End of training
Browse files- README.md +7 -0
- added_tokens.json +0 -14
- config.json +1 -1
- pytorch_model.bin +2 -2
- special_tokens_map.json +1 -15
- tokenizer.json +0 -126
- tokenizer_config.json +1 -127
- training_args.bin +1 -1
README.md
CHANGED
|
@@ -16,6 +16,13 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 16 |
# donut-test-ddt-613img
|
| 17 |
|
| 18 |
This model is a fine-tuned version of [naver-clova-ix/donut-base](https://huggingface.co/naver-clova-ix/donut-base) on the imagefolder dataset.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 19 |
|
| 20 |
## Model description
|
| 21 |
|
|
|
|
| 16 |
# donut-test-ddt-613img
|
| 17 |
|
| 18 |
This model is a fine-tuned version of [naver-clova-ix/donut-base](https://huggingface.co/naver-clova-ix/donut-base) on the imagefolder dataset.
|
| 19 |
+
It achieves the following results on the evaluation set:
|
| 20 |
+
- eval_loss: 0.0929
|
| 21 |
+
- eval_runtime: 154.5449
|
| 22 |
+
- eval_samples_per_second: 0.647
|
| 23 |
+
- eval_steps_per_second: 0.084
|
| 24 |
+
- epoch: 2.75
|
| 25 |
+
- step: 800
|
| 26 |
|
| 27 |
## Model description
|
| 28 |
|
added_tokens.json
CHANGED
|
@@ -1,22 +1,8 @@
|
|
| 1 |
{
|
| 2 |
"</s>": 2,
|
| 3 |
-
"</s_Codice>": 57537,
|
| 4 |
-
"</s_Data>": 57538,
|
| 5 |
-
"</s_Fornitore>": 57532,
|
| 6 |
-
"</s_Numero>": 57529,
|
| 7 |
-
"</s_Oggetti>": 57530,
|
| 8 |
-
"</s_Posizione>": 57535,
|
| 9 |
-
"</s_Quantita>": 57527,
|
| 10 |
"<mask>": 57521,
|
| 11 |
"<pad>": 1,
|
| 12 |
"<s>": 0,
|
| 13 |
-
"<s_Codice>": 57526,
|
| 14 |
-
"<s_Data>": 57525,
|
| 15 |
-
"<s_Fornitore>": 57534,
|
| 16 |
-
"<s_Numero>": 57533,
|
| 17 |
-
"<s_Oggetti>": 57536,
|
| 18 |
-
"<s_Posizione>": 57528,
|
| 19 |
-
"<s_Quantita>": 57531,
|
| 20 |
"<s_iitcdip>": 57523,
|
| 21 |
"<s_synthdog>": 57524,
|
| 22 |
"<sep/>": 57522,
|
|
|
|
| 1 |
{
|
| 2 |
"</s>": 2,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 3 |
"<mask>": 57521,
|
| 4 |
"<pad>": 1,
|
| 5 |
"<s>": 0,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 6 |
"<s_iitcdip>": 57523,
|
| 7 |
"<s_synthdog>": 57524,
|
| 8 |
"<sep/>": 57522,
|
config.json
CHANGED
|
@@ -85,7 +85,7 @@
|
|
| 85 |
"typical_p": 1.0,
|
| 86 |
"use_bfloat16": false,
|
| 87 |
"use_cache": true,
|
| 88 |
-
"vocab_size":
|
| 89 |
},
|
| 90 |
"decoder_start_token_id": 0,
|
| 91 |
"encoder": {
|
|
|
|
| 85 |
"typical_p": 1.0,
|
| 86 |
"use_bfloat16": false,
|
| 87 |
"use_cache": true,
|
| 88 |
+
"vocab_size": 57525
|
| 89 |
},
|
| 90 |
"decoder_start_token_id": 0,
|
| 91 |
"encoder": {
|
pytorch_model.bin
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:a414cf677ce5186dd50d1927cecdff0043f0530e91b75e3667c10ffa1149a9b2
|
| 3 |
+
size 809179350
|
special_tokens_map.json
CHANGED
|
@@ -2,22 +2,8 @@
|
|
| 2 |
"additional_special_tokens": [
|
| 3 |
"<s_iitcdip>",
|
| 4 |
"<s_synthdog>",
|
| 5 |
-
"<s_Data>",
|
| 6 |
-
"<s_Codice>",
|
| 7 |
"</s>",
|
| 8 |
-
"
|
| 9 |
-
"<s_Posizione>",
|
| 10 |
-
"</s_Numero>",
|
| 11 |
-
"</s_Oggetti>",
|
| 12 |
-
"<s_Quantita>",
|
| 13 |
-
"</s_Fornitore>",
|
| 14 |
-
"<s_Numero>",
|
| 15 |
-
"<s_Fornitore>",
|
| 16 |
-
"</s_Posizione>",
|
| 17 |
-
"<s_Oggetti>",
|
| 18 |
-
"<s>",
|
| 19 |
-
"</s_Codice>",
|
| 20 |
-
"</s_Data>"
|
| 21 |
],
|
| 22 |
"bos_token": "<s>",
|
| 23 |
"cls_token": "<s>",
|
|
|
|
| 2 |
"additional_special_tokens": [
|
| 3 |
"<s_iitcdip>",
|
| 4 |
"<s_synthdog>",
|
|
|
|
|
|
|
| 5 |
"</s>",
|
| 6 |
+
"<s>"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 7 |
],
|
| 8 |
"bos_token": "<s>",
|
| 9 |
"cls_token": "<s>",
|
tokenizer.json
CHANGED
|
@@ -88,132 +88,6 @@
|
|
| 88 |
"rstrip": true,
|
| 89 |
"normalized": false,
|
| 90 |
"special": true
|
| 91 |
-
},
|
| 92 |
-
{
|
| 93 |
-
"id": 57525,
|
| 94 |
-
"content": "<s_Data>",
|
| 95 |
-
"single_word": false,
|
| 96 |
-
"lstrip": true,
|
| 97 |
-
"rstrip": true,
|
| 98 |
-
"normalized": false,
|
| 99 |
-
"special": true
|
| 100 |
-
},
|
| 101 |
-
{
|
| 102 |
-
"id": 57526,
|
| 103 |
-
"content": "<s_Codice>",
|
| 104 |
-
"single_word": false,
|
| 105 |
-
"lstrip": true,
|
| 106 |
-
"rstrip": true,
|
| 107 |
-
"normalized": false,
|
| 108 |
-
"special": true
|
| 109 |
-
},
|
| 110 |
-
{
|
| 111 |
-
"id": 57527,
|
| 112 |
-
"content": "</s_Quantita>",
|
| 113 |
-
"single_word": false,
|
| 114 |
-
"lstrip": true,
|
| 115 |
-
"rstrip": true,
|
| 116 |
-
"normalized": false,
|
| 117 |
-
"special": true
|
| 118 |
-
},
|
| 119 |
-
{
|
| 120 |
-
"id": 57528,
|
| 121 |
-
"content": "<s_Posizione>",
|
| 122 |
-
"single_word": false,
|
| 123 |
-
"lstrip": true,
|
| 124 |
-
"rstrip": true,
|
| 125 |
-
"normalized": false,
|
| 126 |
-
"special": true
|
| 127 |
-
},
|
| 128 |
-
{
|
| 129 |
-
"id": 57529,
|
| 130 |
-
"content": "</s_Numero>",
|
| 131 |
-
"single_word": false,
|
| 132 |
-
"lstrip": true,
|
| 133 |
-
"rstrip": true,
|
| 134 |
-
"normalized": false,
|
| 135 |
-
"special": true
|
| 136 |
-
},
|
| 137 |
-
{
|
| 138 |
-
"id": 57530,
|
| 139 |
-
"content": "</s_Oggetti>",
|
| 140 |
-
"single_word": false,
|
| 141 |
-
"lstrip": true,
|
| 142 |
-
"rstrip": true,
|
| 143 |
-
"normalized": false,
|
| 144 |
-
"special": true
|
| 145 |
-
},
|
| 146 |
-
{
|
| 147 |
-
"id": 57531,
|
| 148 |
-
"content": "<s_Quantita>",
|
| 149 |
-
"single_word": false,
|
| 150 |
-
"lstrip": true,
|
| 151 |
-
"rstrip": true,
|
| 152 |
-
"normalized": false,
|
| 153 |
-
"special": true
|
| 154 |
-
},
|
| 155 |
-
{
|
| 156 |
-
"id": 57532,
|
| 157 |
-
"content": "</s_Fornitore>",
|
| 158 |
-
"single_word": false,
|
| 159 |
-
"lstrip": true,
|
| 160 |
-
"rstrip": true,
|
| 161 |
-
"normalized": false,
|
| 162 |
-
"special": true
|
| 163 |
-
},
|
| 164 |
-
{
|
| 165 |
-
"id": 57533,
|
| 166 |
-
"content": "<s_Numero>",
|
| 167 |
-
"single_word": false,
|
| 168 |
-
"lstrip": true,
|
| 169 |
-
"rstrip": true,
|
| 170 |
-
"normalized": false,
|
| 171 |
-
"special": true
|
| 172 |
-
},
|
| 173 |
-
{
|
| 174 |
-
"id": 57534,
|
| 175 |
-
"content": "<s_Fornitore>",
|
| 176 |
-
"single_word": false,
|
| 177 |
-
"lstrip": true,
|
| 178 |
-
"rstrip": true,
|
| 179 |
-
"normalized": false,
|
| 180 |
-
"special": true
|
| 181 |
-
},
|
| 182 |
-
{
|
| 183 |
-
"id": 57535,
|
| 184 |
-
"content": "</s_Posizione>",
|
| 185 |
-
"single_word": false,
|
| 186 |
-
"lstrip": true,
|
| 187 |
-
"rstrip": true,
|
| 188 |
-
"normalized": false,
|
| 189 |
-
"special": true
|
| 190 |
-
},
|
| 191 |
-
{
|
| 192 |
-
"id": 57536,
|
| 193 |
-
"content": "<s_Oggetti>",
|
| 194 |
-
"single_word": false,
|
| 195 |
-
"lstrip": true,
|
| 196 |
-
"rstrip": true,
|
| 197 |
-
"normalized": false,
|
| 198 |
-
"special": true
|
| 199 |
-
},
|
| 200 |
-
{
|
| 201 |
-
"id": 57537,
|
| 202 |
-
"content": "</s_Codice>",
|
| 203 |
-
"single_word": false,
|
| 204 |
-
"lstrip": true,
|
| 205 |
-
"rstrip": true,
|
| 206 |
-
"normalized": false,
|
| 207 |
-
"special": true
|
| 208 |
-
},
|
| 209 |
-
{
|
| 210 |
-
"id": 57538,
|
| 211 |
-
"content": "</s_Data>",
|
| 212 |
-
"single_word": false,
|
| 213 |
-
"lstrip": true,
|
| 214 |
-
"rstrip": true,
|
| 215 |
-
"normalized": false,
|
| 216 |
-
"special": true
|
| 217 |
}
|
| 218 |
],
|
| 219 |
"normalizer": {
|
|
|
|
| 88 |
"rstrip": true,
|
| 89 |
"normalized": false,
|
| 90 |
"special": true
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 91 |
}
|
| 92 |
],
|
| 93 |
"normalizer": {
|
tokenizer_config.json
CHANGED
|
@@ -63,139 +63,13 @@
|
|
| 63 |
"rstrip": true,
|
| 64 |
"single_word": false,
|
| 65 |
"special": true
|
| 66 |
-
},
|
| 67 |
-
"57525": {
|
| 68 |
-
"content": "<s_Data>",
|
| 69 |
-
"lstrip": true,
|
| 70 |
-
"normalized": false,
|
| 71 |
-
"rstrip": true,
|
| 72 |
-
"single_word": false,
|
| 73 |
-
"special": true
|
| 74 |
-
},
|
| 75 |
-
"57526": {
|
| 76 |
-
"content": "<s_Codice>",
|
| 77 |
-
"lstrip": true,
|
| 78 |
-
"normalized": false,
|
| 79 |
-
"rstrip": true,
|
| 80 |
-
"single_word": false,
|
| 81 |
-
"special": true
|
| 82 |
-
},
|
| 83 |
-
"57527": {
|
| 84 |
-
"content": "</s_Quantita>",
|
| 85 |
-
"lstrip": true,
|
| 86 |
-
"normalized": false,
|
| 87 |
-
"rstrip": true,
|
| 88 |
-
"single_word": false,
|
| 89 |
-
"special": true
|
| 90 |
-
},
|
| 91 |
-
"57528": {
|
| 92 |
-
"content": "<s_Posizione>",
|
| 93 |
-
"lstrip": true,
|
| 94 |
-
"normalized": false,
|
| 95 |
-
"rstrip": true,
|
| 96 |
-
"single_word": false,
|
| 97 |
-
"special": true
|
| 98 |
-
},
|
| 99 |
-
"57529": {
|
| 100 |
-
"content": "</s_Numero>",
|
| 101 |
-
"lstrip": true,
|
| 102 |
-
"normalized": false,
|
| 103 |
-
"rstrip": true,
|
| 104 |
-
"single_word": false,
|
| 105 |
-
"special": true
|
| 106 |
-
},
|
| 107 |
-
"57530": {
|
| 108 |
-
"content": "</s_Oggetti>",
|
| 109 |
-
"lstrip": true,
|
| 110 |
-
"normalized": false,
|
| 111 |
-
"rstrip": true,
|
| 112 |
-
"single_word": false,
|
| 113 |
-
"special": true
|
| 114 |
-
},
|
| 115 |
-
"57531": {
|
| 116 |
-
"content": "<s_Quantita>",
|
| 117 |
-
"lstrip": true,
|
| 118 |
-
"normalized": false,
|
| 119 |
-
"rstrip": true,
|
| 120 |
-
"single_word": false,
|
| 121 |
-
"special": true
|
| 122 |
-
},
|
| 123 |
-
"57532": {
|
| 124 |
-
"content": "</s_Fornitore>",
|
| 125 |
-
"lstrip": true,
|
| 126 |
-
"normalized": false,
|
| 127 |
-
"rstrip": true,
|
| 128 |
-
"single_word": false,
|
| 129 |
-
"special": true
|
| 130 |
-
},
|
| 131 |
-
"57533": {
|
| 132 |
-
"content": "<s_Numero>",
|
| 133 |
-
"lstrip": true,
|
| 134 |
-
"normalized": false,
|
| 135 |
-
"rstrip": true,
|
| 136 |
-
"single_word": false,
|
| 137 |
-
"special": true
|
| 138 |
-
},
|
| 139 |
-
"57534": {
|
| 140 |
-
"content": "<s_Fornitore>",
|
| 141 |
-
"lstrip": true,
|
| 142 |
-
"normalized": false,
|
| 143 |
-
"rstrip": true,
|
| 144 |
-
"single_word": false,
|
| 145 |
-
"special": true
|
| 146 |
-
},
|
| 147 |
-
"57535": {
|
| 148 |
-
"content": "</s_Posizione>",
|
| 149 |
-
"lstrip": true,
|
| 150 |
-
"normalized": false,
|
| 151 |
-
"rstrip": true,
|
| 152 |
-
"single_word": false,
|
| 153 |
-
"special": true
|
| 154 |
-
},
|
| 155 |
-
"57536": {
|
| 156 |
-
"content": "<s_Oggetti>",
|
| 157 |
-
"lstrip": true,
|
| 158 |
-
"normalized": false,
|
| 159 |
-
"rstrip": true,
|
| 160 |
-
"single_word": false,
|
| 161 |
-
"special": true
|
| 162 |
-
},
|
| 163 |
-
"57537": {
|
| 164 |
-
"content": "</s_Codice>",
|
| 165 |
-
"lstrip": true,
|
| 166 |
-
"normalized": false,
|
| 167 |
-
"rstrip": true,
|
| 168 |
-
"single_word": false,
|
| 169 |
-
"special": true
|
| 170 |
-
},
|
| 171 |
-
"57538": {
|
| 172 |
-
"content": "</s_Data>",
|
| 173 |
-
"lstrip": true,
|
| 174 |
-
"normalized": false,
|
| 175 |
-
"rstrip": true,
|
| 176 |
-
"single_word": false,
|
| 177 |
-
"special": true
|
| 178 |
}
|
| 179 |
},
|
| 180 |
"additional_special_tokens": [
|
| 181 |
"<s_iitcdip>",
|
| 182 |
"<s_synthdog>",
|
| 183 |
-
"<s_Data>",
|
| 184 |
-
"<s_Codice>",
|
| 185 |
"</s>",
|
| 186 |
-
"
|
| 187 |
-
"<s_Posizione>",
|
| 188 |
-
"</s_Numero>",
|
| 189 |
-
"</s_Oggetti>",
|
| 190 |
-
"<s_Quantita>",
|
| 191 |
-
"</s_Fornitore>",
|
| 192 |
-
"<s_Numero>",
|
| 193 |
-
"<s_Fornitore>",
|
| 194 |
-
"</s_Posizione>",
|
| 195 |
-
"<s_Oggetti>",
|
| 196 |
-
"<s>",
|
| 197 |
-
"</s_Codice>",
|
| 198 |
-
"</s_Data>"
|
| 199 |
],
|
| 200 |
"bos_token": "<s>",
|
| 201 |
"clean_up_tokenization_spaces": true,
|
|
|
|
| 63 |
"rstrip": true,
|
| 64 |
"single_word": false,
|
| 65 |
"special": true
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 66 |
}
|
| 67 |
},
|
| 68 |
"additional_special_tokens": [
|
| 69 |
"<s_iitcdip>",
|
| 70 |
"<s_synthdog>",
|
|
|
|
|
|
|
| 71 |
"</s>",
|
| 72 |
+
"<s>"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 73 |
],
|
| 74 |
"bos_token": "<s>",
|
| 75 |
"clean_up_tokenization_spaces": true,
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4664
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:302bc1cd3c42d4865d7a333486995b0772b5be71d25dc4bbfed33b77458bffee
|
| 3 |
size 4664
|