Hugging Face's logo

fpadovani
/
de_childes_30

Fill-Mask
Transformers
Safetensors
roberta
Generated from Trainer
Model card Files Community
fpadovani commited on
Commit
f2d2758
·
verified ·
1 Parent(s): 2e64ac9

Model save

Browse files
Files changed (3) hide show
  1. README.md +59 -39
  2. generation_config.json +6 -0
  3. model.safetensors +1 -1
README.md CHANGED
@@ -3,18 +3,18 @@ library_name: transformers
3
  tags:
4
  - generated_from_trainer
5
  model-index:
6
- - name: childes_30
7
  results: []
8
  ---
9
 
10
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
11
  should probably proofread and complete it, then remove this comment. -->
12
 
13
- # childes_30
14
 
15
  This model is a fine-tuned version of [](https://huggingface.co/) on an unknown dataset.
16
  It achieves the following results on the evaluation set:
17
- - Loss: 2.8672
18
 
19
  ## Model description
20
 
@@ -33,7 +33,7 @@ More information needed
33
  ### Training hyperparameters
34
 
35
  The following hyperparameters were used during training:
36
- - learning_rate: 5e-05
37
  - train_batch_size: 16
38
  - eval_batch_size: 16
39
  - seed: 30
@@ -41,44 +41,64 @@ The following hyperparameters were used during training:
41
  - total_train_batch_size: 32
42
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
43
  - lr_scheduler_type: linear
44
- - lr_scheduler_warmup_steps: 100000
45
- - training_steps: 400000
 
46
 
47
  ### Training results
48
 
49
- | Training Loss | Epoch | Step | Validation Loss |
50
- |:-------------:|:-------:|:-----:|:---------------:|
51
- | No log | 1.6502 | 2000 | 5.8888 |
52
- | 6.5372 | 3.3003 | 4000 | 5.8455 |
53
- | 6.5372 | 4.9505 | 6000 | 5.7730 |
54
- | 5.7792 | 6.6007 | 8000 | 5.6880 |
55
- | 5.7792 | 8.2508 | 10000 | 5.1237 |
56
- | 5.147 | 9.9010 | 12000 | 4.7844 |
57
- | 5.147 | 11.5512 | 14000 | 4.5138 |
58
- | 4.4762 | 13.2013 | 16000 | 4.0074 |
59
- | 4.4762 | 14.8515 | 18000 | 3.8195 |
60
- | 3.8668 | 16.5017 | 20000 | 3.6909 |
61
- | 3.8668 | 18.1518 | 22000 | 3.5956 |
62
- | 3.6038 | 19.8020 | 24000 | 3.4481 |
63
- | 3.6038 | 21.4521 | 26000 | 3.3742 |
64
- | 3.3867 | 23.1023 | 28000 | 3.2639 |
65
- | 3.3867 | 24.7525 | 30000 | 3.1954 |
66
- | 3.2158 | 26.4026 | 32000 | 3.1180 |
67
- | 3.2158 | 28.0528 | 34000 | 3.0905 |
68
- | 3.0917 | 29.7030 | 36000 | 3.0518 |
69
- | 3.0917 | 31.3531 | 38000 | 3.0318 |
70
- | 3.0074 | 33.0033 | 40000 | 2.9881 |
71
- | 3.0074 | 34.6535 | 42000 | 2.9401 |
72
- | 2.9357 | 36.3036 | 44000 | 2.9188 |
73
- | 2.9357 | 37.9538 | 46000 | 2.9058 |
74
- | 2.8951 | 39.6040 | 48000 | 2.9200 |
75
- | 2.8951 | 41.2541 | 50000 | 2.9114 |
76
- | 2.8637 | 42.9043 | 52000 | 2.8812 |
77
- | 2.8637 | 44.5545 | 54000 | 2.8914 |
78
- | 2.8428 | 46.2046 | 56000 | 2.8537 |
79
- | 2.8428 | 47.8548 | 58000 | 2.8884 |
80
- | 2.8272 | 49.5050 | 60000 | 2.8600 |
81
- | 2.8272 | 51.1551 | 62000 | 2.8672 |
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
82
 
83
 
84
  ### Framework versions
 
3
  tags:
4
  - generated_from_trainer
5
  model-index:
6
+ - name: de_childes_30
7
  results: []
8
  ---
9
 
10
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
11
  should probably proofread and complete it, then remove this comment. -->
12
 
13
+ # de_childes_30
14
 
15
  This model is a fine-tuned version of [](https://huggingface.co/) on an unknown dataset.
16
  It achieves the following results on the evaluation set:
17
+ - Loss: 4.1942
18
 
19
  ## Model description
20
 
 
33
  ### Training hyperparameters
34
 
35
  The following hyperparameters were used during training:
36
+ - learning_rate: 0.0001
37
  - train_batch_size: 16
38
  - eval_batch_size: 16
39
  - seed: 30
 
41
  - total_train_batch_size: 32
42
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
43
  - lr_scheduler_type: linear
44
+ - lr_scheduler_warmup_steps: 40000
45
+ - training_steps: 100000
46
+ - mixed_precision_training: Native AMP
47
 
48
  ### Training results
49
 
50
+ | Training Loss | Epoch | Step | Validation Loss |
51
+ |:-------------:|:-------:|:------:|:---------------:|
52
+ | No log | 1.5021 | 2000 | 7.0910 |
53
+ | 6.9957 | 3.0041 | 4000 | 5.8267 |
54
+ | 6.9957 | 4.5062 | 6000 | 5.4651 |
55
+ | 5.213 | 6.0083 | 8000 | 5.1717 |
56
+ | 5.213 | 7.5103 | 10000 | 4.9590 |
57
+ | 4.7279 | 9.0124 | 12000 | 4.7918 |
58
+ | 4.7279 | 10.5145 | 14000 | 4.6601 |
59
+ | 4.4214 | 12.0165 | 16000 | 4.5524 |
60
+ | 4.4214 | 13.5186 | 18000 | 4.4587 |
61
+ | 4.195 | 15.0207 | 20000 | 4.3684 |
62
+ | 4.195 | 16.5227 | 22000 | 4.2900 |
63
+ | 4.0109 | 18.0248 | 24000 | 4.2277 |
64
+ | 4.0109 | 19.5268 | 26000 | 4.1726 |
65
+ | 3.8596 | 21.0289 | 28000 | 4.1288 |
66
+ | 3.8596 | 22.5310 | 30000 | 4.0892 |
67
+ | 3.7326 | 24.0330 | 32000 | 4.0589 |
68
+ | 3.7326 | 25.5351 | 34000 | 4.0279 |
69
+ | 3.6241 | 27.0372 | 36000 | 4.0029 |
70
+ | 3.6241 | 28.5392 | 38000 | 3.9903 |
71
+ | 3.5294 | 30.0413 | 40000 | 3.9732 |
72
+ | 3.5294 | 31.5434 | 42000 | 3.9699 |
73
+ | 3.4361 | 33.0454 | 44000 | 3.9619 |
74
+ | 3.4361 | 34.5475 | 46000 | 3.9608 |
75
+ | 3.3449 | 36.0496 | 48000 | 3.9666 |
76
+ | 3.3449 | 37.5516 | 50000 | 3.9692 |
77
+ | 3.2664 | 39.0537 | 52000 | 3.9745 |
78
+ | 3.2664 | 40.5558 | 54000 | 3.9862 |
79
+ | 3.1963 | 42.0578 | 56000 | 3.9972 |
80
+ | 3.1963 | 43.5757 | 58000 | 4.0077 |
81
+ | 3.1346 | 45.0777 | 60000 | 4.0204 |
82
+ | 3.1346 | 46.5798 | 62000 | 4.0262 |
83
+ | 3.0792 | 48.0819 | 64000 | 4.0405 |
84
+ | 3.0792 | 49.5839 | 66000 | 4.0522 |
85
+ | 3.0286 | 51.0860 | 68000 | 4.0700 |
86
+ | 3.0286 | 52.5881 | 70000 | 4.0754 |
87
+ | 2.9835 | 54.0901 | 72000 | 4.0929 |
88
+ | 2.9835 | 55.5922 | 74000 | 4.1004 |
89
+ | 2.942 | 57.0943 | 76000 | 4.1144 |
90
+ | 2.942 | 58.5963 | 78000 | 4.1265 |
91
+ | 2.9043 | 60.0984 | 80000 | 4.1325 |
92
+ | 2.9043 | 61.6005 | 82000 | 4.1431 |
93
+ | 2.8693 | 63.1025 | 84000 | 4.1530 |
94
+ | 2.8693 | 64.6046 | 86000 | 4.1612 |
95
+ | 2.8393 | 66.1066 | 88000 | 4.1716 |
96
+ | 2.8393 | 67.6087 | 90000 | 4.1759 |
97
+ | 2.8118 | 69.1108 | 92000 | 4.1815 |
98
+ | 2.8118 | 70.6128 | 94000 | 4.1873 |
99
+ | 2.7874 | 72.1149 | 96000 | 4.1914 |
100
+ | 2.7874 | 73.6170 | 98000 | 4.1935 |
101
+ | 2.7676 | 75.1190 | 100000 | 4.1942 |
102
 
103
 
104
  ### Framework versions
generation_config.json ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 0,
4
+ "eos_token_id": 1,
5
+ "transformers_version": "4.45.2"
6
+ }
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:c4055a337df2fd66617953610f35e72d1d1cbe9aae95395b2369475ef4520440
3
  size 51007160
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:94cc04fdb17016a865f7f7ad4ccf2cf208eb60bf2d25dd7e6628078c8a9e6833
3
  size 51007160