Hugging Face's logo

leocooo
/
v2-camembert-ner-job-ads

Token Classification
Transformers
Safetensors
camembert
Generated from Trainer
Model card Files
xet
 Community
leocooo commited on
Commit
b3d209c
·
verified ·
1 Parent(s): 1062bd2

leocooo/v2-camembert-ner-job-ads

Browse files
Files changed (3) hide show
  1. README.md +30 -24
  2. model.safetensors +1 -1
  3. training_args.bin +1 -1
README.md CHANGED
@@ -20,11 +20,11 @@ should probably proofread and complete it, then remove this comment. -->
20
 
21
  This model is a fine-tuned version of [camembert-base](https://huggingface.co/camembert-base) on the None dataset.
22
  It achieves the following results on the evaluation set:
23
- - Loss: 0.5448
24
- - Precision: 0.1421
25
- - Recall: 0.5632
26
- - F1: 0.2269
27
- - F2: 0.3536
28
 
29
  ## Model description
30
 
@@ -49,31 +49,37 @@ The following hyperparameters were used during training:
49
  - seed: 42
50
  - optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
51
  - lr_scheduler_type: cosine
52
- - lr_scheduler_warmup_steps: 0.1
53
  - num_epochs: 30
54
 
55
  ### Training results
56
 
57
  | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | F2 |
58
  |:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:------:|
59
- | 3.0988 | 1.0 | 121 | 2.9653 | 0.0095 | 0.2909 | 0.0184 | 0.0421 |
60
- | 2.6062 | 2.0 | 242 | 2.1285 | 0.0152 | 0.2569 | 0.0288 | 0.0616 |
61
- | 1.8180 | 3.0 | 363 | 1.6229 | 0.0226 | 0.2101 | 0.0408 | 0.0789 |
62
- | 1.4669 | 4.0 | 484 | 1.2934 | 0.0352 | 0.2480 | 0.0617 | 0.1123 |
63
- | 1.1606 | 5.0 | 605 | 1.1329 | 0.0390 | 0.2738 | 0.0683 | 0.1242 |
64
- | 1.0279 | 6.0 | 726 | 0.9946 | 0.0712 | 0.3979 | 0.1208 | 0.2075 |
65
- | 0.9051 | 7.0 | 847 | 0.9019 | 0.0806 | 0.3962 | 0.1339 | 0.2222 |
66
- | 0.7670 | 8.0 | 968 | 0.8294 | 0.0600 | 0.3529 | 0.1026 | 0.1786 |
67
- | 0.6855 | 9.0 | 1089 | 0.7605 | 0.0956 | 0.4628 | 0.1585 | 0.2618 |
68
- | 0.5873 | 10.0 | 1210 | 0.7084 | 0.0996 | 0.4746 | 0.1646 | 0.2707 |
69
- | 0.5381 | 11.0 | 1331 | 0.6769 | 0.1147 | 0.4981 | 0.1864 | 0.2985 |
70
- | 0.4650 | 12.0 | 1452 | 0.6216 | 0.1155 | 0.5169 | 0.1889 | 0.3050 |
71
- | 0.4258 | 13.0 | 1573 | 0.6069 | 0.1260 | 0.5321 | 0.2038 | 0.3236 |
72
- | 0.3961 | 14.0 | 1694 | 0.5895 | 0.1469 | 0.5526 | 0.2321 | 0.3560 |
73
- | 0.3511 | 15.0 | 1815 | 0.5674 | 0.1439 | 0.5649 | 0.2294 | 0.3564 |
74
- | 0.3246 | 16.0 | 1936 | 0.5544 | 0.1358 | 0.5408 | 0.2171 | 0.3388 |
75
- | 0.3099 | 17.0 | 2057 | 0.5599 | 0.1402 | 0.5607 | 0.2244 | 0.3505 |
76
- | 0.2804 | 18.0 | 2178 | 0.5448 | 0.1421 | 0.5632 | 0.2269 | 0.3536 |
 
 
 
 
 
 
77
 
78
 
79
  ### Framework versions
 
20
 
21
  This model is a fine-tuned version of [camembert-base](https://huggingface.co/camembert-base) on the None dataset.
22
  It achieves the following results on the evaluation set:
23
+ - Loss: 0.5342
24
+ - Precision: 0.3948
25
+ - Recall: 0.7311
26
+ - F1: 0.5127
27
+ - F2: 0.6247
28
 
29
  ## Model description
30
 
 
49
  - seed: 42
50
  - optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
51
  - lr_scheduler_type: cosine
52
+ - lr_scheduler_warmup_steps: 741
53
  - num_epochs: 30
54
 
55
  ### Training results
56
 
57
  | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | F2 |
58
  |:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:------:|
59
+ | 3.0518 | 1.0 | 121 | 2.8715 | 0.0 | 0.0 | 0.0 | 0 |
60
+ | 2.5717 | 2.0 | 242 | 2.2493 | 0.3210 | 0.1778 | 0.2288 | 0.1952 |
61
+ | 2.0235 | 3.0 | 363 | 1.8607 | 0.2572 | 0.4860 | 0.3364 | 0.4126 |
62
+ | 1.7860 | 4.0 | 484 | 1.6046 | 0.2161 | 0.5011 | 0.3019 | 0.3965 |
63
+ | 1.4654 | 5.0 | 605 | 1.3541 | 0.2824 | 0.6345 | 0.3908 | 0.5078 |
64
+ | 1.2812 | 6.0 | 726 | 1.1570 | 0.2624 | 0.6150 | 0.3678 | 0.4847 |
65
+ | 1.0853 | 7.0 | 847 | 0.9783 | 0.2889 | 0.6503 | 0.4001 | 0.5202 |
66
+ | 0.8719 | 8.0 | 968 | 0.8576 | 0.2611 | 0.6552 | 0.3734 | 0.5032 |
67
+ | 0.7604 | 9.0 | 1089 | 0.7621 | 0.3454 | 0.6951 | 0.4615 | 0.5781 |
68
+ | 0.6377 | 10.0 | 1210 | 0.6978 | 0.3358 | 0.6871 | 0.4511 | 0.5682 |
69
+ | 0.5671 | 11.0 | 1331 | 0.6360 | 0.3287 | 0.6903 | 0.4454 | 0.5658 |
70
+ | 0.4867 | 12.0 | 1452 | 0.6012 | 0.3483 | 0.7055 | 0.4663 | 0.5854 |
71
+ | 0.4340 | 13.0 | 1573 | 0.5768 | 0.3330 | 0.7059 | 0.4525 | 0.5767 |
72
+ | 0.4057 | 14.0 | 1694 | 0.5495 | 0.3528 | 0.6907 | 0.4670 | 0.5797 |
73
+ | 0.3688 | 15.0 | 1815 | 0.5370 | 0.3739 | 0.7218 | 0.4926 | 0.6085 |
74
+ | 0.3279 | 16.0 | 1936 | 0.5372 | 0.3589 | 0.7142 | 0.4777 | 0.5961 |
75
+ | 0.3076 | 17.0 | 2057 | 0.5402 | 0.3702 | 0.7285 | 0.4910 | 0.6104 |
76
+ | 0.2874 | 18.0 | 2178 | 0.5484 | 0.3994 | 0.7205 | 0.5139 | 0.6207 |
77
+ | 0.2774 | 19.0 | 2299 | 0.5410 | 0.4021 | 0.7315 | 0.5190 | 0.6285 |
78
+ | 0.2600 | 20.0 | 2420 | 0.5429 | 0.4011 | 0.7340 | 0.5188 | 0.6295 |
79
+ | 0.2505 | 21.0 | 2541 | 0.5525 | 0.4175 | 0.7321 | 0.5317 | 0.6362 |
80
+ | 0.2399 | 22.0 | 2662 | 0.5436 | 0.4165 | 0.7323 | 0.5310 | 0.6359 |
81
+ | 0.2298 | 23.0 | 2783 | 0.5315 | 0.4045 | 0.7326 | 0.5212 | 0.6303 |
82
+ | 0.2266 | 24.0 | 2904 | 0.5342 | 0.3948 | 0.7311 | 0.5127 | 0.6247 |
83
 
84
 
85
  ### Framework versions
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:f47dda8a3e7a975df9367501f4afed5bcc97457441d1743ebc624c4b4503159c
3
  size 440220084
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:59b13e41b246c16dd9f623cc0ff5f3d1d4cd91de00658259bb881a41ae8a7cdc
3
  size 440220084
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:e2f3d32afd30e03453cdcea87a58b52980c05c36a624928bc36ca8a60a9c7147
3
  size 5201
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5dc3c1c2d4f3f20a99d7eb8ad5a896ab48bc0e62b45ab5ea9638d0675d025907
3
  size 5201