Hugging Face's logo

tanoManzo
/
gena-lm-bert-base-t2t-multi_ft_BioS73_1kbpHG19_DHSs_H3K27AC

Safetensors
bert
Generated from Trainer
custom_code
Model card Files Community
tanoManzo commited on
Commit
f0e7a43
·
verified ·
1 Parent(s): cef8c17

End of training

Browse files
Files changed (2) hide show
  1. README.md +45 -32
  2. model.safetensors +1 -1
README.md CHANGED
@@ -18,13 +18,13 @@ should probably proofread and complete it, then remove this comment. -->
18
 
19
  This model is a fine-tuned version of [AIRI-Institute/gena-lm-bert-base-t2t-multi](https://huggingface.co/AIRI-Institute/gena-lm-bert-base-t2t-multi) on the None dataset.
20
  It achieves the following results on the evaluation set:
21
- - Loss: 0.5503
22
- - F1 Score: 0.8650
23
- - Precision: 0.8391
24
- - Recall: 0.8925
25
- - Accuracy: 0.8513
26
- - Auc: 0.8910
27
- - Prc: 0.8501
28
 
29
  ## Model description
30
 
@@ -56,31 +56,44 @@ The following hyperparameters were used during training:
56
 
57
  | Training Loss | Epoch | Step | Validation Loss | F1 Score | Precision | Recall | Accuracy | Auc | Prc |
58
  |:-------------:|:------:|:-----:|:---------------:|:--------:|:---------:|:------:|:--------:|:------:|:------:|
59
- | 0.6862 | 0.1864 | 500 | 0.6364 | 0.7934 | 0.7511 | 0.8408 | 0.7663 | 0.8211 | 0.8070 |
60
- | 0.6114 | 0.3727 | 1000 | 0.5230 | 0.8189 | 0.7812 | 0.8603 | 0.7969 | 0.8235 | 0.7910 |
61
- | 0.4843 | 0.5591 | 1500 | 0.4601 | 0.8377 | 0.7866 | 0.8959 | 0.8148 | 0.8793 | 0.8586 |
62
- | 0.4567 | 0.7454 | 2000 | 0.4643 | 0.8374 | 0.8181 | 0.8575 | 0.8222 | 0.8875 | 0.8761 |
63
- | 0.4294 | 0.9318 | 2500 | 0.4666 | 0.8468 | 0.8109 | 0.8862 | 0.8289 | 0.8878 | 0.8750 |
64
- | 0.4283 | 1.1182 | 3000 | 0.4356 | 0.8457 | 0.8238 | 0.8687 | 0.8308 | 0.8982 | 0.8833 |
65
- | 0.4067 | 1.3045 | 3500 | 0.4572 | 0.8411 | 0.8429 | 0.8394 | 0.8308 | 0.9020 | 0.8824 |
66
- | 0.4167 | 1.4909 | 4000 | 0.4264 | 0.8464 | 0.8259 | 0.8680 | 0.8319 | 0.9039 | 0.8820 |
67
- | 0.3906 | 1.6772 | 4500 | 0.4686 | 0.8525 | 0.8214 | 0.8862 | 0.8364 | 0.8748 | 0.8126 |
68
- | 0.4078 | 1.8636 | 5000 | 0.4441 | 0.8515 | 0.8056 | 0.9029 | 0.8319 | 0.9010 | 0.8761 |
69
- | 0.3771 | 2.0499 | 5500 | 0.5151 | 0.8503 | 0.8275 | 0.8743 | 0.8356 | 0.9025 | 0.8766 |
70
- | 0.4282 | 2.2363 | 6000 | 0.5186 | 0.8528 | 0.8102 | 0.9001 | 0.8341 | 0.9065 | 0.8938 |
71
- | 0.4063 | 2.4227 | 6500 | 0.4397 | 0.8540 | 0.8291 | 0.8806 | 0.8394 | 0.9048 | 0.8749 |
72
- | 0.3998 | 2.6090 | 7000 | 0.4794 | 0.8585 | 0.8350 | 0.8834 | 0.8446 | 0.9008 | 0.8643 |
73
- | 0.3783 | 2.7954 | 7500 | 0.5577 | 0.8473 | 0.8577 | 0.8373 | 0.8390 | 0.9061 | 0.8827 |
74
- | 0.3926 | 2.9817 | 8000 | 0.4794 | 0.8618 | 0.8313 | 0.8946 | 0.8468 | 0.8907 | 0.8481 |
75
- | 0.4023 | 3.1681 | 8500 | 0.5062 | 0.8583 | 0.7932 | 0.9351 | 0.8353 | 0.9041 | 0.8888 |
76
- | 0.382 | 3.3545 | 9000 | 0.5005 | 0.8604 | 0.8193 | 0.9057 | 0.8431 | 0.9086 | 0.8838 |
77
- | 0.407 | 3.5408 | 9500 | 0.5177 | 0.8591 | 0.8399 | 0.8792 | 0.8461 | 0.9087 | 0.8901 |
78
- | 0.3773 | 3.7272 | 10000 | 0.5228 | 0.8654 | 0.8388 | 0.8939 | 0.8517 | 0.9043 | 0.8743 |
79
- | 0.392 | 3.9135 | 10500 | 0.5146 | 0.8649 | 0.8354 | 0.8966 | 0.8505 | 0.8937 | 0.8569 |
80
- | 0.3817 | 4.0999 | 11000 | 0.5353 | 0.8643 | 0.8146 | 0.9204 | 0.8457 | 0.8642 | 0.8163 |
81
- | 0.4062 | 4.2862 | 11500 | 0.5992 | 0.8472 | 0.8732 | 0.8226 | 0.8416 | 0.9122 | 0.8882 |
82
- | 0.3737 | 4.4726 | 12000 | 0.5634 | 0.8645 | 0.8245 | 0.9085 | 0.8479 | 0.8831 | 0.8428 |
83
- | 0.4161 | 4.6590 | 12500 | 0.5503 | 0.8650 | 0.8391 | 0.8925 | 0.8513 | 0.8910 | 0.8501 |
 
 
 
 
 
 
 
 
 
 
 
 
 
84
 
85
 
86
  ### Framework versions
 
18
 
19
  This model is a fine-tuned version of [AIRI-Institute/gena-lm-bert-base-t2t-multi](https://huggingface.co/AIRI-Institute/gena-lm-bert-base-t2t-multi) on the None dataset.
20
  It achieves the following results on the evaluation set:
21
+ - Loss: 0.6584
22
+ - F1 Score: 0.8725
23
+ - Precision: 0.8147
24
+ - Recall: 0.9392
25
+ - Accuracy: 0.8535
26
+ - Auc: 0.8951
27
+ - Prc: 0.8518
28
 
29
  ## Model description
30
 
 
56
 
57
  | Training Loss | Epoch | Step | Validation Loss | F1 Score | Precision | Recall | Accuracy | Auc | Prc |
58
  |:-------------:|:------:|:-----:|:---------------:|:--------:|:---------:|:------:|:--------:|:------:|:------:|
59
+ | 0.6938 | 0.1864 | 500 | 0.6453 | 0.7873 | 0.7876 | 0.7870 | 0.7730 | 0.8514 | 0.8623 |
60
+ | 0.6039 | 0.3727 | 1000 | 0.4934 | 0.8216 | 0.7733 | 0.8764 | 0.7969 | 0.8716 | 0.8609 |
61
+ | 0.4786 | 0.5591 | 1500 | 0.4523 | 0.8358 | 0.7947 | 0.8813 | 0.8151 | 0.8667 | 0.8376 |
62
+ | 0.4456 | 0.7454 | 2000 | 0.4439 | 0.8342 | 0.8304 | 0.8380 | 0.8222 | 0.8845 | 0.8642 |
63
+ | 0.4328 | 0.9318 | 2500 | 0.4393 | 0.8402 | 0.8416 | 0.8387 | 0.8297 | 0.8905 | 0.8884 |
64
+ | 0.4165 | 1.1182 | 3000 | 0.4482 | 0.8541 | 0.7988 | 0.9176 | 0.8327 | 0.8928 | 0.8847 |
65
+ | 0.4133 | 1.3045 | 3500 | 0.4497 | 0.8546 | 0.7966 | 0.9218 | 0.8327 | 0.8546 | 0.7958 |
66
+ | 0.4166 | 1.4909 | 4000 | 0.4378 | 0.8572 | 0.7975 | 0.9267 | 0.8353 | 0.9034 | 0.8878 |
67
+ | 0.3901 | 1.6772 | 4500 | 0.4694 | 0.8544 | 0.8241 | 0.8869 | 0.8386 | 0.8856 | 0.8629 |
68
+ | 0.3914 | 1.8636 | 5000 | 0.4448 | 0.8535 | 0.8075 | 0.9050 | 0.8341 | 0.8829 | 0.8425 |
69
+ | 0.3634 | 2.0499 | 5500 | 0.5303 | 0.8551 | 0.8143 | 0.9001 | 0.8371 | 0.8903 | 0.8633 |
70
+ | 0.4353 | 2.2363 | 6000 | 0.4807 | 0.8573 | 0.8034 | 0.9190 | 0.8367 | 0.8540 | 0.7970 |
71
+ | 0.3747 | 2.4227 | 6500 | 0.4642 | 0.8589 | 0.8117 | 0.9120 | 0.8401 | 0.8839 | 0.8317 |
72
+ | 0.4069 | 2.6090 | 7000 | 0.4868 | 0.8553 | 0.8290 | 0.8834 | 0.8405 | 0.8895 | 0.8494 |
73
+ | 0.379 | 2.7954 | 7500 | 0.4786 | 0.8594 | 0.8217 | 0.9008 | 0.8427 | 0.8977 | 0.8673 |
74
+ | 0.3817 | 2.9817 | 8000 | 0.5133 | 0.8606 | 0.8203 | 0.9050 | 0.8435 | 0.9106 | 0.8982 |
75
+ | 0.3913 | 3.1681 | 8500 | 0.5098 | 0.8602 | 0.815 | 0.9106 | 0.8420 | 0.8636 | 0.8122 |
76
+ | 0.3783 | 3.3545 | 9000 | 0.5139 | 0.8609 | 0.8221 | 0.9036 | 0.8442 | 0.8869 | 0.8522 |
77
+ | 0.4031 | 3.5408 | 9500 | 0.5302 | 0.8628 | 0.8227 | 0.9071 | 0.8461 | 0.9093 | 0.8907 |
78
+ | 0.3905 | 3.7272 | 10000 | 0.5376 | 0.8643 | 0.8109 | 0.9253 | 0.8449 | 0.9103 | 0.8868 |
79
+ | 0.3893 | 3.9135 | 10500 | 0.5243 | 0.8615 | 0.8214 | 0.9057 | 0.8446 | 0.9162 | 0.9067 |
80
+ | 0.3635 | 4.0999 | 11000 | 0.6047 | 0.8614 | 0.8528 | 0.8701 | 0.8505 | 0.9016 | 0.8770 |
81
+ | 0.4137 | 4.2862 | 11500 | 0.5756 | 0.8567 | 0.8511 | 0.8624 | 0.8461 | 0.9134 | 0.8962 |
82
+ | 0.3747 | 4.4726 | 12000 | 0.5890 | 0.8662 | 0.8254 | 0.9113 | 0.8498 | 0.8634 | 0.8047 |
83
+ | 0.42 | 4.6590 | 12500 | 0.5722 | 0.8676 | 0.8267 | 0.9127 | 0.8513 | 0.8865 | 0.8309 |
84
+ | 0.3957 | 4.8453 | 13000 | 0.5824 | 0.8665 | 0.8116 | 0.9295 | 0.8472 | 0.9139 | 0.9004 |
85
+ | 0.3701 | 5.0317 | 13500 | 0.5850 | 0.8697 | 0.8278 | 0.9162 | 0.8535 | 0.8994 | 0.8633 |
86
+ | 0.3653 | 5.2180 | 14000 | 0.6013 | 0.8697 | 0.8098 | 0.9392 | 0.8498 | 0.9012 | 0.8739 |
87
+ | 0.3956 | 5.4044 | 14500 | 0.6008 | 0.8672 | 0.8464 | 0.8890 | 0.8546 | 0.8989 | 0.8667 |
88
+ | 0.3767 | 5.5908 | 15000 | 0.6562 | 0.8594 | 0.8544 | 0.8645 | 0.8490 | 0.9128 | 0.9003 |
89
+ | 0.3733 | 5.7771 | 15500 | 0.6134 | 0.8646 | 0.8472 | 0.8827 | 0.8524 | 0.9076 | 0.8824 |
90
+ | 0.3798 | 5.9635 | 16000 | 0.6094 | 0.8701 | 0.8131 | 0.9358 | 0.8509 | 0.9110 | 0.8878 |
91
+ | 0.3796 | 6.1498 | 16500 | 0.6068 | 0.8737 | 0.8315 | 0.9204 | 0.8580 | 0.8957 | 0.8634 |
92
+ | 0.3674 | 6.3362 | 17000 | 0.6217 | 0.8727 | 0.8269 | 0.9239 | 0.8561 | 0.9120 | 0.8868 |
93
+ | 0.3719 | 6.5225 | 17500 | 0.6346 | 0.8730 | 0.8413 | 0.9071 | 0.8591 | 0.9199 | 0.9018 |
94
+ | 0.3754 | 6.7089 | 18000 | 0.6272 | 0.8719 | 0.8141 | 0.9385 | 0.8528 | 0.9038 | 0.8661 |
95
+ | 0.3674 | 6.8953 | 18500 | 0.6343 | 0.8724 | 0.8285 | 0.9211 | 0.8561 | 0.9069 | 0.8874 |
96
+ | 0.3511 | 7.0816 | 19000 | 0.6584 | 0.8725 | 0.8147 | 0.9392 | 0.8535 | 0.8951 | 0.8518 |
97
 
98
 
99
  ### Framework versions
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:9e5e9422e3539d8323402b2411d95cba1e8e0640fe47476de1c872fe90a3ff5f
3
  size 442503040
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5418cd575e946a68c5ce5f4701e8c95b622bfdd0a91a83c09cdea221945bc1be
3
  size 442503040