bdpc commited on
Commit
6fb0f12
1 Parent(s): 6d59ba9

Saving best model of SciBERT_TwoWayLoss_25K_bs64_P10_N5 to hub

Browse files
README.md ADDED
@@ -0,0 +1,71 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: allenai/scibert_scivocab_uncased
3
+ tags:
4
+ - generated_from_trainer
5
+ metrics:
6
+ - accuracy
7
+ - precision
8
+ - recall
9
+ - f1
10
+ model-index:
11
+ - name: SciBERT_TwoWayLoss_25K_bs64_P10_N5
12
+ results: []
13
+ ---
14
+
15
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
16
+ should probably proofread and complete it, then remove this comment. -->
17
+
18
+ # SciBERT_TwoWayLoss_25K_bs64_P10_N5
19
+
20
+ This model is a fine-tuned version of [allenai/scibert_scivocab_uncased](https://huggingface.co/allenai/scibert_scivocab_uncased) on the None dataset.
21
+ It achieves the following results on the evaluation set:
22
+ - Loss: 15.1250
23
+ - Accuracy: 0.7066
24
+ - Precision: 0.0321
25
+ - Recall: 0.9982
26
+ - F1: 0.0622
27
+ - Hamming: 0.2934
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: 64
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
+ - lr_scheduler_warmup_ratio: 0.1
53
+ - training_steps: 25000
54
+
55
+ ### Training results
56
+
57
+ | Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1 | Hamming |
58
+ |:-------------:|:-----:|:-----:|:---------------:|:--------:|:---------:|:------:|:------:|:-------:|
59
+ | 28.5732 | 0.16 | 5000 | 26.4288 | 0.6945 | 0.0307 | 0.9910 | 0.0595 | 0.3055 |
60
+ | 19.8755 | 0.32 | 10000 | 18.9620 | 0.7010 | 0.0315 | 0.9959 | 0.0610 | 0.2990 |
61
+ | 17.1294 | 0.47 | 15000 | 16.5587 | 0.7021 | 0.0316 | 0.9970 | 0.0613 | 0.2979 |
62
+ | 15.8209 | 0.63 | 20000 | 15.4919 | 0.7053 | 0.0320 | 0.9982 | 0.0620 | 0.2947 |
63
+ | 15.4304 | 0.79 | 25000 | 15.1250 | 0.7066 | 0.0321 | 0.9982 | 0.0622 | 0.2934 |
64
+
65
+
66
+ ### Framework versions
67
+
68
+ - Transformers 4.33.3
69
+ - Pytorch 2.2.0.dev20231002
70
+ - Datasets 2.7.1
71
+ - Tokenizers 0.13.3
all_results.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 0.79,
3
+ "train_loss": 22.11730390625,
4
+ "train_runtime": 25819.8145,
5
+ "train_samples_per_second": 61.968,
6
+ "train_steps_per_second": 0.968
7
+ }
config.json ADDED
@@ -0,0 +1,360 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "allenai/scibert_scivocab_uncased",
3
+ "architectures": [
4
+ "BertForSequenceClassification"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "classifier_dropout": null,
8
+ "hidden_act": "gelu",
9
+ "hidden_dropout_prob": 0.1,
10
+ "hidden_size": 768,
11
+ "id2label": {
12
+ "0": "adap-org",
13
+ "1": "alg-geom",
14
+ "2": "astro-ph",
15
+ "3": "astro-ph.CO",
16
+ "4": "astro-ph.EP",
17
+ "5": "astro-ph.GA",
18
+ "6": "astro-ph.HE",
19
+ "7": "astro-ph.IM",
20
+ "8": "astro-ph.SR",
21
+ "9": "atom-ph",
22
+ "10": "chao-dyn",
23
+ "11": "chem-ph",
24
+ "12": "cmp-lg",
25
+ "13": "comp-gas",
26
+ "14": "cond-mat",
27
+ "15": "cond-mat.dis-nn",
28
+ "16": "cond-mat.mes-hall",
29
+ "17": "cond-mat.mtrl-sci",
30
+ "18": "cond-mat.other",
31
+ "19": "cond-mat.quant-gas",
32
+ "20": "cond-mat.soft",
33
+ "21": "cond-mat.stat-mech",
34
+ "22": "cond-mat.str-el",
35
+ "23": "cond-mat.supr-con",
36
+ "24": "cs.AI",
37
+ "25": "cs.AR",
38
+ "26": "cs.CC",
39
+ "27": "cs.CE",
40
+ "28": "cs.CG",
41
+ "29": "cs.CL",
42
+ "30": "cs.CR",
43
+ "31": "cs.CV",
44
+ "32": "cs.CY",
45
+ "33": "cs.DB",
46
+ "34": "cs.DC",
47
+ "35": "cs.DL",
48
+ "36": "cs.DM",
49
+ "37": "cs.DS",
50
+ "38": "cs.ET",
51
+ "39": "cs.FL",
52
+ "40": "cs.GL",
53
+ "41": "cs.GR",
54
+ "42": "cs.GT",
55
+ "43": "cs.HC",
56
+ "44": "cs.IR",
57
+ "45": "cs.LG",
58
+ "46": "cs.LO",
59
+ "47": "cs.MA",
60
+ "48": "cs.MM",
61
+ "49": "cs.MS",
62
+ "50": "cs.NA",
63
+ "51": "cs.NE",
64
+ "52": "cs.NI",
65
+ "53": "cs.OH",
66
+ "54": "cs.OS",
67
+ "55": "cs.PF",
68
+ "56": "cs.PL",
69
+ "57": "cs.RO",
70
+ "58": "cs.SC",
71
+ "59": "cs.SD",
72
+ "60": "cs.SE",
73
+ "61": "cs.SI",
74
+ "62": "cs.SY",
75
+ "63": "dg-ga",
76
+ "64": "econ.EM",
77
+ "65": "econ.GN",
78
+ "66": "econ.TH",
79
+ "67": "eess.AS",
80
+ "68": "eess.IV",
81
+ "69": "eess.SP",
82
+ "70": "eess.SY",
83
+ "71": "funct-an",
84
+ "72": "gr-qc",
85
+ "73": "hep-ex",
86
+ "74": "hep-lat",
87
+ "75": "hep-ph",
88
+ "76": "hep-th",
89
+ "77": "math.AC",
90
+ "78": "math.AG",
91
+ "79": "math.AP",
92
+ "80": "math.AT",
93
+ "81": "math.CA",
94
+ "82": "math.CO",
95
+ "83": "math.CT",
96
+ "84": "math.CV",
97
+ "85": "math.DG",
98
+ "86": "math.DS",
99
+ "87": "math.FA",
100
+ "88": "math.GM",
101
+ "89": "math.GN",
102
+ "90": "math.GR",
103
+ "91": "math.GT",
104
+ "92": "math.HO",
105
+ "93": "math.KT",
106
+ "94": "math.LO",
107
+ "95": "math.MG",
108
+ "96": "math.NA",
109
+ "97": "math.NT",
110
+ "98": "math.OA",
111
+ "99": "math.OC",
112
+ "100": "math.PR",
113
+ "101": "math.QA",
114
+ "102": "math.RA",
115
+ "103": "math.RT",
116
+ "104": "math.SG",
117
+ "105": "math.SP",
118
+ "106": "mtrl-th",
119
+ "107": "nlin.AO",
120
+ "108": "nlin.CD",
121
+ "109": "nlin.CG",
122
+ "110": "nlin.PS",
123
+ "111": "nlin.SI",
124
+ "112": "nucl-ex",
125
+ "113": "nucl-th",
126
+ "114": "patt-sol",
127
+ "115": "physics.acc-ph",
128
+ "116": "physics.ao-ph",
129
+ "117": "physics.app-ph",
130
+ "118": "physics.atm-clus",
131
+ "119": "physics.atom-ph",
132
+ "120": "physics.bio-ph",
133
+ "121": "physics.chem-ph",
134
+ "122": "physics.class-ph",
135
+ "123": "physics.comp-ph",
136
+ "124": "physics.data-an",
137
+ "125": "physics.ed-ph",
138
+ "126": "physics.flu-dyn",
139
+ "127": "physics.gen-ph",
140
+ "128": "physics.geo-ph",
141
+ "129": "physics.hist-ph",
142
+ "130": "physics.ins-det",
143
+ "131": "physics.med-ph",
144
+ "132": "physics.optics",
145
+ "133": "physics.plasm-ph",
146
+ "134": "physics.pop-ph",
147
+ "135": "physics.soc-ph",
148
+ "136": "physics.space-ph",
149
+ "137": "q-alg",
150
+ "138": "q-bio.BM",
151
+ "139": "q-bio.CB",
152
+ "140": "q-bio.GN",
153
+ "141": "q-bio.MN",
154
+ "142": "q-bio.NC",
155
+ "143": "q-bio.OT",
156
+ "144": "q-bio.PE",
157
+ "145": "q-bio.QM",
158
+ "146": "q-bio.SC",
159
+ "147": "q-bio.TO",
160
+ "148": "q-fin.CP",
161
+ "149": "q-fin.EC",
162
+ "150": "q-fin.GN",
163
+ "151": "q-fin.MF",
164
+ "152": "q-fin.PM",
165
+ "153": "q-fin.PR",
166
+ "154": "q-fin.RM",
167
+ "155": "q-fin.ST",
168
+ "156": "q-fin.TR",
169
+ "157": "quant-ph",
170
+ "158": "solv-int",
171
+ "159": "stat.AP",
172
+ "160": "stat.CO",
173
+ "161": "stat.ME",
174
+ "162": "stat.ML",
175
+ "163": "stat.OT",
176
+ "164": "supr-con"
177
+ },
178
+ "initializer_range": 0.02,
179
+ "intermediate_size": 3072,
180
+ "label2id": {
181
+ "adap-org": 0,
182
+ "alg-geom": 1,
183
+ "astro-ph": 2,
184
+ "astro-ph.CO": 3,
185
+ "astro-ph.EP": 4,
186
+ "astro-ph.GA": 5,
187
+ "astro-ph.HE": 6,
188
+ "astro-ph.IM": 7,
189
+ "astro-ph.SR": 8,
190
+ "atom-ph": 9,
191
+ "chao-dyn": 10,
192
+ "chem-ph": 11,
193
+ "cmp-lg": 12,
194
+ "comp-gas": 13,
195
+ "cond-mat": 14,
196
+ "cond-mat.dis-nn": 15,
197
+ "cond-mat.mes-hall": 16,
198
+ "cond-mat.mtrl-sci": 17,
199
+ "cond-mat.other": 18,
200
+ "cond-mat.quant-gas": 19,
201
+ "cond-mat.soft": 20,
202
+ "cond-mat.stat-mech": 21,
203
+ "cond-mat.str-el": 22,
204
+ "cond-mat.supr-con": 23,
205
+ "cs.AI": 24,
206
+ "cs.AR": 25,
207
+ "cs.CC": 26,
208
+ "cs.CE": 27,
209
+ "cs.CG": 28,
210
+ "cs.CL": 29,
211
+ "cs.CR": 30,
212
+ "cs.CV": 31,
213
+ "cs.CY": 32,
214
+ "cs.DB": 33,
215
+ "cs.DC": 34,
216
+ "cs.DL": 35,
217
+ "cs.DM": 36,
218
+ "cs.DS": 37,
219
+ "cs.ET": 38,
220
+ "cs.FL": 39,
221
+ "cs.GL": 40,
222
+ "cs.GR": 41,
223
+ "cs.GT": 42,
224
+ "cs.HC": 43,
225
+ "cs.IR": 44,
226
+ "cs.LG": 45,
227
+ "cs.LO": 46,
228
+ "cs.MA": 47,
229
+ "cs.MM": 48,
230
+ "cs.MS": 49,
231
+ "cs.NA": 50,
232
+ "cs.NE": 51,
233
+ "cs.NI": 52,
234
+ "cs.OH": 53,
235
+ "cs.OS": 54,
236
+ "cs.PF": 55,
237
+ "cs.PL": 56,
238
+ "cs.RO": 57,
239
+ "cs.SC": 58,
240
+ "cs.SD": 59,
241
+ "cs.SE": 60,
242
+ "cs.SI": 61,
243
+ "cs.SY": 62,
244
+ "dg-ga": 63,
245
+ "econ.EM": 64,
246
+ "econ.GN": 65,
247
+ "econ.TH": 66,
248
+ "eess.AS": 67,
249
+ "eess.IV": 68,
250
+ "eess.SP": 69,
251
+ "eess.SY": 70,
252
+ "funct-an": 71,
253
+ "gr-qc": 72,
254
+ "hep-ex": 73,
255
+ "hep-lat": 74,
256
+ "hep-ph": 75,
257
+ "hep-th": 76,
258
+ "math.AC": 77,
259
+ "math.AG": 78,
260
+ "math.AP": 79,
261
+ "math.AT": 80,
262
+ "math.CA": 81,
263
+ "math.CO": 82,
264
+ "math.CT": 83,
265
+ "math.CV": 84,
266
+ "math.DG": 85,
267
+ "math.DS": 86,
268
+ "math.FA": 87,
269
+ "math.GM": 88,
270
+ "math.GN": 89,
271
+ "math.GR": 90,
272
+ "math.GT": 91,
273
+ "math.HO": 92,
274
+ "math.KT": 93,
275
+ "math.LO": 94,
276
+ "math.MG": 95,
277
+ "math.NA": 96,
278
+ "math.NT": 97,
279
+ "math.OA": 98,
280
+ "math.OC": 99,
281
+ "math.PR": 100,
282
+ "math.QA": 101,
283
+ "math.RA": 102,
284
+ "math.RT": 103,
285
+ "math.SG": 104,
286
+ "math.SP": 105,
287
+ "mtrl-th": 106,
288
+ "nlin.AO": 107,
289
+ "nlin.CD": 108,
290
+ "nlin.CG": 109,
291
+ "nlin.PS": 110,
292
+ "nlin.SI": 111,
293
+ "nucl-ex": 112,
294
+ "nucl-th": 113,
295
+ "patt-sol": 114,
296
+ "physics.acc-ph": 115,
297
+ "physics.ao-ph": 116,
298
+ "physics.app-ph": 117,
299
+ "physics.atm-clus": 118,
300
+ "physics.atom-ph": 119,
301
+ "physics.bio-ph": 120,
302
+ "physics.chem-ph": 121,
303
+ "physics.class-ph": 122,
304
+ "physics.comp-ph": 123,
305
+ "physics.data-an": 124,
306
+ "physics.ed-ph": 125,
307
+ "physics.flu-dyn": 126,
308
+ "physics.gen-ph": 127,
309
+ "physics.geo-ph": 128,
310
+ "physics.hist-ph": 129,
311
+ "physics.ins-det": 130,
312
+ "physics.med-ph": 131,
313
+ "physics.optics": 132,
314
+ "physics.plasm-ph": 133,
315
+ "physics.pop-ph": 134,
316
+ "physics.soc-ph": 135,
317
+ "physics.space-ph": 136,
318
+ "q-alg": 137,
319
+ "q-bio.BM": 138,
320
+ "q-bio.CB": 139,
321
+ "q-bio.GN": 140,
322
+ "q-bio.MN": 141,
323
+ "q-bio.NC": 142,
324
+ "q-bio.OT": 143,
325
+ "q-bio.PE": 144,
326
+ "q-bio.QM": 145,
327
+ "q-bio.SC": 146,
328
+ "q-bio.TO": 147,
329
+ "q-fin.CP": 148,
330
+ "q-fin.EC": 149,
331
+ "q-fin.GN": 150,
332
+ "q-fin.MF": 151,
333
+ "q-fin.PM": 152,
334
+ "q-fin.PR": 153,
335
+ "q-fin.RM": 154,
336
+ "q-fin.ST": 155,
337
+ "q-fin.TR": 156,
338
+ "quant-ph": 157,
339
+ "solv-int": 158,
340
+ "stat.AP": 159,
341
+ "stat.CO": 160,
342
+ "stat.ME": 161,
343
+ "stat.ML": 162,
344
+ "stat.OT": 163,
345
+ "supr-con": 164
346
+ },
347
+ "layer_norm_eps": 1e-12,
348
+ "max_position_embeddings": 512,
349
+ "model_type": "bert",
350
+ "num_attention_heads": 12,
351
+ "num_hidden_layers": 12,
352
+ "pad_token_id": 0,
353
+ "position_embedding_type": "absolute",
354
+ "problem_type": "multi_label_classification",
355
+ "torch_dtype": "float32",
356
+ "transformers_version": "4.33.3",
357
+ "type_vocab_size": 2,
358
+ "use_cache": true,
359
+ "vocab_size": 31090
360
+ }
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:76b22d5c38301f796a37a12d28c707cc3bd4cd5ccf11c6a1c4981da0677e1a93
3
+ size 440250222
special_tokens_map.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "cls_token": "[CLS]",
3
+ "mask_token": "[MASK]",
4
+ "pad_token": "[PAD]",
5
+ "sep_token": "[SEP]",
6
+ "unk_token": "[UNK]"
7
+ }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "clean_up_tokenization_spaces": true,
3
+ "cls_token": "[CLS]",
4
+ "do_basic_tokenize": true,
5
+ "do_lower_case": true,
6
+ "mask_token": "[MASK]",
7
+ "model_max_length": 1000000000000000019884624838656,
8
+ "never_split": null,
9
+ "pad_token": "[PAD]",
10
+ "sep_token": "[SEP]",
11
+ "strip_accents": null,
12
+ "tokenize_chinese_chars": true,
13
+ "tokenizer_class": "BertTokenizer",
14
+ "unk_token": "[UNK]"
15
+ }
train_results.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 0.79,
3
+ "train_loss": 22.11730390625,
4
+ "train_runtime": 25819.8145,
5
+ "train_samples_per_second": 61.968,
6
+ "train_steps_per_second": 0.968
7
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:414c8bef0f21058aec54b38caa81235c491109c7a0706e5bc29155b0799a5c88
3
+ size 4664
vocab.txt ADDED
The diff for this file is too large to render. See raw diff