Training in progress, epoch 1

added_tokens.json

@@ -0,0 +1,3 @@
+{
+  "<mask>": 64000
+}

config.json

@@ -0,0 +1,75 @@
+{
+  "_name_or_path": "vinai/phobert-large",
+  "architectures": [
+    "RobertaForTokenClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "bos_token_id": 0,
+  "classifier_dropout": null,
+  "eos_token_id": 2,
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 1024,
+  "id2label": {
+    "0": "M\u00e3 \u0111\u01a1n",
+    "1": "Kh\u1ed1i l\u01b0\u1ee3ng",
+    "2": "STK",
+    "3": "\u0110\u01a1n v\u1ecb \u0111o",
+    "4": "O",
+    "5": "S\u0110T",
+    "6": "S\u1ea3n ph\u1ea9m c\u1ee5 th\u1ec3",
+    "7": "Ti\u1ec1n",
+    "8": "CCCD - CMT",
+    "9": "T\u00ean ng\u01b0\u1eddi",
+    "10": "S\u1ea3n ph\u1ea9m tr\u1eebu t\u01b0\u1ee3ng",
+    "11": "Ph\u01b0\u01a1ng th\u1ee9c thanh to\u00e1n",
+    "12": "Ca",
+    "13": "Ng\u00e2n h\u00e0ng",
+    "14": "Ng\u00e0y",
+    "15": "\u0110\u1ecba ch\u1ec9 c\u1ee5 th\u1ec3",
+    "16": "Gi\u1edd",
+    "17": "Email",
+    "18": "Ng\u00e0y tr\u1eebu t\u01b0\u1ee3ng",
+    "19": "\u0110\u1ecba ch\u1ec9 tr\u1eebu t\u01b0\u1ee3ng",
+    "20": "S\u0110T tr\u1eebu t\u01b0\u1ee3ng"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 4096,
+  "label2id": {
+    "CCCD - CMT": 8,
+    "Ca": 12,
+    "Email": 17,
+    "Gi\u1edd": 16,
+    "Kh\u1ed1i l\u01b0\u1ee3ng": 1,
+    "M\u00e3 \u0111\u01a1n": 0,
+    "Ng\u00e0y": 14,
+    "Ng\u00e0y tr\u1eebu t\u01b0\u1ee3ng": 18,
+    "Ng\u00e2n h\u00e0ng": 13,
+    "O": 4,
+    "Ph\u01b0\u01a1ng th\u1ee9c thanh to\u00e1n": 11,
+    "STK": 2,
+    "S\u0110T": 5,
+    "S\u0110T tr\u1eebu t\u01b0\u1ee3ng": 20,
+    "S\u1ea3n ph\u1ea9m c\u1ee5 th\u1ec3": 6,
+    "S\u1ea3n ph\u1ea9m tr\u1eebu t\u01b0\u1ee3ng": 10,
+    "Ti\u1ec1n": 7,
+    "T\u00ean ng\u01b0\u1eddi": 9,
+    "\u0110\u01a1n v\u1ecb \u0111o": 3,
+    "\u0110\u1ecba ch\u1ec9 c\u1ee5 th\u1ec3": 15,
+    "\u0110\u1ecba ch\u1ec9 tr\u1eebu t\u01b0\u1ee3ng": 19
+  },
+  "layer_norm_eps": 1e-05,
+  "max_position_embeddings": 258,
+  "model_type": "roberta",
+  "num_attention_heads": 16,
+  "num_hidden_layers": 24,
+  "pad_token_id": 1,
+  "position_embedding_type": "absolute",
+  "tokenizer_class": "PhobertTokenizer",
+  "torch_dtype": "float32",
+  "transformers_version": "4.43.1",
+  "type_vocab_size": 1,
+  "use_cache": true,
+  "vocab_size": 64001
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8f2d9b07fec6cdc668cb14f044c4e4c234a50d3d5afb7a18f49b670ad92f9842
+size 1472588812

runs/Sep05_12-22-50_nmtuet-serv/events.out.tfevents.1725513773.nmtuet-serv.2332847.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:66672bc3fe3b6649110bd8296408f2de75f9e68bffe8d8d2ddeae8fcb0a5e7d3
+size 4184

runs/Sep05_12-24-13_nmtuet-serv/events.out.tfevents.1725513856.nmtuet-serv.2334194.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b8d312a723340c07a93b106f1b2bf4fe868e082cbb5a9655f7da7df47d1b1a71
+size 6724

runs/Sep05_12-26-05_nmtuet-serv/events.out.tfevents.1725513968.nmtuet-serv.2336535.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cade562e47aff33c98d8d8d4bd9aa56ddbddb48c67ee9764bb352caa90ea162e
+size 6515

runs/Sep05_12-26-52_nmtuet-serv/events.out.tfevents.1725514015.nmtuet-serv.2337583.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c3edec4feb0484ffd300ba835cb6b3da263315868d5f806ae80d1ae0bfad4515
+size 6935

runs/Sep05_14-07-21_nmtuet-serv/events.out.tfevents.1725520044.nmtuet-serv.2413982.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7183e530228981e940e569ddd4be6e4de5b9ed3fb23dfa92ef7812322df664b4
+size 6935

runs/Sep05_14-13-50_nmtuet-serv/events.out.tfevents.1725520432.nmtuet-serv.2423513.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ad3e45ec814560f07905a42df51fca1bc3f7cb6ad0f0b3cf1c535c94855cc744
+size 6935

runs/Sep05_14-20-16_nmtuet-serv/events.out.tfevents.1725520819.nmtuet-serv.2430307.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:23fcdb77c915b80cb22ad8c7536cb293fb11e20b88dc7aaef0ce02cc3f39c308
+size 7230

special_tokens_map.json

@@ -0,0 +1,9 @@
+{
+  "bos_token": "<s>",
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "mask_token": "<mask>",
+  "pad_token": "<pad>",
+  "sep_token": "</s>",
+  "unk_token": "<unk>"
+}

tokenizer_config.json

@@ -0,0 +1,54 @@
+{
+  "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
+    },
+    "64000": {
+      "content": "<mask>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<s>",
+  "clean_up_tokenization_spaces": true,
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "mask_token": "<mask>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<pad>",
+  "sep_token": "</s>",
+  "tokenizer_class": "PhobertTokenizer",
+  "unk_token": "<unk>"
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:38ae9524fa597521a5c6fe18345987ba9dc144d9226c9553e91148d80a0eb681
+size 5240