kf_config.json

{
  "library_name": "kerasformers",
  "kerasformers_version": "1.2.1",
  "model_module": "kerasformers.models.dfine",
  "model_class": "DFineDetect",
  "variant": "dfine-large",
  "weights": "model.weights.h5",
  "schema_version": 2,
  "weight_dtype": "float32",
  "model_type": "d_fine",
  "vision_config": {
    "stem_channels": [
      3,
      32,
      48
    ],
    "stage_in_channels": [
      48,
      128,
      512,
      1024
    ],
    "stage_mid_channels": [
      48,
      96,
      192,
      384
    ],
    "stage_out_channels": [
      128,
      512,
      1024,
      2048
    ],
    "stage_num_blocks": [
      1,
      1,
      3,
      1
    ],
    "stage_numb_of_layers": [
      6,
      6,
      6,
      6
    ],
    "use_lab": false,
    "encoder_in_channels": [
      512,
      1024,
      2048
    ],
    "encoder_hidden_dim": 256,
    "encoder_ffn_dim": 1024,
    "encode_proj_layers": [
      2
    ],
    "hidden_expansion": 1.0,
    "ccfm_num_blocks": 3,
    "hidden_dim": 256,
    "decoder_num_layers": 6,
    "decoder_ffn_dim": 1024,
    "decoder_n_points": [
      3,
      6,
      3
    ],
    "num_feature_levels": 3,
    "feat_strides": [
      8,
      16,
      32
    ],
    "num_classes": 80,
    "num_queries": 300,
    "image_size": 640
  }
}