Ruth commited on
Commit
f87b83e
1 Parent(s): b5f2469

Upload README.md

Browse files
Files changed (1) hide show
  1. README.md +46 -25
README.md CHANGED
@@ -1,22 +1,50 @@
1
  ---
 
 
2
  license: mit
3
- tags:
4
- - generated_from_keras_callback
 
 
 
 
 
5
  model-index:
6
- - name: Ruth/gelectra-large-germeval_14
7
- results: []
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
8
  ---
9
 
10
- <!-- This model card has been generated automatically according to the information Keras had access to. You should
11
- probably proofread and complete it, then remove this comment. -->
12
 
13
- # Ruth/gelectra-large-germeval_14
14
 
15
- This model is a fine-tuned version of [deepset/gelectra-large](https://huggingface.co/deepset/gelectra-large) on an unknown dataset.
16
  It achieves the following results on the evaluation set:
17
- - Train Loss: 0.0194
18
- - Validation Loss: 0.1153
19
- - Epoch: 4
 
20
 
21
  ## Model description
22
 
@@ -35,23 +63,16 @@ More information needed
35
  ### Training hyperparameters
36
 
37
  The following hyperparameters were used during training:
38
- - optimizer: {'name': 'AdamWeightDecay', 'learning_rate': {'class_name': 'PolynomialDecay', 'config': {'initial_learning_rate': 2e-05, 'decay_steps': 12750, 'end_learning_rate': 0.0, 'power': 1.0, 'cycle': False, 'name': None}}, 'decay': 0.0, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-08, 'amsgrad': False, 'weight_decay_rate': 0.01}
39
- - training_precision: float32
40
-
41
- ### Training results
42
-
43
- | Train Loss | Validation Loss | Epoch |
44
- |:----------:|:---------------:|:-----:|
45
- | 0.1998 | 0.0986 | 0 |
46
- | 0.0753 | 0.1003 | 1 |
47
- | 0.0448 | 0.1035 | 2 |
48
- | 0.0280 | 0.1137 | 3 |
49
- | 0.0194 | 0.1153 | 4 |
50
-
51
 
52
  ### Framework versions
53
 
54
  - Transformers 4.18.0
55
- - TensorFlow 2.6.2
56
  - Datasets 1.18.0
57
  - Tokenizers 0.12.1
1
  ---
2
+ language:
3
+ - de
4
  license: mit
5
+ datasets:
6
+ - germeval_14
7
+ metrics:
8
+ - precision
9
+ - recall
10
+ - f1
11
+ - accuracy
12
  model-index:
13
+ - name: gelectra-large-germeval_14
14
+ results:
15
+ - task:
16
+ name: Token Classification
17
+ type: token-classification
18
+ dataset:
19
+ name: germeval_14
20
+ type: germeval_14
21
+ args: default
22
+ metrics:
23
+ - name: precision
24
+ type: precision
25
+ value: 0.85778125349123
26
+ - name: recall
27
+ type: recall
28
+ value: 0.8761839552664613
29
+ - name: f1
30
+ type: f1
31
+ value: 0.8668849497572543
32
+ - name: accuracy
33
+ type: accuracy
34
+ value: 0.976306430788049
35
  ---
36
 
37
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
38
+ should probably proofread and complete it, then remove this comment. -->
39
 
40
+ # gelectra-large-germeval_14
41
 
42
+ This model is a fine-tuned version of [deepset/gelectra-large](https://huggingface.co/deepset/gelectra-large) on the germeval_14 dataset.
43
  It achieves the following results on the evaluation set:
44
+ - precision: 0.8578
45
+ - recall: 0.8762
46
+ - f1: 0.8669
47
+ - accuracy: 0.9763
48
 
49
  ## Model description
50
 
63
  ### Training hyperparameters
64
 
65
  The following hyperparameters were used during training:
66
+ - num_train_epochs: 5
67
+ - train_batch_size: 8
68
+ - eval_batch_size: 8
69
+ - learning_rate: 2e-05
70
+ - weight_decay_rate: 0.01
71
+ - num_warmup_steps: 0
72
+ - fp16: True
 
 
 
 
 
 
73
 
74
  ### Framework versions
75
 
76
  - Transformers 4.18.0
 
77
  - Datasets 1.18.0
78
  - Tokenizers 0.12.1