Hugging Face's logo

Aleksandar
/
distilbert-srb-ner

Token Classification
Transformers
PyTorch
Serbian
distilbert
generated_from_trainer
Inference Endpoints
Model card Files Community
1
Aleksandar commited on
Commit
5fdebf6
•
1 Parent(s): 0e09161

add model

Browse files
Files changed (3) hide show
  1. README.md +29 -13
  2. pytorch_model.bin +1 -1
  3. training_args.bin +1 -1
README.md CHANGED
@@ -21,7 +21,7 @@ model_index:
21
  metric:
22
  name: Accuracy
23
  type: accuracy
24
- value: 0.9503250498060186
25
  ---
26
 
27
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -31,11 +31,11 @@ should probably proofread and complete it, then remove this comment. -->
31
 
32
  This model was trained from scratch on the wikiann dataset.
33
  It achieves the following results on the evaluation set:
34
- - Loss: 0.1723
35
- - Precision: 0.8667
36
- - Recall: 0.8860
37
- - F1: 0.8763
38
- - Accuracy: 0.9503
39
 
40
  ## Model description
41
 
@@ -60,16 +60,32 @@ The following hyperparameters were used during training:
60
  - seed: 42
61
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
62
  - lr_scheduler_type: linear
63
- - num_epochs: 4
64
 
65
  ### Training results
66
 
67
- | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
68
- |:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
69
- | 0.3839 | 1.0 | 625 | 0.2204 | 0.8112 | 0.8367 | 0.8238 | 0.9298 |
70
- | 0.2004 | 2.0 | 1250 | 0.1805 | 0.8530 | 0.8676 | 0.8602 | 0.9442 |
71
- | 0.1475 | 3.0 | 1875 | 0.1716 | 0.8536 | 0.8778 | 0.8655 | 0.9467 |
72
- | 0.0943 | 4.0 | 2500 | 0.1723 | 0.8667 | 0.8860 | 0.8763 | 0.9503 |
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
73
 
74
 
75
  ### Framework versions
 
21
  metric:
22
  name: Accuracy
23
  type: accuracy
24
+ value: 0.9576561462374611
25
  ---
26
 
27
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
 
31
 
32
  This model was trained from scratch on the wikiann dataset.
33
  It achieves the following results on the evaluation set:
34
+ - Loss: 0.2972
35
+ - Precision: 0.8871
36
+ - Recall: 0.9100
37
+ - F1: 0.8984
38
+ - Accuracy: 0.9577
39
 
40
  ## Model description
41
 
 
60
  - seed: 42
61
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
62
  - lr_scheduler_type: linear
63
+ - num_epochs: 20
64
 
65
  ### Training results
66
 
67
+ | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
68
+ |:-------------:|:-----:|:-----:|:---------------:|:---------:|:------:|:------:|:--------:|
69
+ | 0.3818 | 1.0 | 625 | 0.2175 | 0.8175 | 0.8370 | 0.8272 | 0.9306 |
70
+ | 0.198 | 2.0 | 1250 | 0.1766 | 0.8551 | 0.8732 | 0.8640 | 0.9458 |
71
+ | 0.1423 | 3.0 | 1875 | 0.1702 | 0.8597 | 0.8763 | 0.8679 | 0.9473 |
72
+ | 0.079 | 4.0 | 2500 | 0.1774 | 0.8674 | 0.8875 | 0.8773 | 0.9515 |
73
+ | 0.0531 | 5.0 | 3125 | 0.2011 | 0.8688 | 0.8965 | 0.8825 | 0.9522 |
74
+ | 0.0429 | 6.0 | 3750 | 0.2082 | 0.8769 | 0.8970 | 0.8868 | 0.9538 |
75
+ | 0.032 | 7.0 | 4375 | 0.2268 | 0.8764 | 0.8916 | 0.8839 | 0.9528 |
76
+ | 0.0204 | 8.0 | 5000 | 0.2423 | 0.8726 | 0.8959 | 0.8841 | 0.9529 |
77
+ | 0.0148 | 9.0 | 5625 | 0.2522 | 0.8774 | 0.8991 | 0.8881 | 0.9538 |
78
+ | 0.0125 | 10.0 | 6250 | 0.2544 | 0.8823 | 0.9024 | 0.8922 | 0.9559 |
79
+ | 0.0108 | 11.0 | 6875 | 0.2592 | 0.8780 | 0.9041 | 0.8909 | 0.9553 |
80
+ | 0.007 | 12.0 | 7500 | 0.2672 | 0.8877 | 0.9056 | 0.8965 | 0.9571 |
81
+ | 0.0048 | 13.0 | 8125 | 0.2714 | 0.8879 | 0.9089 | 0.8982 | 0.9583 |
82
+ | 0.0049 | 14.0 | 8750 | 0.2872 | 0.8873 | 0.9068 | 0.8970 | 0.9573 |
83
+ | 0.0034 | 15.0 | 9375 | 0.2915 | 0.8883 | 0.9114 | 0.8997 | 0.9577 |
84
+ | 0.0027 | 16.0 | 10000 | 0.2890 | 0.8865 | 0.9103 | 0.8983 | 0.9581 |
85
+ | 0.0028 | 17.0 | 10625 | 0.2885 | 0.8877 | 0.9085 | 0.8980 | 0.9576 |
86
+ | 0.0014 | 18.0 | 11250 | 0.2928 | 0.8860 | 0.9073 | 0.8965 | 0.9577 |
87
+ | 0.0013 | 19.0 | 11875 | 0.2963 | 0.8856 | 0.9099 | 0.8976 | 0.9576 |
88
+ | 0.001 | 20.0 | 12500 | 0.2972 | 0.8871 | 0.9100 | 0.8984 | 0.9577 |
89
 
90
 
91
  ### Framework versions
pytorch_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:a123498a940810867bed475247fd308d558c5d916cf6613e424800c6110509f0
3
  size 326138357
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:428067294de68a01e6ad5572659ee165a1bc5f85d0eadc54cbb0d054fb72830a
3
  size 326138357
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:a4c39897724c9cda6d391e41de70c86fd24f73bbe7fcf68bd625d0c7251be61c
3
  size 2671
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:98777bcefc0bbc976193ff0ec06b3f642565a966e26eb8b7ca5948002a377b07
3
  size 2671