Hugging Face's logo

leonong84
/
roberta-tuned

Text Classification
Transformers
TensorFlow
roberta
generated_from_keras_callback
text-embeddings-inference
Model card Files
xet
Community
leonong84 commited on
Commit
d0a6e64
·
1 Parent(s): e2e0c6a

Upload TFRobertaForSequenceClassification

Browse files
Files changed (3) hide show
  1. README.md +30 -12
  2. config.json +9 -9
  3. tf_model.h5 +1 -1
README.md CHANGED
@@ -14,11 +14,11 @@ probably proofread and complete it, then remove this comment. -->
14
 
15
  This model is a fine-tuned version of [roberta-base](https://huggingface.co/roberta-base) on an unknown dataset.
16
  It achieves the following results on the evaluation set:
17
- - Train Loss: 0.5384
18
- - Train Accuracy: 0.7937
19
- - Validation Loss: 1.0730
20
- - Validation Accuracy: 0.6355
21
- - Epoch: 6
22
 
23
  ## Model description
24
 
@@ -44,13 +44,31 @@ The following hyperparameters were used during training:
44
 
45
  | Train Loss | Train Accuracy | Validation Loss | Validation Accuracy | Epoch |
46
  |:----------:|:--------------:|:---------------:|:-------------------:|:-----:|
47
- | 0.8699 | 0.6101 | 0.8097 | 0.6552 | 0 |
48
- | 0.7920 | 0.6621 | 0.8146 | 0.6590 | 1 |
49
- | 0.7564 | 0.6831 | 0.8207 | 0.6558 | 2 |
50
- | 0.7109 | 0.7101 | 0.8552 | 0.6490 | 3 |
51
- | 0.6609 | 0.7345 | 0.9057 | 0.6422 | 4 |
52
- | 0.6041 | 0.7638 | 0.9596 | 0.6392 | 5 |
53
- | 0.5384 | 0.7937 | 1.0730 | 0.6355 | 6 |
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
54
 
55
 
56
  ### Framework versions
 
14
 
15
  This model is a fine-tuned version of [roberta-base](https://huggingface.co/roberta-base) on an unknown dataset.
16
  It achieves the following results on the evaluation set:
17
+ - Train Loss: 0.6007
18
+ - Train Accuracy: 0.7636
19
+ - Validation Loss: 1.0870
20
+ - Validation Accuracy: 0.6223
21
+ - Epoch: 24
22
 
23
  ## Model description
24
 
 
44
 
45
  | Train Loss | Train Accuracy | Validation Loss | Validation Accuracy | Epoch |
46
  |:----------:|:--------------:|:---------------:|:-------------------:|:-----:|
47
+ | 1.0230 | 0.4760 | 0.8721 | 0.6225 | 0 |
48
+ | 0.8559 | 0.6308 | 0.8662 | 0.6365 | 1 |
49
+ | 0.8269 | 0.6457 | 0.8625 | 0.6382 | 2 |
50
+ | 0.8123 | 0.6572 | 0.8630 | 0.6405 | 3 |
51
+ | 0.8068 | 0.6601 | 0.8655 | 0.6417 | 4 |
52
+ | 0.7953 | 0.6642 | 0.8689 | 0.6415 | 5 |
53
+ | 0.7881 | 0.6670 | 0.8824 | 0.6432 | 6 |
54
+ | 0.7816 | 0.6731 | 0.8761 | 0.6440 | 7 |
55
+ | 0.7683 | 0.6779 | 0.8872 | 0.6463 | 8 |
56
+ | 0.7630 | 0.6823 | 0.9060 | 0.6432 | 9 |
57
+ | 0.7562 | 0.6859 | 0.8930 | 0.6453 | 10 |
58
+ | 0.7467 | 0.6908 | 0.9087 | 0.6440 | 11 |
59
+ | 0.7380 | 0.6975 | 0.9209 | 0.6378 | 12 |
60
+ | 0.7302 | 0.6996 | 0.9328 | 0.6375 | 13 |
61
+ | 0.7167 | 0.7047 | 0.9493 | 0.6367 | 14 |
62
+ | 0.7081 | 0.7135 | 0.9467 | 0.6392 | 15 |
63
+ | 0.6989 | 0.7141 | 0.9621 | 0.6370 | 16 |
64
+ | 0.6895 | 0.7211 | 0.9913 | 0.6352 | 17 |
65
+ | 0.6805 | 0.7253 | 0.9848 | 0.6270 | 18 |
66
+ | 0.6657 | 0.7325 | 1.0282 | 0.6327 | 19 |
67
+ | 0.6511 | 0.7394 | 1.0346 | 0.6265 | 20 |
68
+ | 0.6385 | 0.7458 | 1.0475 | 0.6280 | 21 |
69
+ | 0.6278 | 0.7514 | 1.0381 | 0.6240 | 22 |
70
+ | 0.6119 | 0.7604 | 1.0727 | 0.6258 | 23 |
71
+ | 0.6007 | 0.7636 | 1.0870 | 0.6223 | 24 |
72
 
73
 
74
  ### Framework versions
config.json CHANGED
@@ -3,24 +3,24 @@
3
  "architectures": [
4
  "RobertaForSequenceClassification"
5
  ],
6
- "attention_probs_dropout_prob": 0.1,
7
  "bos_token_id": 0,
8
- "classifier_dropout": null,
9
  "eos_token_id": 2,
10
  "hidden_act": "gelu",
11
- "hidden_dropout_prob": 0.1,
12
  "hidden_size": 768,
13
  "id2label": {
14
- "0": "LABEL_0",
15
- "1": "LABEL_1",
16
- "2": "LABEL_2"
17
  },
18
  "initializer_range": 0.02,
19
  "intermediate_size": 3072,
20
  "label2id": {
21
- "LABEL_0": 0,
22
- "LABEL_1": 1,
23
- "LABEL_2": 2
24
  },
25
  "layer_norm_eps": 1e-05,
26
  "max_position_embeddings": 514,
 
3
  "architectures": [
4
  "RobertaForSequenceClassification"
5
  ],
6
+ "attention_probs_dropout_prob": 0.3,
7
  "bos_token_id": 0,
8
+ "classifier_dropout": 0.5,
9
  "eos_token_id": 2,
10
  "hidden_act": "gelu",
11
+ "hidden_dropout_prob": 0.15,
12
  "hidden_size": 768,
13
  "id2label": {
14
+ "0": "negative",
15
+ "1": "neutral",
16
+ "2": "positive"
17
  },
18
  "initializer_range": 0.02,
19
  "intermediate_size": 3072,
20
  "label2id": {
21
+ "negative": 0,
22
+ "neutral": 1,
23
+ "positive": 2
24
  },
25
  "layer_norm_eps": 1e-05,
26
  "max_position_embeddings": 514,
tf_model.h5 CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:ccbd4d8f1ea3bb1926b2dede6d46600e8d516dc47141eadeb772cc2fc7a4db3d
3
  size 498881344
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0a8315d59662395ea10242f167aa245a430b36f31c99d171d3b6a2197f1dd968
3
  size 498881344