End of training
Browse files- README.md +95 -0
- config.json +62 -0
- merges.txt +0 -0
- pytorch_model.bin +3 -0
- special_tokens_map.json +15 -0
- tokenizer.json +0 -0
- tokenizer_config.json +16 -0
- training_args.bin +3 -0
- vocab.json +0 -0
README.md
ADDED
@@ -0,0 +1,95 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: apache-2.0
|
3 |
+
base_model: allenai/longformer-base-4096
|
4 |
+
tags:
|
5 |
+
- generated_from_trainer
|
6 |
+
metrics:
|
7 |
+
- accuracy
|
8 |
+
- f1
|
9 |
+
- precision
|
10 |
+
- recall
|
11 |
+
model-index:
|
12 |
+
- name: framing_classification_longformer_30_augmented_multi_undersampled_second
|
13 |
+
results: []
|
14 |
+
---
|
15 |
+
|
16 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
17 |
+
should probably proofread and complete it, then remove this comment. -->
|
18 |
+
|
19 |
+
# framing_classification_longformer_30_augmented_multi_undersampled_second
|
20 |
+
|
21 |
+
This model is a fine-tuned version of [allenai/longformer-base-4096](https://huggingface.co/allenai/longformer-base-4096) on an unknown dataset.
|
22 |
+
It achieves the following results on the evaluation set:
|
23 |
+
- Loss: 0.3928
|
24 |
+
- Accuracy: 0.9452
|
25 |
+
- F1: 0.9479
|
26 |
+
- Precision: 0.9496
|
27 |
+
- Recall: 0.9468
|
28 |
+
|
29 |
+
## Model description
|
30 |
+
|
31 |
+
More information needed
|
32 |
+
|
33 |
+
## Intended uses & limitations
|
34 |
+
|
35 |
+
More information needed
|
36 |
+
|
37 |
+
## Training and evaluation data
|
38 |
+
|
39 |
+
More information needed
|
40 |
+
|
41 |
+
## Training procedure
|
42 |
+
|
43 |
+
### Training hyperparameters
|
44 |
+
|
45 |
+
The following hyperparameters were used during training:
|
46 |
+
- learning_rate: 2e-05
|
47 |
+
- train_batch_size: 1
|
48 |
+
- eval_batch_size: 1
|
49 |
+
- seed: 42
|
50 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
51 |
+
- lr_scheduler_type: linear
|
52 |
+
- num_epochs: 30
|
53 |
+
|
54 |
+
### Training results
|
55 |
+
|
56 |
+
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 | Precision | Recall |
|
57 |
+
|:-------------:|:-----:|:------:|:---------------:|:--------:|:------:|:---------:|:------:|
|
58 |
+
| 0.7779 | 1.0 | 11979 | 0.7469 | 0.8671 | 0.8587 | 0.8727 | 0.8604 |
|
59 |
+
| 0.5767 | 2.0 | 23958 | 0.6461 | 0.9025 | 0.8972 | 0.9102 | 0.8975 |
|
60 |
+
| 0.4894 | 3.0 | 35937 | 0.5479 | 0.9218 | 0.9208 | 0.9254 | 0.9200 |
|
61 |
+
| 0.5261 | 4.0 | 47916 | 0.4513 | 0.9238 | 0.9238 | 0.9294 | 0.9212 |
|
62 |
+
| 0.4559 | 5.0 | 59895 | 0.4831 | 0.9225 | 0.9219 | 0.9285 | 0.9205 |
|
63 |
+
| 0.4335 | 6.0 | 71874 | 0.6019 | 0.9138 | 0.9132 | 0.9211 | 0.9123 |
|
64 |
+
| 0.4899 | 7.0 | 83853 | 0.4653 | 0.9299 | 0.9295 | 0.9356 | 0.9264 |
|
65 |
+
| 0.5907 | 8.0 | 95832 | 0.4006 | 0.9419 | 0.9437 | 0.9494 | 0.9403 |
|
66 |
+
| 0.5436 | 9.0 | 107811 | 0.4034 | 0.9472 | 0.9494 | 0.9525 | 0.9479 |
|
67 |
+
| 0.3722 | 10.0 | 119790 | 0.4336 | 0.9385 | 0.9391 | 0.9433 | 0.9386 |
|
68 |
+
| 0.443 | 11.0 | 131769 | 0.4519 | 0.9332 | 0.9365 | 0.9375 | 0.9363 |
|
69 |
+
| 0.3867 | 12.0 | 143748 | 0.4657 | 0.9232 | 0.9268 | 0.9242 | 0.9301 |
|
70 |
+
| 0.3362 | 13.0 | 155727 | 0.4739 | 0.9305 | 0.9336 | 0.9327 | 0.9353 |
|
71 |
+
| 0.3106 | 14.0 | 167706 | 0.4533 | 0.9399 | 0.9404 | 0.9445 | 0.9386 |
|
72 |
+
| 0.3265 | 15.0 | 179685 | 0.3928 | 0.9452 | 0.9479 | 0.9496 | 0.9468 |
|
73 |
+
| 0.2352 | 16.0 | 191664 | 0.4770 | 0.9238 | 0.9279 | 0.9251 | 0.9320 |
|
74 |
+
| 0.2962 | 17.0 | 203643 | 0.4575 | 0.9345 | 0.9383 | 0.9390 | 0.9377 |
|
75 |
+
| 0.3453 | 18.0 | 215622 | 0.6071 | 0.9152 | 0.9203 | 0.9174 | 0.9253 |
|
76 |
+
| 0.1919 | 19.0 | 227601 | 0.5484 | 0.9238 | 0.9280 | 0.9260 | 0.9313 |
|
77 |
+
| 0.1667 | 20.0 | 239580 | 0.4992 | 0.9325 | 0.9358 | 0.9359 | 0.9363 |
|
78 |
+
| 0.1111 | 21.0 | 251559 | 0.5727 | 0.9272 | 0.9319 | 0.9292 | 0.9358 |
|
79 |
+
| 0.1705 | 22.0 | 263538 | 0.5028 | 0.9345 | 0.9392 | 0.9385 | 0.9401 |
|
80 |
+
| 0.1745 | 23.0 | 275517 | 0.5244 | 0.9345 | 0.9389 | 0.9387 | 0.9394 |
|
81 |
+
| 0.1366 | 24.0 | 287496 | 0.5165 | 0.9345 | 0.9393 | 0.9370 | 0.9423 |
|
82 |
+
| 0.1031 | 25.0 | 299475 | 0.5479 | 0.9339 | 0.9382 | 0.9368 | 0.9399 |
|
83 |
+
| 0.0851 | 26.0 | 311454 | 0.6036 | 0.9345 | 0.9386 | 0.9378 | 0.9396 |
|
84 |
+
| 0.125 | 27.0 | 323433 | 0.5253 | 0.9385 | 0.9424 | 0.9419 | 0.9430 |
|
85 |
+
| 0.1193 | 28.0 | 335412 | 0.5650 | 0.9352 | 0.9394 | 0.9376 | 0.9415 |
|
86 |
+
| 0.067 | 29.0 | 347391 | 0.5818 | 0.9352 | 0.9395 | 0.9376 | 0.9418 |
|
87 |
+
| 0.0219 | 30.0 | 359370 | 0.5994 | 0.9359 | 0.9402 | 0.9385 | 0.9423 |
|
88 |
+
|
89 |
+
|
90 |
+
### Framework versions
|
91 |
+
|
92 |
+
- Transformers 4.32.0.dev0
|
93 |
+
- Pytorch 2.0.1
|
94 |
+
- Datasets 2.14.4
|
95 |
+
- Tokenizers 0.13.3
|
config.json
ADDED
@@ -0,0 +1,62 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "allenai/longformer-base-4096",
|
3 |
+
"architectures": [
|
4 |
+
"LongformerForSequenceClassification"
|
5 |
+
],
|
6 |
+
"attention_mode": "longformer",
|
7 |
+
"attention_probs_dropout_prob": 0.1,
|
8 |
+
"attention_window": [
|
9 |
+
512,
|
10 |
+
512,
|
11 |
+
512,
|
12 |
+
512,
|
13 |
+
512,
|
14 |
+
512,
|
15 |
+
512,
|
16 |
+
512,
|
17 |
+
512,
|
18 |
+
512,
|
19 |
+
512,
|
20 |
+
512
|
21 |
+
],
|
22 |
+
"bos_token_id": 0,
|
23 |
+
"eos_token_id": 2,
|
24 |
+
"gradient_checkpointing": false,
|
25 |
+
"hidden_act": "gelu",
|
26 |
+
"hidden_dropout_prob": 0.1,
|
27 |
+
"hidden_size": 768,
|
28 |
+
"id2label": {
|
29 |
+
"0": "effects",
|
30 |
+
"1": "politics",
|
31 |
+
"2": "social progress",
|
32 |
+
"3": "economic competitiveness",
|
33 |
+
"4": "morality",
|
34 |
+
"5": "other",
|
35 |
+
"6": "science"
|
36 |
+
},
|
37 |
+
"ignore_attention_mask": false,
|
38 |
+
"initializer_range": 0.02,
|
39 |
+
"intermediate_size": 3072,
|
40 |
+
"label2id": {
|
41 |
+
"economic competitiveness": 3,
|
42 |
+
"effects": 0,
|
43 |
+
"morality": 4,
|
44 |
+
"other": 5,
|
45 |
+
"politics": 1,
|
46 |
+
"science": 6,
|
47 |
+
"social progress": 2
|
48 |
+
},
|
49 |
+
"layer_norm_eps": 1e-05,
|
50 |
+
"max_position_embeddings": 4098,
|
51 |
+
"model_type": "longformer",
|
52 |
+
"num_attention_heads": 12,
|
53 |
+
"num_hidden_layers": 12,
|
54 |
+
"onnx_export": false,
|
55 |
+
"pad_token_id": 1,
|
56 |
+
"problem_type": "single_label_classification",
|
57 |
+
"sep_token_id": 2,
|
58 |
+
"torch_dtype": "float32",
|
59 |
+
"transformers_version": "4.32.0.dev0",
|
60 |
+
"type_vocab_size": 1,
|
61 |
+
"vocab_size": 50265
|
62 |
+
}
|
merges.txt
ADDED
The diff for this file is too large to render.
See raw diff
|
|
pytorch_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:cbe16559881733bce35e2185e5e132ebbeb413692dee6f6a784206e466b520fc
|
3 |
+
size 594754641
|
special_tokens_map.json
ADDED
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"bos_token": "<s>",
|
3 |
+
"cls_token": "<s>",
|
4 |
+
"eos_token": "</s>",
|
5 |
+
"mask_token": {
|
6 |
+
"content": "<mask>",
|
7 |
+
"lstrip": true,
|
8 |
+
"normalized": false,
|
9 |
+
"rstrip": false,
|
10 |
+
"single_word": false
|
11 |
+
},
|
12 |
+
"pad_token": "<pad>",
|
13 |
+
"sep_token": "</s>",
|
14 |
+
"unk_token": "<unk>"
|
15 |
+
}
|
tokenizer.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|
tokenizer_config.json
ADDED
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"add_prefix_space": false,
|
3 |
+
"bos_token": "<s>",
|
4 |
+
"clean_up_tokenization_spaces": true,
|
5 |
+
"cls_token": "<s>",
|
6 |
+
"eos_token": "</s>",
|
7 |
+
"errors": "replace",
|
8 |
+
"mask_token": "<mask>",
|
9 |
+
"max_length": 4096,
|
10 |
+
"model_max_length": 4096,
|
11 |
+
"pad_token": "<pad>",
|
12 |
+
"sep_token": "</s>",
|
13 |
+
"tokenizer_class": "LongformerTokenizer",
|
14 |
+
"trim_offsets": true,
|
15 |
+
"unk_token": "<unk>"
|
16 |
+
}
|
training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:2aca52669a0f0e93fa2dfe21b6325cae64773ed01d7c0848b38c5da4af19789c
|
3 |
+
size 4155
|
vocab.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|