lvis-base_in-l_r50_4x_4x_clip_multi_modal_agg.yaml

_BASE_: "base_r50_4x_clip.yaml"
MODEL:
  ROI_BOX_HEAD:
    USE_ZEROSHOT_CLS: True
    IMAGE_LABEL_LOSS: 'max_size'
    USE_BIAS: -2.0
    ZEROSHOT_WEIGHT_PATH: 'datasets/metadata/lvis_multi-modal_agg_K-005_author.npy'
  WEIGHTS: "output/mm-ovod/lvis-base_r50_4x_clip_multi_modal_agg/model_final.pth"
SOLVER:
  MAX_ITER: 90000
  IMS_PER_BATCH: 64
  BASE_LR: 0.0002
  WARMUP_ITERS: 1000
  WARMUP_FACTOR: 0.001
DATASETS:
  TRAIN: ("lvis_v1_train_norare","imagenet_lvis_v1")
DATALOADER:
  SAMPLER_TRAIN: "MultiDatasetSampler"
  DATASET_RATIO: [1, 4]
  USE_DIFF_BS_SIZE: True
  DATASET_BS: [16, 64]
  DATASET_INPUT_SIZE: [640, 320]
  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: 8
WITH_IMAGE_LABELS: True