{
  "adv_probability": 1.0,
  "adv_probability_finetune": 1.0,
  "adv_probability_pretrain": 0.35,
  "architectures": [
    "OpenViVQAModel"
  ],
  "bootstrap_from_pretrained": true,
  "char_max_num": 50,
  "char_num": 168,
  "clip_vision_name": "openai/clip-vit-base-patch16",
  "contrastive_ignore_value": -1.0,
  "contrastive_label_list": [
    0.9,
    0.9
  ],
  "decoder_start_token_id": 0,
  "do_center_crop": false,
  "do_resize": true,
  "editlen": 2,
  "eos_token_id": 1,
  "generation_max_new_tokens": 56,
  "generation_num_beams": 4,
  "image_mean": [
    0.48145466,
    0.4578275,
    0.40821073
  ],
  "image_std": [
    0.26862954,
    0.26130258,
    0.27577711
  ],
  "local_submodule_dir": null,
  "mlm_ignore_index": -100,
  "model_type": "openvivqa",
  "ocr_cuda_device": "cuda:0",
  "ocr_d_det": 256,
  "ocr_d_rec": 256,
  "ocr_jaccard_dupe": 0.78,
  "ocr_lowercase": true,
  "ocr_max_2d_position_embeddings": 1024,
  "ocr_max_non_alnum_ratio": 0.6,
  "ocr_max_scene_text": 180,
  "ocr_min_text_len_keep": 2,
  "ocr_num_distances": 32,
  "ocrseq_dropout": 0.1,
  "ocrseq_margin_frac": 0.14,
  "ocrseq_pool_size": 3,
  "ocrseq_use_spatial": true,
  "pad_token_id": 0,
  "pretrain": true,
  "pretrain_ablation_mode": "all",
  "pretrain_mask_prob": 0.15,
  "pretrain_mask_seed": 42,
  "qa_clip_d_text": null,
  "term_vocab_path": "/kaggle/input/term-vietnamese-vocab/words.txt",
  "text_max_input_length": 32,
  "text_max_target_length": 56,
  "torch_dtype": "float32",
  "transformers_version": "4.45.2",
  "use_ocr_aug_finetune": true,
  "use_twc": true,
  "vit5_name": "VietAI/vit5-base",
  "vs_backbone": "facebook/convnextv2-tiny-22k-224",
  "vs_dilate_half_patch": true,
  "vs_margin": 0,
  "vs_max_cov": 1.0,
  "vs_min_area": 1,
  "vs_patch_size": 16,
  "vs_percentile": 99,
  "vs_target_size": 224
}