Hugging Face's logo

Sahyus
/
roberta-large-squad2-finetuned-dtc

Question Answering
Transformers
TensorFlow
roberta
generated_from_keras_callback
Inference Endpoints
Model card Files Community
Sahyus commited on
Commit
2f7a9ae
路
1 Parent(s): f5fe9de

Upload TFRobertaForQuestionAnswering

Browse files
Files changed (3) hide show
  1. README.md +98 -0
  2. config.json +29 -0
  3. tf_model.h5 +3 -0
README.md ADDED
@@ -0,0 +1,98 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: cc-by-4.0
3
+ base_model: deepset/roberta-large-squad2
4
+ tags:
5
+ - generated_from_keras_callback
6
+ model-index:
7
+ - name: roberta-large-squad2-finetuned-dtc
8
+ results: []
9
+ ---
10
+
11
+ <!-- This model card has been generated automatically according to the information Keras had access to. You should
12
+ probably proofread and complete it, then remove this comment. -->
13
+
14
+ # roberta-large-squad2-finetuned-dtc
15
+
16
+ This model is a fine-tuned version of [deepset/roberta-large-squad2](https://huggingface.co/deepset/roberta-large-squad2) on an unknown dataset.
17
+ It achieves the following results on the evaluation set:
18
+ - Train Loss: 1.9389
19
+ - Train End Logits Loss: 1.1432
20
+ - Train Start Logits Loss: 0.7957
21
+ - Train End Logits Acc: 0.7392
22
+ - Train Start Logits Acc: 0.8093
23
+ - Validation Loss: 3.7259
24
+ - Validation End Logits Loss: 1.8885
25
+ - Validation Start Logits Loss: 1.8374
26
+ - Validation End Logits Acc: 0.6312
27
+ - Validation Start Logits Acc: 0.7221
28
+ - Epoch: 36
29
+
30
+ ## Model description
31
+
32
+ More information needed
33
+
34
+ ## Intended uses & limitations
35
+
36
+ More information needed
37
+
38
+ ## Training and evaluation data
39
+
40
+ More information needed
41
+
42
+ ## Training procedure
43
+
44
+ ### Training hyperparameters
45
+
46
+ The following hyperparameters were used during training:
47
+ - optimizer: {'name': 'AdamWeightDecay', 'learning_rate': {'module': 'keras.optimizers.schedules', 'class_name': 'PolynomialDecay', 'config': {'initial_learning_rate': 2.4e-05, 'decay_steps': 21400, 'end_learning_rate': 0.0, 'power': 1.0, 'cycle': False, 'name': None}, 'registered_name': None}, 'decay': 0.0, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-08, 'amsgrad': False, 'weight_decay_rate': 0.03}
48
+ - training_precision: float32
49
+
50
+ ### Training results
51
+
52
+ | Train Loss | Train End Logits Loss | Train Start Logits Loss | Train End Logits Acc | Train Start Logits Acc | Validation Loss | Validation End Logits Loss | Validation Start Logits Loss | Validation End Logits Acc | Validation Start Logits Acc | Epoch |
53
+ |:----------:|:---------------------:|:-----------------------:|:--------------------:|:----------------------:|:---------------:|:--------------------------:|:----------------------------:|:-------------------------:|:---------------------------:|:-----:|
54
+ | 5.8888 | 3.0592 | 2.8296 | 0.5456 | 0.5406 | 4.8715 | 2.6861 | 2.1854 | 0.6130 | 0.6182 | 0 |
55
+ | 5.0000 | 2.7063 | 2.2937 | 0.5809 | 0.5762 | 4.0680 | 2.3593 | 1.7087 | 0.6208 | 0.6000 | 1 |
56
+ | 4.7529 | 2.5952 | 2.1576 | 0.5929 | 0.5862 | 4.5767 | 2.7450 | 1.8317 | 0.6208 | 0.6156 | 2 |
57
+ | 4.6181 | 2.5511 | 2.0670 | 0.5984 | 0.5873 | 3.9828 | 2.4125 | 1.5703 | 0.6208 | 0.6052 | 3 |
58
+ | 4.4766 | 2.4920 | 1.9846 | 0.6019 | 0.5946 | 3.7862 | 2.2460 | 1.5402 | 0.6208 | 0.5922 | 4 |
59
+ | 4.5692 | 2.5720 | 1.9972 | 0.6081 | 0.6066 | 3.6069 | 2.1558 | 1.4511 | 0.6208 | 0.6052 | 5 |
60
+ | 4.3098 | 2.4382 | 1.8716 | 0.6016 | 0.5987 | 3.2741 | 1.9275 | 1.3466 | 0.6208 | 0.6364 | 6 |
61
+ | 3.8906 | 2.2240 | 1.6666 | 0.6165 | 0.6256 | 3.3856 | 1.9718 | 1.4138 | 0.6156 | 0.6052 | 7 |
62
+ | 3.7711 | 2.1773 | 1.5939 | 0.6154 | 0.6317 | 3.4381 | 1.7916 | 1.6465 | 0.6182 | 0.4805 | 8 |
63
+ | 3.6331 | 2.1149 | 1.5182 | 0.6177 | 0.6460 | 3.0055 | 1.6855 | 1.3200 | 0.5273 | 0.6338 | 9 |
64
+ | 3.4683 | 2.0212 | 1.4471 | 0.6168 | 0.6545 | 3.3422 | 1.7875 | 1.5547 | 0.4805 | 0.5325 | 10 |
65
+ | 3.3695 | 1.9567 | 1.4129 | 0.6183 | 0.6618 | 2.8283 | 1.5488 | 1.2795 | 0.5455 | 0.6286 | 11 |
66
+ | 3.3125 | 1.9344 | 1.3781 | 0.6215 | 0.6647 | 2.7086 | 1.5124 | 1.1962 | 0.5636 | 0.6338 | 12 |
67
+ | 3.2580 | 1.9282 | 1.3298 | 0.6390 | 0.6852 | 3.0502 | 1.7520 | 1.2982 | 0.6156 | 0.6623 | 13 |
68
+ | 3.2814 | 1.9478 | 1.3336 | 0.6294 | 0.6711 | 2.5437 | 1.4591 | 1.0846 | 0.5948 | 0.6727 | 14 |
69
+ | 3.1027 | 1.8305 | 1.2721 | 0.6370 | 0.6893 | 3.0537 | 1.6897 | 1.3640 | 0.5481 | 0.5922 | 15 |
70
+ | 2.7670 | 1.6628 | 1.1042 | 0.6583 | 0.7217 | 2.4372 | 1.3791 | 1.0581 | 0.6519 | 0.6961 | 16 |
71
+ | 2.7880 | 1.6975 | 1.0905 | 0.6583 | 0.7339 | 2.2441 | 1.2735 | 0.9706 | 0.7039 | 0.7299 | 17 |
72
+ | 2.7786 | 1.6524 | 1.1262 | 0.6606 | 0.7225 | 2.6408 | 1.4267 | 1.2141 | 0.6701 | 0.6831 | 18 |
73
+ | 2.4685 | 1.4862 | 0.9823 | 0.6741 | 0.7447 | 2.7726 | 1.5947 | 1.1779 | 0.6338 | 0.6909 | 19 |
74
+ | 2.4204 | 1.4523 | 0.9682 | 0.6814 | 0.7538 | 2.1115 | 1.1877 | 0.9238 | 0.7429 | 0.7714 | 20 |
75
+ | 2.2158 | 1.3472 | 0.8686 | 0.6939 | 0.7707 | 2.2647 | 1.2382 | 1.0266 | 0.7143 | 0.7532 | 21 |
76
+ | 2.0138 | 1.2461 | 0.7676 | 0.7109 | 0.7994 | 2.1425 | 1.1617 | 0.9808 | 0.7455 | 0.7558 | 22 |
77
+ | 2.0038 | 1.2585 | 0.7453 | 0.7129 | 0.8008 | 1.8952 | 0.9984 | 0.8968 | 0.7688 | 0.7558 | 23 |
78
+ | 1.8391 | 1.1600 | 0.6791 | 0.7231 | 0.8186 | 2.4242 | 1.3208 | 1.1034 | 0.7013 | 0.7039 | 24 |
79
+ | 1.7792 | 1.1060 | 0.6732 | 0.7389 | 0.8248 | 1.8800 | 1.0211 | 0.8588 | 0.7792 | 0.7818 | 25 |
80
+ | 1.6690 | 1.0636 | 0.6054 | 0.7462 | 0.8367 | 2.2503 | 1.2198 | 1.0305 | 0.7325 | 0.7506 | 26 |
81
+ | 1.6197 | 1.0327 | 0.5870 | 0.7591 | 0.8452 | 1.9393 | 0.9581 | 0.9812 | 0.7974 | 0.8052 | 27 |
82
+ | 1.5335 | 0.9795 | 0.5540 | 0.7652 | 0.8595 | 2.2046 | 1.1750 | 1.0296 | 0.7688 | 0.7870 | 28 |
83
+ | 1.4563 | 0.9314 | 0.5249 | 0.7751 | 0.8621 | 1.9638 | 1.0204 | 0.9434 | 0.7403 | 0.7792 | 29 |
84
+ | 1.3903 | 0.9049 | 0.4854 | 0.7772 | 0.8683 | 2.2657 | 1.1569 | 1.1088 | 0.7636 | 0.7896 | 30 |
85
+ | 1.3534 | 0.8813 | 0.4720 | 0.7859 | 0.8744 | 1.9620 | 0.9779 | 0.9840 | 0.7688 | 0.7740 | 31 |
86
+ | 1.4848 | 0.9444 | 0.5405 | 0.7684 | 0.8563 | 2.3368 | 1.1941 | 1.1427 | 0.7299 | 0.7688 | 32 |
87
+ | 1.5092 | 0.9534 | 0.5558 | 0.7550 | 0.8461 | 2.1233 | 1.0956 | 1.0277 | 0.7610 | 0.7740 | 33 |
88
+ | 1.4016 | 0.8789 | 0.5227 | 0.7751 | 0.8624 | 2.4886 | 1.2593 | 1.2294 | 0.7403 | 0.7844 | 34 |
89
+ | 1.8007 | 1.0509 | 0.7498 | 0.7520 | 0.8183 | 2.5730 | 1.3045 | 1.2686 | 0.7195 | 0.7481 | 35 |
90
+ | 1.9389 | 1.1432 | 0.7957 | 0.7392 | 0.8093 | 3.7259 | 1.8885 | 1.8374 | 0.6312 | 0.7221 | 36 |
91
+
92
+
93
+ ### Framework versions
94
+
95
+ - Transformers 4.36.2
96
+ - TensorFlow 2.14.0
97
+ - Datasets 2.16.1
98
+ - Tokenizers 0.15.0
config.json ADDED
@@ -0,0 +1,29 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "deepset/roberta-large-squad2",
3
+ "architectures": [
4
+ "RobertaForQuestionAnswering"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.4,
7
+ "bos_token_id": 0,
8
+ "classifier_dropout": null,
9
+ "eos_token_id": 2,
10
+ "gradient_checkpointing": false,
11
+ "hidden_act": "gelu",
12
+ "hidden_dropout_prob": 0.4,
13
+ "hidden_size": 1024,
14
+ "initializer_range": 0.02,
15
+ "intermediate_size": 4096,
16
+ "language": "english",
17
+ "layer_norm_eps": 1e-15,
18
+ "max_position_embeddings": 514,
19
+ "model_type": "roberta",
20
+ "name": "Roberta",
21
+ "num_attention_heads": 16,
22
+ "num_hidden_layers": 24,
23
+ "pad_token_id": 1,
24
+ "position_embedding_type": "absolute",
25
+ "transformers_version": "4.36.2",
26
+ "type_vocab_size": 1,
27
+ "use_cache": true,
28
+ "vocab_size": 50265
29
+ }
tf_model.h5 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:883577b86090567cfc2ef514cc658deaabf6de26e723a90bef9c9dc72d6eb63c
3
+ size 1417799680