BonySmoke commited on
Commit
1f4f3fd
·
1 Parent(s): cbbbb26

Add support for upper У and В. Also merge and split of не and на

Browse files
README.md CHANGED
@@ -19,11 +19,11 @@ should probably proofread and complete it, then remove this comment. -->
19
 
20
  This model is a fine-tuned version of [youscan/ukr-roberta-base](https://huggingface.co/youscan/ukr-roberta-base) on the None dataset.
21
  It achieves the following results on the evaluation set:
22
- - Loss: 0.6379
23
- - Precision: 0.5083
24
- - Recall: 0.2742
25
- - F1: 0.3562
26
- - Accuracy: 0.9330
27
 
28
  ## Model description
29
 
@@ -54,23 +54,9 @@ The following hyperparameters were used during training:
54
 
55
  | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
56
  |:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
57
- | 0.4387 | 1.0 | 3242 | 0.6192 | 0.5068 | 0.2592 | 0.3430 | 0.9320 |
58
- | 0.3599 | 2.0 | 6484 | 0.6379 | 0.5083 | 0.2742 | 0.3562 | 0.9330 |
59
 
60
- The model was evaluated on the validation dataset provided in the [UNLP 2023 Shared Task in Grammatical Error Correction for Ukrainian](https://github.com/asivokon/unlp-2023-shared-task/tree/main)
61
-
62
- ```
63
- =========== Span-Based Correction ============
64
- TP FP FN Prec Rec F0.5
65
- 433 229 794 0.6541 0.3529 0.5587
66
- ==============================================
67
-
68
-
69
- ============ Span-Based Detection ============
70
- TP FP FN Prec Rec F0.5
71
- 498 163 758 0.7534 0.3965 0.6385
72
- ==============================================
73
- ```
74
 
75
  ### Framework versions
76
 
 
19
 
20
  This model is a fine-tuned version of [youscan/ukr-roberta-base](https://huggingface.co/youscan/ukr-roberta-base) on the None dataset.
21
  It achieves the following results on the evaluation set:
22
+ - Loss: 0.6385
23
+ - Precision: 0.4995
24
+ - Recall: 0.2672
25
+ - F1: 0.3482
26
+ - Accuracy: 0.9325
27
 
28
  ## Model description
29
 
 
54
 
55
  | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
56
  |:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
57
+ | 0.441 | 1.0 | 3242 | 0.6187 | 0.5088 | 0.2602 | 0.3443 | 0.9329 |
58
+ | 0.3626 | 2.0 | 6484 | 0.6385 | 0.4995 | 0.2672 | 0.3482 | 0.9325 |
59
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
60
 
61
  ### Framework versions
62
 
config.json CHANGED
The diff for this file is too large to render. See raw diff
 
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:ea6983d41be8abf7f5883a1713f5e24412c3eef58f866bd7e75b7ab7f3fb1ff1
3
- size 518153680
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:95c479063b5c7e0871d77accbdba1ac96ca4f92a8a177e7b7f1409add6a0d7da
3
+ size 517962968
runs/Nov19_12-22-20_e5e5e07855ce/events.out.tfevents.1700396541.e5e5e07855ce.530.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:818d18fb95a3cca2b5b6d16e1c901d325558418957d783e4cca19f3e34a4e1e1
3
+ size 914163
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:d224e1fb5267cae00cf22a23a901e2d2efaede2678887a21dc77c4b7b33419cd
3
  size 4600
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2a895a9132c03cf0098361016e64e40bf6c58ea99be80269c2bf24e42a60e637
3
  size 4600