TibetanCodexAITeam commited on
Commit
7a5b502
·
verified ·
1 Parent(s): 1505195

Upload DONUT checkpoint checkpoint-154000 via PechaBridge

Browse files
Files changed (1) hide show
  1. config.json +122 -0
config.json ADDED
@@ -0,0 +1,122 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "VisionEncoderDecoderModel"
4
+ ],
5
+ "decoder": {
6
+ "_name_or_path": "",
7
+ "activation_dropout": 0.0,
8
+ "activation_function": "relu",
9
+ "add_cross_attention": true,
10
+ "architectures": null,
11
+ "attention_dropout": 0.0,
12
+ "bos_token_id": 0,
13
+ "chunk_size_feed_forward": 0,
14
+ "classifier_dropout": 0.0,
15
+ "cross_attention_hidden_size": 768,
16
+ "d_model": 1024,
17
+ "decoder_attention_heads": 16,
18
+ "decoder_ffn_dim": 4096,
19
+ "decoder_layerdrop": 0.0,
20
+ "decoder_layers": 12,
21
+ "decoder_start_token_id": 2,
22
+ "dropout": 0.1,
23
+ "dtype": "float32",
24
+ "eos_token_id": 2,
25
+ "finetuning_task": null,
26
+ "id2label": {
27
+ "0": "LABEL_0",
28
+ "1": "LABEL_1"
29
+ },
30
+ "init_std": 0.02,
31
+ "is_decoder": true,
32
+ "is_encoder_decoder": false,
33
+ "label2id": {
34
+ "LABEL_0": 0,
35
+ "LABEL_1": 1
36
+ },
37
+ "layernorm_embedding": false,
38
+ "max_position_embeddings": 1024,
39
+ "model_type": "trocr",
40
+ "output_attentions": false,
41
+ "output_hidden_states": false,
42
+ "pad_token_id": 1,
43
+ "prefix": null,
44
+ "problem_type": null,
45
+ "pruned_heads": {},
46
+ "return_dict": true,
47
+ "scale_embedding": true,
48
+ "sep_token_id": null,
49
+ "task_specific_params": null,
50
+ "tie_encoder_decoder": false,
51
+ "tie_word_embeddings": false,
52
+ "tokenizer_class": null,
53
+ "torchscript": false,
54
+ "use_bfloat16": false,
55
+ "use_cache": false,
56
+ "use_learned_position_embeddings": false,
57
+ "vocab_size": 20008
58
+ },
59
+ "decoder_start_token_id": 20005,
60
+ "dtype": "float32",
61
+ "encoder": {
62
+ "_name_or_path": "",
63
+ "add_cross_attention": false,
64
+ "architectures": null,
65
+ "attention_probs_dropout_prob": 0.0,
66
+ "bos_token_id": null,
67
+ "chunk_size_feed_forward": 0,
68
+ "cross_attention_hidden_size": null,
69
+ "decoder_start_token_id": null,
70
+ "dtype": "float32",
71
+ "encoder_stride": 16,
72
+ "eos_token_id": null,
73
+ "finetuning_task": null,
74
+ "hidden_act": "gelu",
75
+ "hidden_dropout_prob": 0.0,
76
+ "hidden_size": 768,
77
+ "id2label": {
78
+ "0": "LABEL_0",
79
+ "1": "LABEL_1"
80
+ },
81
+ "image_size": 384,
82
+ "initializer_range": 0.02,
83
+ "intermediate_size": 3072,
84
+ "is_decoder": false,
85
+ "is_encoder_decoder": false,
86
+ "label2id": {
87
+ "LABEL_0": 0,
88
+ "LABEL_1": 1
89
+ },
90
+ "layer_norm_eps": 1e-12,
91
+ "model_type": "vit",
92
+ "num_attention_heads": 12,
93
+ "num_channels": 3,
94
+ "num_hidden_layers": 12,
95
+ "output_attentions": false,
96
+ "output_hidden_states": false,
97
+ "pad_token_id": null,
98
+ "patch_size": 16,
99
+ "pooler_act": "tanh",
100
+ "pooler_output_size": 768,
101
+ "prefix": null,
102
+ "problem_type": null,
103
+ "pruned_heads": {},
104
+ "qkv_bias": false,
105
+ "return_dict": true,
106
+ "sep_token_id": null,
107
+ "task_specific_params": null,
108
+ "tie_encoder_decoder": false,
109
+ "tie_word_embeddings": true,
110
+ "tokenizer_class": null,
111
+ "torchscript": false,
112
+ "use_bfloat16": false
113
+ },
114
+ "eos_token_id": 20006,
115
+ "is_encoder_decoder": true,
116
+ "model_type": "vision-encoder-decoder",
117
+ "pad_token_id": 20000,
118
+ "tie_word_embeddings": false,
119
+ "transformers_version": "5.2.0",
120
+ "use_cache": false,
121
+ "vocab_size": 20008
122
+ }