model improved

README.md

@@ -19,7 +19,7 @@ widget:
 
 ## Model Description
 
-This is a DeBERTa(V2) model pre-trained on Japanese Wikipedia and 青空文庫 texts for POS-tagging and dependency-parsing, derived from [deberta-base-japanese-wikipedia](https://huggingface.co/KoichiYasuoka/deberta-base-japanese-wikipedia). Every long-unit-word is tagged by [UPOS](https://universaldependencies.org/u/pos/) (Universal Part-Of-Speech).
+This is a DeBERTa(V2) model pre-trained on Japanese Wikipedia and 青空文庫 texts for POS-tagging and dependency-parsing, derived from [deberta-base-japanese-wikipedia](https://huggingface.co/KoichiYasuoka/deberta-base-japanese-wikipedia). Every long-unit-word is tagged by [UPOS](https://universaldependencies.org/u/pos/) (Universal Part-Of-Speech) and [FEATS](https://universaldependencies.org/u/feat/).
 
 ## How to Use
 

config.json

@@ -14,92 +14,100 @@
     "2": "ADP",
     "3": "ADV",
     "4": "AUX",
-    "5": "AUX+VERB",
-    "6": "B-ADJ",
-    "7": "B-ADJ+ADJ",
-    "8": "B-ADJ+AUX+AUX",
-    "9": "B-ADJ+VERB",
-    "10": "B-ADP",
-    "11": "B-ADP+NOUN",
-    "12": "B-ADV",
-    "13": "B-ADV+ADP",
-    "14": "B-ADV+ADP+VERB",
-    "15": "B-AUX",
-    "16": "B-AUX+AUX+AUX",
-    "17": "B-AUX+AUX+PART",
-    "18": "B-AUX+SCONJ",
-    "19": "B-AUX+VERB",
-    "20": "B-CCONJ",
-    "21": "B-DET",
-    "22": "B-INTJ",
-    "23": "B-NOUN",
-    "24": "B-NOUN+ADJ",
-    "25": "B-NOUN+ADP",
-    "26": "B-NUM",
-    "27": "B-PART",
-    "28": "B-PRON",
-    "29": "B-PROPN",
-    "30": "B-PUNCT",
-    "31": "B-SCONJ",
-    "32": "B-SYM",
-    "33": "B-VERB",
-    "34": "B-VERB+AUX",
-    "35": "B-VERB+AUX+AUX",
-    "36": "B-VERB+AUX+NOUN",
-    "37": "B-VERB+AUX+PART",
-    "38": "B-VERB+AUX+SCONJ",
-    "39": "B-X",
-    "40": "CCONJ",
-    "41": "DET",
-    "42": "I-ADJ",
-    "43": "I-ADJ+ADJ",
-    "44": "I-ADJ+AUX+AUX",
-    "45": "I-ADJ+VERB",
-    "46": "I-ADP",
-    "47": "I-ADP+NOUN",
-    "48": "I-ADV",
-    "49": "I-ADV+ADP",
-    "50": "I-ADV+ADP+VERB",
-    "51": "I-AUX",
-    "52": "I-AUX+AUX+AUX",
-    "53": "I-AUX+AUX+PART",
-    "54": "I-AUX+SCONJ",
-    "55": "I-AUX+VERB",
-    "56": "I-CCONJ",
-    "57": "I-DET",
-    "58": "I-INTJ",
-    "59": "I-NOUN",
-    "60": "I-NOUN+ADJ",
-    "61": "I-NOUN+ADP",
-    "62": "I-NUM",
-    "63": "I-PART",
-    "64": "I-PRON",
-    "65": "I-PROPN",
-    "66": "I-PUNCT",
-    "67": "I-SCONJ",
-    "68": "I-SYM",
-    "69": "I-VERB",
-    "70": "I-VERB+AUX",
-    "71": "I-VERB+AUX+AUX",
-    "72": "I-VERB+AUX+NOUN",
-    "73": "I-VERB+AUX+PART",
-    "74": "I-VERB+AUX+SCONJ",
-    "75": "I-X",
-    "76": "INTJ",
-    "77": "NOUN",
-    "78": "NOUN+ADP",
-    "79": "NOUN+AUX",
-    "80": "NUM",
-    "81": "PART",
-    "82": "PRON",
-    "83": "PRON+ADP",
-    "84": "PROPN",
-    "85": "PUNCT",
-    "86": "SCONJ",
-    "87": "SYM",
-    "88": "VERB",
-    "89": "VERB+AUX",
-    "90": "X"
+    "5": "AUX|Polarity=Neg",
+    "6": "AUX|Polarity=Neg+VERB",
+    "7": "B-ADJ",
+    "8": "B-ADJ+ADJ",
+    "9": "B-ADJ+AUX+AUX",
+    "10": "B-ADJ+VERB",
+    "11": "B-ADP",
+    "12": "B-ADP+NOUN",
+    "13": "B-ADV",
+    "14": "B-ADV+ADP",
+    "15": "B-ADV+ADP+VERB",
+    "16": "B-AUX",
+    "17": "B-AUX+AUX+AUX",
+    "18": "B-AUX+AUX+PART",
+    "19": "B-AUX+SCONJ",
+    "20": "B-AUX|Polarity=Neg",
+    "21": "B-AUX|Polarity=Neg+VERB",
+    "22": "B-CCONJ",
+    "23": "B-DET",
+    "24": "B-INTJ",
+    "25": "B-NOUN",
+    "26": "B-NOUN+ADJ",
+    "27": "B-NOUN+ADP",
+    "28": "B-NOUN|Polarity=Neg",
+    "29": "B-NUM",
+    "30": "B-PART",
+    "31": "B-PRON",
+    "32": "B-PROPN",
+    "33": "B-PUNCT",
+    "34": "B-SCONJ",
+    "35": "B-SYM",
+    "36": "B-VERB",
+    "37": "B-VERB+AUX",
+    "38": "B-VERB+AUX+AUX",
+    "39": "B-VERB+AUX+NOUN",
+    "40": "B-VERB+AUX+PART",
+    "41": "B-VERB+AUX+SCONJ",
+    "42": "B-VERB+AUX|Polarity=Neg",
+    "43": "B-X",
+    "44": "CCONJ",
+    "45": "DET",
+    "46": "I-ADJ",
+    "47": "I-ADJ+ADJ",
+    "48": "I-ADJ+AUX+AUX",
+    "49": "I-ADJ+VERB",
+    "50": "I-ADP",
+    "51": "I-ADP+NOUN",
+    "52": "I-ADV",
+    "53": "I-ADV+ADP",
+    "54": "I-ADV+ADP+VERB",
+    "55": "I-AUX",
+    "56": "I-AUX+AUX+AUX",
+    "57": "I-AUX+AUX+PART",
+    "58": "I-AUX+SCONJ",
+    "59": "I-AUX|Polarity=Neg",
+    "60": "I-AUX|Polarity=Neg+VERB",
+    "61": "I-CCONJ",
+    "62": "I-DET",
+    "63": "I-INTJ",
+    "64": "I-NOUN",
+    "65": "I-NOUN+ADJ",
+    "66": "I-NOUN+ADP",
+    "67": "I-NOUN|Polarity=Neg",
+    "68": "I-NUM",
+    "69": "I-PART",
+    "70": "I-PRON",
+    "71": "I-PROPN",
+    "72": "I-PUNCT",
+    "73": "I-SCONJ",
+    "74": "I-SYM",
+    "75": "I-VERB",
+    "76": "I-VERB+AUX",
+    "77": "I-VERB+AUX+AUX",
+    "78": "I-VERB+AUX+NOUN",
+    "79": "I-VERB+AUX+PART",
+    "80": "I-VERB+AUX+SCONJ",
+    "81": "I-VERB+AUX|Polarity=Neg",
+    "82": "I-X",
+    "83": "INTJ",
+    "84": "NOUN",
+    "85": "NOUN+ADP",
+    "86": "NOUN+AUX",
+    "87": "NUM",
+    "88": "PART",
+    "89": "PRON",
+    "90": "PRON+ADP",
+    "91": "PROPN",
+    "92": "PUNCT",
+    "93": "SCONJ",
+    "94": "SYM",
+    "95": "VERB",
+    "96": "VERB+AUX",
+    "97": "VERB+AUX|Polarity=Neg",
+    "98": "X"
   },
   "initializer_range": 0.02,
   "intermediate_size": 3072,
@@ -109,92 +117,100 @@
     "ADP": 2,
     "ADV": 3,
     "AUX": 4,
-    "AUX+VERB": 5,
-    "B-ADJ": 6,
-    "B-ADJ+ADJ": 7,
-    "B-ADJ+AUX+AUX": 8,
-    "B-ADJ+VERB": 9,
-    "B-ADP": 10,
-    "B-ADP+NOUN": 11,
-    "B-ADV": 12,
-    "B-ADV+ADP": 13,
-    "B-ADV+ADP+VERB": 14,
-    "B-AUX": 15,
-    "B-AUX+AUX+AUX": 16,
-    "B-AUX+AUX+PART": 17,
-    "B-AUX+SCONJ": 18,
-    "B-AUX+VERB": 19,
-    "B-CCONJ": 20,
-    "B-DET": 21,
-    "B-INTJ": 22,
-    "B-NOUN": 23,
-    "B-NOUN+ADJ": 24,
-    "B-NOUN+ADP": 25,
-    "B-NUM": 26,
-    "B-PART": 27,
-    "B-PRON": 28,
-    "B-PROPN": 29,
-    "B-PUNCT": 30,
-    "B-SCONJ": 31,
-    "B-SYM": 32,
-    "B-VERB": 33,
-    "B-VERB+AUX": 34,
-    "B-VERB+AUX+AUX": 35,
-    "B-VERB+AUX+NOUN": 36,
-    "B-VERB+AUX+PART": 37,
-    "B-VERB+AUX+SCONJ": 38,
-    "B-X": 39,
-    "CCONJ": 40,
-    "DET": 41,
-    "I-ADJ": 42,
-    "I-ADJ+ADJ": 43,
-    "I-ADJ+AUX+AUX": 44,
-    "I-ADJ+VERB": 45,
-    "I-ADP": 46,
-    "I-ADP+NOUN": 47,
-    "I-ADV": 48,
-    "I-ADV+ADP": 49,
-    "I-ADV+ADP+VERB": 50,
-    "I-AUX": 51,
-    "I-AUX+AUX+AUX": 52,
-    "I-AUX+AUX+PART": 53,
-    "I-AUX+SCONJ": 54,
-    "I-AUX+VERB": 55,
-    "I-CCONJ": 56,
-    "I-DET": 57,
-    "I-INTJ": 58,
-    "I-NOUN": 59,
-    "I-NOUN+ADJ": 60,
-    "I-NOUN+ADP": 61,
-    "I-NUM": 62,
-    "I-PART": 63,
-    "I-PRON": 64,
-    "I-PROPN": 65,
-    "I-PUNCT": 66,
-    "I-SCONJ": 67,
-    "I-SYM": 68,
-    "I-VERB": 69,
-    "I-VERB+AUX": 70,
-    "I-VERB+AUX+AUX": 71,
-    "I-VERB+AUX+NOUN": 72,
-    "I-VERB+AUX+PART": 73,
-    "I-VERB+AUX+SCONJ": 74,
-    "I-X": 75,
-    "INTJ": 76,
-    "NOUN": 77,
-    "NOUN+ADP": 78,
-    "NOUN+AUX": 79,
-    "NUM": 80,
-    "PART": 81,
-    "PRON": 82,
-    "PRON+ADP": 83,
-    "PROPN": 84,
-    "PUNCT": 85,
-    "SCONJ": 86,
-    "SYM": 87,
-    "VERB": 88,
-    "VERB+AUX": 89,
-    "X": 90
+    "AUX|Polarity=Neg": 5,
+    "AUX|Polarity=Neg+VERB": 6,
+    "B-ADJ": 7,
+    "B-ADJ+ADJ": 8,
+    "B-ADJ+AUX+AUX": 9,
+    "B-ADJ+VERB": 10,
+    "B-ADP": 11,
+    "B-ADP+NOUN": 12,
+    "B-ADV": 13,
+    "B-ADV+ADP": 14,
+    "B-ADV+ADP+VERB": 15,
+    "B-AUX": 16,
+    "B-AUX+AUX+AUX": 17,
+    "B-AUX+AUX+PART": 18,
+    "B-AUX+SCONJ": 19,
+    "B-AUX|Polarity=Neg": 20,
+    "B-AUX|Polarity=Neg+VERB": 21,
+    "B-CCONJ": 22,
+    "B-DET": 23,
+    "B-INTJ": 24,
+    "B-NOUN": 25,
+    "B-NOUN+ADJ": 26,
+    "B-NOUN+ADP": 27,
+    "B-NOUN|Polarity=Neg": 28,
+    "B-NUM": 29,
+    "B-PART": 30,
+    "B-PRON": 31,
+    "B-PROPN": 32,
+    "B-PUNCT": 33,
+    "B-SCONJ": 34,
+    "B-SYM": 35,
+    "B-VERB": 36,
+    "B-VERB+AUX": 37,
+    "B-VERB+AUX+AUX": 38,
+    "B-VERB+AUX+NOUN": 39,
+    "B-VERB+AUX+PART": 40,
+    "B-VERB+AUX+SCONJ": 41,
+    "B-VERB+AUX|Polarity=Neg": 42,
+    "B-X": 43,
+    "CCONJ": 44,
+    "DET": 45,
+    "I-ADJ": 46,
+    "I-ADJ+ADJ": 47,
+    "I-ADJ+AUX+AUX": 48,
+    "I-ADJ+VERB": 49,
+    "I-ADP": 50,
+    "I-ADP+NOUN": 51,
+    "I-ADV": 52,
+    "I-ADV+ADP": 53,
+    "I-ADV+ADP+VERB": 54,
+    "I-AUX": 55,
+    "I-AUX+AUX+AUX": 56,
+    "I-AUX+AUX+PART": 57,
+    "I-AUX+SCONJ": 58,
+    "I-AUX|Polarity=Neg": 59,
+    "I-AUX|Polarity=Neg+VERB": 60,
+    "I-CCONJ": 61,
+    "I-DET": 62,
+    "I-INTJ": 63,
+    "I-NOUN": 64,
+    "I-NOUN+ADJ": 65,
+    "I-NOUN+ADP": 66,
+    "I-NOUN|Polarity=Neg": 67,
+    "I-NUM": 68,
+    "I-PART": 69,
+    "I-PRON": 70,
+    "I-PROPN": 71,
+    "I-PUNCT": 72,
+    "I-SCONJ": 73,
+    "I-SYM": 74,
+    "I-VERB": 75,
+    "I-VERB+AUX": 76,
+    "I-VERB+AUX+AUX": 77,
+    "I-VERB+AUX+NOUN": 78,
+    "I-VERB+AUX+PART": 79,
+    "I-VERB+AUX+SCONJ": 80,
+    "I-VERB+AUX|Polarity=Neg": 81,
+    "I-X": 82,
+    "INTJ": 83,
+    "NOUN": 84,
+    "NOUN+ADP": 85,
+    "NOUN+AUX": 86,
+    "NUM": 87,
+    "PART": 88,
+    "PRON": 89,
+    "PRON+ADP": 90,
+    "PROPN": 91,
+    "PUNCT": 92,
+    "SCONJ": 93,
+    "SYM": 94,
+    "VERB": 95,
+    "VERB+AUX": 96,
+    "VERB+AUX|Polarity=Neg": 97,
+    "X": 98
   },
   "layer_norm_eps": 1e-07,
   "max_position_embeddings": 512,
@@ -292,7 +308,7 @@
           "\u3051\u3069"
         ]
       },
-      "AUX+VERB": {
+      "AUX|Polarity=Neg+VERB": {
         "\u306a\u304f\u306a\u3063": [
           "\u306a\u304f",
           "\u306a\u3063"
@@ -367,10 +383,6 @@
           "\u304a\u3053\u306a\u308f",
           "\u308c"
         ],
-        "\u304b\u307e\u308f\u305a": [
-          "\u304b\u307e\u308f",
-          "\u305a"
-        ],
         "\u3055\u3089\u308f\u308c": [
           "\u3055\u3089\u308f",
           "\u308c"
@@ -395,10 +407,6 @@
           "\u306a\u3058\u307e",
           "\u305b"
         ],
-        "\u306a\u3089\u305a": [
-          "\u306a\u3089",
-          "\u305a"
-        ],
         "\u307f\u306a\u3055\u308c": [
           "\u307f\u306a\u3055",
           "\u308c"
@@ -455,10 +463,6 @@
           "\u548c\u307e",
           "\u305b"
         ],
-        "\u554f\u308f\u305a": [
-          "\u554f\u308f",
-          "\u305a"
-        ],
         "\u554f\u308f\u308c": [
           "\u554f\u308f",
           "\u308c"
@@ -527,10 +531,6 @@
           "\u679c\u305f\u3059",
           "\u3079\u304f"
         ],
-        "\u69cb\u308f\u305a": [
-          "\u69cb\u308f",
-          "\u305a"
-        ],
         "\u6b4c\u308f\u308c": [
           "\u6b4c\u308f",
           "\u308c"
@@ -571,10 +571,6 @@
           "\u884c\u3063",
           "\u3061\u3083\u3044"
         ],
-        "\u884c\u308f\u305a": [
-          "\u884c\u308f",
-          "\u305a"
-        ],
         "\u884c\u308f\u308c": [
           "\u884c\u308f",
           "\u308c"
@@ -710,6 +706,28 @@
           "\u305f",
           "\u305f\u3081\u306b"
         ]
+      },
+      "VERB+AUX|Polarity=Neg": {
+        "\u304b\u307e\u308f\u305a": [
+          "\u304b\u307e\u308f",
+          "\u305a"
+        ],
+        "\u306a\u3089\u305a": [
+          "\u306a\u3089",
+          "\u305a"
+        ],
+        "\u554f\u308f\u305a": [
+          "\u554f\u308f",
+          "\u305a"
+        ],
+        "\u69cb\u308f\u305a": [
+          "\u69cb\u308f",
+          "\u305a"
+        ],
+        "\u884c\u308f\u305a": [
+          "\u884c\u308f",
+          "\u305a"
+        ]
       }
     }
   },

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7d9dc283f344a5962ffc3d73145d7d13aa93061dcffa80c204ca9cda53170536
-size 440451827
+oid sha256:4b5a1be4ff30e5f2537a335a038ef26c5c4e3e8c3df937927415153639d5e5f9
+size 440476467

supar.model

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1418fb8011870c9767cd696aae417a242fa43f12873d299f49a93550c091b56d
-size 488951787
+oid sha256:1e568b444794597ad6c070ed5bd229d942eb6e73edab6bf277c5e8191c830ba5
+size 488951723