tangminhanh commited on
Commit
8b9dc10
1 Parent(s): 177039f

ts_subcate

Browse files
README.md ADDED
@@ -0,0 +1,75 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: mit
3
+ base_model: tangminhanh/ts_cate
4
+ tags:
5
+ - generated_from_trainer
6
+ metrics:
7
+ - accuracy
8
+ - f1
9
+ - precision
10
+ - recall
11
+ model-index:
12
+ - name: ts_subcate
13
+ results: []
14
+ ---
15
+
16
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
17
+ should probably proofread and complete it, then remove this comment. -->
18
+
19
+ # ts_subcate
20
+
21
+ This model is a fine-tuned version of [tangminhanh/ts_cate](https://huggingface.co/tangminhanh/ts_cate) on the None dataset.
22
+ It achieves the following results on the evaluation set:
23
+ - Loss: 0.0084
24
+ - Accuracy: 0.7013
25
+ - F1: 0.7812
26
+ - Precision: 0.8815
27
+ - Recall: 0.7014
28
+
29
+ ## Model description
30
+
31
+ More information needed
32
+
33
+ ## Intended uses & limitations
34
+
35
+ More information needed
36
+
37
+ ## Training and evaluation data
38
+
39
+ More information needed
40
+
41
+ ## Training procedure
42
+
43
+ ### Training hyperparameters
44
+
45
+ The following hyperparameters were used during training:
46
+ - learning_rate: 2e-05
47
+ - train_batch_size: 32
48
+ - eval_batch_size: 64
49
+ - seed: 42
50
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
51
+ - lr_scheduler_type: linear
52
+ - num_epochs: 10
53
+
54
+ ### Training results
55
+
56
+ | Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 | Precision | Recall |
57
+ |:-------------:|:-----:|:----:|:---------------:|:--------:|:------:|:---------:|:------:|
58
+ | No log | 1.0 | 403 | 0.0261 | 0.0 | 0.0 | 0.0 | 0.0 |
59
+ | 0.0805 | 2.0 | 806 | 0.0174 | 0.3328 | 0.4904 | 0.9438 | 0.3313 |
60
+ | 0.0203 | 3.0 | 1209 | 0.0131 | 0.4702 | 0.6217 | 0.9229 | 0.4687 |
61
+ | 0.0138 | 4.0 | 1612 | 0.0111 | 0.5592 | 0.6939 | 0.9170 | 0.5581 |
62
+ | 0.0104 | 5.0 | 2015 | 0.0100 | 0.6297 | 0.7417 | 0.8998 | 0.6308 |
63
+ | 0.0104 | 6.0 | 2418 | 0.0091 | 0.6659 | 0.7628 | 0.8909 | 0.6669 |
64
+ | 0.0084 | 7.0 | 2821 | 0.0088 | 0.6815 | 0.7712 | 0.8867 | 0.6823 |
65
+ | 0.0072 | 8.0 | 3224 | 0.0086 | 0.6889 | 0.7764 | 0.8874 | 0.6900 |
66
+ | 0.0063 | 9.0 | 3627 | 0.0084 | 0.6982 | 0.7803 | 0.8832 | 0.6989 |
67
+ | 0.006 | 10.0 | 4030 | 0.0084 | 0.7013 | 0.7812 | 0.8815 | 0.7014 |
68
+
69
+
70
+ ### Framework versions
71
+
72
+ - Transformers 4.42.4
73
+ - Pytorch 2.3.1+cu121
74
+ - Datasets 2.20.0
75
+ - Tokenizers 0.19.1
added_tokens.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ {
2
+ "[MASK]": 128000
3
+ }
config.json ADDED
@@ -0,0 +1,408 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "tangminhanh/ts_cate",
3
+ "architectures": [
4
+ "DebertaV2ForSequenceClassification"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "hidden_act": "gelu",
8
+ "hidden_dropout_prob": 0.1,
9
+ "hidden_size": 768,
10
+ "id2label": {
11
+ "0": "Access Reactivation",
12
+ "1": "Access Upload",
13
+ "2": "Badge Card Re-assignment",
14
+ "3": "Badge Card Request",
15
+ "4": "Bio Disable",
16
+ "5": "Bio Enrollment",
17
+ "6": "Bio ID Request",
18
+ "7": "Biometric Issue",
19
+ "8": "Device Configuration",
20
+ "9": "General Biometric Services",
21
+ "10": "Logs Report Generation",
22
+ "11": "Sprout Migration",
23
+ "12": "Diagram Creation/Update",
24
+ "13": "Floorplan Creation/Update",
25
+ "14": "Infographics",
26
+ "15": "Knowledgebase",
27
+ "16": "Other Documentations",
28
+ "17": "Process Update",
29
+ "18": "Security Questionnaire",
30
+ "19": "New Joinee",
31
+ "20": "Recurring Ticket Request/Removal",
32
+ "21": "Report Generation",
33
+ "22": "Ticketing Inquiry",
34
+ "23": "Test Ticket",
35
+ "24": "Accountability Form",
36
+ "25": "Disposal",
37
+ "26": "Gatepass Processing",
38
+ "27": "Inventory",
39
+ "28": "Issuance Form",
40
+ "29": "Item Request",
41
+ "30": "Report",
42
+ "31": "Warranty",
43
+ "32": "NT Login Password Reset",
44
+ "33": "Exit Clearance",
45
+ "34": "Employee Leaving",
46
+ "35": "Item Retrieval",
47
+ "36": "Camera Rotate/Disable Request",
48
+ "37": "CCTV Checking",
49
+ "38": "CCTV Footage Extraction",
50
+ "39": "CCTV Issue",
51
+ "40": "CCTV Viewing Assistance",
52
+ "41": "General CCTV Inquiry",
53
+ "42": "Common Area and Conference Room Checking",
54
+ "43": "Daily Checking",
55
+ "44": "Server Room Checking",
56
+ "45": "Server Room Log Sheet Checking",
57
+ "46": "Cable Management",
58
+ "47": "Data Center UPS/PDU Issue",
59
+ "48": "Desktop UPS Issue",
60
+ "49": "Desktop/Laptop Problem",
61
+ "50": "Desktop/Laptop Setup",
62
+ "51": "Device/Equipment Checking",
63
+ "52": "Equipment Loan/Retrieval",
64
+ "53": "Joan Device",
65
+ "54": "Peripherals",
66
+ "55": "Raspberry Pi/Hygrometer",
67
+ "56": "Sonos",
68
+ "57": "Tablet",
69
+ "58": "TV/Projector/Sound System",
70
+ "59": "Video Conference",
71
+ "60": "Data Center Access",
72
+ "61": "General Inquiry",
73
+ "62": "On-site Assistance",
74
+ "63": "Port Checking",
75
+ "64": "Pull-In/Out Assistance",
76
+ "65": "Site Walkthrough / Round Check",
77
+ "66": "Digital Signage",
78
+ "67": "Setup",
79
+ "68": "Removal",
80
+ "69": "Connection Issue",
81
+ "70": "Connection Checking",
82
+ "71": "General Network Activity",
83
+ "72": "Maintenance",
84
+ "73": "Monitoring Alerts",
85
+ "74": "Network Diagram Creation/Update",
86
+ "75": "Network Gear Mount/Unmount",
87
+ "76": "Operational Readiness Test",
88
+ "77": "Port Tracing",
89
+ "78": "Site Build",
90
+ "79": "SSID Creation/Removal",
91
+ "80": "VLAN Change/Create",
92
+ "81": "Website Issue",
93
+ "82": "Whitelist/Blacklist",
94
+ "83": "Wired/Wireless Connection",
95
+ "84": "IP Request",
96
+ "85": "Traffic Engineering",
97
+ "86": "Client High Level Network Design",
98
+ "87": "Client Requirement Discovery",
99
+ "88": "Order",
100
+ "89": "Provisioning",
101
+ "90": "Upgrade",
102
+ "91": "ISP Outage",
103
+ "92": "Network Outage",
104
+ "93": "Power Outage",
105
+ "94": "Server Outage",
106
+ "95": "System Outage",
107
+ "96": "Telephony Outage",
108
+ "97": "High Severity Alert",
109
+ "98": "Low Severity Alert",
110
+ "99": "Vulnerability Assessment",
111
+ "100": "Vulnerability Remediation",
112
+ "101": "New Client",
113
+ "102": "Lease Extension",
114
+ "103": "Emergency Maintenance",
115
+ "104": "ISP Maintenance",
116
+ "105": "Network Maintenance",
117
+ "106": "Power Maintenance",
118
+ "107": "Server Backup Management",
119
+ "108": "Server Deployment and Configuration",
120
+ "109": "Server Health Monitoring",
121
+ "110": "Server Issue",
122
+ "111": "Server Security and Access Control",
123
+ "112": "Server VM Management",
124
+ "113": "SSL License",
125
+ "114": "Server Rebuild",
126
+ "115": "Active Directory Services",
127
+ "116": "Access Permission",
128
+ "117": "Joan Services",
129
+ "118": "Microsoft Dynamics NAV Issue",
130
+ "119": "PaperCut ID",
131
+ "120": "Printer Configuration",
132
+ "121": "Email Account Creation",
133
+ "122": "Email Account Deletion",
134
+ "123": "Email Password Reset/Unlock",
135
+ "124": "NT Login Account Unlock",
136
+ "125": "NT Login Creation",
137
+ "126": "NT Login Disable",
138
+ "127": "NT Login Re-enable",
139
+ "128": "NT Login Information Update",
140
+ "129": "OSP Account Unlock",
141
+ "130": "OSP Access Request",
142
+ "131": "OSP Access Disable",
143
+ "132": "Movement Request",
144
+ "133": "Deployment Request",
145
+ "134": "Access Request",
146
+ "135": "Calendar and Events Management",
147
+ "136": "Distrolist / Mailbox",
148
+ "137": "Email Detail Update",
149
+ "138": "Email Issue",
150
+ "139": "Email Storage",
151
+ "140": "License Request/Removal",
152
+ "141": "MFA Request/Concern",
153
+ "142": "Change Toner",
154
+ "143": "Issue",
155
+ "144": "Printing Services",
156
+ "145": "Set up",
157
+ "146": "Configuration Assistance",
158
+ "147": "Application Issue",
159
+ "148": "Application Installation/Unistallation",
160
+ "149": "Softphone Issue",
161
+ "150": "Hardphone Issue",
162
+ "151": "General Voice Activity/Inquiry",
163
+ "152": "OSP",
164
+ "153": "ERP",
165
+ "154": "Client Portal",
166
+ "155": "CATS",
167
+ "156": "EPICOS",
168
+ "157": "VMS",
169
+ "158": "KMC Hub",
170
+ "159": "New Development Request",
171
+ "160": "Manatal",
172
+ "161": "CASA IT support",
173
+ "162": "Farm-In",
174
+ "163": "Farm-Out",
175
+ "164": "Malicious Email",
176
+ "165": "Phishing Incident",
177
+ "166": "Spam",
178
+ "167": "BOC Clearance",
179
+ "168": "Calculation of Asset Value",
180
+ "169": "General Procurement Inquiry",
181
+ "170": "Government Permit",
182
+ "171": "Payment/Invoices/Liquidation",
183
+ "172": "Purchase Request",
184
+ "173": "Quotation Request",
185
+ "174": "Vendor Coordination",
186
+ "175": "General IT Services",
187
+ "176": "Hardware Purchase",
188
+ "177": "Hardware Rental",
189
+ "178": "IT Storage Rental",
190
+ "179": "Quotation for Room Modification",
191
+ "180": "Quotation for Services",
192
+ "181": "RFID",
193
+ "182": "BCP Testing",
194
+ "183": "Internal"
195
+ },
196
+ "initializer_range": 0.02,
197
+ "intermediate_size": 3072,
198
+ "label2id": {
199
+ "Access Permission": 116,
200
+ "Access Reactivation": 0,
201
+ "Access Request": 134,
202
+ "Access Upload": 1,
203
+ "Accountability Form": 24,
204
+ "Active Directory Services": 115,
205
+ "Application Installation/Unistallation": 148,
206
+ "Application Issue": 147,
207
+ "BCP Testing": 182,
208
+ "BOC Clearance": 167,
209
+ "Badge Card Re-assignment": 2,
210
+ "Badge Card Request": 3,
211
+ "Bio Disable": 4,
212
+ "Bio Enrollment": 5,
213
+ "Bio ID Request": 6,
214
+ "Biometric Issue": 7,
215
+ "CASA IT support": 161,
216
+ "CATS": 155,
217
+ "CCTV Checking": 37,
218
+ "CCTV Footage Extraction": 38,
219
+ "CCTV Issue": 39,
220
+ "CCTV Viewing Assistance": 40,
221
+ "Cable Management": 46,
222
+ "Calculation of Asset Value": 168,
223
+ "Calendar and Events Management": 135,
224
+ "Camera Rotate/Disable Request": 36,
225
+ "Change Toner": 142,
226
+ "Client High Level Network Design": 86,
227
+ "Client Portal": 154,
228
+ "Client Requirement Discovery": 87,
229
+ "Common Area and Conference Room Checking": 42,
230
+ "Configuration Assistance": 146,
231
+ "Connection Checking": 70,
232
+ "Connection Issue": 69,
233
+ "Daily Checking": 43,
234
+ "Data Center Access": 60,
235
+ "Data Center UPS/PDU Issue": 47,
236
+ "Deployment Request": 133,
237
+ "Desktop UPS Issue": 48,
238
+ "Desktop/Laptop Problem": 49,
239
+ "Desktop/Laptop Setup": 50,
240
+ "Device Configuration": 8,
241
+ "Device/Equipment Checking": 51,
242
+ "Diagram Creation/Update": 12,
243
+ "Digital Signage": 66,
244
+ "Disposal": 25,
245
+ "Distrolist / Mailbox": 136,
246
+ "EPICOS": 156,
247
+ "ERP": 153,
248
+ "Email Account Creation": 121,
249
+ "Email Account Deletion": 122,
250
+ "Email Detail Update": 137,
251
+ "Email Issue": 138,
252
+ "Email Password Reset/Unlock": 123,
253
+ "Email Storage": 139,
254
+ "Emergency Maintenance": 103,
255
+ "Employee Leaving": 34,
256
+ "Equipment Loan/Retrieval": 52,
257
+ "Exit Clearance": 33,
258
+ "Farm-In": 162,
259
+ "Farm-Out": 163,
260
+ "Floorplan Creation/Update": 13,
261
+ "Gatepass Processing": 26,
262
+ "General Biometric Services": 9,
263
+ "General CCTV Inquiry": 41,
264
+ "General IT Services": 175,
265
+ "General Inquiry": 61,
266
+ "General Network Activity": 71,
267
+ "General Procurement Inquiry": 169,
268
+ "General Voice Activity/Inquiry": 151,
269
+ "Government Permit": 170,
270
+ "Hardphone Issue": 150,
271
+ "Hardware Purchase": 176,
272
+ "Hardware Rental": 177,
273
+ "High Severity Alert": 97,
274
+ "IP Request": 84,
275
+ "ISP Maintenance": 104,
276
+ "ISP Outage": 91,
277
+ "IT Storage Rental": 178,
278
+ "Infographics": 14,
279
+ "Internal": 183,
280
+ "Inventory": 27,
281
+ "Issuance Form": 28,
282
+ "Issue": 143,
283
+ "Item Request": 29,
284
+ "Item Retrieval": 35,
285
+ "Joan Device": 53,
286
+ "Joan Services": 117,
287
+ "KMC Hub": 158,
288
+ "Knowledgebase": 15,
289
+ "Lease Extension": 102,
290
+ "License Request/Removal": 140,
291
+ "Logs Report Generation": 10,
292
+ "Low Severity Alert": 98,
293
+ "MFA Request/Concern": 141,
294
+ "Maintenance": 72,
295
+ "Malicious Email": 164,
296
+ "Manatal": 160,
297
+ "Microsoft Dynamics NAV Issue": 118,
298
+ "Monitoring Alerts": 73,
299
+ "Movement Request": 132,
300
+ "NT Login Account Unlock": 124,
301
+ "NT Login Creation": 125,
302
+ "NT Login Disable": 126,
303
+ "NT Login Information Update": 128,
304
+ "NT Login Password Reset": 32,
305
+ "NT Login Re-enable": 127,
306
+ "Network Diagram Creation/Update": 74,
307
+ "Network Gear Mount/Unmount": 75,
308
+ "Network Maintenance": 105,
309
+ "Network Outage": 92,
310
+ "New Client": 101,
311
+ "New Development Request": 159,
312
+ "New Joinee": 19,
313
+ "OSP": 152,
314
+ "OSP Access Disable": 131,
315
+ "OSP Access Request": 130,
316
+ "OSP Account Unlock": 129,
317
+ "On-site Assistance": 62,
318
+ "Operational Readiness Test": 76,
319
+ "Order": 88,
320
+ "Other Documentations": 16,
321
+ "PaperCut ID": 119,
322
+ "Payment/Invoices/Liquidation": 171,
323
+ "Peripherals": 54,
324
+ "Phishing Incident": 165,
325
+ "Port Checking": 63,
326
+ "Port Tracing": 77,
327
+ "Power Maintenance": 106,
328
+ "Power Outage": 93,
329
+ "Printer Configuration": 120,
330
+ "Printing Services": 144,
331
+ "Process Update": 17,
332
+ "Provisioning": 89,
333
+ "Pull-In/Out Assistance": 64,
334
+ "Purchase Request": 172,
335
+ "Quotation Request": 173,
336
+ "Quotation for Room Modification": 179,
337
+ "Quotation for Services": 180,
338
+ "RFID": 181,
339
+ "Raspberry Pi/Hygrometer": 55,
340
+ "Recurring Ticket Request/Removal": 20,
341
+ "Removal": 68,
342
+ "Report": 30,
343
+ "Report Generation": 21,
344
+ "SSID Creation/Removal": 79,
345
+ "SSL License": 113,
346
+ "Security Questionnaire": 18,
347
+ "Server Backup Management": 107,
348
+ "Server Deployment and Configuration": 108,
349
+ "Server Health Monitoring": 109,
350
+ "Server Issue": 110,
351
+ "Server Outage": 94,
352
+ "Server Rebuild": 114,
353
+ "Server Room Checking": 44,
354
+ "Server Room Log Sheet Checking": 45,
355
+ "Server Security and Access Control": 111,
356
+ "Server VM Management": 112,
357
+ "Set up": 145,
358
+ "Setup": 67,
359
+ "Site Build": 78,
360
+ "Site Walkthrough / Round Check": 65,
361
+ "Softphone Issue": 149,
362
+ "Sonos": 56,
363
+ "Spam": 166,
364
+ "Sprout Migration": 11,
365
+ "System Outage": 95,
366
+ "TV/Projector/Sound System": 58,
367
+ "Tablet": 57,
368
+ "Telephony Outage": 96,
369
+ "Test Ticket": 23,
370
+ "Ticketing Inquiry": 22,
371
+ "Traffic Engineering": 85,
372
+ "Upgrade": 90,
373
+ "VLAN Change/Create": 80,
374
+ "VMS": 157,
375
+ "Vendor Coordination": 174,
376
+ "Video Conference": 59,
377
+ "Vulnerability Assessment": 99,
378
+ "Vulnerability Remediation": 100,
379
+ "Warranty": 31,
380
+ "Website Issue": 81,
381
+ "Whitelist/Blacklist": 82,
382
+ "Wired/Wireless Connection": 83
383
+ },
384
+ "layer_norm_eps": 1e-07,
385
+ "max_position_embeddings": 512,
386
+ "max_relative_positions": -1,
387
+ "model_type": "deberta-v2",
388
+ "norm_rel_ebd": "layer_norm",
389
+ "num_attention_heads": 12,
390
+ "num_hidden_layers": 6,
391
+ "pad_token_id": 0,
392
+ "pooler_dropout": 0,
393
+ "pooler_hidden_act": "gelu",
394
+ "pooler_hidden_size": 768,
395
+ "pos_att_type": [
396
+ "p2c",
397
+ "c2p"
398
+ ],
399
+ "position_biased_input": false,
400
+ "position_buckets": 256,
401
+ "problem_type": "multi_label_classification",
402
+ "relative_attention": true,
403
+ "share_att_key": true,
404
+ "torch_dtype": "float32",
405
+ "transformers_version": "4.42.4",
406
+ "type_vocab_size": 0,
407
+ "vocab_size": 128100
408
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d55f44784118bf8a803974df029de20086b3400e3c4fa1491787a474385d1613
3
+ size 568158400
runs/Aug06_15-02-40_8286c2bd2340/events.out.tfevents.1722956564.8286c2bd2340.1585.2 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:63c2a2946b2938bba604799429e096c55525050dc3ff42e5fefff3a742f8368b
3
+ size 23228
runs/Aug06_15-02-40_8286c2bd2340/events.out.tfevents.1722959300.8286c2bd2340.1585.3 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ece9763c39ad4723cf55cbc07e467ece5c46912d2732f81d020f809bbf5f9330
3
+ size 560
special_tokens_map.json ADDED
@@ -0,0 +1,51 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "[CLS]",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "cls_token": {
10
+ "content": "[CLS]",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "eos_token": {
17
+ "content": "[SEP]",
18
+ "lstrip": false,
19
+ "normalized": false,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ },
23
+ "mask_token": {
24
+ "content": "[MASK]",
25
+ "lstrip": false,
26
+ "normalized": false,
27
+ "rstrip": false,
28
+ "single_word": false
29
+ },
30
+ "pad_token": {
31
+ "content": "[PAD]",
32
+ "lstrip": false,
33
+ "normalized": false,
34
+ "rstrip": false,
35
+ "single_word": false
36
+ },
37
+ "sep_token": {
38
+ "content": "[SEP]",
39
+ "lstrip": false,
40
+ "normalized": false,
41
+ "rstrip": false,
42
+ "single_word": false
43
+ },
44
+ "unk_token": {
45
+ "content": "[UNK]",
46
+ "lstrip": false,
47
+ "normalized": true,
48
+ "rstrip": false,
49
+ "single_word": false
50
+ }
51
+ }
spm.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c679fbf93643d19aab7ee10c0b99e460bdbc02fedf34b92b05af343b4af586fd
3
+ size 2464616
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,65 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "[PAD]",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "1": {
12
+ "content": "[CLS]",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "2": {
20
+ "content": "[SEP]",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "3": {
28
+ "content": "[UNK]",
29
+ "lstrip": false,
30
+ "normalized": true,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "128000": {
36
+ "content": "[MASK]",
37
+ "lstrip": false,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ }
43
+ },
44
+ "bos_token": "[CLS]",
45
+ "clean_up_tokenization_spaces": true,
46
+ "cls_token": "[CLS]",
47
+ "do_lower_case": false,
48
+ "eos_token": "[SEP]",
49
+ "mask_token": "[MASK]",
50
+ "max_length": 512,
51
+ "model_max_length": 1000000000000000019884624838656,
52
+ "pad_to_multiple_of": null,
53
+ "pad_token": "[PAD]",
54
+ "pad_token_type_id": 0,
55
+ "padding_side": "right",
56
+ "sep_token": "[SEP]",
57
+ "sp_model_kwargs": {},
58
+ "split_by_punct": false,
59
+ "stride": 0,
60
+ "tokenizer_class": "DebertaV2Tokenizer",
61
+ "truncation_side": "right",
62
+ "truncation_strategy": "longest_first",
63
+ "unk_token": "[UNK]",
64
+ "vocab_type": "spm"
65
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ea795bb7e99a64d3970c78db9bd4939955bb3850361dcdf6fa9a49199a682d62
3
+ size 5112