ctrl_vl-bert_base.json

{
  "attention_probs_dropout_prob": 0.1,
  "hidden_act": "gelu",
  "hidden_dropout_prob": 0.1,
  "hidden_size": 768,
  "initializer_range": 0.02,
  "intermediate_size": 3072,
  "max_position_embeddings": 512,
  "num_attention_heads": 12,
  "pooler_size": 1024,
  "type_vocab_size": 3,
  "vocab_size": 30522,
  "bert_model": "bert-base-uncased",
  "do_lower_case": true,
  "num_locs": 5,
  "add_global_imgfeat": "first",
  "v_coordinate_embeddings_dim": 256,
  "image_embeddings": "vl-bert",
  "v_attention_probs_dropout_prob": 0.1,
  "v_hidden_act": "gelu",
  "v_hidden_dropout_prob": 0.1,
  "v_feature_size": 2048,
  "visual_target_weights": {"0": 1.0},
  "v_hidden_size": 768,
  "v_initializer_range": 0.02,
  "v_pooler_size": 1024,
  "v_num_attention_heads": 12,
  "v_intermediate_size": 3072,
  "fusion_method": "mul",
  "image_head_ln": false,
  "clf_hidden_size": 1536,
  "tt_attn_sublayers": [0,2,4,6,8,10,12,14,16,18,20,22],
  "tv_attn_sublayers": [0,2,4,6,8,10,12,14,16,18,20,22],
  "vt_attn_sublayers": [0,2,4,6,8,10,12,14,16,18,20,22],
  "vv_attn_sublayers": [0,2,4,6,8,10,12,14,16,18,20,22],
  "t_ff_sublayers": [1,3,5,7,9,11,13,15,17,19,21,23],
  "v_ff_sublayers": [1,3,5,7,9,11,13,15,17,19,21,23],
  "shared_sublayers": [0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23],
  "single_ln_sublayers": [0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23],
  "sublayer2attn_hidden_size": {},
  "sublayer2num_attention_heads": {},
  "sublayer2intermediate_size": {},
  "sublayer2v_attn_hidden_size": {},
  "sublayer2v_num_attention_heads": {},
  "sublayer2v_intermediate_size": {},
  "bert_layer2attn_sublayer": {
    "0": 0, "1": 2, "2": 4, "3": 6, "4": 8, "5": 10,
    "6": 12, "7": 14, "8": 16, "9": 18, "10": 20, "11": 22
  },
  "bert_layer2ff_sublayer": {
    "0": 1, "1": 3, "2": 5, "3": 7, "4": 9, "5": 11,
    "6": 13, "7": 15, "8": 17, "9": 19, "10": 21, "11": 23
  }
}