Hugging Face's logo

trungpq
/
slac-new-appearance-upsample_replacement

Transformers
Safetensors
bert_model
Generated from Trainer
Model card Files
xet
 Community
trungpq commited on
Commit
e16ae04
·
verified ·
1 Parent(s): e1fae9c

End of training

Browse files
Files changed (4) hide show
  1. README.md +29 -29
  2. config.json +2 -2
  3. model.safetensors +1 -1
  4. training_args.bin +2 -2
README.md CHANGED
@@ -16,12 +16,12 @@ should probably proofread and complete it, then remove this comment. -->
16
 
17
  This model is a fine-tuned version of [](https://huggingface.co/) on an unknown dataset.
18
  It achieves the following results on the evaluation set:
19
- - Loss: 0.2245
20
- - Accuracy: 0.9729
21
- - F1 Macro: 0.9513
22
- - Precision Macro: 0.9555
23
- - Recall Macro: 0.9472
24
- - Total Tf: [1505, 42, 1505, 42]
25
 
26
  ## Model description
27
 
@@ -44,35 +44,35 @@ The following hyperparameters were used during training:
44
  - train_batch_size: 64
45
  - eval_batch_size: 64
46
  - seed: 42
47
- - optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
48
  - lr_scheduler_type: linear
49
- - lr_scheduler_warmup_steps: 313
50
  - num_epochs: 15
51
 
52
  ### Training results
53
 
54
- | Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 Macro | Precision Macro | Recall Macro | Total Tf |
55
- |:-------------:|:-----:|:----:|:---------------:|:--------:|:--------:|:---------------:|:------------:|:--------------------:|
56
- | 0.1053 | 1.0 | 314 | 0.1260 | 0.9651 | 0.9405 | 0.9241 | 0.9592 | [1493, 54, 1493, 54] |
57
- | 0.0389 | 2.0 | 628 | 0.0951 | 0.9735 | 0.9537 | 0.9465 | 0.9613 | [1506, 41, 1506, 41] |
58
- | 0.0216 | 3.0 | 942 | 0.1179 | 0.9735 | 0.9533 | 0.9499 | 0.9567 | [1506, 41, 1506, 41] |
59
- | 0.0162 | 4.0 | 1256 | 0.1303 | 0.9754 | 0.9562 | 0.9576 | 0.9548 | [1509, 38, 1509, 38] |
60
- | 0.0057 | 5.0 | 1570 | 0.1550 | 0.9767 | 0.9583 | 0.9626 | 0.9541 | [1511, 36, 1511, 36] |
61
- | 0.0047 | 6.0 | 1884 | 0.1564 | 0.9748 | 0.9553 | 0.9546 | 0.9560 | [1508, 39, 1508, 39] |
62
- | 0.0034 | 7.0 | 2198 | 0.1683 | 0.9729 | 0.9520 | 0.9494 | 0.9548 | [1505, 42, 1505, 42] |
63
- | 0.0073 | 8.0 | 2512 | 0.1623 | 0.9735 | 0.9524 | 0.9574 | 0.9476 | [1506, 41, 1506, 41] |
64
- | 0.0027 | 9.0 | 2826 | 0.1717 | 0.9761 | 0.9577 | 0.9556 | 0.9598 | [1510, 37, 1510, 37] |
65
- | 0.0032 | 10.0 | 3140 | 0.1760 | 0.9748 | 0.9551 | 0.9558 | 0.9544 | [1508, 39, 1508, 39] |
66
- | 0.0045 | 11.0 | 3454 | 0.1875 | 0.9741 | 0.9540 | 0.9540 | 0.9540 | [1507, 40, 1507, 40] |
67
- | 0.0014 | 12.0 | 3768 | 0.2156 | 0.9722 | 0.9499 | 0.9564 | 0.9438 | [1504, 43, 1504, 43] |
68
- | 0.0011 | 13.0 | 4082 | 0.2069 | 0.9748 | 0.9554 | 0.9534 | 0.9575 | [1508, 39, 1508, 39] |
69
- | 0.0004 | 14.0 | 4396 | 0.2206 | 0.9722 | 0.9501 | 0.9550 | 0.9453 | [1504, 43, 1504, 43] |
70
- | 0.0013 | 15.0 | 4710 | 0.2245 | 0.9729 | 0.9513 | 0.9555 | 0.9472 | [1505, 42, 1505, 42] |
71
 
72
 
73
  ### Framework versions
74
 
75
- - Transformers 4.56.1
76
- - Pytorch 2.8.0+cu128
77
- - Datasets 4.0.0
78
- - Tokenizers 0.22.0
 
16
 
17
  This model is a fine-tuned version of [](https://huggingface.co/) on an unknown dataset.
18
  It achieves the following results on the evaluation set:
19
+ - Loss: 0.2916
20
+ - Accuracy: 0.95
21
+ - F1 Macro: 0.9070
22
+ - Precision Macro: 0.9112
23
+ - Recall Macro: 0.9029
24
+ - Total Tf: [950, 50, 950, 50]
25
 
26
  ## Model description
27
 
 
44
  - train_batch_size: 64
45
  - eval_batch_size: 64
46
  - seed: 42
47
+ - optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
48
  - lr_scheduler_type: linear
49
+ - lr_scheduler_warmup_steps: 25
50
  - num_epochs: 15
51
 
52
  ### Training results
53
 
54
+ | Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 Macro | Precision Macro | Recall Macro | Total Tf |
55
+ |:-------------:|:-----:|:----:|:---------------:|:--------:|:--------:|:---------------:|:------------:|:------------------:|
56
+ | 0.545 | 1.0 | 26 | 0.4612 | 0.904 | 0.8097 | 0.8373 | 0.7884 | [904, 96, 904, 96] |
57
+ | 0.2603 | 2.0 | 52 | 0.2309 | 0.939 | 0.8773 | 0.9188 | 0.8466 | [939, 61, 939, 61] |
58
+ | 0.1088 | 3.0 | 78 | 0.2100 | 0.934 | 0.8820 | 0.8716 | 0.8934 | [934, 66, 934, 66] |
59
+ | 0.0164 | 4.0 | 104 | 0.2163 | 0.944 | 0.8979 | 0.8941 | 0.9019 | [944, 56, 944, 56] |
60
+ | 0.0064 | 5.0 | 130 | 0.2535 | 0.939 | 0.8879 | 0.8870 | 0.8889 | [939, 61, 939, 61] |
61
+ | 0.0039 | 6.0 | 156 | 0.2437 | 0.949 | 0.9049 | 0.9101 | 0.8999 | [949, 51, 949, 51] |
62
+ | 0.0027 | 7.0 | 182 | 0.2676 | 0.95 | 0.9025 | 0.9291 | 0.8805 | [950, 50, 950, 50] |
63
+ | 0.0021 | 8.0 | 208 | 0.2836 | 0.945 | 0.9000 | 0.8952 | 0.9049 | [945, 55, 945, 55] |
64
+ | 0.0018 | 9.0 | 234 | 0.2791 | 0.949 | 0.9039 | 0.9137 | 0.8949 | [949, 51, 949, 51] |
65
+ | 0.0017 | 10.0 | 260 | 0.2909 | 0.948 | 0.9002 | 0.9184 | 0.8843 | [948, 52, 948, 52] |
66
+ | 0.0015 | 11.0 | 286 | 0.2849 | 0.949 | 0.9054 | 0.9085 | 0.9023 | [949, 51, 949, 51] |
67
+ | 0.0013 | 12.0 | 312 | 0.2881 | 0.949 | 0.9054 | 0.9085 | 0.9023 | [949, 51, 949, 51] |
68
+ | 0.0013 | 13.0 | 338 | 0.2899 | 0.95 | 0.9070 | 0.9112 | 0.9029 | [950, 50, 950, 50] |
69
+ | 0.0012 | 14.0 | 364 | 0.2912 | 0.95 | 0.9070 | 0.9112 | 0.9029 | [950, 50, 950, 50] |
70
+ | 0.0012 | 15.0 | 390 | 0.2916 | 0.95 | 0.9070 | 0.9112 | 0.9029 | [950, 50, 950, 50] |
71
 
72
 
73
  ### Framework versions
74
 
75
+ - Transformers 4.52.4
76
+ - Pytorch 2.6.0+cu124
77
+ - Datasets 3.6.0
78
+ - Tokenizers 0.21.2
config.json CHANGED
@@ -2,9 +2,9 @@
2
  "architectures": [
3
  "BERTModel"
4
  ],
5
- "dtype": "float32",
6
  "model_type": "bert_model",
7
  "num_classes": 1,
8
  "pos_weight": null,
9
- "transformers_version": "4.56.1"
 
10
  }
 
2
  "architectures": [
3
  "BERTModel"
4
  ],
 
5
  "model_type": "bert_model",
6
  "num_classes": 1,
7
  "pos_weight": null,
8
+ "torch_dtype": "float32",
9
+ "transformers_version": "4.52.4"
10
  }
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:f233ab2bf11218305352f1407252ca6536d4ff62472293169837c15566d3bf4c
3
  size 437955556
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4327159d600736081f8d4d8eb8748ba28b80d129d395506a7236f6006d937410
3
  size 437955556
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:f2610007ccfbccd5c7fe7bf8518c61220d1a2cf2ab7ca8a8815fd37099d64a1e
3
- size 5841
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:44e5b3b9b59bad12a65c70c6f768ce3c1d7a76ef2dc0ec0418cba1c93e3a19a0
3
+ size 5368