jfthuong/captionning-encoder-decoder

1{2  "architectures": [3    "EncoderDecoderModel"4  ],5  "decoder": {6    "_name_or_path": "gpt2",7    "activation_function": "gelu_new",8    "add_cross_attention": true,9    "architectures": [10      "GPT2LMHeadModel"11    ],12    "attn_pdrop": 0.1,13    "bad_words_ids": null,14    "bos_token_id": 50256,15    "chunk_size_feed_forward": 0,16    "decoder_start_token_id": null,17    "diversity_penalty": 0.0,18    "do_sample": false,19    "early_stopping": false,20    "embd_pdrop": 0.1,21    "encoder_no_repeat_ngram_size": 0,22    "eos_token_id": 50256,23    "finetuning_task": null,24    "forced_bos_token_id": null,25    "forced_eos_token_id": null,26    "id2label": {27      "0": "LABEL_0",28      "1": "LABEL_1"29    },30    "initializer_range": 0.02,31    "is_decoder": true,32    "is_encoder_decoder": false,33    "label2id": {34      "LABEL_0": 0,35      "LABEL_1": 136    },37    "layer_norm_epsilon": 1e-05,38    "length_penalty": 1.0,39    "max_length": 20,40    "min_length": 0,41    "model_type": "gpt2",42    "n_ctx": 1024,43    "n_embd": 768,44    "n_head": 12,45    "n_inner": null,46    "n_layer": 12,47    "n_positions": 1024,48    "no_repeat_ngram_size": 0,49    "num_beam_groups": 1,50    "num_beams": 1,51    "num_return_sequences": 1,52    "output_attentions": false,53    "output_hidden_states": false,54    "output_scores": false,55    "pad_token_id": null,56    "prefix": null,57    "problem_type": null,58    "pruned_heads": {},59    "remove_invalid_values": false,60    "repetition_penalty": 1.0,61    "resid_pdrop": 0.1,62    "return_dict": true,63    "return_dict_in_generate": false,64    "scale_attn_weights": true,65    "sep_token_id": null,66    "summary_activation": null,67    "summary_first_dropout": 0.1,68    "summary_proj_to_labels": true,69    "summary_type": "cls_index",70    "summary_use_proj": true,71    "task_specific_params": {72      "text-generation": {73        "do_sample": true,74        "max_length": 5075      }76    },77    "temperature": 1.0,78    "tie_encoder_decoder": false,79    "tie_word_embeddings": true,80    "tokenizer_class": null,81    "top_k": 50,82    "top_p": 1.0,83    "torch_dtype": null,84    "torchscript": false,85    "transformers_version": "4.11.3",86    "use_bfloat16": false,87    "use_cache": true,88    "vocab_size": 5025789  },90  "encoder": {91    "_name_or_path": "google/vit-base-patch16-224-in21k",92    "add_cross_attention": false,93    "architectures": [94      "ViTModel"95    ],96    "attention_probs_dropout_prob": 0.0,97    "bad_words_ids": null,98    "bos_token_id": null,99    "chunk_size_feed_forward": 0,100    "decoder_start_token_id": null,101    "diversity_penalty": 0.0,102    "do_sample": false,103    "early_stopping": false,104    "encoder_no_repeat_ngram_size": 0,105    "eos_token_id": null,106    "finetuning_task": null,107    "forced_bos_token_id": null,108    "forced_eos_token_id": null,109    "hidden_act": "gelu",110    "hidden_dropout_prob": 0.0,111    "hidden_size": 768,112    "id2label": {113      "0": "LABEL_0",114      "1": "LABEL_1"115    },116    "image_size": 224,117    "initializer_range": 0.02,118    "intermediate_size": 3072,119    "is_decoder": false,120    "is_encoder_decoder": false,121    "label2id": {122      "LABEL_0": 0,123      "LABEL_1": 1124    },125    "layer_norm_eps": 1e-12,126    "length_penalty": 1.0,127    "max_length": 20,128    "min_length": 0,129    "model_type": "vit",130    "no_repeat_ngram_size": 0,131    "num_attention_heads": 12,132    "num_beam_groups": 1,133    "num_beams": 1,134    "num_channels": 3,135    "num_hidden_layers": 12,136    "num_return_sequences": 1,137    "output_attentions": false,138    "output_hidden_states": false,139    "output_scores": false,140    "pad_token_id": null,141    "patch_size": 16,142    "prefix": null,143    "problem_type": null,144    "pruned_heads": {},145    "qkv_bias": true,146    "remove_invalid_values": false,147    "repetition_penalty": 1.0,148    "return_dict": true,149    "return_dict_in_generate": false,150    "sep_token_id": null,151    "task_specific_params": null,152    "temperature": 1.0,153    "tie_encoder_decoder": false,154    "tie_word_embeddings": true,155    "tokenizer_class": null,156    "top_k": 50,157    "top_p": 1.0,158    "torch_dtype": null,159    "torchscript": false,160    "transformers_version": "4.11.3",161    "use_bfloat16": false162  },163  "is_encoder_decoder": true,164  "model_type": "encoder-decoder",165  "torch_dtype": "float32",166  "transformers_version": null167}168