Hugging Face's logo

qanastek
/
pos-french-camembert-flair

Token Classification
Flair
PyTorch
French
sequence-tagger-model
Model card Files Community
qanastek commited on
Commit
76b6de9
•
1 Parent(s): 8fb39ae

Update

Browse files
Files changed (8) hide show
  1. README.md +258 -0
  2. best-model.pt +3 -0
  3. dev.tsv +0 -0
  4. loss.tsv +51 -0
  5. preview.PNG +0 -0
  6. test.tsv +0 -0
  7. training.log +1188 -0
  8. weights.txt +0 -0
README.md ADDED
@@ -0,0 +1,258 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ tags:
3
+ - flair
4
+ - token-classification
5
+ - sequence-tagger-model
6
+ language: fr
7
+ widget:
8
+ - text: "George Washington est allé à Washington"
9
+ ---
10
+
11
+ # POET: A French Extended Part-of-Speech Tagger
12
+
13
+ - Corpora: [ANTILLES](https://github.com/qanastek/ANTILLES)
14
+ - Embeddings: [Contextual String Embeddings for Sequence Labelling](https://aclanthology.org/C18-1139/) + [CamemBERT](https://arxiv.org/abs/1911.03894)
15
+ - Sequence Labelling: [Bi-LSTM-CRF](https://arxiv.org/abs/1011.4088)
16
+ - Number of Epochs: 50
17
+
18
+ **People Involved**
19
+
20
+ * [LABRAK Yanis](https://www.linkedin.com/in/yanis-labrak-8a7412145/) (1)
21
+ * [DUFOUR Richard](https://cv.archives-ouvertes.fr/richard-dufour) (2)
22
+
23
+ **Affiliations**
24
+
25
+ 1. [LIA, NLP team](https://lia.univ-avignon.fr/), Avignon University, Avignon, France.
26
+ 2. [LS2N, TALN team](https://www.ls2n.fr/equipe/taln/), Nantes University, Nantes, France.
27
+
28
+ ## Demo: How to use in Flair
29
+
30
+ Requires [Flair](https://pypi.org/project/flair/): ```pip install flair```
31
+
32
+ ```python
33
+ from flair.data import Sentence
34
+ from flair.models import SequenceTagger
35
+
36
+ # Load the model
37
+ model = SequenceTagger.load("qanastek/pos-french-camembert-flair")
38
+
39
+ sentence = Sentence("George Washington est allé à Washington")
40
+
41
+ # predict tags
42
+ model.predict(sentence)
43
+
44
+ # print predicted pos tags
45
+ print(sentence.to_tagged_string())
46
+ ```
47
+
48
+ Output:
49
+
50
+ ![Preview Output](preview.PNG)
51
+
52
+ ## Training data
53
+
54
+ `ANTILLES` is a part-of-speech tagging corpora based on [UD_French-GSD](https://universaldependencies.org/treebanks/fr_gsd/index.html) which was originally created in 2015 and is based on the [universal dependency treebank v2.0](https://github.com/ryanmcd/uni-dep-tb).
55
+
56
+ Originally, the corpora consists of 400,399 words (16,341 sentences) and had 17 different classes. Now, after applying our tags augmentation we obtain 60 different classes which add linguistic and semantic information such as the gender, number, mood, person, tense or verb form given in the different CoNLL-03 fields from the original corpora.
57
+
58
+ We based our tags on the level of details given by the [LIA_TAGG](http://pageperso.lif.univ-mrs.fr/frederic.bechet/download.html) statistical POS tagger written by [Frédéric Béchet](http://pageperso.lif.univ-mrs.fr/frederic.bechet/index-english.html) in 2001.
59
+
60
+ The corpora used for this model is available on [Github](https://github.com/qanastek/ANTILLES) at the [CoNLL-U format](https://universaldependencies.org/format.html).
61
+
62
+ Training data are fed to the model as free language and doesn't pass a normalization phase. Thus, it's made the model case and punctuation sensitive.
63
+
64
+ ## Original Tags
65
+
66
+ ```plain
67
+ PRON VERB SCONJ ADP CCONJ DET NOUN ADJ AUX ADV PUNCT PROPN NUM SYM PART X INTJ
68
+ ```
69
+
70
+ ## New additional POS tags
71
+
72
+ | Abbreviation | Description | Examples |
73
+ |:--------:|:--------:|:--------:|
74
+ | PREP | Preposition | de |
75
+ | AUX | Auxiliary Verb | est |
76
+ | ADV | Adverb | toujours |
77
+ | COSUB | Subordinating conjunction | que |
78
+ | COCO | Coordinating Conjunction | et |
79
+ | PART | Demonstrative particle | -t |
80
+ | PRON | Pronoun | qui ce quoi |
81
+ | PDEMMS | Demonstrative Pronoun - Singular Masculine | ce |
82
+ | PDEMMP | Demonstrative Pronoun - Plural Masculine | ceux |
83
+ | PDEMFS | Demonstrative Pronoun - Singular Feminine | cette |
84
+ | PDEMFP | Demonstrative Pronoun - Plural Feminine | celles |
85
+ | PINDMS | Indefinite Pronoun - Singular Masculine | tout |
86
+ | PINDMP | Indefinite Pronoun - Plural Masculine | autres |
87
+ | PINDFS | Indefinite Pronoun - Singular Feminine | chacune |
88
+ | PINDFP | Indefinite Pronoun - Plural Feminine | certaines |
89
+ | PROPN | Proper noun | Houston |
90
+ | XFAMIL | Last name | Levy |
91
+ | NUM | Numerical Adjective | trentaine vingtaine |
92
+ | DINTMS | Masculine Numerical Adjective | un |
93
+ | DINTFS | Feminine Numerical Adjective | une |
94
+ | PPOBJMS | Pronoun complements of objects - Singular Masculine | le lui |
95
+ | PPOBJMP | Pronoun complements of objects - Plural Masculine | eux y |
96
+ | PPOBJFS | Pronoun complements of objects - Singular Feminine | moi la |
97
+ | PPOBJFP | Pronoun complements of objects - Plural Feminine | en y |
98
+ | PPER1S | Personal Pronoun First-Person - Singular | je |
99
+ | PPER2S | Personal Pronoun Second-Person - Singular | tu |
100
+ | PPER3MS | Personal Pronoun Third-Person - Singular Masculine | il |
101
+ | PPER3MP | Personal Pronoun Third-Person - Plural Masculine | ils |
102
+ | PPER3FS | Personal Pronoun Third-Person - Singular Feminine | elle |
103
+ | PPER3FP | Personal Pronoun Third-Person - Plural Feminine | elles |
104
+ | PREFS | Reflexive Pronoun First-Person - Singular | me m' |
105
+ | PREF | Reflexive Pronoun Third-Person - Singular | se s' |
106
+ | PREFP | Reflexive Pronoun First / Second-Person - Plural | nous vous |
107
+ | VERB | Verb | obtient |
108
+ | VPPMS | Past Participle - Singular Masculine | formulé |
109
+ | VPPMP | Past Participle - Plural Masculine | classés |
110
+ | VPPFS | Past Participle - Singular Feminine | appelée |
111
+ | VPPFP | Past Participle - Plural Feminine | sanctionnées |
112
+ | DET | Determinant | les l' |
113
+ | DETMS | Determinant - Singular Masculine | les |
114
+ | DETFS | Determinant - Singular Feminine | la |
115
+ | ADJ | Adjective | capable sérieux |
116
+ | ADJMS | Adjective - Singular Masculine | grand important |
117
+ | ADJMP | Adjective - Plural Masculine | grands petits |
118
+ | ADJFS | Adjective - Singular Feminine | française petite |
119
+ | ADJFP | Adjective - Plural Feminine | légères petites |
120
+ | NOUN | Noun | temps |
121
+ | NMS | Noun - Singular Masculine | drapeau |
122
+ | NMP | Noun - Plural Masculine | journalistes |
123
+ | NFS | Noun - Singular Feminine | tête |
124
+ | NFP | Noun - Plural Feminine | ondes |
125
+ | PREL | Relative Pronoun | qui dont |
126
+ | PRELMS | Relative Pronoun - Singular Masculine | lequel |
127
+ | PRELMP | Relative Pronoun - Plural Masculine | lesquels |
128
+ | PRELFS | Relative Pronoun - Singular Feminine | laquelle |
129
+ | PRELFP | Relative Pronoun - Plural Feminine | lesquelles |
130
+ | INTJ | Interjection | merci bref |
131
+ | CHIF | Numbers | 1979 10 |
132
+ | SYM | Symbol | € % |
133
+ | YPFOR | Endpoint | . |
134
+ | PUNCT | Ponctuation | : , |
135
+ | MOTINC | Unknown words | Technology Lady |
136
+ | X | Typos & others | sfeir 3D statu |
137
+
138
+ ## Evaluation results
139
+
140
+ The test corpora used for this evaluation is available on [Github](https://github.com/qanastek/ANTILLES/blob/main/ANTILLES/test.conllu).
141
+
142
+ ```plain
143
+ Results:
144
+ - F-score (micro) 0.9797
145
+ - F-score (macro) 0.9178
146
+ - Accuracy 0.9797
147
+
148
+ By class:
149
+ precision recall f1-score support
150
+
151
+ PREP 0.9966 0.9987 0.9976 1483
152
+ PUNCT 1.0000 1.0000 1.0000 833
153
+ NMS 0.9634 0.9801 0.9717 753
154
+ DET 0.9923 0.9984 0.9954 645
155
+ VERB 0.9913 0.9811 0.9862 583
156
+ NFS 0.9667 0.9839 0.9752 560
157
+ ADV 0.9940 0.9821 0.9880 504
158
+ PROPN 0.9541 0.8937 0.9229 395
159
+ DETMS 1.0000 1.0000 1.0000 362
160
+ AUX 0.9860 0.9915 0.9888 355
161
+ YPFOR 1.0000 1.0000 1.0000 353
162
+ NMP 0.9666 0.9475 0.9570 305
163
+ COCO 0.9959 1.0000 0.9980 245
164
+ ADJMS 0.9463 0.9385 0.9424 244
165
+ DETFS 1.0000 1.0000 1.0000 240
166
+ CHIF 0.9648 0.9865 0.9755 222
167
+ NFP 0.9515 0.9849 0.9679 199
168
+ ADJFS 0.9657 0.9286 0.9468 182
169
+ VPPMS 0.9387 0.9745 0.9563 157
170
+ COSUB 1.0000 0.9844 0.9921 128
171
+ DINTMS 0.9918 0.9918 0.9918 122
172
+ XFAMIL 0.9298 0.9217 0.9258 115
173
+ PPER3MS 1.0000 1.0000 1.0000 87
174
+ ADJMP 0.9294 0.9634 0.9461 82
175
+ PDEMMS 1.0000 1.0000 1.0000 75
176
+ ADJFP 0.9861 0.9342 0.9595 76
177
+ PREL 0.9859 1.0000 0.9929 70
178
+ DINTFS 0.9839 1.0000 0.9919 61
179
+ PREF 1.0000 1.0000 1.0000 52
180
+ PPOBJMS 0.9565 0.9362 0.9462 47
181
+ PREFP 0.9778 1.0000 0.9888 44
182
+ PINDMS 1.0000 0.9773 0.9885 44
183
+ VPPFS 0.8298 0.9750 0.8966 40
184
+ PPER1S 1.0000 1.0000 1.0000 42
185
+ SYM 1.0000 0.9474 0.9730 38
186
+ NOUN 0.8824 0.7692 0.8219 39
187
+ PRON 1.0000 0.9677 0.9836 31
188
+ PDEMFS 1.0000 1.0000 1.0000 29
189
+ VPPMP 0.9286 1.0000 0.9630 26
190
+ ADJ 0.9524 0.9091 0.9302 22
191
+ PPER3MP 1.0000 1.0000 1.0000 20
192
+ VPPFP 1.0000 1.0000 1.0000 19
193
+ PPER3FS 1.0000 1.0000 1.0000 18
194
+ MOTINC 0.3333 0.4000 0.3636 15
195
+ PREFS 1.0000 1.0000 1.0000 10
196
+ PPOBJMP 1.0000 0.8000 0.8889 10
197
+ PPOBJFS 0.6250 0.8333 0.7143 6
198
+ INTJ 0.5000 0.6667 0.5714 6
199
+ PART 1.0000 1.0000 1.0000 4
200
+ PDEMMP 1.0000 1.0000 1.0000 3
201
+ PDEMFP 1.0000 1.0000 1.0000 3
202
+ PPER3FP 1.0000 1.0000 1.0000 2
203
+ NUM 1.0000 0.3333 0.5000 3
204
+ PPER2S 1.0000 1.0000 1.0000 2
205
+ PPOBJFP 0.5000 0.5000 0.5000 2
206
+ PRELMS 1.0000 1.0000 1.0000 2
207
+ PINDFS 0.5000 1.0000 0.6667 1
208
+ PINDMP 1.0000 1.0000 1.0000 1
209
+ X 0.0000 0.0000 0.0000 1
210
+ PINDFP 1.0000 1.0000 1.0000 1
211
+
212
+ micro avg 0.9797 0.9797 0.9797 10019
213
+ macro avg 0.9228 0.9230 0.9178 10019
214
+ weighted avg 0.9802 0.9797 0.9798 10019
215
+ samples avg 0.9797 0.9797 0.9797 10019
216
+ ```
217
+
218
+ ## BibTeX Citations
219
+
220
+ Please cite the following paper when using this model.
221
+
222
+ UD_French-GSD corpora:
223
+
224
+ ```latex
225
+ @misc{
226
+ universaldependencies,
227
+ title={UniversalDependencies/UD_French-GSD},
228
+ url={https://github.com/UniversalDependencies/UD_French-GSD}, journal={GitHub},
229
+ author={UniversalDependencies}
230
+ }
231
+ ```
232
+
233
+ LIA TAGG:
234
+
235
+ ```latex
236
+ @techreport{LIA_TAGG,
237
+ author = {Frédéric Béchet},
238
+ title = {LIA_TAGG: a statistical POS tagger + syntactic bracketer},
239
+ institution = {Aix-Marseille University & CNRS},
240
+ year = {2001}
241
+ }
242
+ ```
243
+
244
+ Flair Embeddings:
245
+
246
+ ```latex
247
+ @inproceedings{akbik2018coling,
248
+ title={Contextual String Embeddings for Sequence Labeling},
249
+ author={Akbik, Alan and Blythe, Duncan and Vollgraf, Roland},
250
+ booktitle = {{COLING} 2018, 27th International Conference on Computational Linguistics},
251
+ pages = {1638--1649},
252
+ year = {2018}
253
+ }
254
+ ```
255
+
256
+ ## Acknowledgment
257
+
258
+ This work was financially supported by [Zenidoc](https://zenidoc.fr/)
best-model.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:81250a3c3c5e6639a92f0a5c782ca1533c886a4ae9d48dccf7fffb85c8c27794
3
+ size 539091093
dev.tsv ADDED
The diff for this file is too large to render. See raw diff
 
loss.tsv ADDED
@@ -0,0 +1,51 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ EPOCH TIMESTAMP BAD_EPOCHS LEARNING_RATE TRAIN_LOSS DEV_LOSS DEV_PRECISION DEV_RECALL DEV_F1 DEV_ACCURACY
2
+ 1 08:40:38 0 0.1000 0.4139037357786823 0.09867297857999802 0.9723 0.9723 0.9723 0.9723
3
+ 2 08:43:59 0 0.1000 0.18683743789460058 0.08219591528177261 0.9761 0.9761 0.9761 0.9761
4
+ 3 08:47:19 0 0.1000 0.15330191376146995 0.07821641117334366 0.9771 0.9771 0.9771 0.9771
5
+ 4 08:50:41 0 0.1000 0.13679069024466697 0.07048774510622025 0.9784 0.9784 0.9784 0.9784
6
+ 5 08:54:05 0 0.1000 0.12696925415918978 0.06857253611087799 0.9795 0.9795 0.9795 0.9795
7
+ 6 08:57:27 0 0.1000 0.11950518270160057 0.06588418781757355 0.9805 0.9805 0.9805 0.9805
8
+ 7 09:00:50 0 0.1000 0.11493925645982497 0.06450950354337692 0.981 0.981 0.981 0.981
9
+ 8 09:04:17 1 0.1000 0.10857167749940388 0.06390747427940369 0.9805 0.9805 0.9805 0.9805
10
+ 9 09:07:32 0 0.1000 0.10607049356155288 0.06607701629400253 0.9814 0.9814 0.9814 0.9814
11
+ 10 09:10:55 1 0.1000 0.10357679251794805 0.06536506861448288 0.9811 0.9811 0.9811 0.9811
12
+ 11 09:14:14 2 0.1000 0.09934903912638607 0.06659943610429764 0.9811 0.9811 0.9811 0.9811
13
+ 12 09:17:31 0 0.1000 0.09835840644128474 0.06410104781389236 0.9816 0.9816 0.9816 0.9816
14
+ 13 09:20:54 1 0.1000 0.09667944757963298 0.06427688896656036 0.9816 0.9816 0.9816 0.9816
15
+ 14 09:24:13 0 0.1000 0.09310611423129937 0.06639766693115234 0.9817 0.9817 0.9817 0.9817
16
+ 15 09:27:36 0 0.1000 0.09273757020644302 0.06283392012119293 0.982 0.982 0.982 0.982
17
+ 16 09:30:58 1 0.1000 0.0906242242911817 0.06354553997516632 0.982 0.982 0.982 0.982
18
+ 17 09:34:16 0 0.1000 0.08953603486075622 0.06361010670661926 0.9823 0.9823 0.9823 0.9823
19
+ 18 09:37:38 1 0.1000 0.08716175395396096 0.06376409530639648 0.982 0.982 0.982 0.982
20
+ 19 09:40:54 0 0.1000 0.08630291839682559 0.06360483914613724 0.9824 0.9824 0.9824 0.9824
21
+ 20 09:44:18 1 0.1000 0.08576645481590557 0.06494450569152832 0.982 0.982 0.982 0.982
22
+ 21 09:47:35 0 0.1000 0.08385216420677111 0.06328344345092773 0.9827 0.9827 0.9827 0.9827
23
+ 22 09:50:58 1 0.1000 0.08442341949455835 0.06346500664949417 0.9815 0.9815 0.9815 0.9815
24
+ 23 09:54:16 2 0.1000 0.08142570236260006 0.06540019810199738 0.9821 0.9821 0.9821 0.9821
25
+ 24 09:57:35 3 0.1000 0.0822403573790078 0.06453310698270798 0.9819 0.9819 0.9819 0.9819
26
+ 25 10:00:51 4 0.1000 0.08115838012320148 0.06579063087701797 0.9817 0.9817 0.9817 0.9817
27
+ 26 10:04:08 1 0.0500 0.07444606900847728 0.06646668165922165 0.9822 0.9822 0.9822 0.9822
28
+ 27 10:07:27 2 0.0500 0.0712278272039567 0.06514652073383331 0.9823 0.9823 0.9823 0.9823
29
+ 28 10:10:43 0 0.0500 0.07007554484263678 0.06285692006349564 0.9828 0.9828 0.9828 0.9828
30
+ 29 10:14:05 0 0.0500 0.06775021975879568 0.06288447976112366 0.9831 0.9831 0.9831 0.9831
31
+ 30 10:17:28 1 0.0500 0.06664810656288497 0.06311798095703125 0.9824 0.9824 0.9824 0.9824
32
+ 31 10:20:44 2 0.0500 0.06655944385465427 0.06285466253757477 0.9829 0.9829 0.9829 0.9829
33
+ 32 10:24:00 3 0.0500 0.06484422466931324 0.062373436987400055 0.9827 0.9827 0.9827 0.9827
34
+ 33 10:27:18 4 0.0500 0.0640099294991078 0.06352584064006805 0.983 0.983 0.983 0.983
35
+ 34 10:30:35 0 0.0250 0.060174371477019914 0.06348917633295059 0.9835 0.9835 0.9835 0.9835
36
+ 35 10:33:57 1 0.0250 0.06001775798271323 0.06338120251893997 0.9829 0.9829 0.9829 0.9829
37
+ 36 10:37:17 2 0.0250 0.05871860721139249 0.06424003839492798 0.9835 0.9835 0.9835 0.9835
38
+ 37 10:40:35 3 0.0250 0.058616780999994414 0.06326954811811447 0.9831 0.9831 0.9831 0.9831
39
+ 38 10:43:52 4 0.0250 0.05700768598441678 0.06343492120504379 0.9831 0.9831 0.9831 0.9831
40
+ 39 10:47:11 1 0.0125 0.05521906535375693 0.06419230252504349 0.9829 0.9829 0.9829 0.9829
41
+ 40 10:50:26 2 0.0125 0.0545923078244546 0.06343018263578415 0.9829 0.9829 0.9829 0.9829
42
+ 41 10:53:43 3 0.0125 0.05370077676597374 0.06420625746250153 0.9831 0.9831 0.9831 0.9831
43
+ 42 10:57:03 4 0.0125 0.053218689249659105 0.06362675130367279 0.9831 0.9831 0.9831 0.9831
44
+ 43 11:00:20 1 0.0063 0.052319793331815405 0.06449297815561295 0.983 0.983 0.983 0.983
45
+ 44 11:03:36 2 0.0063 0.052714465782813934 0.06455685943365097 0.9831 0.9831 0.9831 0.9831
46
+ 45 11:06:56 3 0.0063 0.052344465870703995 0.06413871794939041 0.983 0.983 0.983 0.983
47
+ 46 11:10:13 4 0.0063 0.05212448329383334 0.0644669309258461 0.983 0.983 0.983 0.983
48
+ 47 11:13:29 1 0.0031 0.05113414183996207 0.06470787525177002 0.9829 0.9829 0.9829 0.9829
49
+ 48 11:16:48 2 0.0031 0.05100923420506556 0.06484530121088028 0.983 0.983 0.983 0.983
50
+ 49 11:20:04 3 0.0031 0.05102771810317176 0.06486314535140991 0.983 0.983 0.983 0.983
51
+ 50 11:23:20 4 0.0031 0.05190557099563212 0.06452730298042297 0.9831 0.9831 0.9831 0.9831
preview.PNG ADDED
test.tsv ADDED
The diff for this file is too large to render. See raw diff
 
training.log ADDED
@@ -0,0 +1,1188 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2021-12-31 08:35:07,676 ----------------------------------------------------------------------------------------------------
2
+ 2021-12-31 08:35:07,680 Model: "SequenceTagger(
3
+ (embeddings): StackedEmbeddings(
4
+ (list_embedding_0): FlairEmbeddings(
5
+ (lm): LanguageModel(
6
+ (drop): Dropout(p=0.5, inplace=False)
7
+ (encoder): Embedding(275, 100)
8
+ (rnn): LSTM(100, 1024)
9
+ (decoder): Linear(in_features=1024, out_features=275, bias=True)
10
+ )
11
+ )
12
+ (list_embedding_1): FlairEmbeddings(
13
+ (lm): LanguageModel(
14
+ (drop): Dropout(p=0.5, inplace=False)
15
+ (encoder): Embedding(275, 100)
16
+ (rnn): LSTM(100, 1024)
17
+ (decoder): Linear(in_features=1024, out_features=275, bias=True)
18
+ )
19
+ )
20
+ (list_embedding_2): TransformerWordEmbeddings(
21
+ (model): CamembertModel(
22
+ (embeddings): RobertaEmbeddings(
23
+ (word_embeddings): Embedding(32005, 768, padding_idx=1)
24
+ (position_embeddings): Embedding(514, 768, padding_idx=1)
25
+ (token_type_embeddings): Embedding(1, 768)
26
+ (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
27
+ (dropout): Dropout(p=0.1, inplace=False)
28
+ )
29
+ (encoder): RobertaEncoder(
30
+ (layer): ModuleList(
31
+ (0): RobertaLayer(
32
+ (attention): RobertaAttention(
33
+ (self): RobertaSelfAttention(
34
+ (query): Linear(in_features=768, out_features=768, bias=True)
35
+ (key): Linear(in_features=768, out_features=768, bias=True)
36
+ (value): Linear(in_features=768, out_features=768, bias=True)
37
+ (dropout): Dropout(p=0.1, inplace=False)
38
+ )
39
+ (output): RobertaSelfOutput(
40
+ (dense): Linear(in_features=768, out_features=768, bias=True)
41
+ (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
42
+ (dropout): Dropout(p=0.1, inplace=False)
43
+ )
44
+ )
45
+ (intermediate): RobertaIntermediate(
46
+ (dense): Linear(in_features=768, out_features=3072, bias=True)
47
+ )
48
+ (output): RobertaOutput(
49
+ (dense): Linear(in_features=3072, out_features=768, bias=True)
50
+ (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
51
+ (dropout): Dropout(p=0.1, inplace=False)
52
+ )
53
+ )
54
+ (1): RobertaLayer(
55
+ (attention): RobertaAttention(
56
+ (self): RobertaSelfAttention(
57
+ (query): Linear(in_features=768, out_features=768, bias=True)
58
+ (key): Linear(in_features=768, out_features=768, bias=True)
59
+ (value): Linear(in_features=768, out_features=768, bias=True)
60
+ (dropout): Dropout(p=0.1, inplace=False)
61
+ )
62
+ (output): RobertaSelfOutput(
63
+ (dense): Linear(in_features=768, out_features=768, bias=True)
64
+ (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
65
+ (dropout): Dropout(p=0.1, inplace=False)
66
+ )
67
+ )
68
+ (intermediate): RobertaIntermediate(
69
+ (dense): Linear(in_features=768, out_features=3072, bias=True)
70
+ )
71
+ (output): RobertaOutput(
72
+ (dense): Linear(in_features=3072, out_features=768, bias=True)
73
+ (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
74
+ (dropout): Dropout(p=0.1, inplace=False)
75
+ )
76
+ )
77
+ (2): RobertaLayer(
78
+ (attention): RobertaAttention(
79
+ (self): RobertaSelfAttention(
80
+ (query): Linear(in_features=768, out_features=768, bias=True)
81
+ (key): Linear(in_features=768, out_features=768, bias=True)
82
+ (value): Linear(in_features=768, out_features=768, bias=True)
83
+ (dropout): Dropout(p=0.1, inplace=False)
84
+ )
85
+ (output): RobertaSelfOutput(
86
+ (dense): Linear(in_features=768, out_features=768, bias=True)
87
+ (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
88
+ (dropout): Dropout(p=0.1, inplace=False)
89
+ )
90
+ )
91
+ (intermediate): RobertaIntermediate(
92
+ (dense): Linear(in_features=768, out_features=3072, bias=True)
93
+ )
94
+ (output): RobertaOutput(
95
+ (dense): Linear(in_features=3072, out_features=768, bias=True)
96
+ (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
97
+ (dropout): Dropout(p=0.1, inplace=False)
98
+ )
99
+ )
100
+ (3): RobertaLayer(
101
+ (attention): RobertaAttention(
102
+ (self): RobertaSelfAttention(
103
+ (query): Linear(in_features=768, out_features=768, bias=True)
104
+ (key): Linear(in_features=768, out_features=768, bias=True)
105
+ (value): Linear(in_features=768, out_features=768, bias=True)
106
+ (dropout): Dropout(p=0.1, inplace=False)
107
+ )
108
+ (output): RobertaSelfOutput(
109
+ (dense): Linear(in_features=768, out_features=768, bias=True)
110
+ (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
111
+ (dropout): Dropout(p=0.1, inplace=False)
112
+ )
113
+ )
114
+ (intermediate): RobertaIntermediate(
115
+ (dense): Linear(in_features=768, out_features=3072, bias=True)
116
+ )
117
+ (output): RobertaOutput(
118
+ (dense): Linear(in_features=3072, out_features=768, bias=True)
119
+ (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
120
+ (dropout): Dropout(p=0.1, inplace=False)
121
+ )
122
+ )
123
+ (4): RobertaLayer(
124
+ (attention): RobertaAttention(
125
+ (self): RobertaSelfAttention(
126
+ (query): Linear(in_features=768, out_features=768, bias=True)
127
+ (key): Linear(in_features=768, out_features=768, bias=True)
128
+ (value): Linear(in_features=768, out_features=768, bias=True)
129
+ (dropout): Dropout(p=0.1, inplace=False)
130
+ )
131
+ (output): RobertaSelfOutput(
132
+ (dense): Linear(in_features=768, out_features=768, bias=True)
133
+ (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
134
+ (dropout): Dropout(p=0.1, inplace=False)
135
+ )
136
+ )
137
+ (intermediate): RobertaIntermediate(
138
+ (dense): Linear(in_features=768, out_features=3072, bias=True)
139
+ )
140
+ (output): RobertaOutput(
141
+ (dense): Linear(in_features=3072, out_features=768, bias=True)
142
+ (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
143
+ (dropout): Dropout(p=0.1, inplace=False)
144
+ )
145
+ )
146
+ (5): RobertaLayer(
147
+ (attention): RobertaAttention(
148
+ (self): RobertaSelfAttention(
149
+ (query): Linear(in_features=768, out_features=768, bias=True)
150
+ (key): Linear(in_features=768, out_features=768, bias=True)
151
+ (value): Linear(in_features=768, out_features=768, bias=True)
152
+ (dropout): Dropout(p=0.1, inplace=False)
153
+ )
154
+ (output): RobertaSelfOutput(
155
+ (dense): Linear(in_features=768, out_features=768, bias=True)
156
+ (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
157
+ (dropout): Dropout(p=0.1, inplace=False)
158
+ )
159
+ )
160
+ (intermediate): RobertaIntermediate(
161
+ (dense): Linear(in_features=768, out_features=3072, bias=True)
162
+ )
163
+ (output): RobertaOutput(
164
+ (dense): Linear(in_features=3072, out_features=768, bias=True)
165
+ (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
166
+ (dropout): Dropout(p=0.1, inplace=False)
167
+ )
168
+ )
169
+ (6): RobertaLayer(
170
+ (attention): RobertaAttention(
171
+ (self): RobertaSelfAttention(
172
+ (query): Linear(in_features=768, out_features=768, bias=True)
173
+ (key): Linear(in_features=768, out_features=768, bias=True)
174
+ (value): Linear(in_features=768, out_features=768, bias=True)
175
+ (dropout): Dropout(p=0.1, inplace=False)
176
+ )
177
+ (output): RobertaSelfOutput(
178
+ (dense): Linear(in_features=768, out_features=768, bias=True)
179
+ (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
180
+ (dropout): Dropout(p=0.1, inplace=False)
181
+ )
182
+ )
183
+ (intermediate): RobertaIntermediate(
184
+ (dense): Linear(in_features=768, out_features=3072, bias=True)
185
+ )
186
+ (output): RobertaOutput(
187
+ (dense): Linear(in_features=3072, out_features=768, bias=True)
188
+ (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
189
+ (dropout): Dropout(p=0.1, inplace=False)
190
+ )
191
+ )
192
+ (7): RobertaLayer(
193
+ (attention): RobertaAttention(
194
+ (self): RobertaSelfAttention(
195
+ (query): Linear(in_features=768, out_features=768, bias=True)
196
+ (key): Linear(in_features=768, out_features=768, bias=True)
197
+ (value): Linear(in_features=768, out_features=768, bias=True)
198
+ (dropout): Dropout(p=0.1, inplace=False)
199
+ )
200
+ (output): RobertaSelfOutput(
201
+ (dense): Linear(in_features=768, out_features=768, bias=True)
202
+ (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
203
+ (dropout): Dropout(p=0.1, inplace=False)
204
+ )
205
+ )
206
+ (intermediate): RobertaIntermediate(
207
+ (dense): Linear(in_features=768, out_features=3072, bias=True)
208
+ )
209
+ (output): RobertaOutput(
210
+ (dense): Linear(in_features=3072, out_features=768, bias=True)
211
+ (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
212
+ (dropout): Dropout(p=0.1, inplace=False)
213
+ )
214
+ )
215
+ (8): RobertaLayer(
216
+ (attention): RobertaAttention(
217
+ (self): RobertaSelfAttention(
218
+ (query): Linear(in_features=768, out_features=768, bias=True)
219
+ (key): Linear(in_features=768, out_features=768, bias=True)
220
+ (value): Linear(in_features=768, out_features=768, bias=True)
221
+ (dropout): Dropout(p=0.1, inplace=False)
222
+ )
223
+ (output): RobertaSelfOutput(
224
+ (dense): Linear(in_features=768, out_features=768, bias=True)
225
+ (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
226
+ (dropout): Dropout(p=0.1, inplace=False)
227
+ )
228
+ )
229
+ (intermediate): RobertaIntermediate(
230
+ (dense): Linear(in_features=768, out_features=3072, bias=True)
231
+ )
232
+ (output): RobertaOutput(
233
+ (dense): Linear(in_features=3072, out_features=768, bias=True)
234
+ (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
235
+ (dropout): Dropout(p=0.1, inplace=False)
236
+ )
237
+ )
238
+ (9): RobertaLayer(
239
+ (attention): RobertaAttention(
240
+ (self): RobertaSelfAttention(
241
+ (query): Linear(in_features=768, out_features=768, bias=True)
242
+ (key): Linear(in_features=768, out_features=768, bias=True)
243
+ (value): Linear(in_features=768, out_features=768, bias=True)
244
+ (dropout): Dropout(p=0.1, inplace=False)
245
+ )
246
+ (output): RobertaSelfOutput(
247
+ (dense): Linear(in_features=768, out_features=768, bias=True)
248
+ (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
249
+ (dropout): Dropout(p=0.1, inplace=False)
250
+ )
251
+ )
252
+ (intermediate): RobertaIntermediate(
253
+ (dense): Linear(in_features=768, out_features=3072, bias=True)
254
+ )
255
+ (output): RobertaOutput(
256
+ (dense): Linear(in_features=3072, out_features=768, bias=True)
257
+ (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
258
+ (dropout): Dropout(p=0.1, inplace=False)
259
+ )
260
+ )
261
+ (10): RobertaLayer(
262
+ (attention): RobertaAttention(
263
+ (self): RobertaSelfAttention(
264
+ (query): Linear(in_features=768, out_features=768, bias=True)
265
+ (key): Linear(in_features=768, out_features=768, bias=True)
266
+ (value): Linear(in_features=768, out_features=768, bias=True)
267
+ (dropout): Dropout(p=0.1, inplace=False)
268
+ )
269
+ (output): RobertaSelfOutput(
270
+ (dense): Linear(in_features=768, out_features=768, bias=True)
271
+ (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
272
+ (dropout): Dropout(p=0.1, inplace=False)
273
+ )
274
+ )
275
+ (intermediate): RobertaIntermediate(
276
+ (dense): Linear(in_features=768, out_features=3072, bias=True)
277
+ )
278
+ (output): RobertaOutput(
279
+ (dense): Linear(in_features=3072, out_features=768, bias=True)
280
+ (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
281
+ (dropout): Dropout(p=0.1, inplace=False)
282
+ )
283
+ )
284
+ (11): RobertaLayer(
285
+ (attention): RobertaAttention(
286
+ (self): RobertaSelfAttention(
287
+ (query): Linear(in_features=768, out_features=768, bias=True)
288
+ (key): Linear(in_features=768, out_features=768, bias=True)
289
+ (value): Linear(in_features=768, out_features=768, bias=True)
290
+ (dropout): Dropout(p=0.1, inplace=False)
291
+ )
292
+ (output): RobertaSelfOutput(
293
+ (dense): Linear(in_features=768, out_features=768, bias=True)
294
+ (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
295
+ (dropout): Dropout(p=0.1, inplace=False)
296
+ )
297
+ )
298
+ (intermediate): RobertaIntermediate(
299
+ (dense): Linear(in_features=768, out_features=3072, bias=True)
300
+ )
301
+ (output): RobertaOutput(
302
+ (dense): Linear(in_features=3072, out_features=768, bias=True)
303
+ (LayerNorm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
304
+ (dropout): Dropout(p=0.1, inplace=False)
305
+ )
306
+ )
307
+ )
308
+ )
309
+ (pooler): RobertaPooler(
310
+ (dense): Linear(in_features=768, out_features=768, bias=True)
311
+ (activation): Tanh()
312
+ )
313
+ )
314
+ )
315
+ )
316
+ (word_dropout): WordDropout(p=0.05)
317
+ (locked_dropout): LockedDropout(p=0.5)
318
+ (embedding2nn): Linear(in_features=2816, out_features=2816, bias=True)
319
+ (rnn): LSTM(2816, 256, batch_first=True, bidirectional=True)
320
+ (linear): Linear(in_features=512, out_features=68, bias=True)
321
+ (beta): 1.0
322
+ (weights): None
323
+ (weight_tensor) None
324
+ )"
325
+ 2021-12-31 08:35:07,680 ----------------------------------------------------------------------------------------------------
326
+ 2021-12-31 08:35:07,681 Corpus: "Corpus: 14449 train + 1476 dev + 416 test sentences"
327
+ 2021-12-31 08:35:07,681 ----------------------------------------------------------------------------------------------------
328
+ 2021-12-31 08:35:07,681 Parameters:
329
+ 2021-12-31 08:35:07,681 - learning_rate: "0.1"
330
+ 2021-12-31 08:35:07,681 - mini_batch_size: "8"
331
+ 2021-12-31 08:35:07,681 - patience: "3"
332
+ 2021-12-31 08:35:07,681 - anneal_factor: "0.5"
333
+ 2021-12-31 08:35:07,681 - max_epochs: "50"
334
+ 2021-12-31 08:35:07,681 - shuffle: "True"
335
+ 2021-12-31 08:35:07,681 - train_with_dev: "False"
336
+ 2021-12-31 08:35:07,681 - batch_growth_annealing: "False"
337
+ 2021-12-31 08:35:07,681 ----------------------------------------------------------------------------------------------------
338
+ 2021-12-31 08:35:07,681 Model training base path: "models/UPOS_UD_FRENCH_GSD_PLUS_Flair-Embeddings_50_2021-12-31-08:34:44"
339
+ 2021-12-31 08:35:07,681 ----------------------------------------------------------------------------------------------------
340
+ 2021-12-31 08:35:07,682 Device: cuda:0
341
+ 2021-12-31 08:35:07,682 ----------------------------------------------------------------------------------------------------
342
+ 2021-12-31 08:35:07,682 Embeddings storage mode: cpu
343
+ 2021-12-31 08:35:07,686 ----------------------------------------------------------------------------------------------------
344
+ 2021-12-31 08:35:35,600 epoch 1 - iter 180/1807 - loss 1.43338722 - samples/sec: 51.63 - lr: 0.100000
345
+ 2021-12-31 08:36:03,642 epoch 1 - iter 360/1807 - loss 0.97278560 - samples/sec: 51.39 - lr: 0.100000
346
+ 2021-12-31 08:36:31,448 epoch 1 - iter 540/1807 - loss 0.77628898 - samples/sec: 51.83 - lr: 0.100000
347
+ 2021-12-31 08:37:00,007 epoch 1 - iter 720/1807 - loss 0.66122431 - samples/sec: 50.46 - lr: 0.100000
348
+ 2021-12-31 08:37:29,449 epoch 1 - iter 900/1807 - loss 0.58637716 - samples/sec: 48.94 - lr: 0.100000
349
+ 2021-12-31 08:37:57,842 epoch 1 - iter 1080/1807 - loss 0.53261867 - samples/sec: 50.75 - lr: 0.100000
350
+ 2021-12-31 08:38:27,836 epoch 1 - iter 1260/1807 - loss 0.49236809 - samples/sec: 48.04 - lr: 0.100000
351
+ 2021-12-31 08:38:56,177 epoch 1 - iter 1440/1807 - loss 0.46224064 - samples/sec: 50.84 - lr: 0.100000
352
+ 2021-12-31 08:39:25,301 epoch 1 - iter 1620/1807 - loss 0.43700232 - samples/sec: 49.48 - lr: 0.100000
353
+ 2021-12-31 08:39:53,843 epoch 1 - iter 1800/1807 - loss 0.41459922 - samples/sec: 50.49 - lr: 0.100000
354
+ 2021-12-31 08:39:54,850 ----------------------------------------------------------------------------------------------------
355
+ 2021-12-31 08:39:54,851 EPOCH 1 done: loss 0.4139 - lr 0.1000000
356
+ 2021-12-31 08:40:38,186 DEV : loss 0.09867297857999802 - f1-score (micro avg) 0.9723
357
+ 2021-12-31 08:40:38,373 BAD EPOCHS (no improvement): 0
358
+ 2021-12-31 08:40:38,375 saving best model
359
+ 2021-12-31 08:40:43,945 ----------------------------------------------------------------------------------------------------
360
+ 2021-12-31 08:40:59,809 epoch 2 - iter 180/1807 - loss 0.20282785 - samples/sec: 90.92 - lr: 0.100000
361
+ 2021-12-31 08:41:15,798 epoch 2 - iter 360/1807 - loss 0.20600484 - samples/sec: 90.20 - lr: 0.100000
362
+ 2021-12-31 08:41:31,824 epoch 2 - iter 540/1807 - loss 0.20352355 - samples/sec: 89.99 - lr: 0.100000
363
+ 2021-12-31 08:41:47,291 epoch 2 - iter 720/1807 - loss 0.19945298 - samples/sec: 93.24 - lr: 0.100000
364
+ 2021-12-31 08:42:03,389 epoch 2 - iter 900/1807 - loss 0.19672769 - samples/sec: 89.58 - lr: 0.100000
365
+ 2021-12-31 08:42:19,546 epoch 2 - iter 1080/1807 - loss 0.19404584 - samples/sec: 89.25 - lr: 0.100000
366
+ 2021-12-31 08:42:35,186 epoch 2 - iter 1260/1807 - loss 0.19211776 - samples/sec: 92.22 - lr: 0.100000
367
+ 2021-12-31 08:42:51,014 epoch 2 - iter 1440/1807 - loss 0.19040930 - samples/sec: 91.11 - lr: 0.100000
368
+ 2021-12-31 08:43:07,108 epoch 2 - iter 1620/1807 - loss 0.18835936 - samples/sec: 89.60 - lr: 0.100000
369
+ 2021-12-31 08:43:22,664 epoch 2 - iter 1800/1807 - loss 0.18684498 - samples/sec: 92.71 - lr: 0.100000
370
+ 2021-12-31 08:43:23,166 ----------------------------------------------------------------------------------------------------
371
+ 2021-12-31 08:43:23,166 EPOCH 2 done: loss 0.1868 - lr 0.1000000
372
+ 2021-12-31 08:43:59,411 DEV : loss 0.08219591528177261 - f1-score (micro avg) 0.9761
373
+ 2021-12-31 08:43:59,601 BAD EPOCHS (no improvement): 0
374
+ 2021-12-31 08:43:59,602 saving best model
375
+ 2021-12-31 08:44:04,994 ----------------------------------------------------------------------------------------------------
376
+ 2021-12-31 08:44:21,188 epoch 3 - iter 180/1807 - loss 0.16248988 - samples/sec: 89.06 - lr: 0.100000
377
+ 2021-12-31 08:44:37,143 epoch 3 - iter 360/1807 - loss 0.16012805 - samples/sec: 90.38 - lr: 0.100000
378
+ 2021-12-31 08:44:53,240 epoch 3 - iter 540/1807 - loss 0.15771573 - samples/sec: 89.59 - lr: 0.100000
379
+ 2021-12-31 08:45:08,820 epoch 3 - iter 720/1807 - loss 0.15678918 - samples/sec: 92.57 - lr: 0.100000
380
+ 2021-12-31 08:45:24,447 epoch 3 - iter 900/1807 - loss 0.15583330 - samples/sec: 92.28 - lr: 0.100000
381
+ 2021-12-31 08:45:40,453 epoch 3 - iter 1080/1807 - loss 0.15551694 - samples/sec: 90.10 - lr: 0.100000
382
+ 2021-12-31 08:45:56,421 epoch 3 - iter 1260/1807 - loss 0.15503272 - samples/sec: 90.32 - lr: 0.100000
383
+ 2021-12-31 08:46:12,207 epoch 3 - iter 1440/1807 - loss 0.15478837 - samples/sec: 91.35 - lr: 0.100000
384
+ 2021-12-31 08:46:28,067 epoch 3 - iter 1620/1807 - loss 0.15437671 - samples/sec: 90.93 - lr: 0.100000
385
+ 2021-12-31 08:46:44,096 epoch 3 - iter 1800/1807 - loss 0.15334210 - samples/sec: 89.96 - lr: 0.100000
386
+ 2021-12-31 08:46:44,638 ----------------------------------------------------------------------------------------------------
387
+ 2021-12-31 08:46:44,638 EPOCH 3 done: loss 0.1533 - lr 0.1000000
388
+ 2021-12-31 08:47:19,364 DEV : loss 0.07821641117334366 - f1-score (micro avg) 0.9771
389
+ 2021-12-31 08:47:19,574 BAD EPOCHS (no improvement): 0
390
+ 2021-12-31 08:47:19,576 saving best model
391
+ 2021-12-31 08:47:25,807 ----------------------------------------------------------------------------------------------------
392
+ 2021-12-31 08:47:42,295 epoch 4 - iter 180/1807 - loss 0.14078583 - samples/sec: 87.48 - lr: 0.100000
393
+ 2021-12-31 08:47:58,394 epoch 4 - iter 360/1807 - loss 0.14084079 - samples/sec: 89.58 - lr: 0.100000
394
+ 2021-12-31 08:48:14,377 epoch 4 - iter 540/1807 - loss 0.13969043 - samples/sec: 90.22 - lr: 0.100000
395
+ 2021-12-31 08:48:30,411 epoch 4 - iter 720/1807 - loss 0.13901425 - samples/sec: 89.95 - lr: 0.100000
396
+ 2021-12-31 08:48:45,985 epoch 4 - iter 900/1807 - loss 0.13965987 - samples/sec: 92.60 - lr: 0.100000
397
+ 2021-12-31 08:49:01,706 epoch 4 - iter 1080/1807 - loss 0.13942263 - samples/sec: 91.73 - lr: 0.100000
398
+ 2021-12-31 08:49:17,833 epoch 4 - iter 1260/1807 - loss 0.13931213 - samples/sec: 89.42 - lr: 0.100000
399
+ 2021-12-31 08:49:33,693 epoch 4 - iter 1440/1807 - loss 0.13835426 - samples/sec: 90.94 - lr: 0.100000
400
+ 2021-12-31 08:49:49,444 epoch 4 - iter 1620/1807 - loss 0.13722078 - samples/sec: 91.56 - lr: 0.100000
401
+ 2021-12-31 08:50:05,233 epoch 4 - iter 1800/1807 - loss 0.13680325 - samples/sec: 91.33 - lr: 0.100000
402
+ 2021-12-31 08:50:05,825 ----------------------------------------------------------------------------------------------------
403
+ 2021-12-31 08:50:05,826 EPOCH 4 done: loss 0.1368 - lr 0.1000000
404
+ 2021-12-31 08:50:40,951 DEV : loss 0.07048774510622025 - f1-score (micro avg) 0.9784
405
+ 2021-12-31 08:50:41,121 BAD EPOCHS (no improvement): 0
406
+ 2021-12-31 08:50:41,123 saving best model
407
+ 2021-12-31 08:50:46,985 ----------------------------------------------------------------------------------------------------
408
+ 2021-12-31 08:51:03,480 epoch 5 - iter 180/1807 - loss 0.12576483 - samples/sec: 87.44 - lr: 0.100000
409
+ 2021-12-31 08:51:19,312 epoch 5 - iter 360/1807 - loss 0.12838224 - samples/sec: 91.10 - lr: 0.100000
410
+ 2021-12-31 08:51:35,140 epoch 5 - iter 540/1807 - loss 0.13027925 - samples/sec: 91.11 - lr: 0.100000
411
+ 2021-12-31 08:51:51,382 epoch 5 - iter 720/1807 - loss 0.13001079 - samples/sec: 88.78 - lr: 0.100000
412
+ 2021-12-31 08:52:07,009 epoch 5 - iter 900/1807 - loss 0.12990639 - samples/sec: 92.28 - lr: 0.100000
413
+ 2021-12-31 08:52:22,749 epoch 5 - iter 1080/1807 - loss 0.12927608 - samples/sec: 91.63 - lr: 0.100000
414
+ 2021-12-31 08:52:38,459 epoch 5 - iter 1260/1807 - loss 0.12839810 - samples/sec: 91.79 - lr: 0.100000
415
+ 2021-12-31 08:52:54,183 epoch 5 - iter 1440/1807 - loss 0.12750076 - samples/sec: 91.71 - lr: 0.100000
416
+ 2021-12-31 08:53:09,782 epoch 5 - iter 1620/1807 - loss 0.12744081 - samples/sec: 92.45 - lr: 0.100000
417
+ 2021-12-31 08:53:26,181 epoch 5 - iter 1800/1807 - loss 0.12697954 - samples/sec: 87.94 - lr: 0.100000
418
+ 2021-12-31 08:53:26,718 ----------------------------------------------------------------------------------------------------
419
+ 2021-12-31 08:53:26,718 EPOCH 5 done: loss 0.1270 - lr 0.1000000
420
+ 2021-12-31 08:54:05,303 DEV : loss 0.06857253611087799 - f1-score (micro avg) 0.9795
421
+ 2021-12-31 08:54:05,490 BAD EPOCHS (no improvement): 0
422
+ 2021-12-31 08:54:05,491 saving best model
423
+ 2021-12-31 08:54:11,317 ----------------------------------------------------------------------------------------------------
424
+ 2021-12-31 08:54:27,729 epoch 6 - iter 180/1807 - loss 0.12012197 - samples/sec: 87.88 - lr: 0.100000
425
+ 2021-12-31 08:54:43,570 epoch 6 - iter 360/1807 - loss 0.12134345 - samples/sec: 91.04 - lr: 0.100000
426
+ 2021-12-31 08:54:59,298 epoch 6 - iter 540/1807 - loss 0.12010472 - samples/sec: 91.70 - lr: 0.100000
427
+ 2021-12-31 08:55:14,710 epoch 6 - iter 720/1807 - loss 0.11985671 - samples/sec: 93.58 - lr: 0.100000
428
+ 2021-12-31 08:55:30,873 epoch 6 - iter 900/1807 - loss 0.12032070 - samples/sec: 89.22 - lr: 0.100000
429
+ 2021-12-31 08:55:46,705 epoch 6 - iter 1080/1807 - loss 0.11976455 - samples/sec: 91.08 - lr: 0.100000
430
+ 2021-12-31 08:56:02,915 epoch 6 - iter 1260/1807 - loss 0.11964832 - samples/sec: 88.97 - lr: 0.100000
431
+ 2021-12-31 08:56:18,616 epoch 6 - iter 1440/1807 - loss 0.11958148 - samples/sec: 91.86 - lr: 0.100000
432
+ 2021-12-31 08:56:34,478 epoch 6 - iter 1620/1807 - loss 0.12003314 - samples/sec: 90.91 - lr: 0.100000
433
+ 2021-12-31 08:56:50,548 epoch 6 - iter 1800/1807 - loss 0.11950787 - samples/sec: 89.75 - lr: 0.100000
434
+ 2021-12-31 08:56:51,070 ----------------------------------------------------------------------------------------------------
435
+ 2021-12-31 08:56:51,070 EPOCH 6 done: loss 0.1195 - lr 0.1000000
436
+ 2021-12-31 08:57:26,881 DEV : loss 0.06588418781757355 - f1-score (micro avg) 0.9805
437
+ 2021-12-31 08:57:27,077 BAD EPOCHS (no improvement): 0
438
+ 2021-12-31 08:57:27,079 saving best model
439
+ 2021-12-31 08:57:32,878 ----------------------------------------------------------------------------------------------------
440
+ 2021-12-31 08:57:49,222 epoch 7 - iter 180/1807 - loss 0.11622596 - samples/sec: 88.27 - lr: 0.100000
441
+ 2021-12-31 08:58:05,154 epoch 7 - iter 360/1807 - loss 0.11182908 - samples/sec: 90.52 - lr: 0.100000
442
+ 2021-12-31 08:58:21,316 epoch 7 - iter 540/1807 - loss 0.11325284 - samples/sec: 89.23 - lr: 0.100000
443
+ 2021-12-31 08:58:37,501 epoch 7 - iter 720/1807 - loss 0.11356510 - samples/sec: 89.11 - lr: 0.100000
444
+ 2021-12-31 08:58:53,437 epoch 7 - iter 900/1807 - loss 0.11375009 - samples/sec: 90.50 - lr: 0.100000
445
+ 2021-12-31 08:59:09,683 epoch 7 - iter 1080/1807 - loss 0.11424006 - samples/sec: 88.76 - lr: 0.100000
446
+ 2021-12-31 08:59:25,513 epoch 7 - iter 1260/1807 - loss 0.11502991 - samples/sec: 91.10 - lr: 0.100000
447
+ 2021-12-31 08:59:41,355 epoch 7 - iter 1440/1807 - loss 0.11465724 - samples/sec: 91.04 - lr: 0.100000
448
+ 2021-12-31 08:59:57,048 epoch 7 - iter 1620/1807 - loss 0.11489345 - samples/sec: 91.91 - lr: 0.100000
449
+ 2021-12-31 09:00:13,626 epoch 7 - iter 1800/1807 - loss 0.11495780 - samples/sec: 86.99 - lr: 0.100000
450
+ 2021-12-31 09:00:14,225 ----------------------------------------------------------------------------------------------------
451
+ 2021-12-31 09:00:14,225 EPOCH 7 done: loss 0.1149 - lr 0.1000000
452
+ 2021-12-31 09:00:50,356 DEV : loss 0.06450950354337692 - f1-score (micro avg) 0.981
453
+ 2021-12-31 09:00:50,566 BAD EPOCHS (no improvement): 0
454
+ 2021-12-31 09:00:50,572 saving best model
455
+ 2021-12-31 09:00:56,353 ----------------------------------------------------------------------------------------------------
456
+ 2021-12-31 09:01:12,703 epoch 8 - iter 180/1807 - loss 0.10372694 - samples/sec: 88.23 - lr: 0.100000
457
+ 2021-12-31 09:01:28,785 epoch 8 - iter 360/1807 - loss 0.10507104 - samples/sec: 89.68 - lr: 0.100000
458
+ 2021-12-31 09:01:45,134 epoch 8 - iter 540/1807 - loss 0.10666062 - samples/sec: 88.21 - lr: 0.100000
459
+ 2021-12-31 09:02:01,507 epoch 8 - iter 720/1807 - loss 0.10750728 - samples/sec: 88.08 - lr: 0.100000
460
+ 2021-12-31 09:02:17,626 epoch 8 - iter 900/1807 - loss 0.10760637 - samples/sec: 89.47 - lr: 0.100000
461
+ 2021-12-31 09:02:33,374 epoch 8 - iter 1080/1807 - loss 0.10788257 - samples/sec: 91.58 - lr: 0.100000
462
+ 2021-12-31 09:02:49,200 epoch 8 - iter 1260/1807 - loss 0.10808589 - samples/sec: 91.12 - lr: 0.100000
463
+ 2021-12-31 09:03:05,738 epoch 8 - iter 1440/1807 - loss 0.10815170 - samples/sec: 87.20 - lr: 0.100000
464
+ 2021-12-31 09:03:21,442 epoch 8 - iter 1620/1807 - loss 0.10840840 - samples/sec: 91.84 - lr: 0.100000
465
+ 2021-12-31 09:03:37,709 epoch 8 - iter 1800/1807 - loss 0.10855634 - samples/sec: 88.66 - lr: 0.100000
466
+ 2021-12-31 09:03:38,280 ----------------------------------------------------------------------------------------------------
467
+ 2021-12-31 09:03:38,280 EPOCH 8 done: loss 0.1086 - lr 0.1000000
468
+ 2021-12-31 09:04:17,043 DEV : loss 0.06390747427940369 - f1-score (micro avg) 0.9805
469
+ 2021-12-31 09:04:17,194 BAD EPOCHS (no improvement): 1
470
+ 2021-12-31 09:04:17,196 ----------------------------------------------------------------------------------------------------
471
+ 2021-12-31 09:04:33,331 epoch 9 - iter 180/1807 - loss 0.10260778 - samples/sec: 89.39 - lr: 0.100000
472
+ 2021-12-31 09:04:49,336 epoch 9 - iter 360/1807 - loss 0.10566575 - samples/sec: 90.11 - lr: 0.100000
473
+ 2021-12-31 09:05:05,083 epoch 9 - iter 540/1807 - loss 0.10556216 - samples/sec: 91.59 - lr: 0.100000
474
+ 2021-12-31 09:05:21,004 epoch 9 - iter 720/1807 - loss 0.10506801 - samples/sec: 90.58 - lr: 0.100000
475
+ 2021-12-31 09:05:37,109 epoch 9 - iter 900/1807 - loss 0.10596338 - samples/sec: 89.54 - lr: 0.100000
476
+ 2021-12-31 09:05:52,784 epoch 9 - iter 1080/1807 - loss 0.10577668 - samples/sec: 92.02 - lr: 0.100000
477
+ 2021-12-31 09:06:08,937 epoch 9 - iter 1260/1807 - loss 0.10613509 - samples/sec: 89.28 - lr: 0.100000
478
+ 2021-12-31 09:06:24,601 epoch 9 - iter 1440/1807 - loss 0.10637150 - samples/sec: 92.06 - lr: 0.100000
479
+ 2021-12-31 09:06:40,409 epoch 9 - iter 1620/1807 - loss 0.10629708 - samples/sec: 91.23 - lr: 0.100000
480
+ 2021-12-31 09:06:55,972 epoch 9 - iter 1800/1807 - loss 0.10610710 - samples/sec: 92.67 - lr: 0.100000
481
+ 2021-12-31 09:06:56,557 ----------------------------------------------------------------------------------------------------
482
+ 2021-12-31 09:06:56,557 EPOCH 9 done: loss 0.1061 - lr 0.1000000
483
+ 2021-12-31 09:07:32,784 DEV : loss 0.06607701629400253 - f1-score (micro avg) 0.9814
484
+ 2021-12-31 09:07:32,970 BAD EPOCHS (no improvement): 0
485
+ 2021-12-31 09:07:32,972 saving best model
486
+ 2021-12-31 09:07:38,755 ----------------------------------------------------------------------------------------------------
487
+ 2021-12-31 09:07:55,004 epoch 10 - iter 180/1807 - loss 0.10366226 - samples/sec: 88.76 - lr: 0.100000
488
+ 2021-12-31 09:08:11,104 epoch 10 - iter 360/1807 - loss 0.10828055 - samples/sec: 89.58 - lr: 0.100000
489
+ 2021-12-31 09:08:26,748 epoch 10 - iter 540/1807 - loss 0.10589800 - samples/sec: 92.20 - lr: 0.100000
490
+ 2021-12-31 09:08:42,772 epoch 10 - iter 720/1807 - loss 0.10467961 - samples/sec: 90.00 - lr: 0.100000
491
+ 2021-12-31 09:08:58,992 epoch 10 - iter 900/1807 - loss 0.10355149 - samples/sec: 88.91 - lr: 0.100000
492
+ 2021-12-31 09:09:14,753 epoch 10 - iter 1080/1807 - loss 0.10313717 - samples/sec: 91.50 - lr: 0.100000
493
+ 2021-12-31 09:09:30,631 epoch 10 - iter 1260/1807 - loss 0.10353533 - samples/sec: 90.84 - lr: 0.100000
494
+ 2021-12-31 09:09:46,654 epoch 10 - iter 1440/1807 - loss 0.10386166 - samples/sec: 90.02 - lr: 0.100000
495
+ 2021-12-31 09:10:02,791 epoch 10 - iter 1620/1807 - loss 0.10346798 - samples/sec: 89.36 - lr: 0.100000
496
+ 2021-12-31 09:10:18,970 epoch 10 - iter 1800/1807 - loss 0.10358051 - samples/sec: 89.14 - lr: 0.100000
497
+ 2021-12-31 09:10:19,492 ----------------------------------------------------------------------------------------------------
498
+ 2021-12-31 09:10:19,492 EPOCH 10 done: loss 0.1036 - lr 0.1000000
499
+ 2021-12-31 09:10:55,557 DEV : loss 0.06536506861448288 - f1-score (micro avg) 0.9811
500
+ 2021-12-31 09:10:55,753 BAD EPOCHS (no improvement): 1
501
+ 2021-12-31 09:10:55,756 ----------------------------------------------------------------------------------------------------
502
+ 2021-12-31 09:11:12,024 epoch 11 - iter 180/1807 - loss 0.10182872 - samples/sec: 88.66 - lr: 0.100000
503
+ 2021-12-31 09:11:28,246 epoch 11 - iter 360/1807 - loss 0.10175535 - samples/sec: 88.90 - lr: 0.100000
504
+ 2021-12-31 09:11:43,844 epoch 11 - iter 540/1807 - loss 0.10107946 - samples/sec: 92.46 - lr: 0.100000
505
+ 2021-12-31 09:11:59,559 epoch 11 - iter 720/1807 - loss 0.10053922 - samples/sec: 91.77 - lr: 0.100000
506
+ 2021-12-31 09:12:15,490 epoch 11 - iter 900/1807 - loss 0.10047028 - samples/sec: 90.54 - lr: 0.100000
507
+ 2021-12-31 09:12:31,195 epoch 11 - iter 1080/1807 - loss 0.09993958 - samples/sec: 91.82 - lr: 0.100000
508
+ 2021-12-31 09:12:47,013 epoch 11 - iter 1260/1807 - loss 0.09996914 - samples/sec: 91.17 - lr: 0.100000
509
+ 2021-12-31 09:13:03,156 epoch 11 - iter 1440/1807 - loss 0.09980985 - samples/sec: 89.35 - lr: 0.100000
510
+ 2021-12-31 09:13:18,852 epoch 11 - iter 1620/1807 - loss 0.09941318 - samples/sec: 91.88 - lr: 0.100000
511
+ 2021-12-31 09:13:35,014 epoch 11 - iter 1800/1807 - loss 0.09934768 - samples/sec: 89.23 - lr: 0.100000
512
+ 2021-12-31 09:13:35,650 ----------------------------------------------------------------------------------------------------
513
+ 2021-12-31 09:13:35,650 EPOCH 11 done: loss 0.0993 - lr 0.1000000
514
+ 2021-12-31 09:14:14,419 DEV : loss 0.06659943610429764 - f1-score (micro avg) 0.9811
515
+ 2021-12-31 09:14:14,622 BAD EPOCHS (no improvement): 2
516
+ 2021-12-31 09:14:14,623 ----------------------------------------------------------------------------------------------------
517
+ 2021-12-31 09:14:30,892 epoch 12 - iter 180/1807 - loss 0.09334718 - samples/sec: 88.66 - lr: 0.100000
518
+ 2021-12-31 09:14:46,737 epoch 12 - iter 360/1807 - loss 0.09477923 - samples/sec: 91.02 - lr: 0.100000
519
+ 2021-12-31 09:15:02,926 epoch 12 - iter 540/1807 - loss 0.09677398 - samples/sec: 89.09 - lr: 0.100000
520
+ 2021-12-31 09:15:19,177 epoch 12 - iter 720/1807 - loss 0.09825518 - samples/sec: 88.74 - lr: 0.100000
521
+ 2021-12-31 09:15:34,958 epoch 12 - iter 900/1807 - loss 0.09910665 - samples/sec: 91.38 - lr: 0.100000
522
+ 2021-12-31 09:15:51,056 epoch 12 - iter 1080/1807 - loss 0.09820501 - samples/sec: 89.59 - lr: 0.100000
523
+ 2021-12-31 09:16:07,231 epoch 12 - iter 1260/1807 - loss 0.09858638 - samples/sec: 89.16 - lr: 0.100000
524
+ 2021-12-31 09:16:22,988 epoch 12 - iter 1440/1807 - loss 0.09845736 - samples/sec: 91.52 - lr: 0.100000
525
+ 2021-12-31 09:16:38,631 epoch 12 - iter 1620/1807 - loss 0.09859390 - samples/sec: 92.21 - lr: 0.100000
526
+ 2021-12-31 09:16:54,209 epoch 12 - iter 1800/1807 - loss 0.09847298 - samples/sec: 92.58 - lr: 0.100000
527
+ 2021-12-31 09:16:54,729 ----------------------------------------------------------------------------------------------------
528
+ 2021-12-31 09:16:54,730 EPOCH 12 done: loss 0.0984 - lr 0.1000000
529
+ 2021-12-31 09:17:31,308 DEV : loss 0.06410104781389236 - f1-score (micro avg) 0.9816
530
+ 2021-12-31 09:17:31,487 BAD EPOCHS (no improvement): 0
531
+ 2021-12-31 09:17:31,489 saving best model
532
+ 2021-12-31 09:17:37,260 ----------------------------------------------------------------------------------------------------
533
+ 2021-12-31 09:17:54,060 epoch 13 - iter 180/1807 - loss 0.10013605 - samples/sec: 85.86 - lr: 0.100000
534
+ 2021-12-31 09:18:09,827 epoch 13 - iter 360/1807 - loss 0.09881566 - samples/sec: 91.47 - lr: 0.100000
535
+ 2021-12-31 09:18:25,218 epoch 13 - iter 540/1807 - loss 0.09860664 - samples/sec: 93.71 - lr: 0.100000
536
+ 2021-12-31 09:18:41,246 epoch 13 - iter 720/1807 - loss 0.09768065 - samples/sec: 89.97 - lr: 0.100000
537
+ 2021-12-31 09:18:57,306 epoch 13 - iter 900/1807 - loss 0.09766501 - samples/sec: 89.79 - lr: 0.100000
538
+ 2021-12-31 09:19:12,914 epoch 13 - iter 1080/1807 - loss 0.09767968 - samples/sec: 92.41 - lr: 0.100000
539
+ 2021-12-31 09:19:29,144 epoch 13 - iter 1260/1807 - loss 0.09667902 - samples/sec: 88.86 - lr: 0.100000
540
+ 2021-12-31 09:19:45,573 epoch 13 - iter 1440/1807 - loss 0.09670686 - samples/sec: 87.78 - lr: 0.100000
541
+ 2021-12-31 09:20:01,566 epoch 13 - iter 1620/1807 - loss 0.09672936 - samples/sec: 90.18 - lr: 0.100000
542
+ 2021-12-31 09:20:17,572 epoch 13 - iter 1800/1807 - loss 0.09666135 - samples/sec: 90.10 - lr: 0.100000
543
+ 2021-12-31 09:20:18,200 ----------------------------------------------------------------------------------------------------
544
+ 2021-12-31 09:20:18,200 EPOCH 13 done: loss 0.0967 - lr 0.1000000
545
+ 2021-12-31 09:20:54,147 DEV : loss 0.06427688896656036 - f1-score (micro avg) 0.9816
546
+ 2021-12-31 09:20:54,334 BAD EPOCHS (no improvement): 1
547
+ 2021-12-31 09:20:54,335 ----------------------------------------------------------------------------------------------------
548
+ 2021-12-31 09:21:10,174 epoch 14 - iter 180/1807 - loss 0.09391481 - samples/sec: 91.06 - lr: 0.100000
549
+ 2021-12-31 09:21:26,400 epoch 14 - iter 360/1807 - loss 0.09267418 - samples/sec: 88.88 - lr: 0.100000
550
+ 2021-12-31 09:21:42,313 epoch 14 - iter 540/1807 - loss 0.09273735 - samples/sec: 90.64 - lr: 0.100000
551
+ 2021-12-31 09:21:58,477 epoch 14 - iter 720/1807 - loss 0.09237732 - samples/sec: 89.22 - lr: 0.100000
552
+ 2021-12-31 09:22:14,088 epoch 14 - iter 900/1807 - loss 0.09290387 - samples/sec: 92.38 - lr: 0.100000
553
+ 2021-12-31 09:22:29,793 epoch 14 - iter 1080/1807 - loss 0.09305725 - samples/sec: 91.82 - lr: 0.100000
554
+ 2021-12-31 09:22:45,455 epoch 14 - iter 1260/1807 - loss 0.09321173 - samples/sec: 92.09 - lr: 0.100000
555
+ 2021-12-31 09:23:01,412 epoch 14 - iter 1440/1807 - loss 0.09321459 - samples/sec: 90.38 - lr: 0.100000
556
+ 2021-12-31 09:23:17,629 epoch 14 - iter 1620/1807 - loss 0.09332877 - samples/sec: 88.93 - lr: 0.100000
557
+ 2021-12-31 09:23:33,527 epoch 14 - iter 1800/1807 - loss 0.09313892 - samples/sec: 90.71 - lr: 0.100000
558
+ 2021-12-31 09:23:34,165 ----------------------------------------------------------------------------------------------------
559
+ 2021-12-31 09:23:34,165 EPOCH 14 done: loss 0.0931 - lr 0.1000000
560
+ 2021-12-31 09:24:12,840 DEV : loss 0.06639766693115234 - f1-score (micro avg) 0.9817
561
+ 2021-12-31 09:24:13,034 BAD EPOCHS (no improvement): 0
562
+ 2021-12-31 09:24:13,036 saving best model
563
+ 2021-12-31 09:24:18,822 ----------------------------------------------------------------------------------------------------
564
+ 2021-12-31 09:24:34,568 epoch 15 - iter 180/1807 - loss 0.09134784 - samples/sec: 91.60 - lr: 0.100000
565
+ 2021-12-31 09:24:50,712 epoch 15 - iter 360/1807 - loss 0.09119751 - samples/sec: 89.33 - lr: 0.100000
566
+ 2021-12-31 09:25:07,155 epoch 15 - iter 540/1807 - loss 0.08993505 - samples/sec: 87.70 - lr: 0.100000
567
+ 2021-12-31 09:25:23,092 epoch 15 - iter 720/1807 - loss 0.09062331 - samples/sec: 90.50 - lr: 0.100000
568
+ 2021-12-31 09:25:39,643 epoch 15 - iter 900/1807 - loss 0.09054947 - samples/sec: 87.13 - lr: 0.100000
569
+ 2021-12-31 09:25:56,080 epoch 15 - iter 1080/1807 - loss 0.09120586 - samples/sec: 87.73 - lr: 0.100000
570
+ 2021-12-31 09:26:12,023 epoch 15 - iter 1260/1807 - loss 0.09202164 - samples/sec: 90.49 - lr: 0.100000
571
+ 2021-12-31 09:26:27,452 epoch 15 - iter 1440/1807 - loss 0.09257595 - samples/sec: 93.48 - lr: 0.100000
572
+ 2021-12-31 09:26:43,293 epoch 15 - iter 1620/1807 - loss 0.09296868 - samples/sec: 91.04 - lr: 0.100000
573
+ 2021-12-31 09:26:59,412 epoch 15 - iter 1800/1807 - loss 0.09272942 - samples/sec: 89.47 - lr: 0.100000
574
+ 2021-12-31 09:26:59,991 ----------------------------------------------------------------------------------------------------
575
+ 2021-12-31 09:26:59,991 EPOCH 15 done: loss 0.0927 - lr 0.1000000
576
+ 2021-12-31 09:27:36,227 DEV : loss 0.06283392012119293 - f1-score (micro avg) 0.982
577
+ 2021-12-31 09:27:36,433 BAD EPOCHS (no improvement): 0
578
+ 2021-12-31 09:27:36,435 saving best model
579
+ 2021-12-31 09:27:42,216 ----------------------------------------------------------------------------------------------------
580
+ 2021-12-31 09:27:58,274 epoch 16 - iter 180/1807 - loss 0.08868552 - samples/sec: 89.83 - lr: 0.100000
581
+ 2021-12-31 09:28:14,083 epoch 16 - iter 360/1807 - loss 0.08898795 - samples/sec: 91.23 - lr: 0.100000
582
+ 2021-12-31 09:28:30,428 epoch 16 - iter 540/1807 - loss 0.08723848 - samples/sec: 88.23 - lr: 0.100000
583
+ 2021-12-31 09:28:46,065 epoch 16 - iter 720/1807 - loss 0.08840922 - samples/sec: 92.21 - lr: 0.100000
584
+ 2021-12-31 09:29:01,697 epoch 16 - iter 900/1807 - loss 0.08907246 - samples/sec: 92.26 - lr: 0.100000
585
+ 2021-12-31 09:29:17,387 epoch 16 - iter 1080/1807 - loss 0.09016391 - samples/sec: 91.91 - lr: 0.100000
586
+ 2021-12-31 09:29:33,637 epoch 16 - iter 1260/1807 - loss 0.09090909 - samples/sec: 88.74 - lr: 0.100000
587
+ 2021-12-31 09:29:49,596 epoch 16 - iter 1440/1807 - loss 0.09079363 - samples/sec: 90.36 - lr: 0.100000
588
+ 2021-12-31 09:30:05,085 epoch 16 - iter 1620/1807 - loss 0.09144623 - samples/sec: 93.12 - lr: 0.100000
589
+ 2021-12-31 09:30:21,000 epoch 16 - iter 1800/1807 - loss 0.09062250 - samples/sec: 90.62 - lr: 0.100000
590
+ 2021-12-31 09:30:21,608 ----------------------------------------------------------------------------------------------------
591
+ 2021-12-31 09:30:21,608 EPOCH 16 done: loss 0.0906 - lr 0.1000000
592
+ 2021-12-31 09:30:58,333 DEV : loss 0.06354553997516632 - f1-score (micro avg) 0.982
593
+ 2021-12-31 09:30:58,512 BAD EPOCHS (no improvement): 1
594
+ 2021-12-31 09:30:58,514 ----------------------------------------------------------------------------------------------------
595
+ 2021-12-31 09:31:14,847 epoch 17 - iter 180/1807 - loss 0.08390522 - samples/sec: 88.30 - lr: 0.100000
596
+ 2021-12-31 09:31:30,522 epoch 17 - iter 360/1807 - loss 0.08649584 - samples/sec: 92.01 - lr: 0.100000
597
+ 2021-12-31 09:31:46,288 epoch 17 - iter 540/1807 - loss 0.08940335 - samples/sec: 91.48 - lr: 0.100000
598
+ 2021-12-31 09:32:02,118 epoch 17 - iter 720/1807 - loss 0.09059873 - samples/sec: 91.09 - lr: 0.100000
599
+ 2021-12-31 09:32:17,806 epoch 17 - iter 900/1807 - loss 0.09026440 - samples/sec: 91.93 - lr: 0.100000
600
+ 2021-12-31 09:32:33,488 epoch 17 - iter 1080/1807 - loss 0.09038711 - samples/sec: 91.96 - lr: 0.100000
601
+ 2021-12-31 09:32:49,442 epoch 17 - iter 1260/1807 - loss 0.08978670 - samples/sec: 90.39 - lr: 0.100000
602
+ 2021-12-31 09:33:05,170 epoch 17 - iter 1440/1807 - loss 0.08929018 - samples/sec: 91.69 - lr: 0.100000
603
+ 2021-12-31 09:33:21,122 epoch 17 - iter 1620/1807 - loss 0.08920206 - samples/sec: 90.40 - lr: 0.100000
604
+ 2021-12-31 09:33:36,598 epoch 17 - iter 1800/1807 - loss 0.08958801 - samples/sec: 93.18 - lr: 0.100000
605
+ 2021-12-31 09:33:37,149 ----------------------------------------------------------------------------------------------------
606
+ 2021-12-31 09:33:37,149 EPOCH 17 done: loss 0.0895 - lr 0.1000000
607
+ 2021-12-31 09:34:16,446 DEV : loss 0.06361010670661926 - f1-score (micro avg) 0.9823
608
+ 2021-12-31 09:34:16,599 BAD EPOCHS (no improvement): 0
609
+ 2021-12-31 09:34:16,600 saving best model
610
+ 2021-12-31 09:34:22,434 ----------------------------------------------------------------------------------------------------
611
+ 2021-12-31 09:34:38,419 epoch 18 - iter 180/1807 - loss 0.08343062 - samples/sec: 90.22 - lr: 0.100000
612
+ 2021-12-31 09:34:54,655 epoch 18 - iter 360/1807 - loss 0.08575852 - samples/sec: 88.82 - lr: 0.100000
613
+ 2021-12-31 09:35:10,385 epoch 18 - iter 540/1807 - loss 0.08392644 - samples/sec: 91.68 - lr: 0.100000
614
+ 2021-12-31 09:35:26,310 epoch 18 - iter 720/1807 - loss 0.08351999 - samples/sec: 90.57 - lr: 0.100000
615
+ 2021-12-31 09:35:41,876 epoch 18 - iter 900/1807 - loss 0.08509375 - samples/sec: 92.64 - lr: 0.100000
616
+ 2021-12-31 09:35:57,882 epoch 18 - iter 1080/1807 - loss 0.08493115 - samples/sec: 90.10 - lr: 0.100000
617
+ 2021-12-31 09:36:13,926 epoch 18 - iter 1260/1807 - loss 0.08609299 - samples/sec: 89.88 - lr: 0.100000
618
+ 2021-12-31 09:36:30,070 epoch 18 - iter 1440/1807 - loss 0.08644835 - samples/sec: 89.34 - lr: 0.100000
619
+ 2021-12-31 09:36:45,689 epoch 18 - iter 1620/1807 - loss 0.08698449 - samples/sec: 92.33 - lr: 0.100000
620
+ 2021-12-31 09:37:01,595 epoch 18 - iter 1800/1807 - loss 0.08715385 - samples/sec: 90.66 - lr: 0.100000
621
+ 2021-12-31 09:37:02,116 ----------------------------------------------------------------------------------------------------
622
+ 2021-12-31 09:37:02,116 EPOCH 18 done: loss 0.0872 - lr 0.1000000
623
+ 2021-12-31 09:37:38,287 DEV : loss 0.06376409530639648 - f1-score (micro avg) 0.982
624
+ 2021-12-31 09:37:38,491 BAD EPOCHS (no improvement): 1
625
+ 2021-12-31 09:37:38,492 ----------------------------------------------------------------------------------------------------
626
+ 2021-12-31 09:37:54,464 epoch 19 - iter 180/1807 - loss 0.07802257 - samples/sec: 90.31 - lr: 0.100000
627
+ 2021-12-31 09:38:10,256 epoch 19 - iter 360/1807 - loss 0.07892620 - samples/sec: 91.32 - lr: 0.100000
628
+ 2021-12-31 09:38:26,632 epoch 19 - iter 540/1807 - loss 0.08133170 - samples/sec: 88.06 - lr: 0.100000
629
+ 2021-12-31 09:38:42,673 epoch 19 - iter 720/1807 - loss 0.08367885 - samples/sec: 89.91 - lr: 0.100000
630
+ 2021-12-31 09:38:58,503 epoch 19 - iter 900/1807 - loss 0.08447871 - samples/sec: 91.11 - lr: 0.100000
631
+ 2021-12-31 09:39:14,461 epoch 19 - iter 1080/1807 - loss 0.08413767 - samples/sec: 90.37 - lr: 0.100000
632
+ 2021-12-31 09:39:30,176 epoch 19 - iter 1260/1807 - loss 0.08455665 - samples/sec: 91.77 - lr: 0.100000
633
+ 2021-12-31 09:39:46,325 epoch 19 - iter 1440/1807 - loss 0.08578599 - samples/sec: 89.30 - lr: 0.100000
634
+ 2021-12-31 09:40:02,191 epoch 19 - iter 1620/1807 - loss 0.08628902 - samples/sec: 90.90 - lr: 0.100000
635
+ 2021-12-31 09:40:18,069 epoch 19 - iter 1800/1807 - loss 0.08634962 - samples/sec: 90.82 - lr: 0.100000
636
+ 2021-12-31 09:40:18,635 ----------------------------------------------------------------------------------------------------
637
+ 2021-12-31 09:40:18,636 EPOCH 19 done: loss 0.0863 - lr 0.1000000
638
+ 2021-12-31 09:40:54,638 DEV : loss 0.06360483914613724 - f1-score (micro avg) 0.9824
639
+ 2021-12-31 09:40:54,809 BAD EPOCHS (no improvement): 0
640
+ 2021-12-31 09:40:54,812 saving best model
641
+ 2021-12-31 09:41:00,532 ----------------------------------------------------------------------------------------------------
642
+ 2021-12-31 09:41:16,605 epoch 20 - iter 180/1807 - loss 0.08580796 - samples/sec: 89.75 - lr: 0.100000
643
+ 2021-12-31 09:41:32,626 epoch 20 - iter 360/1807 - loss 0.08441046 - samples/sec: 90.02 - lr: 0.100000
644
+ 2021-12-31 09:41:48,195 epoch 20 - iter 540/1807 - loss 0.08457436 - samples/sec: 92.63 - lr: 0.100000
645
+ 2021-12-31 09:42:03,884 epoch 20 - iter 720/1807 - loss 0.08433505 - samples/sec: 91.92 - lr: 0.100000
646
+ 2021-12-31 09:42:19,662 epoch 20 - iter 900/1807 - loss 0.08465375 - samples/sec: 91.40 - lr: 0.100000
647
+ 2021-12-31 09:42:35,290 epoch 20 - iter 1080/1807 - loss 0.08384813 - samples/sec: 92.28 - lr: 0.100000
648
+ 2021-12-31 09:42:50,667 epoch 20 - iter 1260/1807 - loss 0.08437448 - samples/sec: 93.79 - lr: 0.100000
649
+ 2021-12-31 09:43:06,838 epoch 20 - iter 1440/1807 - loss 0.08483000 - samples/sec: 89.18 - lr: 0.100000
650
+ 2021-12-31 09:43:23,128 epoch 20 - iter 1620/1807 - loss 0.08554680 - samples/sec: 88.52 - lr: 0.100000
651
+ 2021-12-31 09:43:38,996 epoch 20 - iter 1800/1807 - loss 0.08579345 - samples/sec: 90.89 - lr: 0.100000
652
+ 2021-12-31 09:43:39,520 ----------------------------------------------------------------------------------------------------
653
+ 2021-12-31 09:43:39,520 EPOCH 20 done: loss 0.0858 - lr 0.1000000
654
+ 2021-12-31 09:44:18,433 DEV : loss 0.06494450569152832 - f1-score (micro avg) 0.982
655
+ 2021-12-31 09:44:18,588 BAD EPOCHS (no improvement): 1
656
+ 2021-12-31 09:44:18,590 ----------------------------------------------------------------------------------------------------
657
+ 2021-12-31 09:44:34,495 epoch 21 - iter 180/1807 - loss 0.08058450 - samples/sec: 90.65 - lr: 0.100000
658
+ 2021-12-31 09:44:50,061 epoch 21 - iter 360/1807 - loss 0.08169987 - samples/sec: 92.62 - lr: 0.100000
659
+ 2021-12-31 09:45:05,780 epoch 21 - iter 540/1807 - loss 0.08147401 - samples/sec: 91.76 - lr: 0.100000
660
+ 2021-12-31 09:45:21,869 epoch 21 - iter 720/1807 - loss 0.08235327 - samples/sec: 89.64 - lr: 0.100000
661
+ 2021-12-31 09:45:38,316 epoch 21 - iter 900/1807 - loss 0.08324710 - samples/sec: 87.67 - lr: 0.100000
662
+ 2021-12-31 09:45:54,314 epoch 21 - iter 1080/1807 - loss 0.08294963 - samples/sec: 90.14 - lr: 0.100000
663
+ 2021-12-31 09:46:10,369 epoch 21 - iter 1260/1807 - loss 0.08355307 - samples/sec: 89.83 - lr: 0.100000
664
+ 2021-12-31 09:46:26,469 epoch 21 - iter 1440/1807 - loss 0.08343050 - samples/sec: 89.57 - lr: 0.100000
665
+ 2021-12-31 09:46:42,401 epoch 21 - iter 1620/1807 - loss 0.08414815 - samples/sec: 90.52 - lr: 0.100000
666
+ 2021-12-31 09:46:58,257 epoch 21 - iter 1800/1807 - loss 0.08376554 - samples/sec: 90.95 - lr: 0.100000
667
+ 2021-12-31 09:46:58,880 ----------------------------------------------------------------------------------------------------
668
+ 2021-12-31 09:46:58,880 EPOCH 21 done: loss 0.0839 - lr 0.1000000
669
+ 2021-12-31 09:47:35,248 DEV : loss 0.06328344345092773 - f1-score (micro avg) 0.9827
670
+ 2021-12-31 09:47:35,446 BAD EPOCHS (no improvement): 0
671
+ 2021-12-31 09:47:35,448 saving best model
672
+ 2021-12-31 09:47:41,248 ----------------------------------------------------------------------------------------------------
673
+ 2021-12-31 09:47:57,255 epoch 22 - iter 180/1807 - loss 0.08050373 - samples/sec: 90.12 - lr: 0.100000
674
+ 2021-12-31 09:48:13,186 epoch 22 - iter 360/1807 - loss 0.08239139 - samples/sec: 90.52 - lr: 0.100000
675
+ 2021-12-31 09:48:29,067 epoch 22 - iter 540/1807 - loss 0.08228212 - samples/sec: 90.81 - lr: 0.100000
676
+ 2021-12-31 09:48:45,039 epoch 22 - iter 720/1807 - loss 0.08279713 - samples/sec: 90.30 - lr: 0.100000
677
+ 2021-12-31 09:49:00,510 epoch 22 - iter 900/1807 - loss 0.08334789 - samples/sec: 93.22 - lr: 0.100000
678
+ 2021-12-31 09:49:16,362 epoch 22 - iter 1080/1807 - loss 0.08342389 - samples/sec: 90.97 - lr: 0.100000
679
+ 2021-12-31 09:49:32,567 epoch 22 - iter 1260/1807 - loss 0.08349166 - samples/sec: 88.99 - lr: 0.100000
680
+ 2021-12-31 09:49:48,320 epoch 22 - iter 1440/1807 - loss 0.08427908 - samples/sec: 91.55 - lr: 0.100000
681
+ 2021-12-31 09:50:04,570 epoch 22 - iter 1620/1807 - loss 0.08465300 - samples/sec: 88.75 - lr: 0.100000
682
+ 2021-12-31 09:50:20,943 epoch 22 - iter 1800/1807 - loss 0.08437528 - samples/sec: 88.07 - lr: 0.100000
683
+ 2021-12-31 09:50:21,480 ----------------------------------------------------------------------------------------------------
684
+ 2021-12-31 09:50:21,480 EPOCH 22 done: loss 0.0844 - lr 0.1000000
685
+ 2021-12-31 09:50:58,771 DEV : loss 0.06346500664949417 - f1-score (micro avg) 0.9815
686
+ 2021-12-31 09:50:58,967 BAD EPOCHS (no improvement): 1
687
+ 2021-12-31 09:50:58,969 ----------------------------------------------------------------------------------------------------
688
+ 2021-12-31 09:51:15,272 epoch 23 - iter 180/1807 - loss 0.07857499 - samples/sec: 88.47 - lr: 0.100000
689
+ 2021-12-31 09:51:31,123 epoch 23 - iter 360/1807 - loss 0.07736816 - samples/sec: 91.00 - lr: 0.100000
690
+ 2021-12-31 09:51:47,441 epoch 23 - iter 540/1807 - loss 0.07865886 - samples/sec: 88.38 - lr: 0.100000
691
+ 2021-12-31 09:52:03,508 epoch 23 - iter 720/1807 - loss 0.08053686 - samples/sec: 89.75 - lr: 0.100000
692
+ 2021-12-31 09:52:19,618 epoch 23 - iter 900/1807 - loss 0.08084826 - samples/sec: 89.52 - lr: 0.100000
693
+ 2021-12-31 09:52:35,467 epoch 23 - iter 1080/1807 - loss 0.08116025 - samples/sec: 91.00 - lr: 0.100000
694
+ 2021-12-31 09:52:51,307 epoch 23 - iter 1260/1807 - loss 0.08137722 - samples/sec: 91.04 - lr: 0.100000
695
+ 2021-12-31 09:53:07,605 epoch 23 - iter 1440/1807 - loss 0.08168418 - samples/sec: 88.48 - lr: 0.100000
696
+ 2021-12-31 09:53:23,242 epoch 23 - iter 1620/1807 - loss 0.08161521 - samples/sec: 92.22 - lr: 0.100000
697
+ 2021-12-31 09:53:38,917 epoch 23 - iter 1800/1807 - loss 0.08147531 - samples/sec: 92.01 - lr: 0.100000
698
+ 2021-12-31 09:53:39,396 ----------------------------------------------------------------------------------------------------
699
+ 2021-12-31 09:53:39,396 EPOCH 23 done: loss 0.0814 - lr 0.1000000
700
+ 2021-12-31 09:54:15,841 DEV : loss 0.06540019810199738 - f1-score (micro avg) 0.9821
701
+ 2021-12-31 09:54:16,023 BAD EPOCHS (no improvement): 2
702
+ 2021-12-31 09:54:16,025 ----------------------------------------------------------------------------------------------------
703
+ 2021-12-31 09:54:32,334 epoch 24 - iter 180/1807 - loss 0.07795468 - samples/sec: 88.43 - lr: 0.100000
704
+ 2021-12-31 09:54:48,084 epoch 24 - iter 360/1807 - loss 0.07908717 - samples/sec: 91.57 - lr: 0.100000
705
+ 2021-12-31 09:55:04,326 epoch 24 - iter 540/1807 - loss 0.08004992 - samples/sec: 88.79 - lr: 0.100000
706
+ 2021-12-31 09:55:20,651 epoch 24 - iter 720/1807 - loss 0.08100541 - samples/sec: 88.34 - lr: 0.100000
707
+ 2021-12-31 09:55:36,785 epoch 24 - iter 900/1807 - loss 0.08142507 - samples/sec: 89.38 - lr: 0.100000
708
+ 2021-12-31 09:55:52,742 epoch 24 - iter 1080/1807 - loss 0.08232817 - samples/sec: 90.38 - lr: 0.100000
709
+ 2021-12-31 09:56:08,164 epoch 24 - iter 1260/1807 - loss 0.08188184 - samples/sec: 93.53 - lr: 0.100000
710
+ 2021-12-31 09:56:24,063 epoch 24 - iter 1440/1807 - loss 0.08243719 - samples/sec: 90.71 - lr: 0.100000
711
+ 2021-12-31 09:56:40,384 epoch 24 - iter 1620/1807 - loss 0.08222346 - samples/sec: 88.35 - lr: 0.100000
712
+ 2021-12-31 09:56:56,011 epoch 24 - iter 1800/1807 - loss 0.08229498 - samples/sec: 92.29 - lr: 0.100000
713
+ 2021-12-31 09:56:56,616 ----------------------------------------------------------------------------------------------------
714
+ 2021-12-31 09:56:56,616 EPOCH 24 done: loss 0.0822 - lr 0.1000000
715
+ 2021-12-31 09:57:35,721 DEV : loss 0.06453310698270798 - f1-score (micro avg) 0.9819
716
+ 2021-12-31 09:57:35,917 BAD EPOCHS (no improvement): 3
717
+ 2021-12-31 09:57:35,919 ----------------------------------------------------------------------------------------------------
718
+ 2021-12-31 09:57:52,048 epoch 25 - iter 180/1807 - loss 0.07765362 - samples/sec: 89.42 - lr: 0.100000
719
+ 2021-12-31 09:58:07,956 epoch 25 - iter 360/1807 - loss 0.07932940 - samples/sec: 90.65 - lr: 0.100000
720
+ 2021-12-31 09:58:23,863 epoch 25 - iter 540/1807 - loss 0.08046614 - samples/sec: 90.65 - lr: 0.100000
721
+ 2021-12-31 09:58:39,725 epoch 25 - iter 720/1807 - loss 0.07941669 - samples/sec: 90.92 - lr: 0.100000
722
+ 2021-12-31 09:58:55,303 epoch 25 - iter 900/1807 - loss 0.08092722 - samples/sec: 92.57 - lr: 0.100000
723
+ 2021-12-31 09:59:11,794 epoch 25 - iter 1080/1807 - loss 0.08150485 - samples/sec: 87.44 - lr: 0.100000
724
+ 2021-12-31 09:59:27,795 epoch 25 - iter 1260/1807 - loss 0.08118184 - samples/sec: 90.13 - lr: 0.100000
725
+ 2021-12-31 09:59:43,595 epoch 25 - iter 1440/1807 - loss 0.08068256 - samples/sec: 91.28 - lr: 0.100000
726
+ 2021-12-31 09:59:59,146 epoch 25 - iter 1620/1807 - loss 0.08113371 - samples/sec: 92.74 - lr: 0.100000
727
+ 2021-12-31 10:00:14,684 epoch 25 - iter 1800/1807 - loss 0.08112289 - samples/sec: 92.81 - lr: 0.100000
728
+ 2021-12-31 10:00:15,230 ----------------------------------------------------------------------------------------------------
729
+ 2021-12-31 10:00:15,230 EPOCH 25 done: loss 0.0812 - lr 0.1000000
730
+ 2021-12-31 10:00:51,681 DEV : loss 0.06579063087701797 - f1-score (micro avg) 0.9817
731
+ 2021-12-31 10:00:51,872 BAD EPOCHS (no improvement): 4
732
+ 2021-12-31 10:00:51,874 ----------------------------------------------------------------------------------------------------
733
+ 2021-12-31 10:01:08,252 epoch 26 - iter 180/1807 - loss 0.07473820 - samples/sec: 88.06 - lr: 0.050000
734
+ 2021-12-31 10:01:24,095 epoch 26 - iter 360/1807 - loss 0.07741051 - samples/sec: 91.03 - lr: 0.050000
735
+ 2021-12-31 10:01:40,042 epoch 26 - iter 540/1807 - loss 0.07612793 - samples/sec: 90.43 - lr: 0.050000
736
+ 2021-12-31 10:01:55,977 epoch 26 - iter 720/1807 - loss 0.07597233 - samples/sec: 90.49 - lr: 0.050000
737
+ 2021-12-31 10:02:12,264 epoch 26 - iter 900/1807 - loss 0.07560347 - samples/sec: 88.55 - lr: 0.050000
738
+ 2021-12-31 10:02:28,030 epoch 26 - iter 1080/1807 - loss 0.07626889 - samples/sec: 91.47 - lr: 0.050000
739
+ 2021-12-31 10:02:43,691 epoch 26 - iter 1260/1807 - loss 0.07613186 - samples/sec: 92.08 - lr: 0.050000
740
+ 2021-12-31 10:02:59,223 epoch 26 - iter 1440/1807 - loss 0.07558384 - samples/sec: 92.85 - lr: 0.050000
741
+ 2021-12-31 10:03:15,259 epoch 26 - iter 1620/1807 - loss 0.07503334 - samples/sec: 89.93 - lr: 0.050000
742
+ 2021-12-31 10:03:31,614 epoch 26 - iter 1800/1807 - loss 0.07448614 - samples/sec: 88.18 - lr: 0.050000
743
+ 2021-12-31 10:03:32,151 ----------------------------------------------------------------------------------------------------
744
+ 2021-12-31 10:03:32,151 EPOCH 26 done: loss 0.0744 - lr 0.0500000
745
+ 2021-12-31 10:04:08,767 DEV : loss 0.06646668165922165 - f1-score (micro avg) 0.9822
746
+ 2021-12-31 10:04:08,949 BAD EPOCHS (no improvement): 1
747
+ 2021-12-31 10:04:08,950 ----------------------------------------------------------------------------------------------------
748
+ 2021-12-31 10:04:25,529 epoch 27 - iter 180/1807 - loss 0.06581114 - samples/sec: 86.99 - lr: 0.050000
749
+ 2021-12-31 10:04:41,436 epoch 27 - iter 360/1807 - loss 0.06857834 - samples/sec: 90.66 - lr: 0.050000
750
+ 2021-12-31 10:04:57,191 epoch 27 - iter 540/1807 - loss 0.07081005 - samples/sec: 91.54 - lr: 0.050000
751
+ 2021-12-31 10:05:13,183 epoch 27 - iter 720/1807 - loss 0.07198836 - samples/sec: 90.18 - lr: 0.050000
752
+ 2021-12-31 10:05:29,131 epoch 27 - iter 900/1807 - loss 0.07153264 - samples/sec: 90.42 - lr: 0.050000
753
+ 2021-12-31 10:05:44,864 epoch 27 - iter 1080/1807 - loss 0.07164274 - samples/sec: 91.66 - lr: 0.050000
754
+ 2021-12-31 10:06:00,643 epoch 27 - iter 1260/1807 - loss 0.07167991 - samples/sec: 91.40 - lr: 0.050000
755
+ 2021-12-31 10:06:15,929 epoch 27 - iter 1440/1807 - loss 0.07130117 - samples/sec: 94.34 - lr: 0.050000
756
+ 2021-12-31 10:06:32,208 epoch 27 - iter 1620/1807 - loss 0.07137995 - samples/sec: 88.59 - lr: 0.050000
757
+ 2021-12-31 10:06:48,072 epoch 27 - iter 1800/1807 - loss 0.07123898 - samples/sec: 90.90 - lr: 0.050000
758
+ 2021-12-31 10:06:48,616 ----------------------------------------------------------------------------------------------------
759
+ 2021-12-31 10:06:48,616 EPOCH 27 done: loss 0.0712 - lr 0.0500000
760
+ 2021-12-31 10:07:27,769 DEV : loss 0.06514652073383331 - f1-score (micro avg) 0.9823
761
+ 2021-12-31 10:07:27,967 BAD EPOCHS (no improvement): 2
762
+ 2021-12-31 10:07:27,968 ----------------------------------------------------------------------------------------------------
763
+ 2021-12-31 10:07:43,921 epoch 28 - iter 180/1807 - loss 0.06865415 - samples/sec: 90.41 - lr: 0.050000
764
+ 2021-12-31 10:08:00,073 epoch 28 - iter 360/1807 - loss 0.06855531 - samples/sec: 89.28 - lr: 0.050000
765
+ 2021-12-31 10:08:16,259 epoch 28 - iter 540/1807 - loss 0.06891820 - samples/sec: 89.09 - lr: 0.050000
766
+ 2021-12-31 10:08:31,981 epoch 28 - iter 720/1807 - loss 0.06951336 - samples/sec: 91.73 - lr: 0.050000
767
+ 2021-12-31 10:08:47,429 epoch 28 - iter 900/1807 - loss 0.07014278 - samples/sec: 93.35 - lr: 0.050000
768
+ 2021-12-31 10:09:03,024 epoch 28 - iter 1080/1807 - loss 0.07071541 - samples/sec: 92.47 - lr: 0.050000
769
+ 2021-12-31 10:09:18,974 epoch 28 - iter 1260/1807 - loss 0.07012373 - samples/sec: 90.41 - lr: 0.050000
770
+ 2021-12-31 10:09:34,620 epoch 28 - iter 1440/1807 - loss 0.07028479 - samples/sec: 92.17 - lr: 0.050000
771
+ 2021-12-31 10:09:50,427 epoch 28 - iter 1620/1807 - loss 0.07017402 - samples/sec: 91.23 - lr: 0.050000
772
+ 2021-12-31 10:10:05,997 epoch 28 - iter 1800/1807 - loss 0.07002142 - samples/sec: 92.62 - lr: 0.050000
773
+ 2021-12-31 10:10:06,547 ----------------------------------------------------------------------------------------------------
774
+ 2021-12-31 10:10:06,548 EPOCH 28 done: loss 0.0701 - lr 0.0500000
775
+ 2021-12-31 10:10:43,342 DEV : loss 0.06285692006349564 - f1-score (micro avg) 0.9828
776
+ 2021-12-31 10:10:43,549 BAD EPOCHS (no improvement): 0
777
+ 2021-12-31 10:10:43,550 saving best model
778
+ 2021-12-31 10:10:49,346 ----------------------------------------------------------------------------------------------------
779
+ 2021-12-31 10:11:05,893 epoch 29 - iter 180/1807 - loss 0.06749112 - samples/sec: 87.17 - lr: 0.050000
780
+ 2021-12-31 10:11:21,660 epoch 29 - iter 360/1807 - loss 0.06704871 - samples/sec: 91.46 - lr: 0.050000
781
+ 2021-12-31 10:11:37,404 epoch 29 - iter 540/1807 - loss 0.06846136 - samples/sec: 91.60 - lr: 0.050000
782
+ 2021-12-31 10:11:53,397 epoch 29 - iter 720/1807 - loss 0.06901632 - samples/sec: 90.17 - lr: 0.050000
783
+ 2021-12-31 10:12:09,257 epoch 29 - iter 900/1807 - loss 0.06809349 - samples/sec: 90.93 - lr: 0.050000
784
+ 2021-12-31 10:12:24,599 epoch 29 - iter 1080/1807 - loss 0.06824897 - samples/sec: 94.00 - lr: 0.050000
785
+ 2021-12-31 10:12:40,447 epoch 29 - iter 1260/1807 - loss 0.06782382 - samples/sec: 91.00 - lr: 0.050000
786
+ 2021-12-31 10:12:56,595 epoch 29 - iter 1440/1807 - loss 0.06808796 - samples/sec: 89.30 - lr: 0.050000
787
+ 2021-12-31 10:13:12,755 epoch 29 - iter 1620/1807 - loss 0.06798634 - samples/sec: 89.24 - lr: 0.050000
788
+ 2021-12-31 10:13:28,701 epoch 29 - iter 1800/1807 - loss 0.06777472 - samples/sec: 90.44 - lr: 0.050000
789
+ 2021-12-31 10:13:29,227 ----------------------------------------------------------------------------------------------------
790
+ 2021-12-31 10:13:29,228 EPOCH 29 done: loss 0.0678 - lr 0.0500000
791
+ 2021-12-31 10:14:05,041 DEV : loss 0.06288447976112366 - f1-score (micro avg) 0.9831
792
+ 2021-12-31 10:14:05,221 BAD EPOCHS (no improvement): 0
793
+ 2021-12-31 10:14:05,222 saving best model
794
+ 2021-12-31 10:14:10,675 ----------------------------------------------------------------------------------------------------
795
+ 2021-12-31 10:14:26,845 epoch 30 - iter 180/1807 - loss 0.06615046 - samples/sec: 89.20 - lr: 0.050000
796
+ 2021-12-31 10:14:42,781 epoch 30 - iter 360/1807 - loss 0.06701908 - samples/sec: 90.50 - lr: 0.050000
797
+ 2021-12-31 10:14:58,746 epoch 30 - iter 540/1807 - loss 0.06748578 - samples/sec: 90.33 - lr: 0.050000
798
+ 2021-12-31 10:15:14,479 epoch 30 - iter 720/1807 - loss 0.06796474 - samples/sec: 91.66 - lr: 0.050000
799
+ 2021-12-31 10:15:30,280 epoch 30 - iter 900/1807 - loss 0.06739311 - samples/sec: 91.26 - lr: 0.050000
800
+ 2021-12-31 10:15:45,933 epoch 30 - iter 1080/1807 - loss 0.06699810 - samples/sec: 92.13 - lr: 0.050000
801
+ 2021-12-31 10:16:01,690 epoch 30 - iter 1260/1807 - loss 0.06745951 - samples/sec: 91.53 - lr: 0.050000
802
+ 2021-12-31 10:16:17,453 epoch 30 - iter 1440/1807 - loss 0.06704309 - samples/sec: 91.49 - lr: 0.050000
803
+ 2021-12-31 10:16:33,233 epoch 30 - iter 1620/1807 - loss 0.06649743 - samples/sec: 91.38 - lr: 0.050000
804
+ 2021-12-31 10:16:49,143 epoch 30 - iter 1800/1807 - loss 0.06655280 - samples/sec: 90.65 - lr: 0.050000
805
+ 2021-12-31 10:16:49,685 ----------------------------------------------------------------------------------------------------
806
+ 2021-12-31 10:16:49,685 EPOCH 30 done: loss 0.0666 - lr 0.0500000
807
+ 2021-12-31 10:17:28,240 DEV : loss 0.06311798095703125 - f1-score (micro avg) 0.9824
808
+ 2021-12-31 10:17:28,433 BAD EPOCHS (no improvement): 1
809
+ 2021-12-31 10:17:28,434 ----------------------------------------------------------------------------------------------------
810
+ 2021-12-31 10:17:44,966 epoch 31 - iter 180/1807 - loss 0.06627745 - samples/sec: 87.24 - lr: 0.050000
811
+ 2021-12-31 10:18:00,662 epoch 31 - iter 360/1807 - loss 0.06286711 - samples/sec: 91.88 - lr: 0.050000
812
+ 2021-12-31 10:18:16,307 epoch 31 - iter 540/1807 - loss 0.06454841 - samples/sec: 92.17 - lr: 0.050000
813
+ 2021-12-31 10:18:32,243 epoch 31 - iter 720/1807 - loss 0.06465161 - samples/sec: 90.50 - lr: 0.050000
814
+ 2021-12-31 10:18:47,799 epoch 31 - iter 900/1807 - loss 0.06488043 - samples/sec: 92.70 - lr: 0.050000
815
+ 2021-12-31 10:19:03,602 epoch 31 - iter 1080/1807 - loss 0.06501278 - samples/sec: 91.26 - lr: 0.050000
816
+ 2021-12-31 10:19:19,610 epoch 31 - iter 1260/1807 - loss 0.06524649 - samples/sec: 90.08 - lr: 0.050000
817
+ 2021-12-31 10:19:35,038 epoch 31 - iter 1440/1807 - loss 0.06554492 - samples/sec: 93.48 - lr: 0.050000
818
+ 2021-12-31 10:19:51,164 epoch 31 - iter 1620/1807 - loss 0.06599922 - samples/sec: 89.43 - lr: 0.050000
819
+ 2021-12-31 10:20:07,078 epoch 31 - iter 1800/1807 - loss 0.06644678 - samples/sec: 90.61 - lr: 0.050000
820
+ 2021-12-31 10:20:07,640 ----------------------------------------------------------------------------------------------------
821
+ 2021-12-31 10:20:07,640 EPOCH 31 done: loss 0.0666 - lr 0.0500000
822
+ 2021-12-31 10:20:43,927 DEV : loss 0.06285466253757477 - f1-score (micro avg) 0.9829
823
+ 2021-12-31 10:20:44,123 BAD EPOCHS (no improvement): 2
824
+ 2021-12-31 10:20:44,125 ----------------------------------------------------------------------------------------------------
825
+ 2021-12-31 10:21:00,298 epoch 32 - iter 180/1807 - loss 0.06077116 - samples/sec: 89.18 - lr: 0.050000
826
+ 2021-12-31 10:21:16,393 epoch 32 - iter 360/1807 - loss 0.06270324 - samples/sec: 89.60 - lr: 0.050000
827
+ 2021-12-31 10:21:32,158 epoch 32 - iter 540/1807 - loss 0.06340224 - samples/sec: 91.47 - lr: 0.050000
828
+ 2021-12-31 10:21:48,183 epoch 32 - iter 720/1807 - loss 0.06267842 - samples/sec: 89.99 - lr: 0.050000
829
+ 2021-12-31 10:22:03,949 epoch 32 - iter 900/1807 - loss 0.06345792 - samples/sec: 91.50 - lr: 0.050000
830
+ 2021-12-31 10:22:19,674 epoch 32 - iter 1080/1807 - loss 0.06439376 - samples/sec: 91.71 - lr: 0.050000
831
+ 2021-12-31 10:22:35,414 epoch 32 - iter 1260/1807 - loss 0.06437464 - samples/sec: 91.63 - lr: 0.050000
832
+ 2021-12-31 10:22:51,702 epoch 32 - iter 1440/1807 - loss 0.06435182 - samples/sec: 88.53 - lr: 0.050000
833
+ 2021-12-31 10:23:07,918 epoch 32 - iter 1620/1807 - loss 0.06467809 - samples/sec: 88.93 - lr: 0.050000
834
+ 2021-12-31 10:23:23,880 epoch 32 - iter 1800/1807 - loss 0.06484923 - samples/sec: 90.35 - lr: 0.050000
835
+ 2021-12-31 10:23:24,513 ----------------------------------------------------------------------------------------------------
836
+ 2021-12-31 10:23:24,513 EPOCH 32 done: loss 0.0648 - lr 0.0500000
837
+ 2021-12-31 10:24:00,678 DEV : loss 0.062373436987400055 - f1-score (micro avg) 0.9827
838
+ 2021-12-31 10:24:00,863 BAD EPOCHS (no improvement): 3
839
+ 2021-12-31 10:24:00,865 ----------------------------------------------------------------------------------------------------
840
+ 2021-12-31 10:24:17,368 epoch 33 - iter 180/1807 - loss 0.06511517 - samples/sec: 87.39 - lr: 0.050000
841
+ 2021-12-31 10:24:33,869 epoch 33 - iter 360/1807 - loss 0.06359714 - samples/sec: 87.39 - lr: 0.050000
842
+ 2021-12-31 10:24:49,974 epoch 33 - iter 540/1807 - loss 0.06324776 - samples/sec: 89.54 - lr: 0.050000
843
+ 2021-12-31 10:25:05,411 epoch 33 - iter 720/1807 - loss 0.06296883 - samples/sec: 93.42 - lr: 0.050000
844
+ 2021-12-31 10:25:21,477 epoch 33 - iter 900/1807 - loss 0.06304943 - samples/sec: 89.76 - lr: 0.050000
845
+ 2021-12-31 10:25:37,062 epoch 33 - iter 1080/1807 - loss 0.06266940 - samples/sec: 92.52 - lr: 0.050000
846
+ 2021-12-31 10:25:52,743 epoch 33 - iter 1260/1807 - loss 0.06359599 - samples/sec: 91.97 - lr: 0.050000
847
+ 2021-12-31 10:26:08,521 epoch 33 - iter 1440/1807 - loss 0.06353058 - samples/sec: 91.40 - lr: 0.050000
848
+ 2021-12-31 10:26:24,080 epoch 33 - iter 1620/1807 - loss 0.06366170 - samples/sec: 92.69 - lr: 0.050000
849
+ 2021-12-31 10:26:39,568 epoch 33 - iter 1800/1807 - loss 0.06405823 - samples/sec: 93.11 - lr: 0.050000
850
+ 2021-12-31 10:26:40,121 ----------------------------------------------------------------------------------------------------
851
+ 2021-12-31 10:26:40,121 EPOCH 33 done: loss 0.0640 - lr 0.0500000
852
+ 2021-12-31 10:27:18,678 DEV : loss 0.06352584064006805 - f1-score (micro avg) 0.983
853
+ 2021-12-31 10:27:18,875 BAD EPOCHS (no improvement): 4
854
+ 2021-12-31 10:27:18,877 ----------------------------------------------------------------------------------------------------
855
+ 2021-12-31 10:27:34,632 epoch 34 - iter 180/1807 - loss 0.05738992 - samples/sec: 91.55 - lr: 0.025000
856
+ 2021-12-31 10:27:50,783 epoch 34 - iter 360/1807 - loss 0.05964139 - samples/sec: 89.29 - lr: 0.025000
857
+ 2021-12-31 10:28:06,956 epoch 34 - iter 540/1807 - loss 0.05950577 - samples/sec: 89.16 - lr: 0.025000
858
+ 2021-12-31 10:28:23,264 epoch 34 - iter 720/1807 - loss 0.06033373 - samples/sec: 88.43 - lr: 0.025000
859
+ 2021-12-31 10:28:38,762 epoch 34 - iter 900/1807 - loss 0.06053852 - samples/sec: 93.06 - lr: 0.025000
860
+ 2021-12-31 10:28:54,790 epoch 34 - iter 1080/1807 - loss 0.06008683 - samples/sec: 89.97 - lr: 0.025000
861
+ 2021-12-31 10:29:10,752 epoch 34 - iter 1260/1807 - loss 0.06017032 - samples/sec: 90.34 - lr: 0.025000
862
+ 2021-12-31 10:29:26,533 epoch 34 - iter 1440/1807 - loss 0.06026720 - samples/sec: 91.39 - lr: 0.025000
863
+ 2021-12-31 10:29:41,962 epoch 34 - iter 1620/1807 - loss 0.06023939 - samples/sec: 93.47 - lr: 0.025000
864
+ 2021-12-31 10:29:57,974 epoch 34 - iter 1800/1807 - loss 0.06024915 - samples/sec: 90.06 - lr: 0.025000
865
+ 2021-12-31 10:29:58,641 ----------------------------------------------------------------------------------------------------
866
+ 2021-12-31 10:29:58,642 EPOCH 34 done: loss 0.0602 - lr 0.0250000
867
+ 2021-12-31 10:30:34,901 DEV : loss 0.06348917633295059 - f1-score (micro avg) 0.9835
868
+ 2021-12-31 10:30:35,087 BAD EPOCHS (no improvement): 0
869
+ 2021-12-31 10:30:35,089 saving best model
870
+ 2021-12-31 10:30:40,883 ----------------------------------------------------------------------------------------------------
871
+ 2021-12-31 10:30:57,202 epoch 35 - iter 180/1807 - loss 0.05878333 - samples/sec: 88.38 - lr: 0.025000
872
+ 2021-12-31 10:31:12,996 epoch 35 - iter 360/1807 - loss 0.05795906 - samples/sec: 91.32 - lr: 0.025000
873
+ 2021-12-31 10:31:29,079 epoch 35 - iter 540/1807 - loss 0.05935994 - samples/sec: 89.67 - lr: 0.025000
874
+ 2021-12-31 10:31:45,084 epoch 35 - iter 720/1807 - loss 0.05982168 - samples/sec: 90.10 - lr: 0.025000
875
+ 2021-12-31 10:32:00,692 epoch 35 - iter 900/1807 - loss 0.05928538 - samples/sec: 92.39 - lr: 0.025000
876
+ 2021-12-31 10:32:16,615 epoch 35 - iter 1080/1807 - loss 0.05961166 - samples/sec: 90.58 - lr: 0.025000
877
+ 2021-12-31 10:32:32,475 epoch 35 - iter 1260/1807 - loss 0.06019352 - samples/sec: 90.93 - lr: 0.025000
878
+ 2021-12-31 10:32:48,494 epoch 35 - iter 1440/1807 - loss 0.06020781 - samples/sec: 90.02 - lr: 0.025000
879
+ 2021-12-31 10:33:04,244 epoch 35 - iter 1620/1807 - loss 0.05999299 - samples/sec: 91.57 - lr: 0.025000
880
+ 2021-12-31 10:33:20,684 epoch 35 - iter 1800/1807 - loss 0.05998842 - samples/sec: 87.72 - lr: 0.025000
881
+ 2021-12-31 10:33:21,238 ----------------------------------------------------------------------------------------------------
882
+ 2021-12-31 10:33:21,238 EPOCH 35 done: loss 0.0600 - lr 0.0250000
883
+ 2021-12-31 10:33:57,434 DEV : loss 0.06338120251893997 - f1-score (micro avg) 0.9829
884
+ 2021-12-31 10:33:57,624 BAD EPOCHS (no improvement): 1
885
+ 2021-12-31 10:33:57,626 ----------------------------------------------------------------------------------------------------
886
+ 2021-12-31 10:34:13,768 epoch 36 - iter 180/1807 - loss 0.06028850 - samples/sec: 89.35 - lr: 0.025000
887
+ 2021-12-31 10:34:29,556 epoch 36 - iter 360/1807 - loss 0.05827195 - samples/sec: 91.34 - lr: 0.025000
888
+ 2021-12-31 10:34:46,060 epoch 36 - iter 540/1807 - loss 0.05947832 - samples/sec: 87.38 - lr: 0.025000
889
+ 2021-12-31 10:35:02,018 epoch 36 - iter 720/1807 - loss 0.05898679 - samples/sec: 90.38 - lr: 0.025000
890
+ 2021-12-31 10:35:18,203 epoch 36 - iter 900/1807 - loss 0.05910041 - samples/sec: 89.10 - lr: 0.025000
891
+ 2021-12-31 10:35:34,254 epoch 36 - iter 1080/1807 - loss 0.05973540 - samples/sec: 89.84 - lr: 0.025000
892
+ 2021-12-31 10:35:50,256 epoch 36 - iter 1260/1807 - loss 0.05924335 - samples/sec: 90.13 - lr: 0.025000
893
+ 2021-12-31 10:36:06,236 epoch 36 - iter 1440/1807 - loss 0.05881263 - samples/sec: 90.25 - lr: 0.025000
894
+ 2021-12-31 10:36:22,117 epoch 36 - iter 1620/1807 - loss 0.05885928 - samples/sec: 90.80 - lr: 0.025000
895
+ 2021-12-31 10:36:38,208 epoch 36 - iter 1800/1807 - loss 0.05867245 - samples/sec: 89.62 - lr: 0.025000
896
+ 2021-12-31 10:36:38,763 ----------------------------------------------------------------------------------------------------
897
+ 2021-12-31 10:36:38,763 EPOCH 36 done: loss 0.0587 - lr 0.0250000
898
+ 2021-12-31 10:37:17,552 DEV : loss 0.06424003839492798 - f1-score (micro avg) 0.9835
899
+ 2021-12-31 10:37:17,751 BAD EPOCHS (no improvement): 2
900
+ 2021-12-31 10:37:17,752 ----------------------------------------------------------------------------------------------------
901
+ 2021-12-31 10:37:33,804 epoch 37 - iter 180/1807 - loss 0.05692650 - samples/sec: 89.85 - lr: 0.025000
902
+ 2021-12-31 10:37:50,368 epoch 37 - iter 360/1807 - loss 0.05616469 - samples/sec: 87.06 - lr: 0.025000
903
+ 2021-12-31 10:38:06,389 epoch 37 - iter 540/1807 - loss 0.05662717 - samples/sec: 90.01 - lr: 0.025000
904
+ 2021-12-31 10:38:22,399 epoch 37 - iter 720/1807 - loss 0.05716632 - samples/sec: 90.08 - lr: 0.025000
905
+ 2021-12-31 10:38:37,783 epoch 37 - iter 900/1807 - loss 0.05713545 - samples/sec: 93.74 - lr: 0.025000
906
+ 2021-12-31 10:38:53,871 epoch 37 - iter 1080/1807 - loss 0.05764661 - samples/sec: 89.64 - lr: 0.025000
907
+ 2021-12-31 10:39:10,031 epoch 37 - iter 1260/1807 - loss 0.05713711 - samples/sec: 89.23 - lr: 0.025000
908
+ 2021-12-31 10:39:25,737 epoch 37 - iter 1440/1807 - loss 0.05769197 - samples/sec: 91.83 - lr: 0.025000
909
+ 2021-12-31 10:39:41,486 epoch 37 - iter 1620/1807 - loss 0.05788084 - samples/sec: 91.57 - lr: 0.025000
910
+ 2021-12-31 10:39:57,218 epoch 37 - iter 1800/1807 - loss 0.05864320 - samples/sec: 91.67 - lr: 0.025000
911
+ 2021-12-31 10:39:57,747 ----------------------------------------------------------------------------------------------------
912
+ 2021-12-31 10:39:57,748 EPOCH 37 done: loss 0.0586 - lr 0.0250000
913
+ 2021-12-31 10:40:34,869 DEV : loss 0.06326954811811447 - f1-score (micro avg) 0.9831
914
+ 2021-12-31 10:40:35,052 BAD EPOCHS (no improvement): 3
915
+ 2021-12-31 10:40:35,054 ----------------------------------------------------------------------------------------------------
916
+ 2021-12-31 10:40:51,312 epoch 38 - iter 180/1807 - loss 0.05496563 - samples/sec: 88.71 - lr: 0.025000
917
+ 2021-12-31 10:41:07,088 epoch 38 - iter 360/1807 - loss 0.05435886 - samples/sec: 91.42 - lr: 0.025000
918
+ 2021-12-31 10:41:22,841 epoch 38 - iter 540/1807 - loss 0.05464384 - samples/sec: 91.55 - lr: 0.025000
919
+ 2021-12-31 10:41:38,398 epoch 38 - iter 720/1807 - loss 0.05548335 - samples/sec: 92.69 - lr: 0.025000
920
+ 2021-12-31 10:41:54,754 epoch 38 - iter 900/1807 - loss 0.05628518 - samples/sec: 88.18 - lr: 0.025000
921
+ 2021-12-31 10:42:10,229 epoch 38 - iter 1080/1807 - loss 0.05604961 - samples/sec: 93.19 - lr: 0.025000
922
+ 2021-12-31 10:42:26,417 epoch 38 - iter 1260/1807 - loss 0.05594531 - samples/sec: 89.09 - lr: 0.025000
923
+ 2021-12-31 10:42:42,839 epoch 38 - iter 1440/1807 - loss 0.05651329 - samples/sec: 87.81 - lr: 0.025000
924
+ 2021-12-31 10:42:58,889 epoch 38 - iter 1620/1807 - loss 0.05695998 - samples/sec: 89.85 - lr: 0.025000
925
+ 2021-12-31 10:43:15,043 epoch 38 - iter 1800/1807 - loss 0.05706783 - samples/sec: 89.27 - lr: 0.025000
926
+ 2021-12-31 10:43:15,590 ----------------------------------------------------------------------------------------------------
927
+ 2021-12-31 10:43:15,590 EPOCH 38 done: loss 0.0570 - lr 0.0250000
928
+ 2021-12-31 10:43:52,423 DEV : loss 0.06343492120504379 - f1-score (micro avg) 0.9831
929
+ 2021-12-31 10:43:52,610 BAD EPOCHS (no improvement): 4
930
+ 2021-12-31 10:43:52,612 ----------------------------------------------------------------------------------------------------
931
+ 2021-12-31 10:44:08,739 epoch 39 - iter 180/1807 - loss 0.05834451 - samples/sec: 89.43 - lr: 0.012500
932
+ 2021-12-31 10:44:24,462 epoch 39 - iter 360/1807 - loss 0.05496382 - samples/sec: 91.72 - lr: 0.012500
933
+ 2021-12-31 10:44:40,570 epoch 39 - iter 540/1807 - loss 0.05537094 - samples/sec: 89.53 - lr: 0.012500
934
+ 2021-12-31 10:44:56,434 epoch 39 - iter 720/1807 - loss 0.05546561 - samples/sec: 90.90 - lr: 0.012500
935
+ 2021-12-31 10:45:12,338 epoch 39 - iter 900/1807 - loss 0.05527723 - samples/sec: 90.67 - lr: 0.012500
936
+ 2021-12-31 10:45:27,903 epoch 39 - iter 1080/1807 - loss 0.05518412 - samples/sec: 92.65 - lr: 0.012500
937
+ 2021-12-31 10:45:43,777 epoch 39 - iter 1260/1807 - loss 0.05540916 - samples/sec: 90.86 - lr: 0.012500
938
+ 2021-12-31 10:45:59,259 epoch 39 - iter 1440/1807 - loss 0.05568263 - samples/sec: 93.15 - lr: 0.012500
939
+ 2021-12-31 10:46:15,024 epoch 39 - iter 1620/1807 - loss 0.05532678 - samples/sec: 91.47 - lr: 0.012500
940
+ 2021-12-31 10:46:30,975 epoch 39 - iter 1800/1807 - loss 0.05524694 - samples/sec: 90.40 - lr: 0.012500
941
+ 2021-12-31 10:46:31,584 ----------------------------------------------------------------------------------------------------
942
+ 2021-12-31 10:46:31,585 EPOCH 39 done: loss 0.0552 - lr 0.0125000
943
+ 2021-12-31 10:47:10,908 DEV : loss 0.06419230252504349 - f1-score (micro avg) 0.9829
944
+ 2021-12-31 10:47:11,105 BAD EPOCHS (no improvement): 1
945
+ 2021-12-31 10:47:11,106 ----------------------------------------------------------------------------------------------------
946
+ 2021-12-31 10:47:26,949 epoch 40 - iter 180/1807 - loss 0.05824543 - samples/sec: 91.06 - lr: 0.012500
947
+ 2021-12-31 10:47:42,913 epoch 40 - iter 360/1807 - loss 0.05527233 - samples/sec: 90.33 - lr: 0.012500
948
+ 2021-12-31 10:47:59,224 epoch 40 - iter 540/1807 - loss 0.05570769 - samples/sec: 88.41 - lr: 0.012500
949
+ 2021-12-31 10:48:14,703 epoch 40 - iter 720/1807 - loss 0.05485811 - samples/sec: 93.17 - lr: 0.012500
950
+ 2021-12-31 10:48:30,458 epoch 40 - iter 900/1807 - loss 0.05502772 - samples/sec: 91.54 - lr: 0.012500
951
+ 2021-12-31 10:48:46,369 epoch 40 - iter 1080/1807 - loss 0.05487373 - samples/sec: 90.63 - lr: 0.012500
952
+ 2021-12-31 10:49:01,734 epoch 40 - iter 1260/1807 - loss 0.05438047 - samples/sec: 93.85 - lr: 0.012500
953
+ 2021-12-31 10:49:17,649 epoch 40 - iter 1440/1807 - loss 0.05459548 - samples/sec: 90.61 - lr: 0.012500
954
+ 2021-12-31 10:49:33,390 epoch 40 - iter 1620/1807 - loss 0.05450567 - samples/sec: 91.62 - lr: 0.012500
955
+ 2021-12-31 10:49:49,353 epoch 40 - iter 1800/1807 - loss 0.05462945 - samples/sec: 90.34 - lr: 0.012500
956
+ 2021-12-31 10:49:49,959 ----------------------------------------------------------------------------------------------------
957
+ 2021-12-31 10:49:49,959 EPOCH 40 done: loss 0.0546 - lr 0.0125000
958
+ 2021-12-31 10:50:26,216 DEV : loss 0.06343018263578415 - f1-score (micro avg) 0.9829
959
+ 2021-12-31 10:50:26,401 BAD EPOCHS (no improvement): 2
960
+ 2021-12-31 10:50:26,402 ----------------------------------------------------------------------------------------------------
961
+ 2021-12-31 10:50:42,801 epoch 41 - iter 180/1807 - loss 0.04923909 - samples/sec: 87.95 - lr: 0.012500
962
+ 2021-12-31 10:50:58,898 epoch 41 - iter 360/1807 - loss 0.05125288 - samples/sec: 89.59 - lr: 0.012500
963
+ 2021-12-31 10:51:14,501 epoch 41 - iter 540/1807 - loss 0.05242298 - samples/sec: 92.43 - lr: 0.012500
964
+ 2021-12-31 10:51:30,244 epoch 41 - iter 720/1807 - loss 0.05272643 - samples/sec: 91.60 - lr: 0.012500
965
+ 2021-12-31 10:51:46,266 epoch 41 - iter 900/1807 - loss 0.05277145 - samples/sec: 90.01 - lr: 0.012500
966
+ 2021-12-31 10:52:02,535 epoch 41 - iter 1080/1807 - loss 0.05329680 - samples/sec: 88.64 - lr: 0.012500
967
+ 2021-12-31 10:52:18,362 epoch 41 - iter 1260/1807 - loss 0.05349535 - samples/sec: 91.12 - lr: 0.012500
968
+ 2021-12-31 10:52:34,324 epoch 41 - iter 1440/1807 - loss 0.05371268 - samples/sec: 90.35 - lr: 0.012500
969
+ 2021-12-31 10:52:50,154 epoch 41 - iter 1620/1807 - loss 0.05362217 - samples/sec: 91.09 - lr: 0.012500
970
+ 2021-12-31 10:53:06,114 epoch 41 - iter 1800/1807 - loss 0.05361560 - samples/sec: 90.36 - lr: 0.012500
971
+ 2021-12-31 10:53:06,648 ----------------------------------------------------------------------------------------------------
972
+ 2021-12-31 10:53:06,649 EPOCH 41 done: loss 0.0537 - lr 0.0125000
973
+ 2021-12-31 10:53:42,920 DEV : loss 0.06420625746250153 - f1-score (micro avg) 0.9831
974
+ 2021-12-31 10:53:43,107 BAD EPOCHS (no improvement): 3
975
+ 2021-12-31 10:53:43,108 ----------------------------------------------------------------------------------------------------
976
+ 2021-12-31 10:53:59,320 epoch 42 - iter 180/1807 - loss 0.04886676 - samples/sec: 88.96 - lr: 0.012500
977
+ 2021-12-31 10:54:15,301 epoch 42 - iter 360/1807 - loss 0.05210812 - samples/sec: 90.24 - lr: 0.012500
978
+ 2021-12-31 10:54:31,014 epoch 42 - iter 540/1807 - loss 0.05220145 - samples/sec: 91.78 - lr: 0.012500
979
+ 2021-12-31 10:54:46,930 epoch 42 - iter 720/1807 - loss 0.05239133 - samples/sec: 90.61 - lr: 0.012500
980
+ 2021-12-31 10:55:02,977 epoch 42 - iter 900/1807 - loss 0.05260141 - samples/sec: 89.87 - lr: 0.012500
981
+ 2021-12-31 10:55:19,228 epoch 42 - iter 1080/1807 - loss 0.05260187 - samples/sec: 88.74 - lr: 0.012500
982
+ 2021-12-31 10:55:35,215 epoch 42 - iter 1260/1807 - loss 0.05242910 - samples/sec: 90.21 - lr: 0.012500
983
+ 2021-12-31 10:55:51,163 epoch 42 - iter 1440/1807 - loss 0.05265492 - samples/sec: 90.43 - lr: 0.012500
984
+ 2021-12-31 10:56:07,328 epoch 42 - iter 1620/1807 - loss 0.05317972 - samples/sec: 89.21 - lr: 0.012500
985
+ 2021-12-31 10:56:23,405 epoch 42 - iter 1800/1807 - loss 0.05319734 - samples/sec: 89.70 - lr: 0.012500
986
+ 2021-12-31 10:56:23,951 ----------------------------------------------------------------------------------------------------
987
+ 2021-12-31 10:56:23,951 EPOCH 42 done: loss 0.0532 - lr 0.0125000
988
+ 2021-12-31 10:57:03,168 DEV : loss 0.06362675130367279 - f1-score (micro avg) 0.9831
989
+ 2021-12-31 10:57:03,368 BAD EPOCHS (no improvement): 4
990
+ 2021-12-31 10:57:03,370 ----------------------------------------------------------------------------------------------------
991
+ 2021-12-31 10:57:19,009 epoch 43 - iter 180/1807 - loss 0.05496817 - samples/sec: 92.23 - lr: 0.006250
992
+ 2021-12-31 10:57:34,952 epoch 43 - iter 360/1807 - loss 0.05262157 - samples/sec: 90.45 - lr: 0.006250
993
+ 2021-12-31 10:57:51,104 epoch 43 - iter 540/1807 - loss 0.05252708 - samples/sec: 89.28 - lr: 0.006250
994
+ 2021-12-31 10:58:06,630 epoch 43 - iter 720/1807 - loss 0.05258453 - samples/sec: 92.89 - lr: 0.006250
995
+ 2021-12-31 10:58:22,297 epoch 43 - iter 900/1807 - loss 0.05170441 - samples/sec: 92.05 - lr: 0.006250
996
+ 2021-12-31 10:58:38,636 epoch 43 - iter 1080/1807 - loss 0.05199907 - samples/sec: 88.26 - lr: 0.006250
997
+ 2021-12-31 10:58:54,582 epoch 43 - iter 1260/1807 - loss 0.05289598 - samples/sec: 90.42 - lr: 0.006250
998
+ 2021-12-31 10:59:10,756 epoch 43 - iter 1440/1807 - loss 0.05239565 - samples/sec: 89.17 - lr: 0.006250
999
+ 2021-12-31 10:59:26,756 epoch 43 - iter 1620/1807 - loss 0.05245197 - samples/sec: 90.14 - lr: 0.006250
1000
+ 2021-12-31 10:59:43,140 epoch 43 - iter 1800/1807 - loss 0.05236153 - samples/sec: 88.01 - lr: 0.006250
1001
+ 2021-12-31 10:59:43,734 ----------------------------------------------------------------------------------------------------
1002
+ 2021-12-31 10:59:43,734 EPOCH 43 done: loss 0.0523 - lr 0.0062500
1003
+ 2021-12-31 11:00:19,875 DEV : loss 0.06449297815561295 - f1-score (micro avg) 0.983
1004
+ 2021-12-31 11:00:20,058 BAD EPOCHS (no improvement): 1
1005
+ 2021-12-31 11:00:20,060 ----------------------------------------------------------------------------------------------------
1006
+ 2021-12-31 11:00:36,054 epoch 44 - iter 180/1807 - loss 0.05668095 - samples/sec: 90.17 - lr: 0.006250
1007
+ 2021-12-31 11:00:51,879 epoch 44 - iter 360/1807 - loss 0.05376107 - samples/sec: 91.13 - lr: 0.006250
1008
+ 2021-12-31 11:01:07,774 epoch 44 - iter 540/1807 - loss 0.05410164 - samples/sec: 90.73 - lr: 0.006250
1009
+ 2021-12-31 11:01:23,539 epoch 44 - iter 720/1807 - loss 0.05349578 - samples/sec: 91.47 - lr: 0.006250
1010
+ 2021-12-31 11:01:39,511 epoch 44 - iter 900/1807 - loss 0.05316904 - samples/sec: 90.29 - lr: 0.006250
1011
+ 2021-12-31 11:01:55,495 epoch 44 - iter 1080/1807 - loss 0.05360298 - samples/sec: 90.23 - lr: 0.006250
1012
+ 2021-12-31 11:02:11,974 epoch 44 - iter 1260/1807 - loss 0.05360002 - samples/sec: 87.52 - lr: 0.006250
1013
+ 2021-12-31 11:02:27,697 epoch 44 - iter 1440/1807 - loss 0.05333331 - samples/sec: 91.72 - lr: 0.006250
1014
+ 2021-12-31 11:02:43,120 epoch 44 - iter 1620/1807 - loss 0.05286587 - samples/sec: 93.50 - lr: 0.006250
1015
+ 2021-12-31 11:02:58,798 epoch 44 - iter 1800/1807 - loss 0.05270956 - samples/sec: 91.99 - lr: 0.006250
1016
+ 2021-12-31 11:02:59,351 ----------------------------------------------------------------------------------------------------
1017
+ 2021-12-31 11:02:59,352 EPOCH 44 done: loss 0.0527 - lr 0.0062500
1018
+ 2021-12-31 11:03:35,832 DEV : loss 0.06455685943365097 - f1-score (micro avg) 0.9831
1019
+ 2021-12-31 11:03:36,019 BAD EPOCHS (no improvement): 2
1020
+ 2021-12-31 11:03:36,021 ----------------------------------------------------------------------------------------------------
1021
+ 2021-12-31 11:03:52,202 epoch 45 - iter 180/1807 - loss 0.05063292 - samples/sec: 89.13 - lr: 0.006250
1022
+ 2021-12-31 11:04:08,225 epoch 45 - iter 360/1807 - loss 0.05171673 - samples/sec: 90.00 - lr: 0.006250
1023
+ 2021-12-31 11:04:24,263 epoch 45 - iter 540/1807 - loss 0.05167432 - samples/sec: 89.93 - lr: 0.006250
1024
+ 2021-12-31 11:04:40,362 epoch 45 - iter 720/1807 - loss 0.05121190 - samples/sec: 89.58 - lr: 0.006250
1025
+ 2021-12-31 11:04:56,274 epoch 45 - iter 900/1807 - loss 0.05221446 - samples/sec: 90.63 - lr: 0.006250
1026
+ 2021-12-31 11:05:12,479 epoch 45 - iter 1080/1807 - loss 0.05188940 - samples/sec: 88.99 - lr: 0.006250
1027
+ 2021-12-31 11:05:28,572 epoch 45 - iter 1260/1807 - loss 0.05237022 - samples/sec: 89.62 - lr: 0.006250
1028
+ 2021-12-31 11:05:44,476 epoch 45 - iter 1440/1807 - loss 0.05180768 - samples/sec: 90.68 - lr: 0.006250
1029
+ 2021-12-31 11:06:00,356 epoch 45 - iter 1620/1807 - loss 0.05176296 - samples/sec: 90.81 - lr: 0.006250
1030
+ 2021-12-31 11:06:16,343 epoch 45 - iter 1800/1807 - loss 0.05236414 - samples/sec: 90.20 - lr: 0.006250
1031
+ 2021-12-31 11:06:16,948 ----------------------------------------------------------------------------------------------------
1032
+ 2021-12-31 11:06:16,949 EPOCH 45 done: loss 0.0523 - lr 0.0062500
1033
+ 2021-12-31 11:06:56,269 DEV : loss 0.06413871794939041 - f1-score (micro avg) 0.983
1034
+ 2021-12-31 11:06:56,425 BAD EPOCHS (no improvement): 3
1035
+ 2021-12-31 11:06:56,427 ----------------------------------------------------------------------------------------------------
1036
+ 2021-12-31 11:07:12,359 epoch 46 - iter 180/1807 - loss 0.04909660 - samples/sec: 90.52 - lr: 0.006250
1037
+ 2021-12-31 11:07:27,933 epoch 46 - iter 360/1807 - loss 0.04990439 - samples/sec: 92.58 - lr: 0.006250
1038
+ 2021-12-31 11:07:44,036 epoch 46 - iter 540/1807 - loss 0.05183261 - samples/sec: 89.55 - lr: 0.006250
1039
+ 2021-12-31 11:07:59,808 epoch 46 - iter 720/1807 - loss 0.05108367 - samples/sec: 91.44 - lr: 0.006250
1040
+ 2021-12-31 11:08:16,323 epoch 46 - iter 900/1807 - loss 0.05156129 - samples/sec: 87.33 - lr: 0.006250
1041
+ 2021-12-31 11:08:32,181 epoch 46 - iter 1080/1807 - loss 0.05164911 - samples/sec: 90.93 - lr: 0.006250
1042
+ 2021-12-31 11:08:48,124 epoch 46 - iter 1260/1807 - loss 0.05241189 - samples/sec: 90.45 - lr: 0.006250
1043
+ 2021-12-31 11:09:04,600 epoch 46 - iter 1440/1807 - loss 0.05209220 - samples/sec: 87.53 - lr: 0.006250
1044
+ 2021-12-31 11:09:20,227 epoch 46 - iter 1620/1807 - loss 0.05187081 - samples/sec: 92.29 - lr: 0.006250
1045
+ 2021-12-31 11:09:36,191 epoch 46 - iter 1800/1807 - loss 0.05205935 - samples/sec: 90.34 - lr: 0.006250
1046
+ 2021-12-31 11:09:36,782 ----------------------------------------------------------------------------------------------------
1047
+ 2021-12-31 11:09:36,782 EPOCH 46 done: loss 0.0521 - lr 0.0062500
1048
+ 2021-12-31 11:10:13,201 DEV : loss 0.0644669309258461 - f1-score (micro avg) 0.983
1049
+ 2021-12-31 11:10:13,398 BAD EPOCHS (no improvement): 4
1050
+ 2021-12-31 11:10:13,399 ----------------------------------------------------------------------------------------------------
1051
+ 2021-12-31 11:10:29,417 epoch 47 - iter 180/1807 - loss 0.05250873 - samples/sec: 90.04 - lr: 0.003125
1052
+ 2021-12-31 11:10:45,589 epoch 47 - iter 360/1807 - loss 0.05160928 - samples/sec: 89.18 - lr: 0.003125
1053
+ 2021-12-31 11:11:01,280 epoch 47 - iter 540/1807 - loss 0.05161492 - samples/sec: 91.91 - lr: 0.003125
1054
+ 2021-12-31 11:11:17,277 epoch 47 - iter 720/1807 - loss 0.05136337 - samples/sec: 90.15 - lr: 0.003125
1055
+ 2021-12-31 11:11:33,230 epoch 47 - iter 900/1807 - loss 0.05023989 - samples/sec: 90.40 - lr: 0.003125
1056
+ 2021-12-31 11:11:49,156 epoch 47 - iter 1080/1807 - loss 0.05064277 - samples/sec: 90.55 - lr: 0.003125
1057
+ 2021-12-31 11:12:04,959 epoch 47 - iter 1260/1807 - loss 0.05089925 - samples/sec: 91.25 - lr: 0.003125
1058
+ 2021-12-31 11:12:21,092 epoch 47 - iter 1440/1807 - loss 0.05071923 - samples/sec: 89.39 - lr: 0.003125
1059
+ 2021-12-31 11:12:36,949 epoch 47 - iter 1620/1807 - loss 0.05083516 - samples/sec: 90.95 - lr: 0.003125
1060
+ 2021-12-31 11:12:52,744 epoch 47 - iter 1800/1807 - loss 0.05106443 - samples/sec: 91.31 - lr: 0.003125
1061
+ 2021-12-31 11:12:53,321 ----------------------------------------------------------------------------------------------------
1062
+ 2021-12-31 11:12:53,321 EPOCH 47 done: loss 0.0511 - lr 0.0031250
1063
+ 2021-12-31 11:13:29,490 DEV : loss 0.06470787525177002 - f1-score (micro avg) 0.9829
1064
+ 2021-12-31 11:13:29,672 BAD EPOCHS (no improvement): 1
1065
+ 2021-12-31 11:13:29,674 ----------------------------------------------------------------------------------------------------
1066
+ 2021-12-31 11:13:45,987 epoch 48 - iter 180/1807 - loss 0.05119727 - samples/sec: 88.41 - lr: 0.003125
1067
+ 2021-12-31 11:14:02,271 epoch 48 - iter 360/1807 - loss 0.05026057 - samples/sec: 88.57 - lr: 0.003125
1068
+ 2021-12-31 11:14:18,202 epoch 48 - iter 540/1807 - loss 0.04968790 - samples/sec: 90.53 - lr: 0.003125
1069
+ 2021-12-31 11:14:33,834 epoch 48 - iter 720/1807 - loss 0.05040465 - samples/sec: 92.25 - lr: 0.003125
1070
+ 2021-12-31 11:14:49,709 epoch 48 - iter 900/1807 - loss 0.05065504 - samples/sec: 90.84 - lr: 0.003125
1071
+ 2021-12-31 11:15:05,727 epoch 48 - iter 1080/1807 - loss 0.05037297 - samples/sec: 90.02 - lr: 0.003125
1072
+ 2021-12-31 11:15:21,077 epoch 48 - iter 1260/1807 - loss 0.05063199 - samples/sec: 93.96 - lr: 0.003125
1073
+ 2021-12-31 11:15:36,587 epoch 48 - iter 1440/1807 - loss 0.05076731 - samples/sec: 92.98 - lr: 0.003125
1074
+ 2021-12-31 11:15:52,489 epoch 48 - iter 1620/1807 - loss 0.05082260 - samples/sec: 90.68 - lr: 0.003125
1075
+ 2021-12-31 11:16:08,520 epoch 48 - iter 1800/1807 - loss 0.05101165 - samples/sec: 89.96 - lr: 0.003125
1076
+ 2021-12-31 11:16:09,115 ----------------------------------------------------------------------------------------------------
1077
+ 2021-12-31 11:16:09,116 EPOCH 48 done: loss 0.0510 - lr 0.0031250
1078
+ 2021-12-31 11:16:48,035 DEV : loss 0.06484530121088028 - f1-score (micro avg) 0.983
1079
+ 2021-12-31 11:16:48,189 BAD EPOCHS (no improvement): 2
1080
+ 2021-12-31 11:16:48,191 ----------------------------------------------------------------------------------------------------
1081
+ 2021-12-31 11:17:03,775 epoch 49 - iter 180/1807 - loss 0.04706234 - samples/sec: 92.51 - lr: 0.003125
1082
+ 2021-12-31 11:17:19,604 epoch 49 - iter 360/1807 - loss 0.04796051 - samples/sec: 91.07 - lr: 0.003125
1083
+ 2021-12-31 11:17:35,506 epoch 49 - iter 540/1807 - loss 0.04820802 - samples/sec: 90.67 - lr: 0.003125
1084
+ 2021-12-31 11:17:51,301 epoch 49 - iter 720/1807 - loss 0.04872061 - samples/sec: 91.31 - lr: 0.003125
1085
+ 2021-12-31 11:18:06,963 epoch 49 - iter 900/1807 - loss 0.04900955 - samples/sec: 92.08 - lr: 0.003125
1086
+ 2021-12-31 11:18:22,961 epoch 49 - iter 1080/1807 - loss 0.04952427 - samples/sec: 90.14 - lr: 0.003125
1087
+ 2021-12-31 11:18:39,172 epoch 49 - iter 1260/1807 - loss 0.04981242 - samples/sec: 88.96 - lr: 0.003125
1088
+ 2021-12-31 11:18:55,485 epoch 49 - iter 1440/1807 - loss 0.05015633 - samples/sec: 88.41 - lr: 0.003125
1089
+ 2021-12-31 11:19:11,166 epoch 49 - iter 1620/1807 - loss 0.05076498 - samples/sec: 91.97 - lr: 0.003125
1090
+ 2021-12-31 11:19:27,065 epoch 49 - iter 1800/1807 - loss 0.05104387 - samples/sec: 90.71 - lr: 0.003125
1091
+ 2021-12-31 11:19:27,675 ----------------------------------------------------------------------------------------------------
1092
+ 2021-12-31 11:19:27,675 EPOCH 49 done: loss 0.0510 - lr 0.0031250
1093
+ 2021-12-31 11:20:04,021 DEV : loss 0.06486314535140991 - f1-score (micro avg) 0.983
1094
+ 2021-12-31 11:20:04,217 BAD EPOCHS (no improvement): 3
1095
+ 2021-12-31 11:20:04,218 ----------------------------------------------------------------------------------------------------
1096
+ 2021-12-31 11:20:20,650 epoch 50 - iter 180/1807 - loss 0.05726933 - samples/sec: 87.77 - lr: 0.003125
1097
+ 2021-12-31 11:20:36,455 epoch 50 - iter 360/1807 - loss 0.05538766 - samples/sec: 91.25 - lr: 0.003125
1098
+ 2021-12-31 11:20:52,012 epoch 50 - iter 540/1807 - loss 0.05444601 - samples/sec: 92.69 - lr: 0.003125
1099
+ 2021-12-31 11:21:07,973 epoch 50 - iter 720/1807 - loss 0.05313637 - samples/sec: 90.35 - lr: 0.003125
1100
+ 2021-12-31 11:21:23,983 epoch 50 - iter 900/1807 - loss 0.05290526 - samples/sec: 90.08 - lr: 0.003125
1101
+ 2021-12-31 11:21:39,924 epoch 50 - iter 1080/1807 - loss 0.05235234 - samples/sec: 90.47 - lr: 0.003125
1102
+ 2021-12-31 11:21:55,732 epoch 50 - iter 1260/1807 - loss 0.05207690 - samples/sec: 91.23 - lr: 0.003125
1103
+ 2021-12-31 11:22:11,663 epoch 50 - iter 1440/1807 - loss 0.05205514 - samples/sec: 90.52 - lr: 0.003125
1104
+ 2021-12-31 11:22:27,392 epoch 50 - iter 1620/1807 - loss 0.05173851 - samples/sec: 91.69 - lr: 0.003125
1105
+ 2021-12-31 11:22:43,193 epoch 50 - iter 1800/1807 - loss 0.05189058 - samples/sec: 91.27 - lr: 0.003125
1106
+ 2021-12-31 11:22:43,750 ----------------------------------------------------------------------------------------------------
1107
+ 2021-12-31 11:22:43,750 EPOCH 50 done: loss 0.0519 - lr 0.0031250
1108
+ 2021-12-31 11:23:20,432 DEV : loss 0.06452730298042297 - f1-score (micro avg) 0.9831
1109
+ 2021-12-31 11:23:20,619 BAD EPOCHS (no improvement): 4
1110
+ 2021-12-31 11:23:25,890 ----------------------------------------------------------------------------------------------------
1111
+ 2021-12-31 11:23:25,893 loading file models/UPOS_UD_FRENCH_GSD_PLUS_Flair-Embeddings_50_2021-12-31-08:34:44/best-model.pt
1112
+ 2021-12-31 11:23:43,354 0.9797 0.9797 0.9797 0.9797
1113
+ 2021-12-31 11:23:43,354
1114
+ Results:
1115
+ - F-score (micro) 0.9797
1116
+ - F-score (macro) 0.9178
1117
+ - Accuracy 0.9797
1118
+
1119
+ By class:
1120
+ precision recall f1-score support
1121
+
1122
+ PREP 0.9966 0.9987 0.9976 1483
1123
+ PUNCT 1.0000 1.0000 1.0000 833
1124
+ NMS 0.9634 0.9801 0.9717 753
1125
+ DET 0.9923 0.9984 0.9954 645
1126
+ VERB 0.9913 0.9811 0.9862 583
1127
+ NFS 0.9667 0.9839 0.9752 560
1128
+ ADV 0.9940 0.9821 0.9880 504
1129
+ PROPN 0.9541 0.8937 0.9229 395
1130
+ DETMS 1.0000 1.0000 1.0000 362
1131
+ AUX 0.9860 0.9915 0.9888 355
1132
+ YPFOR 1.0000 1.0000 1.0000 353
1133
+ NMP 0.9666 0.9475 0.9570 305
1134
+ COCO 0.9959 1.0000 0.9980 245
1135
+ ADJMS 0.9463 0.9385 0.9424 244
1136
+ DETFS 1.0000 1.0000 1.0000 240
1137
+ CHIF 0.9648 0.9865 0.9755 222
1138
+ NFP 0.9515 0.9849 0.9679 199
1139
+ ADJFS 0.9657 0.9286 0.9468 182
1140
+ VPPMS 0.9387 0.9745 0.9563 157
1141
+ COSUB 1.0000 0.9844 0.9921 128
1142
+ DINTMS 0.9918 0.9918 0.9918 122
1143
+ XFAMIL 0.9298 0.9217 0.9258 115
1144
+ PPER3MS 1.0000 1.0000 1.0000 87
1145
+ ADJMP 0.9294 0.9634 0.9461 82
1146
+ PDEMMS 1.0000 1.0000 1.0000 75
1147
+ ADJFP 0.9861 0.9342 0.9595 76
1148
+ PREL 0.9859 1.0000 0.9929 70
1149
+ DINTFS 0.9839 1.0000 0.9919 61
1150
+ PREF 1.0000 1.0000 1.0000 52
1151
+ PPOBJMS 0.9565 0.9362 0.9462 47
1152
+ PREFP 0.9778 1.0000 0.9888 44
1153
+ PINDMS 1.0000 0.9773 0.9885 44
1154
+ VPPFS 0.8298 0.9750 0.8966 40
1155
+ PPER1S 1.0000 1.0000 1.0000 42
1156
+ SYM 1.0000 0.9474 0.9730 38
1157
+ NOUN 0.8824 0.7692 0.8219 39
1158
+ PRON 1.0000 0.9677 0.9836 31
1159
+ PDEMFS 1.0000 1.0000 1.0000 29
1160
+ VPPMP 0.9286 1.0000 0.9630 26
1161
+ ADJ 0.9524 0.9091 0.9302 22
1162
+ PPER3MP 1.0000 1.0000 1.0000 20
1163
+ VPPFP 1.0000 1.0000 1.0000 19
1164
+ PPER3FS 1.0000 1.0000 1.0000 18
1165
+ MOTINC 0.3333 0.4000 0.3636 15
1166
+ PREFS 1.0000 1.0000 1.0000 10
1167
+ PPOBJMP 1.0000 0.8000 0.8889 10
1168
+ PPOBJFS 0.6250 0.8333 0.7143 6
1169
+ INTJ 0.5000 0.6667 0.5714 6
1170
+ PART 1.0000 1.0000 1.0000 4
1171
+ PDEMMP 1.0000 1.0000 1.0000 3
1172
+ PDEMFP 1.0000 1.0000 1.0000 3
1173
+ PPER3FP 1.0000 1.0000 1.0000 2
1174
+ NUM 1.0000 0.3333 0.5000 3
1175
+ PPER2S 1.0000 1.0000 1.0000 2
1176
+ PPOBJFP 0.5000 0.5000 0.5000 2
1177
+ PRELMS 1.0000 1.0000 1.0000 2
1178
+ PINDFS 0.5000 1.0000 0.6667 1
1179
+ PINDMP 1.0000 1.0000 1.0000 1
1180
+ X 0.0000 0.0000 0.0000 1
1181
+ PINDFP 1.0000 1.0000 1.0000 1
1182
+
1183
+ micro avg 0.9797 0.9797 0.9797 10019
1184
+ macro avg 0.9228 0.9230 0.9178 10019
1185
+ weighted avg 0.9802 0.9797 0.9798 10019
1186
+ samples avg 0.9797 0.9797 0.9797 10019
1187
+
1188
+ 2021-12-31 11:23:43,354 ----------------------------------------------------------------------------------------------------
weights.txt ADDED
File without changes