Detic / configs /Detic_LCOCOI21k_CLIP_SwinB_896b32_4x_ft4x_max-size.yaml
AK391
files
159f437
_BASE_: "Base-C2_L_R5021k_640b64_4x.yaml"
MODEL:
WEIGHTS: "models/BoxSup-C2_LCOCO_CLIP_SwinB_896b32_4x.pth"
DYNAMIC_CLASSIFIER: True
ROI_BOX_HEAD:
USE_ZEROSHOT_CLS: True
IMAGE_LABEL_LOSS: 'max_size'
ZEROSHOT_WEIGHT_PATH: 'datasets/metadata/lvis-21k_clip_a+cname.npy'
USE_FED_LOSS: False # Federated loss is enabled when DYNAMIC_CLASSIFIER is on
ROI_HEADS:
NUM_CLASSES: 22047
BACKBONE:
NAME: build_swintransformer_fpn_backbone
SWIN:
SIZE: B-22k
FPN:
IN_FEATURES: ["swin1", "swin2", "swin3"]
RESET_CLS_TESTS: True
TEST_CLASSIFIERS: ("datasets/metadata/oid_clip_a+cname.npy","datasets/metadata/o365_clip_a+cnamefix.npy")
TEST_NUM_CLASSES: [500, 365]
SOLVER:
MAX_ITER: 180000
IMS_PER_BATCH: 32
BASE_LR: 0.0001
WARMUP_ITERS: 1000
WARMUP_FACTOR: 0.001
DATASETS:
TRAIN: ("lvis_v1_train+coco","imagenet_lvis-22k")
TEST: ('oid_val_expanded', 'objects365_v2_val')
DATALOADER:
SAMPLER_TRAIN: "MultiDatasetSampler"
DATASET_RATIO: [1, 16]
USE_DIFF_BS_SIZE: True
DATASET_BS: [4, 16]
DATASET_INPUT_SIZE: [896, 448]
USE_RFS: [True, False]
DATASET_INPUT_SCALE: [[0.1, 2.0], [0.5, 1.5]]
FILTER_EMPTY_ANNOTATIONS: False
MULTI_DATASET_GROUPING: True
DATASET_ANN: ['box', 'image']
NUM_WORKERS: 4
USE_TAR_DATASET: True
WITH_IMAGE_LABELS: True