Update spaCy pipeline

README.md

@@ -13,18 +13,18 @@ model-index:
     metrics:
     - name: NER Precision
       type: precision
-      value: 0.8918918919
+      value: 0.9045226131
     - name: NER Recall
       type: recall
-      value: 0.8823529412
+      value: 0.8955223881
     - name: NER F Score
       type: f_score
-      value: 0.8870967742
+      value: 0.9
 ---
 | Feature | Description |
 | --- | --- |
 | **Name** | `vi_ner_task` |
-| **Version** | `1.0.0` |
+| **Version** | `1.0.1` |
 | **spaCy** | `>=3.7.5,<3.8.0` |
 | **Default Pipeline** | `tok2vec`, `ner` |
 | **Components** | `tok2vec`, `ner` |
@@ -49,8 +49,8 @@ model-index:
 
 | Type | Score |
 | --- | --- |
-| `ENTS_F` | 88.71 |
-| `ENTS_P` | 89.19 |
-| `ENTS_R` | 88.24 |
-| `TOK2VEC_LOSS` | 27395.24 |
-| `NER_LOSS` | 12711.07 |
+| `ENTS_F` | 90.00 |
+| `ENTS_P` | 90.45 |
+| `ENTS_R` | 89.55 |
+| `TOK2VEC_LOSS` | 121354.85 |
+| `NER_LOSS` | 16964.64 |

config.cfg

@@ -9,7 +9,7 @@ gpu_allocator = null
 seed = 0
 
 [nlp]
-lang = "vi"
+lang = "xx"
 pipeline = ["tok2vec","ner"]
 batch_size = 1000
 disabled = []

meta.json

@@ -1,7 +1,7 @@
 {
   "lang":"vi",
   "name":"ner_task",
-  "version":"1.0.0",
+  "version":"1.0.1",
   "description":"",
   "author":"Ch\u00e1nh H\u1ef7",
   "email":"hlchy11997@gmail.com",
@@ -38,33 +38,33 @@
 
   ],
   "performance":{
-    "ents_f":0.8870967742,
-    "ents_p":0.8918918919,
-    "ents_r":0.8823529412,
+    "ents_f":0.9,
+    "ents_p":0.9045226131,
+    "ents_r":0.8955223881,
     "ents_per_type":{
-      "TIME":{
-        "p":0.9512195122,
-        "r":0.9069767442,
-        "f":0.9285714286
-      },
       "DATE":{
-        "p":0.8888888889,
-        "r":0.8888888889,
-        "f":0.8888888889
+        "p":0.8392857143,
+        "r":0.8545454545,
+        "f":0.8468468468
       },
       "TASK":{
-        "p":0.8103448276,
-        "r":0.8103448276,
-        "f":0.8103448276
+        "p":0.875,
+        "r":0.8615384615,
+        "f":0.8682170543
+      },
+      "TIME":{
+        "p":0.9795918367,
+        "r":0.9230769231,
+        "f":0.9504950495
       },
       "PERSON":{
-        "p":0.96875,
-        "r":0.96875,
-        "f":0.96875
+        "p":0.9666666667,
+        "r":1.0,
+        "f":0.9830508475
       }
     },
-    "tok2vec_loss":273.9524151373,
-    "ner_loss":127.1106786892
+    "tok2vec_loss":1213.5484849717,
+    "ner_loss":169.6463696686
   },
   "requirements":[
 

ner/model

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4a9ea7ea2de49d995def50a6dfd9cedef1bd6d73a5c21ff5b7f0a9f1c714c4f8
+oid sha256:9cfe9e0d5d1dc13c7b773a0b0210deff0258bcf1dd25d5261ab2f5ea2a45da0b
 size 130628

ner/moves

@@ -1 +1 @@
-��moves��{"0":{},"1":{"TASK":2657,"DATE":771,"TIME":597,"PERSON":360},"2":{"TASK":2657,"DATE":771,"TIME":597,"PERSON":360},"3":{"TASK":2657,"DATE":771,"TIME":597,"PERSON":360},"4":{"TASK":2657,"DATE":771,"TIME":597,"PERSON":360,"":1},"5":{"":1}}�cfg��neg_key�
+��moves��{"0":{},"1":{"TASK":3050,"DATE":987,"TIME":783,"PERSON":405},"2":{"TASK":3050,"DATE":987,"TIME":783,"PERSON":405},"3":{"TASK":3050,"DATE":987,"TIME":783,"PERSON":405},"4":{"TASK":3050,"DATE":987,"TIME":783,"PERSON":405,"":1},"5":{"":1}}�cfg��neg_key�

tok2vec/model

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ab4bab442ff6d338b1ac05318823db887132498a5fc58a8759ebf1cd34c84df9
+oid sha256:801482097833d619649bf0599975a6dbe63471c569f75ab74facb6d099437483
 size 6009091

vi_ner_task-any-py3-none-any.whl

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a6a4ccaeb1717a48b13fc36d5909bee451dbcb9f23aa9514e7dcca667e315d93
-size 5670015
+oid sha256:4ee49645855b8ed6b597ff20033b30efabc72df4ba3ac344f3d49cf034e15aec
+size 5677973

vocab/strings.json

@@ -68,10 +68,16 @@
   "0.o",
   "0/1",
   "0/9",
+  "025",
+  "026",
+  "027",
+  "028",
   "0_0",
   "0_o",
   "1",
+  "1/1",
   "1/10",
+  "1/11",
   "1/12",
   "10",
   "10/10",
@@ -82,8 +88,10 @@
   "11h30",
   "12",
   "12/10",
+  "12/8",
   "12h",
   "12h00",
+  "13h",
   "13h15",
   "14",
   "14:00",
@@ -91,10 +99,13 @@
   "14h",
   "14h30",
   "15",
+  "15/1",
   "15/10",
   "15/11",
+  "15/3",
   "15h",
   "15h00",
+  "16h",
   "16h45",
   "17",
   "17/10",
@@ -112,20 +123,29 @@
   "2",
   "2/1",
   "2/10",
+  "2/8",
   "20/1",
   "20/10",
   "20/11",
   "20/12",
+  "2025",
+  "2026",
+  "2027",
+  "2028",
   "20h",
   "20h00",
   "21h",
   "21h00",
+  "22",
   "22/11",
   "22h00",
   "23/12",
   "23h59",
+  "24",
   "25/10",
   "25/11",
+  "28/10",
+  "2h",
   "3",
   "3/10",
   "30",
@@ -140,6 +160,9 @@
   "45",
   "4h",
   "5",
+  "5/1",
+  "5/3",
+  "5/5",
   "5h",
   "6",
   "6h",
@@ -289,6 +312,8 @@
   "Ch\u1ecb",
   "Ch\u1ee7",
   "Cu\u1ed1i",
+  "Cu\u1ed9c",
+  "C\u00f4ng",
   "C\u01b0\u1eddng",
   "C\u1ea3",
   "C\u1ea7n",
@@ -311,6 +336,7 @@
   "G",
   "G\u00f2n",
   "G\u1ea5p",
+  "G\u1eedi",
   "H",
   "Hai",
   "Hoa",
@@ -319,6 +345,7 @@
   "Ho\u00e1",
   "Huy",
   "H\u00e0",
+  "H\u00e3y",
   "H\u00f2a",
   "H\u00f3a",
   "H\u00f4m",
@@ -330,6 +357,7 @@
   "H\u1ea3i",
   "H\u1eb1ng",
   "H\u1ecdc",
+  "H\u1ee3p",
   "I",
   "II",
   "IME",
@@ -343,12 +371,14 @@
   "Kh\u1ea9n",
   "Ki\u00ean",
   "Ki\u1ec1u",
+  "K\u1ebf",
   "L",
   "Lan",
   "Linh",
   "L\u00e0m",
   "L\u00e1t",
   "L\u00e2m",
+  "L\u00ean",
   "L\u00fac",
   "L\u1ea1t",
   "L\u1ecbch",
@@ -359,6 +389,7 @@
   "Mobile",
   "M\u00f9ng",
   "M\u1ed1t",
+  "M\u1ee5c",
   "N",
   "NTT",
   "Nam",
@@ -371,6 +402,7 @@
   "Ng\u1eef",
   "Nh\u00f3m",
   "Nh\u1eadt",
+  "Nh\u1eafc",
   "Nh\u1edb",
   "N\u0103m",
   "N\u1ea5u",
@@ -418,6 +450,7 @@
   "Th\u00fay",
   "Th\u1ea3o",
   "Th\u1eafng",
+  "Th\u1eddi",
   "Th\u1ee7",
   "Th\u1ee9",
   "Tin",
@@ -426,6 +459,7 @@
   "Tr\u00e2m",
   "Tr\u01b0a",
   "Tr\u01b0\u1edbc",
+  "Tr\u01b0\u1edfng",
   "Tu\u1ea5n",
   "Tu\u1ea7n",
   "T\u00e0i",
@@ -439,7 +473,9 @@
   "VIP",
   "V_V",
   "Vi\u1ebft",
+  "Vi\u1ec7c",
   "Vi\u1ec7t",
+  "Vui",
   "V\u00e2n",
   "V\u0103n",
   "V\u0169ng",
@@ -544,6 +580,7 @@
   "b\u1ea3o",
   "b\u1ea3y",
   "b\u1eaft",
+  "b\u1eb1ng",
   "b\u1ec7nh",
   "b\u1ecb",
   "b\u1ed1",
@@ -576,10 +613,12 @@
   "ch\u01a1i",
   "ch\u01b0\u01a1ng",
   "ch\u1ea1y",
+  "ch\u1ec9",
   "ch\u1ec9nh",
   "ch\u1ecb",
   "ch\u1ed1t",
   "ch\u1ed7",
+  "ch\u1edd",
   "ch\u1ee3",
   "ch\u1ee7",
   "ch\u1ee9c",
@@ -598,6 +637,7 @@
   "c\u00e0y",
   "c\u00e1",
   "c\u00e1c",
+  "c\u00e1ch",
   "c\u00e1o",
   "c\u00e2y",
   "c\u00f2n",
@@ -642,6 +682,7 @@
   "dd/dd",
   "dd:dd",
   "ddd",
+  "dddd",
   "ddx",
   "ddxdd",
   "deadline",
@@ -650,12 +691,14 @@
   "design",
   "dev",
   "devops",
+  "di\u1ec5n",
   "di\u1ec7t",
   "du",
   "dung",
   "duy\u1ec7t",
   "dx",
   "dxdd",
+  "d\u00e0i",
   "d\u00e3",
   "d\u0169ng",
   "d\u01b0\u1ee1ng",
@@ -717,6 +760,7 @@
   "g\u1eb7p",
   "g\u1ecdc",
   "g\u1ecdi",
+  "g\u1ee1",
   "g\u1eedi",
   "h",
   "h.",
@@ -730,10 +774,13 @@
   "h59",
   "hai",
   "ham",
+  "hao",
   "hay",
   "him",
   "hip",
+  "hi\u1ec3m",
   "hi\u1ec3u",
+  "hi\u1ec7n",
   "hi\u1ec7u",
   "hoa",
   "hoan",
@@ -745,13 +792,16 @@
   "ho\u1ea1ch",
   "ho\u1ea1t",
   "huy",
+  "hu\u1ebf",
   "h\u00e0",
   "h\u00e0ng",
+  "h\u00e0nh",
   "h\u00e1m",
   "h\u00e1p",
   "h\u00e1t",
   "h\u00e1u",
   "h\u00e2n",
+  "h\u00e3y",
   "h\u00eam",
   "h\u00ecnh",
   "h\u00f2a",
@@ -784,6 +834,7 @@
   "h\u1ea9n",
   "h\u1ea9u",
   "h\u1eadn",
+  "h\u1eadp",
   "h\u1eadt",
   "h\u1eafc",
   "h\u1eafn",
@@ -839,6 +890,7 @@
   "i\u1ec3m",
   "i\u1ec3n",
   "i\u1ec3u",
+  "i\u1ec5n",
   "i\u1ec7c",
   "i\u1ec7m",
   "i\u1ec7n",
@@ -875,14 +927,17 @@
   "km",
   "kup",
   "k\u00e8m",
+  "k\u00e9o",
   "k\u00ea",
   "k\u00fd",
   "k\u1ebf",
   "k\u1ebft",
+  "k\u1ec3",
   "k\u1ef3",
   "k\u1ef9",
   "l",
   "l.",
+  "lab",
   "lan",
   "lao",
   "lax",
@@ -903,6 +958,7 @@
   "l\u00e1t",
   "l\u00e2m",
   "l\u00ean",
+  "l\u00f2ng",
   "l\u00f4ng",
   "l\u00fac",
   "l\u00fd",
@@ -917,6 +973,7 @@
   "l\u1edbn",
   "l\u1edbp",
   "l\u1eddi",
+  "l\u1ef1c",
   "m",
   "m.",
   "mai",
@@ -937,6 +994,7 @@
   "m\u1ea5t",
   "m\u1eabu",
   "m\u1eadt",
+  "m\u1eb7t",
   "m\u1eb9",
   "m\u1ec1m",
   "m\u1ed1t",
@@ -977,11 +1035,13 @@
   "nh\u1ea1c",
   "nh\u1ea5t",
   "nh\u1eadn",
+  "nh\u1eadp",
   "nh\u1eadt",
   "nh\u1eafc",
   "nh\u1eafn",
   "nh\u1ecf",
   "nh\u1edb",
+  "nh\u1edf",
   "nic",
   "nit",
   "ntt",
@@ -993,6 +1053,7 @@
   "n\u01b0\u1edbc",
   "n\u1ea5u",
   "n\u1eb5ng",
+  "n\u1ebfu",
   "n\u1ed9i",
   "n\u1ed9p",
   "n\u1eefa",
@@ -1029,8 +1090,10 @@
   "phim",
   "phi\u00ean",
   "ph\u00e1p",
+  "ph\u00e1t",
   "ph\u00e2n",
   "ph\u00ea",
+  "ph\u00ed",
   "ph\u00f2ng",
   "ph\u00fac",
   "ph\u00fat",
@@ -1057,12 +1120,14 @@
   "quang",
   "quanh",
   "quen",
+  "quy",
   "qu\u00e0",
   "qu\u00e1n",
   "qu\u00e2n",
   "qu\u00ean",
   "qu\u00fd",
   "qu\u1ea3",
+  "qu\u1ea3n",
   "qu\u1ea3ng",
   "qu\u1ea7n",
   "qu\u1ed1c",
@@ -1147,6 +1212,7 @@
   "thu",
   "thuy\u1ebft",
   "thu\u1eadt",
+  "thu\u1ebf",
   "thu\u1ed1c",
   "th\u00e0nh",
   "th\u00e1ng",
@@ -1161,6 +1227,7 @@
   "th\u1ea3o",
   "th\u1ea7u",
   "th\u1ea7y",
+  "th\u1eadp",
   "th\u1eafng",
   "th\u1ec3",
   "th\u1ecb",
@@ -1168,6 +1235,7 @@
   "th\u1eddi",
   "th\u1ee7",
   "th\u1ee9",
+  "th\u1ee9c",
   "th\u1eed",
   "th\u1ef1c",
   "time",
@@ -1201,6 +1269,7 @@
   "tr\u01b0a",
   "tr\u01b0\u1edbc",
   "tr\u01b0\u1eddng",
+  "tr\u01b0\u1edfng",
   "tr\u1ea3",
   "tr\u1ecdng",
   "tr\u1ee3",
@@ -1239,6 +1308,7 @@
   "t\u1ed5ng",
   "t\u1edbi",
   "t\u1ee5",
+  "t\u1ee5c",
   "t\u1ee7",
   "t\u1ee9c",
   "t\u1eeb",
@@ -1256,6 +1326,7 @@
   "u\u00e1n",
   "u\u00e2n",
   "u\u00ean",
+  "u\u1ea3n",
   "u\u1ea5n",
   "u\u1ea5t",
   "u\u1ea7n",
@@ -1283,6 +1354,7 @@
   "vi\u1ec7c",
   "vi\u1ec7n",
   "vi\u1ec7t",
+  "vui",
   "v\u00e0",
   "v\u00e0o",
   "v\u00e2n",
@@ -1290,6 +1362,7 @@
   "v\u0169ng",
   "v\u01b0\u1eddn",
   "v\u1ea5n",
+  "v\u1eadn",
   "v\u1eadt",
   "v\u1ebd",
   "v\u1ec1",
@@ -1315,11 +1388,13 @@
   "xe",
   "xem",
   "xin",
+  "xong",
   "xu\u1ea5t",
   "xx",
   "xxx",
   "xxxx",
   "x\u00e1c",
+  "x\u00e9t",
   "x\u1ea1",
   "x\u1ebfp",
   "x\u1eed",
@@ -1395,6 +1470,7 @@
   "\u0110\u1eebng",
   "\u0111",
   "\u0111a",
+  "\u0111ang",
   "\u0111em",
   "\u0111i",
   "\u0111i\u1ec7n",
@@ -1404,12 +1480,15 @@
   "\u0111\u00e1",
   "\u0111\u00e1nh",
   "\u0111\u00e3",
+  "\u0111\u00eam",
   "\u0111\u00ecnh",
+  "\u0111\u00f3",
   "\u0111\u00f3n",
   "\u0111\u00f3ng",
   "\u0111\u0103ng",
   "\u0111\u01a1n",
   "\u0111\u01b0a",
+  "\u0111\u01b0\u1ee3c",
   "\u0111\u1ea1i",
   "\u0111\u1ea7u",
   "\u0111\u1ea7y",
@@ -1444,6 +1523,7 @@
   "\u01b0\u1edbi",
   "\u01b0\u1eddi",
   "\u01b0\u1eddn",
+  "\u01b0\u1ee3c",
   "\u0ca0",
   "\u0ca0_\u0ca0",
   "\u0ca0\ufe35\u0ca0",
@@ -1472,6 +1552,7 @@
   "\u1edbng",
   "\u1eddng",
   "\u1edf",
+  "\u1edfng",
   "\u1ee1ng",
   "\u1ee5ng",
   "\u1ee9",