
distance_similarity: { | ||
distance_norm: 1.414 # match range | ||
with_rotation: false | ||
rotation_alpha: 0.0 # rot error contribution | ||
} | ||
} | ||
} | ||
anchor_generators: { | ||
anchor_generator_range: { | ||
# sizes: [2.87427237, 12.01320693, 3.81509561] # wlh | ||
sizes: [3, 15, 3.8, 2, 3, 3.8] | ||
anchor_ranges: [-50, -50, 0.22228277, 50, 50, 0.22228277] | ||
rotations: [0, 1.57] # DON'T modify this unless you are very familiar with my code. | ||
matched_threshold : 0.5 | ||
unmatched_threshold : 0.35 | ||
class_name: "trailer" | ||
} | ||
region_similarity_calculator: { | ||
nearest_iou_similarity: {} | ||
} | ||
} | ||
anchor_generators: { | ||
anchor_generator_range: { | ||
sizes: [2.4560939, 6.73778078, 2.73004906] # wlh | ||
anchor_ranges: [-50, -50, -0.37937912, 50, 50, -0.37937912] | ||
rotations: [0, 1.57] # DON'T modify this unless you are very familiar with my code. | ||
matched_threshold : 0.5 | ||
unmatched_threshold : 0.35 | ||
class_name: "truck" | ||
} | ||
region_similarity_calculator: { | ||
nearest_iou_similarity: {} | ||
} | ||
} | ||
anchor_generators: { | ||
anchor_generator_range: { | ||
sizes: [2.49008838, 0.48578221, 0.98297065] # wlh | ||
anchor_ranges: [-50, -50, -1.27247968, 50, 50, -1.27247968] | ||
rotations: [0, 1.57] # DON'T modify this unless you are very familiar with my code. | ||
matched_threshold : 0.3 | ||
unmatched_threshold : 0.2 | ||
class_name: "barrier" | ||
} | ||
region_similarity_calculator: { | ||
nearest_iou_similarity: {} | ||
} | ||
} | ||
sample_positive_fraction : -1 | ||
sample_size : 512 | ||
} | ||
} | ||
} | ||
|
||
train_input_reader: { | ||
dataset: { | ||
dataset_class_name: "NuScenesDatasetD2" | ||
kitti_info_path: "/media/yy/960evo/datasets/nuscene/v1.0-trainval/infos_train.pkl" | ||
kitti_root_path: "/media/yy/960evo/datasets/nuscene/v1.0-trainval" | ||
} | ||
|
||
batch_size: 3 | ||
preprocess: { | ||
max_number_of_voxels: 25000 | ||
shuffle_points: false | ||
num_workers: 3 | ||
groundtruth_localization_noise_std: [0, 0, 0] | ||
groundtruth_rotation_uniform_noise: [0, 0] | ||
# groundtruth_localization_noise_std: [0.25, 0.25, 0.25] | ||
# groundtruth_rotation_uniform_noise: [-0.15707963267, 0.15707963267] | ||
global_rotation_uniform_noise: [0, 0] | ||
global_scaling_uniform_noise: [0.95, 1.05] | ||
global_random_rotation_range_per_object: [0, 0] | ||
global_translate_noise_std: [0.2, 0.2, 0.2] | ||
anchor_area_threshold: -1 | ||
remove_points_after_sample: true | ||
groundtruth_points_drop_percentage: 0.0 | ||
groundtruth_drop_max_keep_points: 15 | ||
remove_unknown_examples: false | ||
remove_environment: false | ||
database_sampler { | ||
} | ||
} | ||
} | ||
|
||
train_config: { | ||
optimizer: { | ||
adam_optimizer: { | ||
learning_rate: { | ||
one_cycle: { | ||
lr_max: 3e-3 | ||
moms: [0.95, 0.85] | ||
div_factor: 10.0 | ||
pct_start: 0.4 | ||
} | ||
} | ||
weight_decay: 0.01 | ||
} | ||
fixed_weight_decay: true | ||
use_moving_average: false | ||
} | ||
steps: 58650 # 14065 * 20 (28130 // 2 ) | ||
steps_per_eval: 5865 # 14065 * 2 | ||
save_checkpoints_secs : 1800 # half hour | ||
save_summary_steps : 10 | ||
enable_mixed_precision: false | ||
loss_scale_factor: -1 | ||
clear_metrics_every_epoch: true | ||
} | ||
|
||
eval_input_reader: { | ||
dataset: { | ||
dataset_class_name: "NuScenesDataset" | ||
kitti_info_path: "/media/yy/960evo/datasets/nuscene/v1.0-trainval/infos_val.pkl" | ||
kitti_root_path: "/media/yy/960evo/datasets/nuscene/v1.0-trainval" | ||
} | ||
batch_size: 1 | ||
|
||
preprocess: { | ||
max_number_of_voxels: 30000 | ||
shuffle_points: false | ||
num_workers: 2 | ||
anchor_area_threshold: -1 | ||
remove_environment: false | ||
} | ||
} |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Oops, something went wrong.