Upload tokenizer

added_tokens.json

@@ -0,0 +1,128 @@
+{
+  "</s_Acceso>": 57609,
+  "</s_Avance_Obra>": 57543,
+  "</s_CP>": 57539,
+  "</s_Calle>": 57535,
+  "</s_Cedula_arqui_cont>": 57569,
+  "</s_Cedula_arqui_val>": 57561,
+  "</s_Cedula_cont>": 57567,
+  "</s_Cedula_val>": 57563,
+  "</s_Ciudad>": 57537,
+  "</s_Colonia>": 57531,
+  "</s_Construcion_M2>": 57579,
+  "</s_Consultora>": 57611,
+  "</s_Costo_insta>": 57639,
+  "</s_Costo_neto_repo>": 57547,
+  "</s_Costo_neto_repo_dem>": 57553,
+  "</s_Costo_uni_neto_repo>": 57545,
+  "</s_Deterioro>": 57555,
+  "</s_Dis_val_com_homo>": 57627,
+  "</s_Dis_val_com_pre_homo>": 57615,
+  "</s_Edad_const>": 57589,
+  "</s_Equipo_Urbano>": 57631,
+  "</s_Escritura>": 57599,
+  "</s_Estado>": 57541,
+  "</s_Factor>": 57623,
+  "</s_Fecha_del_avaluo>": 57597,
+  "</s_Georeferencia>": 57527,
+  "</s_ID>": 57613,
+  "</s_Indice_fisico>": 57551,
+  "</s_Infraestructura>": 57573,
+  "</s_Interior>": 57637,
+  "</s_Lote>": 57525,
+  "</s_Manzana>": 57533,
+  "</s_Notas>": 57643,
+  "</s_Numero_de_avaluo>": 57595,
+  "</s_Numero_exterior>": 57529,
+  "</s_Ocupacion>": 57603,
+  "</s_Predial>": 57605,
+  "</s_Propietario>": 57601,
+  "</s_Proposito>": 57593,
+  "</s_Prototipo>": 57585,
+  "</s_Relacion_valor>": 57619,
+  "</s_SHF>": 57645,
+  "</s_Servicios_muni>": 57571,
+  "</s_Solicitante>": 57641,
+  "</s_Tipo_de_inmueble>": 57607,
+  "</s_Val_comp_merc>": 57621,
+  "</s_Val_hip>": 57629,
+  "</s_Val_uni_con_dete>": 57625,
+  "</s_Val_uni_sin_dete>": 57617,
+  "</s_Valor_concluido>": 57633,
+  "</s_Valor_fisico>": 57557,
+  "</s_Valor_homo_M2>": 57549,
+  "</s_contro>": 57565,
+  "</s_letra>": 57635,
+  "</s_no_banos>": 57575,
+  "</s_no_estacionamiento>": 57583,
+  "</s_no_niv_UV>": 57591,
+  "</s_no_recamaras>": 57587,
+  "</s_terreno_M2>": 57577,
+  "</s_valu>": 57559,
+  "</s_vendible_M2>": 57581,
+  "<s_Acceso>": 57610,
+  "<s_Avance_Obra>": 57544,
+  "<s_CP>": 57540,
+  "<s_Calle>": 57536,
+  "<s_Cedula_arqui_cont>": 57570,
+  "<s_Cedula_arqui_val>": 57562,
+  "<s_Cedula_cont>": 57568,
+  "<s_Cedula_val>": 57564,
+  "<s_Ciudad>": 57538,
+  "<s_Colonia>": 57532,
+  "<s_Construcion_M2>": 57580,
+  "<s_Consultora>": 57612,
+  "<s_Costo_insta>": 57640,
+  "<s_Costo_neto_repo>": 57548,
+  "<s_Costo_neto_repo_dem>": 57554,
+  "<s_Costo_uni_neto_repo>": 57546,
+  "<s_Deterioro>": 57556,
+  "<s_Dis_val_com_homo>": 57628,
+  "<s_Dis_val_com_pre_homo>": 57616,
+  "<s_Edad_const>": 57590,
+  "<s_Equipo_Urbano>": 57632,
+  "<s_Escritura>": 57600,
+  "<s_Estado>": 57542,
+  "<s_Factor>": 57624,
+  "<s_Fecha_del_avaluo>": 57598,
+  "<s_Georeferencia>": 57528,
+  "<s_ID>": 57614,
+  "<s_Indice_fisico>": 57552,
+  "<s_Infraestructura>": 57574,
+  "<s_Interior>": 57638,
+  "<s_Lote>": 57526,
+  "<s_Manzana>": 57534,
+  "<s_Notas>": 57644,
+  "<s_Numero_de_avaluo>": 57596,
+  "<s_Numero_exterior>": 57530,
+  "<s_Ocupacion>": 57604,
+  "<s_Predial>": 57606,
+  "<s_Propietario>": 57602,
+  "<s_Proposito>": 57594,
+  "<s_Prototipo>": 57586,
+  "<s_Relacion_valor>": 57620,
+  "<s_SHF>": 57646,
+  "<s_Servicios_muni>": 57572,
+  "<s_Solicitante>": 57642,
+  "<s_Tipo_de_inmueble>": 57608,
+  "<s_Val_comp_merc>": 57622,
+  "<s_Val_hip>": 57630,
+  "<s_Val_uni_con_dete>": 57626,
+  "<s_Val_uni_sin_dete>": 57618,
+  "<s_Valor_concluido>": 57634,
+  "<s_Valor_fisico>": 57558,
+  "<s_Valor_homo_M2>": 57550,
+  "<s_avaluo>": 57647,
+  "<s_contro>": 57566,
+  "<s_iitcdip>": 57523,
+  "<s_letra>": 57636,
+  "<s_no_banos>": 57576,
+  "<s_no_estacionamiento>": 57584,
+  "<s_no_niv_UV>": 57592,
+  "<s_no_recamaras>": 57588,
+  "<s_synthdog>": 57524,
+  "<s_terreno_M2>": 57578,
+  "<s_valu>": 57560,
+  "<s_vendible_M2>": 57582,
+  "<sep/>": 57522
+}

sentencepiece.bpe.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cb9e3dce4c326195d08fc3dd0f7e2eee1da8595c847bf4c1a9c78b7a82d47e2d
+size 1296245

special_tokens_map.json

@@ -0,0 +1,18 @@
+{
+  "additional_special_tokens": [
+    "<sep/>"
+  ],
+  "bos_token": "<s>",
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "mask_token": {
+    "content": "<mask>",
+    "lstrip": true,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<pad>",
+  "sep_token": "</s>",
+  "unk_token": "<unk>"
+}

tokenizer_config.json

@@ -0,0 +1,23 @@
+{
+  "bos_token": "<s>",
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "from_slow": true,
+  "mask_token": {
+    "__type": "AddedToken",
+    "content": "<mask>",
+    "lstrip": true,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "model_max_length": 1000000000000000019884624838656,
+  "name_or_path": "C:\\Users\\Jorge Tellso\\Documents\\OCR\\prueba\\codigo\\result\\avaluo\\20240801_123101",
+  "pad_token": "<pad>",
+  "processor_class": "DonutProcessor",
+  "sep_token": "</s>",
+  "sp_model_kwargs": {},
+  "special_tokens_map_file": null,
+  "tokenizer_class": "XLMRobertaTokenizer",
+  "unk_token": "<unk>"
+}