Upload model
Browse files- config.json +16 -16
- model.safetensors +2 -2
config.json
CHANGED
|
@@ -4,16 +4,16 @@
|
|
| 4 |
],
|
| 5 |
"decoder": {
|
| 6 |
"activation_dropout": 0.0,
|
| 7 |
-
"activation_function": "
|
| 8 |
"add_cross_attention": true,
|
| 9 |
"attention_dropout": 0.0,
|
| 10 |
"classifier_dropout": 0.0,
|
| 11 |
-
"cross_attention_hidden_size":
|
| 12 |
-
"d_model":
|
| 13 |
-
"decoder_attention_heads":
|
| 14 |
-
"decoder_ffn_dim":
|
| 15 |
"decoder_layerdrop": 0.0,
|
| 16 |
-
"decoder_layers":
|
| 17 |
"dropout": 0.1,
|
| 18 |
"dtype": "float32",
|
| 19 |
"init_std": 0.02,
|
|
@@ -21,11 +21,10 @@
|
|
| 21 |
"layernorm_embedding": true,
|
| 22 |
"max_position_embeddings": 512,
|
| 23 |
"model_type": "trocr",
|
| 24 |
-
"scale_embedding":
|
| 25 |
-
"tie_word_embeddings": false,
|
| 26 |
"use_cache": false,
|
| 27 |
"use_learned_position_embeddings": true,
|
| 28 |
-
"vocab_size":
|
| 29 |
},
|
| 30 |
"decoder_start_token_id": 0,
|
| 31 |
"dtype": "float32",
|
|
@@ -36,19 +35,19 @@
|
|
| 36 |
"encoder_stride": 16,
|
| 37 |
"hidden_act": "gelu",
|
| 38 |
"hidden_dropout_prob": 0.0,
|
| 39 |
-
"hidden_size":
|
| 40 |
"image_size": 384,
|
| 41 |
"initializer_range": 0.02,
|
| 42 |
-
"intermediate_size":
|
| 43 |
"layer_norm_eps": 1e-12,
|
| 44 |
-
"model_type": "
|
| 45 |
-
"num_attention_heads":
|
| 46 |
"num_channels": 3,
|
| 47 |
"num_hidden_layers": 12,
|
| 48 |
"patch_size": 16,
|
| 49 |
"pooler_act": "tanh",
|
| 50 |
-
"pooler_output_size":
|
| 51 |
-
"qkv_bias":
|
| 52 |
},
|
| 53 |
"eos_token_id": 2,
|
| 54 |
"is_encoder_decoder": true,
|
|
@@ -58,7 +57,8 @@
|
|
| 58 |
"no_repeat_ngram_size": null,
|
| 59 |
"num_beams": null,
|
| 60 |
"pad_token_id": 1,
|
|
|
|
| 61 |
"tie_word_embeddings": false,
|
| 62 |
"transformers_version": "4.57.3",
|
| 63 |
-
"vocab_size":
|
| 64 |
}
|
|
|
|
| 4 |
],
|
| 5 |
"decoder": {
|
| 6 |
"activation_dropout": 0.0,
|
| 7 |
+
"activation_function": "gelu",
|
| 8 |
"add_cross_attention": true,
|
| 9 |
"attention_dropout": 0.0,
|
| 10 |
"classifier_dropout": 0.0,
|
| 11 |
+
"cross_attention_hidden_size": 768,
|
| 12 |
+
"d_model": 1024,
|
| 13 |
+
"decoder_attention_heads": 16,
|
| 14 |
+
"decoder_ffn_dim": 4096,
|
| 15 |
"decoder_layerdrop": 0.0,
|
| 16 |
+
"decoder_layers": 12,
|
| 17 |
"dropout": 0.1,
|
| 18 |
"dtype": "float32",
|
| 19 |
"init_std": 0.02,
|
|
|
|
| 21 |
"layernorm_embedding": true,
|
| 22 |
"max_position_embeddings": 512,
|
| 23 |
"model_type": "trocr",
|
| 24 |
+
"scale_embedding": false,
|
|
|
|
| 25 |
"use_cache": false,
|
| 26 |
"use_learned_position_embeddings": true,
|
| 27 |
+
"vocab_size": 50265
|
| 28 |
},
|
| 29 |
"decoder_start_token_id": 0,
|
| 30 |
"dtype": "float32",
|
|
|
|
| 35 |
"encoder_stride": 16,
|
| 36 |
"hidden_act": "gelu",
|
| 37 |
"hidden_dropout_prob": 0.0,
|
| 38 |
+
"hidden_size": 768,
|
| 39 |
"image_size": 384,
|
| 40 |
"initializer_range": 0.02,
|
| 41 |
+
"intermediate_size": 3072,
|
| 42 |
"layer_norm_eps": 1e-12,
|
| 43 |
+
"model_type": "vit",
|
| 44 |
+
"num_attention_heads": 12,
|
| 45 |
"num_channels": 3,
|
| 46 |
"num_hidden_layers": 12,
|
| 47 |
"patch_size": 16,
|
| 48 |
"pooler_act": "tanh",
|
| 49 |
+
"pooler_output_size": 768,
|
| 50 |
+
"qkv_bias": false
|
| 51 |
},
|
| 52 |
"eos_token_id": 2,
|
| 53 |
"is_encoder_decoder": true,
|
|
|
|
| 57 |
"no_repeat_ngram_size": null,
|
| 58 |
"num_beams": null,
|
| 59 |
"pad_token_id": 1,
|
| 60 |
+
"processor_class": "TrOCRProcessor",
|
| 61 |
"tie_word_embeddings": false,
|
| 62 |
"transformers_version": "4.57.3",
|
| 63 |
+
"vocab_size": 50265
|
| 64 |
}
|
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a3114b341d514aab5c6bc422545fe0b05eeb64f2de0beb47dfdad8d57378da51
|
| 3 |
+
size 1335747032
|