{ "pred_head_dropout_prob": 0.1, "attention_probs_dropout_prob": 0.1, "finetuning_task": null, "hidden_act": "gelu", "hidden_dropout_prob": 0.1, "hidden_size": 768, "image_feat_size": 768, "image_prob_size": 1000, "angle_feat_size": 4, "obj_feat_size": 2048, "obj_prob_size": 1601, "img_feature_type": "imagenet", "initializer_range": 0.02, "intermediate_size": 3072, "num_l_layers": 9, "num_x_layers": 4, "num_pano_layers": 2, "layer_norm_eps": 1e-12, "max_position_embeddings": 512, "max_action_steps": 100, "num_attention_heads": 12, "num_hidden_layers": 12, "num_labels": 2, "output_attentions": false, "output_hidden_states": false, "pruned_heads": {}, "torchscript": false, "type_vocab_size": 2, "update_lang_bert": true, "vocab_size": 30522, "use_lang2visn_attn": true, "graph_sprels": true, "glocal_fuse": true, "lang_bert_name": "bert-base-uncased" }