-
Notifications
You must be signed in to change notification settings - Fork 97
/
Copy pathpipeline_config.py
166 lines (158 loc) · 9.22 KB
/
pipeline_config.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
import os
from attrdict import AttrDict
from .utils import read_config, check_env_vars
check_env_vars()
config = read_config(config_path=os.getenv('CONFIG_PATH'))
params = config.parameters
SIZE_COLUMNS = ['height', 'width']
X_COLUMNS = ['file_path_image']
Y_COLUMNS = ['file_path_mask_eroded_0_dilated_0']
Y_COLUMNS_SCORING = ['ImageId']
SEED = 1234
CATEGORY_IDS = [None, 100]
CATEGORY_LAYERS = [1, 1] # thresholds, 1 means [0.5], 19 means [0.05, ... 0.95] use only with second layer model
MEAN = [0.485, 0.456, 0.406]
STD = [0.229, 0.224, 0.225]
GLOBAL_CONFIG = {'exp_root': params.experiment_dir,
'load_in_memory': params.load_in_memory,
'num_workers': params.num_workers,
'num_classes': 2,
'img_H-W': (params.image_h, params.image_w),
'batch_size_train': params.batch_size_train,
'batch_size_inference': params.batch_size_inference,
'loader_mode': params.loader_mode,
'stream_mode': params.stream_mode
}
SOLUTION_CONFIG = AttrDict({
'env': {'cache_dirpath': params.experiment_dir},
'execution': GLOBAL_CONFIG,
'xy_splitter': {'x_columns': X_COLUMNS,
'y_columns': Y_COLUMNS,
},
'reader_single': {'x_columns': X_COLUMNS,
'y_columns': Y_COLUMNS,
},
'loader': {'dataset_params': {'h_pad': params.h_pad,
'w_pad': params.w_pad,
'h': params.image_h,
'w': params.image_w,
'pad_method': params.pad_method
},
'loader_params': {'training': {'batch_size': params.batch_size_train,
'shuffle': True,
'num_workers': params.num_workers,
'pin_memory': params.pin_memory
},
'inference': {'batch_size': params.batch_size_inference,
'shuffle': False,
'num_workers': params.num_workers,
'pin_memory': params.pin_memory
},
},
},
'unet': {
'architecture_config': {'model_params': {'n_filters': params.n_filters,
'conv_kernel': params.conv_kernel,
'pool_kernel': params.pool_kernel,
'pool_stride': params.pool_stride,
'repeat_blocks': params.repeat_blocks,
'batch_norm': params.use_batch_norm,
'dropout': params.dropout_conv,
'in_channels': params.image_channels,
'out_channels': params.channels_per_output,
'nr_outputs': params.nr_unet_outputs,
'encoder': params.encoder
},
'optimizer_params': {'lr': params.lr,
},
'regularizer_params': {'regularize': True,
'weight_decay_conv2d': params.l2_reg_conv,
},
'weights_init': {'function': 'he',
},
'loss_weights': {'bce_mask': params.bce_mask,
'dice_mask': params.dice_mask,
},
'weighted_cross_entropy': {'w0': params.w0,
'sigma': params.sigma,
'imsize': (params.image_h, params.image_w)},
'dice': {'smooth': params.dice_smooth,
'dice_activation': params.dice_activation},
},
'training_config': {'epochs': params.epochs_nr,
},
'callbacks_config': {
'model_checkpoint': {
'filepath': os.path.join(GLOBAL_CONFIG['exp_root'], 'checkpoints', 'unet', 'best.torch'),
'epoch_every': 1,
'minimize': not params.validate_with_map
},
'exp_lr_scheduler': {'gamma': params.gamma,
'epoch_every': 1},
'plateau_lr_scheduler': {'lr_factor': params.lr_factor,
'lr_patience': params.lr_patience,
'epoch_every': 1},
'training_monitor': {'batch_every': 1,
'epoch_every': 1},
'experiment_timing': {'batch_every': 10,
'epoch_every': 1},
'validation_monitor': {
'epoch_every': 1,
'data_dir': params.data_dir,
'validate_with_map': params.validate_with_map,
'small_annotations_size': params.small_annotations_size,
},
'neptune_monitor': {'model_name': 'unet',
'image_nr': 16,
'image_resize': 0.2,
'outputs_to_plot': params.unet_outputs_to_plot},
'early_stopping': {'patience': params.patience,
'minimize': not params.validate_with_map},
},
},
'tta_generator': {'flip_ud': True,
'flip_lr': True,
'rotation': True,
'color_shift_runs': False},
'tta_aggregator': {'method': params.tta_aggregation_method,
'num_threads': params.num_threads
},
'postprocessor': {'mask_dilation': {'dilate_selem_size': params.dilate_selem_size
},
'mask_erosion': {'erode_selem_size': params.erode_selem_size
},
'prediction_crop': {'h_crop': params.crop_image_h,
'w_crop': params.crop_image_w
},
'scoring_model': params.scoring_model,
'lightGBM': {'model_params': {'learning_rate': params.lgbm__learning_rate,
'boosting_type': 'gbdt',
'objective': 'regression',
'metric': 'regression_l2',
'sub_feature': 1.0,
'num_leaves': params.lgbm__num_leaves,
'min_data': params.lgbm__min_data,
'max_depth': params.lgbm__max_depth,
'num_threads': params.num_threads},
'training_params': {'number_boosting_rounds': params.lgbm__number_of_trees,
'early_stopping_rounds': params.lgbm__early_stopping},
'train_size': params.lgbm__train_size,
'target': params.lgbm__target
},
'random_forest': {'train_size': params.lgbm__train_size,
'target': params.lgbm__target,
'model_params': {'n_estimators': params.rf__n_estimators,
'criterion': params.rf__criterion,
'max_depth': params.rf__max_depth,
'min_samples_split': params.rf__min_samples_split,
'min_samples_leaf': params.rf__min_samples_leaf,
'max_features': params.rf__max_features,
'max_leaf_nodes': params.rf__max_leaf_nodes,
'n_jobs': params.rf__n_jobs,
'verbose': params.rf__verbose,
}
},
'nms': {'iou_threshold': params.nms__iou_threshold,
'num_threads': params.num_threads},
}
})