model improved

config.json

@@ -21,27 +21,27 @@
     "9": "B-CCONJ",
     "10": "B-DET",
     "11": "B-DET+NOUN",
-    "12": "B-DET+VERB",
-    "13": "B-INFR.EV",
-    "14": "B-INTJ",
-    "15": "B-NOUN",
-    "16": "B-NOUN+ADP",
-    "17": "B-NOUN+ADP+NOUN",
+    "12": "B-INFR.EV",
+    "13": "B-INTJ",
+    "14": "B-NOUN",
+    "15": "B-NOUN+ADP",
+    "16": "B-NOUN+ADP+NOUN",
+    "17": "B-NOUN+ADP+VERB",
     "18": "B-NOUN+ADV",
     "19": "B-NOUN+NOUN",
     "20": "B-NOUN+VERB",
     "21": "B-NUM",
     "22": "B-NUM+NOUN",
     "23": "B-PART",
-    "24": "B-PART+NOUN",
-    "25": "B-PART+VERB",
-    "26": "B-PRON",
-    "27": "B-PROPN",
-    "28": "B-PUNCT",
-    "29": "B-SCONJ",
-    "30": "B-SCONJ+ADV",
-    "31": "B-VERB",
-    "32": "B-VERB+AUX",
+    "24": "B-PART+AUX",
+    "25": "B-PART+NOUN",
+    "26": "B-PART+VERB",
+    "27": "B-PRON",
+    "28": "B-PROPN",
+    "29": "B-PUNCT",
+    "30": "B-SCONJ",
+    "31": "B-SCONJ+ADV",
+    "32": "B-VERB",
     "33": "B-VERB+NOUN",
     "34": "B-VERB+PART",
     "35": "B-VERB+SCONJ",
@@ -59,27 +59,27 @@
     "47": "I-CCONJ",
     "48": "I-DET",
     "49": "I-DET+NOUN",
-    "50": "I-DET+VERB",
-    "51": "I-INFR.EV",
-    "52": "I-INTJ",
-    "53": "I-NOUN",
-    "54": "I-NOUN+ADP",
-    "55": "I-NOUN+ADP+NOUN",
+    "50": "I-INFR.EV",
+    "51": "I-INTJ",
+    "52": "I-NOUN",
+    "53": "I-NOUN+ADP",
+    "54": "I-NOUN+ADP+NOUN",
+    "55": "I-NOUN+ADP+VERB",
     "56": "I-NOUN+ADV",
     "57": "I-NOUN+NOUN",
     "58": "I-NOUN+VERB",
     "59": "I-NUM",
     "60": "I-NUM+NOUN",
     "61": "I-PART",
-    "62": "I-PART+NOUN",
-    "63": "I-PART+VERB",
-    "64": "I-PRON",
-    "65": "I-PROPN",
-    "66": "I-PUNCT",
-    "67": "I-SCONJ",
-    "68": "I-SCONJ+ADV",
-    "69": "I-VERB",
-    "70": "I-VERB+AUX",
+    "62": "I-PART+AUX",
+    "63": "I-PART+NOUN",
+    "64": "I-PART+VERB",
+    "65": "I-PRON",
+    "66": "I-PROPN",
+    "67": "I-PUNCT",
+    "68": "I-SCONJ",
+    "69": "I-SCONJ+ADV",
+    "70": "I-VERB",
     "71": "I-VERB+NOUN",
     "72": "I-VERB+PART",
     "73": "I-VERB+SCONJ",
@@ -91,21 +91,19 @@
     "79": "NOUN+NOUN",
     "80": "NOUN+VERB",
     "81": "NUM",
-    "82": "NUM+VERB+NOUN",
-    "83": "PART",
-    "84": "PART+NOUN",
-    "85": "PART+VERB",
-    "86": "PROPN",
-    "87": "PUNCT",
-    "88": "SCONJ",
-    "89": "SYM",
-    "90": "VERB",
-    "91": "VERB+AUX",
-    "92": "VERB+NOUN",
-    "93": "VERB+PART",
-    "94": "VERB+VERB",
-    "95": "VERT",
-    "96": "X"
+    "82": "PART",
+    "83": "PART+VERB",
+    "84": "PROPN",
+    "85": "PUNCT",
+    "86": "SCONJ",
+    "87": "SYM",
+    "88": "VERB",
+    "89": "VERB+AUX",
+    "90": "VERB+NOUN",
+    "91": "VERB+PART",
+    "92": "VERB+VERB",
+    "93": "VERT",
+    "94": "X"
   },
   "initializer_range": 0.02,
   "intermediate_size": 3072,
@@ -122,27 +120,27 @@
     "B-CCONJ": 9,
     "B-DET": 10,
     "B-DET+NOUN": 11,
-    "B-DET+VERB": 12,
-    "B-INFR.EV": 13,
-    "B-INTJ": 14,
-    "B-NOUN": 15,
-    "B-NOUN+ADP": 16,
-    "B-NOUN+ADP+NOUN": 17,
+    "B-INFR.EV": 12,
+    "B-INTJ": 13,
+    "B-NOUN": 14,
+    "B-NOUN+ADP": 15,
+    "B-NOUN+ADP+NOUN": 16,
+    "B-NOUN+ADP+VERB": 17,
     "B-NOUN+ADV": 18,
     "B-NOUN+NOUN": 19,
     "B-NOUN+VERB": 20,
     "B-NUM": 21,
     "B-NUM+NOUN": 22,
     "B-PART": 23,
-    "B-PART+NOUN": 24,
-    "B-PART+VERB": 25,
-    "B-PRON": 26,
-    "B-PROPN": 27,
-    "B-PUNCT": 28,
-    "B-SCONJ": 29,
-    "B-SCONJ+ADV": 30,
-    "B-VERB": 31,
-    "B-VERB+AUX": 32,
+    "B-PART+AUX": 24,
+    "B-PART+NOUN": 25,
+    "B-PART+VERB": 26,
+    "B-PRON": 27,
+    "B-PROPN": 28,
+    "B-PUNCT": 29,
+    "B-SCONJ": 30,
+    "B-SCONJ+ADV": 31,
+    "B-VERB": 32,
     "B-VERB+NOUN": 33,
     "B-VERB+PART": 34,
     "B-VERB+SCONJ": 35,
@@ -160,27 +158,27 @@
     "I-CCONJ": 47,
     "I-DET": 48,
     "I-DET+NOUN": 49,
-    "I-DET+VERB": 50,
-    "I-INFR.EV": 51,
-    "I-INTJ": 52,
-    "I-NOUN": 53,
-    "I-NOUN+ADP": 54,
-    "I-NOUN+ADP+NOUN": 55,
+    "I-INFR.EV": 50,
+    "I-INTJ": 51,
+    "I-NOUN": 52,
+    "I-NOUN+ADP": 53,
+    "I-NOUN+ADP+NOUN": 54,
+    "I-NOUN+ADP+VERB": 55,
     "I-NOUN+ADV": 56,
     "I-NOUN+NOUN": 57,
     "I-NOUN+VERB": 58,
     "I-NUM": 59,
     "I-NUM+NOUN": 60,
     "I-PART": 61,
-    "I-PART+NOUN": 62,
-    "I-PART+VERB": 63,
-    "I-PRON": 64,
-    "I-PROPN": 65,
-    "I-PUNCT": 66,
-    "I-SCONJ": 67,
-    "I-SCONJ+ADV": 68,
-    "I-VERB": 69,
-    "I-VERB+AUX": 70,
+    "I-PART+AUX": 62,
+    "I-PART+NOUN": 63,
+    "I-PART+VERB": 64,
+    "I-PRON": 65,
+    "I-PROPN": 66,
+    "I-PUNCT": 67,
+    "I-SCONJ": 68,
+    "I-SCONJ+ADV": 69,
+    "I-VERB": 70,
     "I-VERB+NOUN": 71,
     "I-VERB+PART": 72,
     "I-VERB+SCONJ": 73,
@@ -192,21 +190,19 @@
     "NOUN+NOUN": 79,
     "NOUN+VERB": 80,
     "NUM": 81,
-    "NUM+VERB+NOUN": 82,
-    "PART": 83,
-    "PART+NOUN": 84,
-    "PART+VERB": 85,
-    "PROPN": 86,
-    "PUNCT": 87,
-    "SCONJ": 88,
-    "SYM": 89,
-    "VERB": 90,
-    "VERB+AUX": 91,
-    "VERB+NOUN": 92,
-    "VERB+PART": 93,
-    "VERB+VERB": 94,
-    "VERT": 95,
-    "X": 96
+    "PART": 82,
+    "PART+VERB": 83,
+    "PROPN": 84,
+    "PUNCT": 85,
+    "SCONJ": 86,
+    "SYM": 87,
+    "VERB": 88,
+    "VERB+AUX": 89,
+    "VERB+NOUN": 90,
+    "VERB+PART": 91,
+    "VERB+VERB": 92,
+    "VERT": 93,
+    "X": 94
   },
   "layer_norm_eps": 1e-07,
   "max_position_embeddings": 512,
@@ -258,13 +254,9 @@
           "Tam",
           "pa"
         ],
-        "oararke": [
-          "oar",
-          "arke"
-        ],
-        "oararkehe": [
-          "oar",
-          "arkehe"
+        "tanpa": [
+          "tan",
+          "pa"
         ],
         "tanto": [
           "tan",
@@ -283,41 +275,55 @@
           "an"
         ]
       },
-      "DET+VERB": {
-        "iyorun": [
-          "iyor",
-          "un"
-        ]
-      },
       "NOUN+ADP": {
         "Kunneiwano": [
           "Kunnei",
           "wano"
         ],
+        "Orota": [
+          "Oro",
+          "ta"
+        ],
         "Orowano": [
           "Oro",
           "wano"
         ],
-        "Pet-samaketa": [
-          "Pet-samake",
+        "Oshmaketa": [
+          "Oshmake",
           "ta"
         ],
-        "Shoita": [
-          "Shoi",
+        "Pet-samaketa": [
+          "Pet-samake",
           "ta"
         ],
         "Soita": [
           "Soi",
           "ta"
         ],
+        "cheppone": [
+          "cheppo",
+          "ne"
+        ],
         "keseta": [
           "kese",
           "ta"
         ],
+        "kesta": [
+          "kes",
+          "ta"
+        ],
         "kunneywano": [
           "kunney",
           "wano"
         ],
+        "neyta": [
+          "ney",
+          "ta"
+        ],
+        "orota": [
+          "oro",
+          "ta"
+        ],
         "orowa": [
           "oro",
           "wa"
@@ -326,16 +332,20 @@
           "oro",
           "wano"
         ],
+        "oshmaketa": [
+          "oshmake",
+          "ta"
+        ],
         "otta": [
           "ot",
           "ta"
         ],
-        "samaketa": [
-          "samake",
+        "petsamaketa": [
+          "petsamake",
           "ta"
         ],
-        "shoita": [
-          "shoi",
+        "samaketa": [
+          "samake",
           "ta"
         ],
         "soyta": [
@@ -345,10 +355,6 @@
         "tomta": [
           "tom",
           "ta"
-        ],
-        "tumukeheta": [
-          "tumukehe",
-          "ta"
         ]
       },
       "NOUN+ADP+NOUN": {
@@ -363,6 +369,13 @@
           "puray"
         ]
       },
+      "NOUN+ADP+VERB": {
+        "soytaarpa": [
+          "soy",
+          "ta",
+          "arpa"
+        ]
+      },
       "NOUN+ADV": {
         "Tambeta ne": [
           "Tambe",
@@ -370,22 +383,18 @@
         ]
       },
       "NOUN+NOUN": {
-        "Hinakoro": [
-          "Hinak",
-          "oro"
-        ],
         "Petetoko": [
           "Pet",
           "etoko"
         ],
+        "Shirokanipe": [
+          "Shirokani",
+          "pe"
+        ],
         "hekattar": [
           "hekat",
           "tar"
         ],
-        "hinakoro": [
-          "hinak",
-          "oro"
-        ],
         "inaanpe": [
           "inaan",
           "pe"
@@ -406,6 +415,10 @@
           "kamuy",
           "nis"
         ],
+        "konkanipe": [
+          "konkani",
+          "pe"
+        ],
         "petetok": [
           "pet",
           "etok"
@@ -413,6 +426,10 @@
         "petetoko": [
           "pet",
           "etoko"
+        ],
+        "sirokanipe": [
+          "sirokani",
+          "pe"
         ]
       },
       "NOUN+VERB": {
@@ -479,33 +496,19 @@
           "to"
         ]
       },
-      "NUM+VERB+NOUN": {
-        "Shineanto": [
-          "Shine",
-          "an",
-          "to"
-        ],
-        "sineanto": [
-          "sine",
-          "an",
-          "to"
+      "PART+AUX": {
+        "chine": [
+          "chi",
+          "ne"
         ]
       },
       "PART+NOUN": {
         "=anpe": [
           "=an",
           "pe"
-        ],
-        "shichorpok": [
-          "shi",
-          "chorpok"
         ]
       },
       "PART+VERB": {
-        "Chirushka": [
-          "Chi",
-          "rushka"
-        ],
         "ainu-wap": [
           "a",
           "inu-wap"
@@ -529,10 +532,6 @@
         "karapa": [
           "k",
           "arapa"
-        ],
-        "shiokote": [
-          "shi",
-          "okote"
         ]
       },
       "SCONJ+ADV": {
@@ -542,14 +541,6 @@
         ]
       },
       "VERB+AUX": {
-        "poppeta ashinnangoro": [
-          "poppeta ashin",
-          "nangoro"
-        ],
-        "poppetaasinnankor": [
-          "poppetaasin",
-          "nankor"
-        ],
         "sattek": [
           "sat",
           "tek"
@@ -568,13 +559,9 @@
           "an",
           "pe"
         ],
-        "ashbe": [
-          "ash",
-          "be"
-        ],
-        "aspe": [
-          "as",
-          "pe"
+        "anto": [
+          "an",
+          "to"
         ],
         "h\u00e9sep\u00e1ha": [
           "h\u00e9se",
@@ -584,10 +571,6 @@
           "kar",
           "i"
         ],
-        "ohasiri": [
-          "oha",
-          "siri"
-        ],
         "wenpuri": [
           "wen",
           "puri"
@@ -605,6 +588,10 @@
         "sapash": [
           "sap",
           "ash"
+        ],
+        "shinotash": [
+          "shinot",
+          "ash"
         ]
       },
       "VERB+SCONJ": {
@@ -633,5 +620,5 @@
   "torch_dtype": "float32",
   "transformers_version": "4.22.1",
   "type_vocab_size": 0,
-  "vocab_size": 5092
+  "vocab_size": 6143
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:540dae77de84baec491ec433af635a0a345f9ceb4a7c0fe9845948fa5181efbf
-size 416095379
+oid sha256:6b288c1d7f545b97c411dedfe9ad50d32805f0d3ac04e74b9e9059010a5cf0fe
+size 419317843

supar.model

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9af6a91919dc6e9ba6390a1933911dd5ce8c8c44d9f84e56a26e21cd7e8f84cc
-size 461045771
+oid sha256:cef6c2b683eee67b07578529bbb3849ede3b772becc7e15e271f2d26795a4bf0
+size 464297035