Mejor modelo hasta epoch 0 (val_edit_distance=0.7593764066696167)
Browse files
es-digital-seq/adapter_model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 93378688
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f7d787b192262ea1162030a05e3cad765edc0cb1899cbf7669715791e28c6beb
|
| 3 |
size 93378688
|
es-digital-seq/added_tokens.json
CHANGED
|
@@ -1,15 +1,17 @@
|
|
| 1 |
{
|
| 2 |
-
"</s_1_de_bachillerato>":
|
| 3 |
-
"</s_3_de_la_eso>":
|
| 4 |
-
"</s_4_de_la_eso>":
|
| 5 |
-
"</s_grade>":
|
| 6 |
-
"</
|
|
|
|
| 7 |
"<end_of_utterance>": 32002,
|
| 8 |
"<fake_token_around_image>": 32000,
|
| 9 |
"<image>": 32001,
|
| 10 |
-
"<s_1_de_bachillerato>":
|
| 11 |
-
"<s_3_de_la_eso>":
|
| 12 |
-
"<s_4_de_la_eso>":
|
| 13 |
-
"<s_grade>":
|
| 14 |
-
"<
|
|
|
|
| 15 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"</s_1_de_bachillerato>": 32014,
|
| 3 |
+
"</s_3_de_la_eso>": 32006,
|
| 4 |
+
"</s_4_de_la_eso>": 32012,
|
| 5 |
+
"</s_grade>": 32010,
|
| 6 |
+
"</s_gt_parse>": 32004,
|
| 7 |
+
"</s_subject>": 32008,
|
| 8 |
"<end_of_utterance>": 32002,
|
| 9 |
"<fake_token_around_image>": 32000,
|
| 10 |
"<image>": 32001,
|
| 11 |
+
"<s_1_de_bachillerato>": 32013,
|
| 12 |
+
"<s_3_de_la_eso>": 32005,
|
| 13 |
+
"<s_4_de_la_eso>": 32011,
|
| 14 |
+
"<s_grade>": 32009,
|
| 15 |
+
"<s_gt_parse>": 32003,
|
| 16 |
+
"<s_subject>": 32007
|
| 17 |
}
|
es-digital-seq/tokenizer.json
CHANGED
|
@@ -59,7 +59,7 @@
|
|
| 59 |
},
|
| 60 |
{
|
| 61 |
"id": 32003,
|
| 62 |
-
"content": "<
|
| 63 |
"single_word": false,
|
| 64 |
"lstrip": false,
|
| 65 |
"rstrip": false,
|
|
@@ -68,7 +68,7 @@
|
|
| 68 |
},
|
| 69 |
{
|
| 70 |
"id": 32004,
|
| 71 |
-
"content": "</
|
| 72 |
"single_word": false,
|
| 73 |
"lstrip": false,
|
| 74 |
"rstrip": false,
|
|
@@ -77,7 +77,7 @@
|
|
| 77 |
},
|
| 78 |
{
|
| 79 |
"id": 32005,
|
| 80 |
-
"content": "<
|
| 81 |
"single_word": false,
|
| 82 |
"lstrip": false,
|
| 83 |
"rstrip": false,
|
|
@@ -86,7 +86,7 @@
|
|
| 86 |
},
|
| 87 |
{
|
| 88 |
"id": 32006,
|
| 89 |
-
"content": "</
|
| 90 |
"single_word": false,
|
| 91 |
"lstrip": false,
|
| 92 |
"rstrip": false,
|
|
@@ -95,7 +95,7 @@
|
|
| 95 |
},
|
| 96 |
{
|
| 97 |
"id": 32007,
|
| 98 |
-
"content": "<
|
| 99 |
"single_word": false,
|
| 100 |
"lstrip": false,
|
| 101 |
"rstrip": false,
|
|
@@ -104,7 +104,7 @@
|
|
| 104 |
},
|
| 105 |
{
|
| 106 |
"id": 32008,
|
| 107 |
-
"content": "</
|
| 108 |
"single_word": false,
|
| 109 |
"lstrip": false,
|
| 110 |
"rstrip": false,
|
|
@@ -113,7 +113,7 @@
|
|
| 113 |
},
|
| 114 |
{
|
| 115 |
"id": 32009,
|
| 116 |
-
"content": "<
|
| 117 |
"single_word": false,
|
| 118 |
"lstrip": false,
|
| 119 |
"rstrip": false,
|
|
@@ -122,7 +122,7 @@
|
|
| 122 |
},
|
| 123 |
{
|
| 124 |
"id": 32010,
|
| 125 |
-
"content": "</
|
| 126 |
"single_word": false,
|
| 127 |
"lstrip": false,
|
| 128 |
"rstrip": false,
|
|
@@ -131,7 +131,7 @@
|
|
| 131 |
},
|
| 132 |
{
|
| 133 |
"id": 32011,
|
| 134 |
-
"content": "<
|
| 135 |
"single_word": false,
|
| 136 |
"lstrip": false,
|
| 137 |
"rstrip": false,
|
|
@@ -140,6 +140,24 @@
|
|
| 140 |
},
|
| 141 |
{
|
| 142 |
"id": 32012,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 143 |
"content": "</s_1_de_bachillerato>",
|
| 144 |
"single_word": false,
|
| 145 |
"lstrip": false,
|
|
|
|
| 59 |
},
|
| 60 |
{
|
| 61 |
"id": 32003,
|
| 62 |
+
"content": "<s_gt_parse>",
|
| 63 |
"single_word": false,
|
| 64 |
"lstrip": false,
|
| 65 |
"rstrip": false,
|
|
|
|
| 68 |
},
|
| 69 |
{
|
| 70 |
"id": 32004,
|
| 71 |
+
"content": "</s_gt_parse>",
|
| 72 |
"single_word": false,
|
| 73 |
"lstrip": false,
|
| 74 |
"rstrip": false,
|
|
|
|
| 77 |
},
|
| 78 |
{
|
| 79 |
"id": 32005,
|
| 80 |
+
"content": "<s_3_de_la_eso>",
|
| 81 |
"single_word": false,
|
| 82 |
"lstrip": false,
|
| 83 |
"rstrip": false,
|
|
|
|
| 86 |
},
|
| 87 |
{
|
| 88 |
"id": 32006,
|
| 89 |
+
"content": "</s_3_de_la_eso>",
|
| 90 |
"single_word": false,
|
| 91 |
"lstrip": false,
|
| 92 |
"rstrip": false,
|
|
|
|
| 95 |
},
|
| 96 |
{
|
| 97 |
"id": 32007,
|
| 98 |
+
"content": "<s_subject>",
|
| 99 |
"single_word": false,
|
| 100 |
"lstrip": false,
|
| 101 |
"rstrip": false,
|
|
|
|
| 104 |
},
|
| 105 |
{
|
| 106 |
"id": 32008,
|
| 107 |
+
"content": "</s_subject>",
|
| 108 |
"single_word": false,
|
| 109 |
"lstrip": false,
|
| 110 |
"rstrip": false,
|
|
|
|
| 113 |
},
|
| 114 |
{
|
| 115 |
"id": 32009,
|
| 116 |
+
"content": "<s_grade>",
|
| 117 |
"single_word": false,
|
| 118 |
"lstrip": false,
|
| 119 |
"rstrip": false,
|
|
|
|
| 122 |
},
|
| 123 |
{
|
| 124 |
"id": 32010,
|
| 125 |
+
"content": "</s_grade>",
|
| 126 |
"single_word": false,
|
| 127 |
"lstrip": false,
|
| 128 |
"rstrip": false,
|
|
|
|
| 131 |
},
|
| 132 |
{
|
| 133 |
"id": 32011,
|
| 134 |
+
"content": "<s_4_de_la_eso>",
|
| 135 |
"single_word": false,
|
| 136 |
"lstrip": false,
|
| 137 |
"rstrip": false,
|
|
|
|
| 140 |
},
|
| 141 |
{
|
| 142 |
"id": 32012,
|
| 143 |
+
"content": "</s_4_de_la_eso>",
|
| 144 |
+
"single_word": false,
|
| 145 |
+
"lstrip": false,
|
| 146 |
+
"rstrip": false,
|
| 147 |
+
"normalized": true,
|
| 148 |
+
"special": false
|
| 149 |
+
},
|
| 150 |
+
{
|
| 151 |
+
"id": 32013,
|
| 152 |
+
"content": "<s_1_de_bachillerato>",
|
| 153 |
+
"single_word": false,
|
| 154 |
+
"lstrip": false,
|
| 155 |
+
"rstrip": false,
|
| 156 |
+
"normalized": true,
|
| 157 |
+
"special": false
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"id": 32014,
|
| 161 |
"content": "</s_1_de_bachillerato>",
|
| 162 |
"single_word": false,
|
| 163 |
"lstrip": false,
|
es-digital-seq/tokenizer_config.json
CHANGED
|
@@ -52,7 +52,7 @@
|
|
| 52 |
"special": true
|
| 53 |
},
|
| 54 |
"32003": {
|
| 55 |
-
"content": "<
|
| 56 |
"lstrip": false,
|
| 57 |
"normalized": true,
|
| 58 |
"rstrip": false,
|
|
@@ -60,7 +60,7 @@
|
|
| 60 |
"special": false
|
| 61 |
},
|
| 62 |
"32004": {
|
| 63 |
-
"content": "</
|
| 64 |
"lstrip": false,
|
| 65 |
"normalized": true,
|
| 66 |
"rstrip": false,
|
|
@@ -68,7 +68,7 @@
|
|
| 68 |
"special": false
|
| 69 |
},
|
| 70 |
"32005": {
|
| 71 |
-
"content": "<
|
| 72 |
"lstrip": false,
|
| 73 |
"normalized": true,
|
| 74 |
"rstrip": false,
|
|
@@ -76,7 +76,7 @@
|
|
| 76 |
"special": false
|
| 77 |
},
|
| 78 |
"32006": {
|
| 79 |
-
"content": "</
|
| 80 |
"lstrip": false,
|
| 81 |
"normalized": true,
|
| 82 |
"rstrip": false,
|
|
@@ -84,7 +84,7 @@
|
|
| 84 |
"special": false
|
| 85 |
},
|
| 86 |
"32007": {
|
| 87 |
-
"content": "<
|
| 88 |
"lstrip": false,
|
| 89 |
"normalized": true,
|
| 90 |
"rstrip": false,
|
|
@@ -92,7 +92,7 @@
|
|
| 92 |
"special": false
|
| 93 |
},
|
| 94 |
"32008": {
|
| 95 |
-
"content": "</
|
| 96 |
"lstrip": false,
|
| 97 |
"normalized": true,
|
| 98 |
"rstrip": false,
|
|
@@ -100,7 +100,7 @@
|
|
| 100 |
"special": false
|
| 101 |
},
|
| 102 |
"32009": {
|
| 103 |
-
"content": "<
|
| 104 |
"lstrip": false,
|
| 105 |
"normalized": true,
|
| 106 |
"rstrip": false,
|
|
@@ -108,7 +108,7 @@
|
|
| 108 |
"special": false
|
| 109 |
},
|
| 110 |
"32010": {
|
| 111 |
-
"content": "</
|
| 112 |
"lstrip": false,
|
| 113 |
"normalized": true,
|
| 114 |
"rstrip": false,
|
|
@@ -116,7 +116,7 @@
|
|
| 116 |
"special": false
|
| 117 |
},
|
| 118 |
"32011": {
|
| 119 |
-
"content": "<
|
| 120 |
"lstrip": false,
|
| 121 |
"normalized": true,
|
| 122 |
"rstrip": false,
|
|
@@ -124,6 +124,22 @@
|
|
| 124 |
"special": false
|
| 125 |
},
|
| 126 |
"32012": {
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 127 |
"content": "</s_1_de_bachillerato>",
|
| 128 |
"lstrip": false,
|
| 129 |
"normalized": true,
|
|
|
|
| 52 |
"special": true
|
| 53 |
},
|
| 54 |
"32003": {
|
| 55 |
+
"content": "<s_gt_parse>",
|
| 56 |
"lstrip": false,
|
| 57 |
"normalized": true,
|
| 58 |
"rstrip": false,
|
|
|
|
| 60 |
"special": false
|
| 61 |
},
|
| 62 |
"32004": {
|
| 63 |
+
"content": "</s_gt_parse>",
|
| 64 |
"lstrip": false,
|
| 65 |
"normalized": true,
|
| 66 |
"rstrip": false,
|
|
|
|
| 68 |
"special": false
|
| 69 |
},
|
| 70 |
"32005": {
|
| 71 |
+
"content": "<s_3_de_la_eso>",
|
| 72 |
"lstrip": false,
|
| 73 |
"normalized": true,
|
| 74 |
"rstrip": false,
|
|
|
|
| 76 |
"special": false
|
| 77 |
},
|
| 78 |
"32006": {
|
| 79 |
+
"content": "</s_3_de_la_eso>",
|
| 80 |
"lstrip": false,
|
| 81 |
"normalized": true,
|
| 82 |
"rstrip": false,
|
|
|
|
| 84 |
"special": false
|
| 85 |
},
|
| 86 |
"32007": {
|
| 87 |
+
"content": "<s_subject>",
|
| 88 |
"lstrip": false,
|
| 89 |
"normalized": true,
|
| 90 |
"rstrip": false,
|
|
|
|
| 92 |
"special": false
|
| 93 |
},
|
| 94 |
"32008": {
|
| 95 |
+
"content": "</s_subject>",
|
| 96 |
"lstrip": false,
|
| 97 |
"normalized": true,
|
| 98 |
"rstrip": false,
|
|
|
|
| 100 |
"special": false
|
| 101 |
},
|
| 102 |
"32009": {
|
| 103 |
+
"content": "<s_grade>",
|
| 104 |
"lstrip": false,
|
| 105 |
"normalized": true,
|
| 106 |
"rstrip": false,
|
|
|
|
| 108 |
"special": false
|
| 109 |
},
|
| 110 |
"32010": {
|
| 111 |
+
"content": "</s_grade>",
|
| 112 |
"lstrip": false,
|
| 113 |
"normalized": true,
|
| 114 |
"rstrip": false,
|
|
|
|
| 116 |
"special": false
|
| 117 |
},
|
| 118 |
"32011": {
|
| 119 |
+
"content": "<s_4_de_la_eso>",
|
| 120 |
"lstrip": false,
|
| 121 |
"normalized": true,
|
| 122 |
"rstrip": false,
|
|
|
|
| 124 |
"special": false
|
| 125 |
},
|
| 126 |
"32012": {
|
| 127 |
+
"content": "</s_4_de_la_eso>",
|
| 128 |
+
"lstrip": false,
|
| 129 |
+
"normalized": true,
|
| 130 |
+
"rstrip": false,
|
| 131 |
+
"single_word": false,
|
| 132 |
+
"special": false
|
| 133 |
+
},
|
| 134 |
+
"32013": {
|
| 135 |
+
"content": "<s_1_de_bachillerato>",
|
| 136 |
+
"lstrip": false,
|
| 137 |
+
"normalized": true,
|
| 138 |
+
"rstrip": false,
|
| 139 |
+
"single_word": false,
|
| 140 |
+
"special": false
|
| 141 |
+
},
|
| 142 |
+
"32014": {
|
| 143 |
"content": "</s_1_de_bachillerato>",
|
| 144 |
"lstrip": false,
|
| 145 |
"normalized": true,
|