Upload 5 files

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+tokenizer.json filter=lfs diff=lfs merge=lfs -text

config.json

@@ -0,0 +1,298 @@
+{
+  "_name_or_path": "FacebookAI/xlm-roberta-large",
+  "architectures": [
+    "XLMRobertaForTokenClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "bos_token_id": 0,
+  "classifier_dropout": null,
+  "eos_token_id": 2,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 1024,
+  "id2label": {
+    "0": "B-art-broadcastprogram",
+    "1": "B-art-film",
+    "2": "B-art-music",
+    "3": "B-art-other",
+    "4": "B-art-painting",
+    "5": "B-art-writtenart",
+    "6": "B-building-airport",
+    "7": "B-building-hospital",
+    "8": "B-building-hotel",
+    "9": "B-building-library",
+    "10": "B-building-other",
+    "11": "B-building-restaurant",
+    "12": "B-building-sportsfacility",
+    "13": "B-building-theater",
+    "14": "B-event-attack/battle/war/militaryconflict",
+    "15": "B-event-disaster",
+    "16": "B-event-election",
+    "17": "B-event-other",
+    "18": "B-event-protest",
+    "19": "B-event-sportsevent",
+    "20": "B-location-GPE",
+    "21": "B-location-bodiesofwater",
+    "22": "B-location-island",
+    "23": "B-location-mountain",
+    "24": "B-location-other",
+    "25": "B-location-park",
+    "26": "B-location-road/railway/highway/transit",
+    "27": "B-organization-company",
+    "28": "B-organization-education",
+    "29": "B-organization-government/governmentagency",
+    "30": "B-organization-media/newspaper",
+    "31": "B-organization-other",
+    "32": "B-organization-politicalparty",
+    "33": "B-organization-religion",
+    "34": "B-organization-showorganization",
+    "35": "B-organization-sportsleague",
+    "36": "B-organization-sportsteam",
+    "37": "B-other-astronomything",
+    "38": "B-other-award",
+    "39": "B-other-biologything",
+    "40": "B-other-chemicalthing",
+    "41": "B-other-currency",
+    "42": "B-other-disease",
+    "43": "B-other-educationaldegree",
+    "44": "B-other-god",
+    "45": "B-other-language",
+    "46": "B-other-law",
+    "47": "B-other-livingthing",
+    "48": "B-other-medical",
+    "49": "B-person-actor",
+    "50": "B-person-artist/author",
+    "51": "B-person-athlete",
+    "52": "B-person-director",
+    "53": "B-person-other",
+    "54": "B-person-politician",
+    "55": "B-person-scholar",
+    "56": "B-person-soldier",
+    "57": "B-product-airplane",
+    "58": "B-product-car",
+    "59": "B-product-food",
+    "60": "B-product-game",
+    "61": "B-product-other",
+    "62": "B-product-ship",
+    "63": "B-product-software",
+    "64": "B-product-train",
+    "65": "B-product-weapon",
+    "66": "I-art-broadcastprogram",
+    "67": "I-art-film",
+    "68": "I-art-music",
+    "69": "I-art-other",
+    "70": "I-art-painting",
+    "71": "I-art-writtenart",
+    "72": "I-building-airport",
+    "73": "I-building-hospital",
+    "74": "I-building-hotel",
+    "75": "I-building-library",
+    "76": "I-building-other",
+    "77": "I-building-restaurant",
+    "78": "I-building-sportsfacility",
+    "79": "I-building-theater",
+    "80": "I-event-attack/battle/war/militaryconflict",
+    "81": "I-event-disaster",
+    "82": "I-event-election",
+    "83": "I-event-other",
+    "84": "I-event-protest",
+    "85": "I-event-sportsevent",
+    "86": "I-location-GPE",
+    "87": "I-location-bodiesofwater",
+    "88": "I-location-island",
+    "89": "I-location-mountain",
+    "90": "I-location-other",
+    "91": "I-location-park",
+    "92": "I-location-road/railway/highway/transit",
+    "93": "I-organization-company",
+    "94": "I-organization-education",
+    "95": "I-organization-government/governmentagency",
+    "96": "I-organization-media/newspaper",
+    "97": "I-organization-other",
+    "98": "I-organization-politicalparty",
+    "99": "I-organization-religion",
+    "100": "I-organization-showorganization",
+    "101": "I-organization-sportsleague",
+    "102": "I-organization-sportsteam",
+    "103": "I-other-astronomything",
+    "104": "I-other-award",
+    "105": "I-other-biologything",
+    "106": "I-other-chemicalthing",
+    "107": "I-other-currency",
+    "108": "I-other-disease",
+    "109": "I-other-educationaldegree",
+    "110": "I-other-god",
+    "111": "I-other-language",
+    "112": "I-other-law",
+    "113": "I-other-livingthing",
+    "114": "I-other-medical",
+    "115": "I-person-actor",
+    "116": "I-person-artist/author",
+    "117": "I-person-athlete",
+    "118": "I-person-director",
+    "119": "I-person-other",
+    "120": "I-person-politician",
+    "121": "I-person-scholar",
+    "122": "I-person-soldier",
+    "123": "I-product-airplane",
+    "124": "I-product-car",
+    "125": "I-product-food",
+    "126": "I-product-game",
+    "127": "I-product-other",
+    "128": "I-product-ship",
+    "129": "I-product-software",
+    "130": "I-product-train",
+    "131": "I-product-weapon",
+    "132": "O"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 4096,
+  "label2id": {
+    "B-art-broadcastprogram": 0,
+    "B-art-film": 1,
+    "B-art-music": 2,
+    "B-art-other": 3,
+    "B-art-painting": 4,
+    "B-art-writtenart": 5,
+    "B-building-airport": 6,
+    "B-building-hospital": 7,
+    "B-building-hotel": 8,
+    "B-building-library": 9,
+    "B-building-other": 10,
+    "B-building-restaurant": 11,
+    "B-building-sportsfacility": 12,
+    "B-building-theater": 13,
+    "B-event-attack/battle/war/militaryconflict": 14,
+    "B-event-disaster": 15,
+    "B-event-election": 16,
+    "B-event-other": 17,
+    "B-event-protest": 18,
+    "B-event-sportsevent": 19,
+    "B-location-GPE": 20,
+    "B-location-bodiesofwater": 21,
+    "B-location-island": 22,
+    "B-location-mountain": 23,
+    "B-location-other": 24,
+    "B-location-park": 25,
+    "B-location-road/railway/highway/transit": 26,
+    "B-organization-company": 27,
+    "B-organization-education": 28,
+    "B-organization-government/governmentagency": 29,
+    "B-organization-media/newspaper": 30,
+    "B-organization-other": 31,
+    "B-organization-politicalparty": 32,
+    "B-organization-religion": 33,
+    "B-organization-showorganization": 34,
+    "B-organization-sportsleague": 35,
+    "B-organization-sportsteam": 36,
+    "B-other-astronomything": 37,
+    "B-other-award": 38,
+    "B-other-biologything": 39,
+    "B-other-chemicalthing": 40,
+    "B-other-currency": 41,
+    "B-other-disease": 42,
+    "B-other-educationaldegree": 43,
+    "B-other-god": 44,
+    "B-other-language": 45,
+    "B-other-law": 46,
+    "B-other-livingthing": 47,
+    "B-other-medical": 48,
+    "B-person-actor": 49,
+    "B-person-artist/author": 50,
+    "B-person-athlete": 51,
+    "B-person-director": 52,
+    "B-person-other": 53,
+    "B-person-politician": 54,
+    "B-person-scholar": 55,
+    "B-person-soldier": 56,
+    "B-product-airplane": 57,
+    "B-product-car": 58,
+    "B-product-food": 59,
+    "B-product-game": 60,
+    "B-product-other": 61,
+    "B-product-ship": 62,
+    "B-product-software": 63,
+    "B-product-train": 64,
+    "B-product-weapon": 65,
+    "I-art-broadcastprogram": 66,
+    "I-art-film": 67,
+    "I-art-music": 68,
+    "I-art-other": 69,
+    "I-art-painting": 70,
+    "I-art-writtenart": 71,
+    "I-building-airport": 72,
+    "I-building-hospital": 73,
+    "I-building-hotel": 74,
+    "I-building-library": 75,
+    "I-building-other": 76,
+    "I-building-restaurant": 77,
+    "I-building-sportsfacility": 78,
+    "I-building-theater": 79,
+    "I-event-attack/battle/war/militaryconflict": 80,
+    "I-event-disaster": 81,
+    "I-event-election": 82,
+    "I-event-other": 83,
+    "I-event-protest": 84,
+    "I-event-sportsevent": 85,
+    "I-location-GPE": 86,
+    "I-location-bodiesofwater": 87,
+    "I-location-island": 88,
+    "I-location-mountain": 89,
+    "I-location-other": 90,
+    "I-location-park": 91,
+    "I-location-road/railway/highway/transit": 92,
+    "I-organization-company": 93,
+    "I-organization-education": 94,
+    "I-organization-government/governmentagency": 95,
+    "I-organization-media/newspaper": 96,
+    "I-organization-other": 97,
+    "I-organization-politicalparty": 98,
+    "I-organization-religion": 99,
+    "I-organization-showorganization": 100,
+    "I-organization-sportsleague": 101,
+    "I-organization-sportsteam": 102,
+    "I-other-astronomything": 103,
+    "I-other-award": 104,
+    "I-other-biologything": 105,
+    "I-other-chemicalthing": 106,
+    "I-other-currency": 107,
+    "I-other-disease": 108,
+    "I-other-educationaldegree": 109,
+    "I-other-god": 110,
+    "I-other-language": 111,
+    "I-other-law": 112,
+    "I-other-livingthing": 113,
+    "I-other-medical": 114,
+    "I-person-actor": 115,
+    "I-person-artist/author": 116,
+    "I-person-athlete": 117,
+    "I-person-director": 118,
+    "I-person-other": 119,
+    "I-person-politician": 120,
+    "I-person-scholar": 121,
+    "I-person-soldier": 122,
+    "I-product-airplane": 123,
+    "I-product-car": 124,
+    "I-product-food": 125,
+    "I-product-game": 126,
+    "I-product-other": 127,
+    "I-product-ship": 128,
+    "I-product-software": 129,
+    "I-product-train": 130,
+    "I-product-weapon": 131,
+    "O": 132
+  },
+  "layer_norm_eps": 1e-05,
+  "max_position_embeddings": 514,
+  "model_type": "xlm-roberta",
+  "num_attention_heads": 16,
+  "num_hidden_layers": 24,
+  "output_past": true,
+  "pad_token_id": 1,
+  "position_embedding_type": "absolute",
+  "torch_dtype": "float32",
+  "transformers_version": "4.47.0",
+  "type_vocab_size": 1,
+  "use_cache": true,
+  "vocab_size": 250002
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ab8bbf9bf16848810af5aadc53ab20ae8b93c1c8c519b4c10b12c38f28e0ffa7
+size 2235957164

special_tokens_map.json

@@ -0,0 +1,15 @@
+{
+  "bos_token": "<s>",
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "mask_token": {
+    "content": "<mask>",
+    "lstrip": true,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<pad>",
+  "sep_token": "</s>",
+  "unk_token": "<unk>"
+}

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8373f9cd3d27591e1924426bcc1c8799bc5a9affc4fc857982c5d66668dd1f41
+size 17082832

tokenizer_config.json

@@ -0,0 +1,56 @@
+{
+  "add_prefix_space": true,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250001": {
+      "content": "<mask>",
+      "lstrip": true,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<s>",
+  "clean_up_tokenization_spaces": false,
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "extra_special_tokens": {},
+  "mask_token": "<mask>",
+  "model_max_length": 512,
+  "pad_token": "<pad>",
+  "sep_token": "</s>",
+  "tokenizer_class": "XLMRobertaTokenizer",
+  "unk_token": "<unk>"
+}