Hugging Face's logo

ICFNext
/
EYY-Categorisation

Text Classification
Transformers
PyTorch
TensorBoard
Safetensors
distilbert
Inference Endpoints
Model card Files Metrics Community
1
ebrigham commited on
Commit
6c71162
•
1 Parent(s): 78c33d0

First model version

Browse files
This view is limited to 50 files because it contains too many changes.   See raw diff
Files changed (50) hide show
  1. checkpoint-1000/config.json +58 -0
  2. checkpoint-1000/optimizer.pt +3 -0
  3. checkpoint-1000/pytorch_model.bin +3 -0
  4. checkpoint-1000/rng_state.pth +3 -0
  5. checkpoint-1000/scheduler.pt +3 -0
  6. checkpoint-1000/special_tokens_map.json +1 -0
  7. checkpoint-1000/tokenizer.json +0 -0
  8. checkpoint-1000/tokenizer_config.json +1 -0
  9. checkpoint-1000/trainer_state.json +28 -0
  10. checkpoint-1000/training_args.bin +3 -0
  11. checkpoint-1000/vocab.txt +0 -0
  12. checkpoint-1500/config.json +58 -0
  13. checkpoint-1500/optimizer.pt +3 -0
  14. checkpoint-1500/pytorch_model.bin +3 -0
  15. checkpoint-1500/rng_state.pth +3 -0
  16. checkpoint-1500/scheduler.pt +3 -0
  17. checkpoint-1500/special_tokens_map.json +1 -0
  18. checkpoint-1500/tokenizer.json +0 -0
  19. checkpoint-1500/tokenizer_config.json +1 -0
  20. checkpoint-1500/trainer_state.json +34 -0
  21. checkpoint-1500/training_args.bin +3 -0
  22. checkpoint-1500/vocab.txt +0 -0
  23. checkpoint-2000/config.json +58 -0
  24. checkpoint-2000/optimizer.pt +3 -0
  25. checkpoint-2000/pytorch_model.bin +3 -0
  26. checkpoint-2000/rng_state.pth +3 -0
  27. checkpoint-2000/scheduler.pt +3 -0
  28. checkpoint-2000/special_tokens_map.json +1 -0
  29. checkpoint-2000/tokenizer.json +0 -0
  30. checkpoint-2000/tokenizer_config.json +1 -0
  31. checkpoint-2000/trainer_state.json +40 -0
  32. checkpoint-2000/training_args.bin +3 -0
  33. checkpoint-2000/vocab.txt +0 -0
  34. checkpoint-2500/config.json +58 -0
  35. checkpoint-2500/optimizer.pt +3 -0
  36. checkpoint-2500/pytorch_model.bin +3 -0
  37. checkpoint-2500/rng_state.pth +3 -0
  38. checkpoint-2500/scheduler.pt +3 -0
  39. checkpoint-2500/special_tokens_map.json +1 -0
  40. checkpoint-2500/tokenizer.json +0 -0
  41. checkpoint-2500/tokenizer_config.json +1 -0
  42. checkpoint-2500/trainer_state.json +46 -0
  43. checkpoint-2500/training_args.bin +3 -0
  44. checkpoint-2500/vocab.txt +0 -0
  45. checkpoint-3000/config.json +58 -0
  46. checkpoint-3000/optimizer.pt +3 -0
  47. checkpoint-3000/pytorch_model.bin +3 -0
  48. checkpoint-3000/rng_state.pth +3 -0
  49. checkpoint-3000/scheduler.pt +3 -0
  50. checkpoint-3000/special_tokens_map.json +1 -0
checkpoint-1000/config.json ADDED
@@ -0,0 +1,58 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "distilbert-base-uncased",
3
+ "activation": "gelu",
4
+ "architectures": [
5
+ "DistilBertForSequenceClassification"
6
+ ],
7
+ "attention_dropout": 0.1,
8
+ "dim": 768,
9
+ "dropout": 0.1,
10
+ "hidden_dim": 3072,
11
+ "id2label": {
12
+ "0": "n/a",
13
+ "1": "culture",
14
+ "2": "digital",
15
+ "3": "employment",
16
+ "4": "renewable energy",
17
+ "5": "natural sustainability",
18
+ "6": "studying abroad",
19
+ "7": "policy dialogues",
20
+ "8": "youth and the world",
21
+ "9": "education and training",
22
+ "10": "disinformation and fake news",
23
+ "11": "health and well-being",
24
+ "12": "research and innovation",
25
+ "13": "participation and engagement",
26
+ "14": "inclusion and democratic values"
27
+ },
28
+ "initializer_range": 0.02,
29
+ "label2id": {
30
+ "culture": 1,
31
+ "digital": 2,
32
+ "disinformation and fake news": 10,
33
+ "education and training": 9,
34
+ "employment": 3,
35
+ "health and well-being": 11,
36
+ "inclusion and democratic values": 14,
37
+ "n/a": 0,
38
+ "natural sustainability": 5,
39
+ "participation and engagement": 13,
40
+ "policy dialogues": 7,
41
+ "renewable energy": 4,
42
+ "research and innovation": 12,
43
+ "studying abroad": 6,
44
+ "youth and the world": 8
45
+ },
46
+ "max_position_embeddings": 512,
47
+ "model_type": "distilbert",
48
+ "n_heads": 12,
49
+ "n_layers": 6,
50
+ "pad_token_id": 0,
51
+ "qa_dropout": 0.1,
52
+ "seq_classif_dropout": 0.2,
53
+ "sinusoidal_pos_embds": false,
54
+ "tie_weights_": true,
55
+ "torch_dtype": "float32",
56
+ "transformers_version": "4.17.0",
57
+ "vocab_size": 30522
58
+ }
checkpoint-1000/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:09984568656eb1f9140c17b97ce363ea7e1e1bec162cf2ada71cc6af46895f04
3
+ size 535779937
checkpoint-1000/pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0a146d0f2aae9e489127e0815c3464901087fe12ae649c459990644c19731e95
3
+ size 267900465
checkpoint-1000/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9fef4daa5a15555136f8017a0acdcf11d42550007acf44bcf5c3a6bdd3adb4af
3
+ size 14503
checkpoint-1000/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c4f83269c532c4e9a8ba2ecf1bcab12e0feb2f4e34cf92313a53653115587887
3
+ size 623
checkpoint-1000/special_tokens_map.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]"}
checkpoint-1000/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-1000/tokenizer_config.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"do_lower_case": true, "unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]", "tokenize_chinese_chars": true, "strip_accents": null, "model_max_length": 512, "special_tokens_map_file": null, "name_or_path": "distilbert-base-uncased", "tokenizer_class": "DistilBertTokenizer"}
checkpoint-1000/trainer_state.json ADDED
@@ -0,0 +1,28 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 0.26666666666666666,
5
+ "global_step": 1000,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.13,
12
+ "learning_rate": 4.3333333333333334e-05,
13
+ "loss": 2.4835,
14
+ "step": 500
15
+ },
16
+ {
17
+ "epoch": 0.27,
18
+ "learning_rate": 3.6666666666666666e-05,
19
+ "loss": 2.4544,
20
+ "step": 1000
21
+ }
22
+ ],
23
+ "max_steps": 3750,
24
+ "num_train_epochs": 1,
25
+ "total_flos": 971567230864320.0,
26
+ "trial_name": null,
27
+ "trial_params": null
28
+ }
checkpoint-1000/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:833718e3d44c252c8b3b2743e40695c847337730d454e2957cfab6ded5799dc5
3
+ size 3055
checkpoint-1000/vocab.txt ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-1500/config.json ADDED
@@ -0,0 +1,58 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "distilbert-base-uncased",
3
+ "activation": "gelu",
4
+ "architectures": [
5
+ "DistilBertForSequenceClassification"
6
+ ],
7
+ "attention_dropout": 0.1,
8
+ "dim": 768,
9
+ "dropout": 0.1,
10
+ "hidden_dim": 3072,
11
+ "id2label": {
12
+ "0": "n/a",
13
+ "1": "culture",
14
+ "2": "digital",
15
+ "3": "employment",
16
+ "4": "renewable energy",
17
+ "5": "natural sustainability",
18
+ "6": "studying abroad",
19
+ "7": "policy dialogues",
20
+ "8": "youth and the world",
21
+ "9": "education and training",
22
+ "10": "disinformation and fake news",
23
+ "11": "health and well-being",
24
+ "12": "research and innovation",
25
+ "13": "participation and engagement",
26
+ "14": "inclusion and democratic values"
27
+ },
28
+ "initializer_range": 0.02,
29
+ "label2id": {
30
+ "culture": 1,
31
+ "digital": 2,
32
+ "disinformation and fake news": 10,
33
+ "education and training": 9,
34
+ "employment": 3,
35
+ "health and well-being": 11,
36
+ "inclusion and democratic values": 14,
37
+ "n/a": 0,
38
+ "natural sustainability": 5,
39
+ "participation and engagement": 13,
40
+ "policy dialogues": 7,
41
+ "renewable energy": 4,
42
+ "research and innovation": 12,
43
+ "studying abroad": 6,
44
+ "youth and the world": 8
45
+ },
46
+ "max_position_embeddings": 512,
47
+ "model_type": "distilbert",
48
+ "n_heads": 12,
49
+ "n_layers": 6,
50
+ "pad_token_id": 0,
51
+ "qa_dropout": 0.1,
52
+ "seq_classif_dropout": 0.2,
53
+ "sinusoidal_pos_embds": false,
54
+ "tie_weights_": true,
55
+ "torch_dtype": "float32",
56
+ "transformers_version": "4.17.0",
57
+ "vocab_size": 30522
58
+ }
checkpoint-1500/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:99c582666243bdf25c9d347db25cda6a1addfa51d5804bf9d85322e267cbfba2
3
+ size 535779937
checkpoint-1500/pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1a08792612ed491f795523fc5ad603d8973f30a870d847089df1e598b9077d3f
3
+ size 267900465
checkpoint-1500/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2b70d16f31c6acc0b13971f0b78e9dcd0d5ed1f43dd68df23c7161ddf1d06f86
3
+ size 14503
checkpoint-1500/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0448fa8959e682269aceffad7c77d85f2d35fb1b603085f58ee828052f2267a5
3
+ size 623
checkpoint-1500/special_tokens_map.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]"}
checkpoint-1500/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-1500/tokenizer_config.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"do_lower_case": true, "unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]", "tokenize_chinese_chars": true, "strip_accents": null, "model_max_length": 512, "special_tokens_map_file": null, "name_or_path": "distilbert-base-uncased", "tokenizer_class": "DistilBertTokenizer"}
checkpoint-1500/trainer_state.json ADDED
@@ -0,0 +1,34 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 0.4,
5
+ "global_step": 1500,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.13,
12
+ "learning_rate": 4.3333333333333334e-05,
13
+ "loss": 2.4835,
14
+ "step": 500
15
+ },
16
+ {
17
+ "epoch": 0.27,
18
+ "learning_rate": 3.6666666666666666e-05,
19
+ "loss": 2.4544,
20
+ "step": 1000
21
+ },
22
+ {
23
+ "epoch": 0.4,
24
+ "learning_rate": 3e-05,
25
+ "loss": 2.4483,
26
+ "step": 1500
27
+ }
28
+ ],
29
+ "max_steps": 3750,
30
+ "num_train_epochs": 1,
31
+ "total_flos": 1452659585109120.0,
32
+ "trial_name": null,
33
+ "trial_params": null
34
+ }
checkpoint-1500/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:833718e3d44c252c8b3b2743e40695c847337730d454e2957cfab6ded5799dc5
3
+ size 3055
checkpoint-1500/vocab.txt ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-2000/config.json ADDED
@@ -0,0 +1,58 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "distilbert-base-uncased",
3
+ "activation": "gelu",
4
+ "architectures": [
5
+ "DistilBertForSequenceClassification"
6
+ ],
7
+ "attention_dropout": 0.1,
8
+ "dim": 768,
9
+ "dropout": 0.1,
10
+ "hidden_dim": 3072,
11
+ "id2label": {
12
+ "0": "n/a",
13
+ "1": "culture",
14
+ "2": "digital",
15
+ "3": "employment",
16
+ "4": "renewable energy",
17
+ "5": "natural sustainability",
18
+ "6": "studying abroad",
19
+ "7": "policy dialogues",
20
+ "8": "youth and the world",
21
+ "9": "education and training",
22
+ "10": "disinformation and fake news",
23
+ "11": "health and well-being",
24
+ "12": "research and innovation",
25
+ "13": "participation and engagement",
26
+ "14": "inclusion and democratic values"
27
+ },
28
+ "initializer_range": 0.02,
29
+ "label2id": {
30
+ "culture": 1,
31
+ "digital": 2,
32
+ "disinformation and fake news": 10,
33
+ "education and training": 9,
34
+ "employment": 3,
35
+ "health and well-being": 11,
36
+ "inclusion and democratic values": 14,
37
+ "n/a": 0,
38
+ "natural sustainability": 5,
39
+ "participation and engagement": 13,
40
+ "policy dialogues": 7,
41
+ "renewable energy": 4,
42
+ "research and innovation": 12,
43
+ "studying abroad": 6,
44
+ "youth and the world": 8
45
+ },
46
+ "max_position_embeddings": 512,
47
+ "model_type": "distilbert",
48
+ "n_heads": 12,
49
+ "n_layers": 6,
50
+ "pad_token_id": 0,
51
+ "qa_dropout": 0.1,
52
+ "seq_classif_dropout": 0.2,
53
+ "sinusoidal_pos_embds": false,
54
+ "tie_weights_": true,
55
+ "torch_dtype": "float32",
56
+ "transformers_version": "4.17.0",
57
+ "vocab_size": 30522
58
+ }
checkpoint-2000/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:60a410a19fed6a5e2ed8988db4c779365319f66117c96a9fdfaf8d3a370bdaa9
3
+ size 535779937
checkpoint-2000/pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3797a9d385b51b0b4b04c7578259807ad59a7d122e6089f74fdf105b8d10a21a
3
+ size 267900465
checkpoint-2000/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d42da5f1df304e6b0252701fceaa84057bc75e55946f88656b60745112313184
3
+ size 14503
checkpoint-2000/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:eb3e3e15e91c6931f61158f085bbf18d801bf4aa96563c336e19563a458f23ff
3
+ size 623
checkpoint-2000/special_tokens_map.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]"}
checkpoint-2000/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-2000/tokenizer_config.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"do_lower_case": true, "unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]", "tokenize_chinese_chars": true, "strip_accents": null, "model_max_length": 512, "special_tokens_map_file": null, "name_or_path": "distilbert-base-uncased", "tokenizer_class": "DistilBertTokenizer"}
checkpoint-2000/trainer_state.json ADDED
@@ -0,0 +1,40 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 0.5333333333333333,
5
+ "global_step": 2000,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.13,
12
+ "learning_rate": 4.3333333333333334e-05,
13
+ "loss": 2.4835,
14
+ "step": 500
15
+ },
16
+ {
17
+ "epoch": 0.27,
18
+ "learning_rate": 3.6666666666666666e-05,
19
+ "loss": 2.4544,
20
+ "step": 1000
21
+ },
22
+ {
23
+ "epoch": 0.4,
24
+ "learning_rate": 3e-05,
25
+ "loss": 2.4483,
26
+ "step": 1500
27
+ },
28
+ {
29
+ "epoch": 0.53,
30
+ "learning_rate": 2.3333333333333336e-05,
31
+ "loss": 2.4441,
32
+ "step": 2000
33
+ }
34
+ ],
35
+ "max_steps": 3750,
36
+ "num_train_epochs": 1,
37
+ "total_flos": 1923127247203200.0,
38
+ "trial_name": null,
39
+ "trial_params": null
40
+ }
checkpoint-2000/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:833718e3d44c252c8b3b2743e40695c847337730d454e2957cfab6ded5799dc5
3
+ size 3055
checkpoint-2000/vocab.txt ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-2500/config.json ADDED
@@ -0,0 +1,58 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "distilbert-base-uncased",
3
+ "activation": "gelu",
4
+ "architectures": [
5
+ "DistilBertForSequenceClassification"
6
+ ],
7
+ "attention_dropout": 0.1,
8
+ "dim": 768,
9
+ "dropout": 0.1,
10
+ "hidden_dim": 3072,
11
+ "id2label": {
12
+ "0": "n/a",
13
+ "1": "culture",
14
+ "2": "digital",
15
+ "3": "employment",
16
+ "4": "renewable energy",
17
+ "5": "natural sustainability",
18
+ "6": "studying abroad",
19
+ "7": "policy dialogues",
20
+ "8": "youth and the world",
21
+ "9": "education and training",
22
+ "10": "disinformation and fake news",
23
+ "11": "health and well-being",
24
+ "12": "research and innovation",
25
+ "13": "participation and engagement",
26
+ "14": "inclusion and democratic values"
27
+ },
28
+ "initializer_range": 0.02,
29
+ "label2id": {
30
+ "culture": 1,
31
+ "digital": 2,
32
+ "disinformation and fake news": 10,
33
+ "education and training": 9,
34
+ "employment": 3,
35
+ "health and well-being": 11,
36
+ "inclusion and democratic values": 14,
37
+ "n/a": 0,
38
+ "natural sustainability": 5,
39
+ "participation and engagement": 13,
40
+ "policy dialogues": 7,
41
+ "renewable energy": 4,
42
+ "research and innovation": 12,
43
+ "studying abroad": 6,
44
+ "youth and the world": 8
45
+ },
46
+ "max_position_embeddings": 512,
47
+ "model_type": "distilbert",
48
+ "n_heads": 12,
49
+ "n_layers": 6,
50
+ "pad_token_id": 0,
51
+ "qa_dropout": 0.1,
52
+ "seq_classif_dropout": 0.2,
53
+ "sinusoidal_pos_embds": false,
54
+ "tie_weights_": true,
55
+ "torch_dtype": "float32",
56
+ "transformers_version": "4.17.0",
57
+ "vocab_size": 30522
58
+ }
checkpoint-2500/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a22b36e80a8d5076cfe1085324a9bd5e7b8f8734f83f29eea506a2f92e08a013
3
+ size 535779937
checkpoint-2500/pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8f39b22e177fb0f8dd140fff19e4f82bf1934044b9c367860cf4ab947b19e7e2
3
+ size 267900465
checkpoint-2500/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2952355ad4e46086847dab0a7b91185afd1f274ad431221d3a5448e29d558b0c
3
+ size 14503
checkpoint-2500/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b4c13494368e217239013918c949ddcfa2b6fbb317a7b57308ea9ba0e0c5dfb4
3
+ size 623
checkpoint-2500/special_tokens_map.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]"}
checkpoint-2500/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-2500/tokenizer_config.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"do_lower_case": true, "unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]", "tokenize_chinese_chars": true, "strip_accents": null, "model_max_length": 512, "special_tokens_map_file": null, "name_or_path": "distilbert-base-uncased", "tokenizer_class": "DistilBertTokenizer"}
checkpoint-2500/trainer_state.json ADDED
@@ -0,0 +1,46 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 0.6666666666666666,
5
+ "global_step": 2500,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.13,
12
+ "learning_rate": 4.3333333333333334e-05,
13
+ "loss": 2.4835,
14
+ "step": 500
15
+ },
16
+ {
17
+ "epoch": 0.27,
18
+ "learning_rate": 3.6666666666666666e-05,
19
+ "loss": 2.4544,
20
+ "step": 1000
21
+ },
22
+ {
23
+ "epoch": 0.4,
24
+ "learning_rate": 3e-05,
25
+ "loss": 2.4483,
26
+ "step": 1500
27
+ },
28
+ {
29
+ "epoch": 0.53,
30
+ "learning_rate": 2.3333333333333336e-05,
31
+ "loss": 2.4441,
32
+ "step": 2000
33
+ },
34
+ {
35
+ "epoch": 0.67,
36
+ "learning_rate": 1.6666666666666667e-05,
37
+ "loss": 2.4412,
38
+ "step": 2500
39
+ }
40
+ ],
41
+ "max_steps": 3750,
42
+ "num_train_epochs": 1,
43
+ "total_flos": 2386771256661120.0,
44
+ "trial_name": null,
45
+ "trial_params": null
46
+ }
checkpoint-2500/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:833718e3d44c252c8b3b2743e40695c847337730d454e2957cfab6ded5799dc5
3
+ size 3055
checkpoint-2500/vocab.txt ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-3000/config.json ADDED
@@ -0,0 +1,58 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "distilbert-base-uncased",
3
+ "activation": "gelu",
4
+ "architectures": [
5
+ "DistilBertForSequenceClassification"
6
+ ],
7
+ "attention_dropout": 0.1,
8
+ "dim": 768,
9
+ "dropout": 0.1,
10
+ "hidden_dim": 3072,
11
+ "id2label": {
12
+ "0": "n/a",
13
+ "1": "culture",
14
+ "2": "digital",
15
+ "3": "employment",
16
+ "4": "renewable energy",
17
+ "5": "natural sustainability",
18
+ "6": "studying abroad",
19
+ "7": "policy dialogues",
20
+ "8": "youth and the world",
21
+ "9": "education and training",
22
+ "10": "disinformation and fake news",
23
+ "11": "health and well-being",
24
+ "12": "research and innovation",
25
+ "13": "participation and engagement",
26
+ "14": "inclusion and democratic values"
27
+ },
28
+ "initializer_range": 0.02,
29
+ "label2id": {
30
+ "culture": 1,
31
+ "digital": 2,
32
+ "disinformation and fake news": 10,
33
+ "education and training": 9,
34
+ "employment": 3,
35
+ "health and well-being": 11,
36
+ "inclusion and democratic values": 14,
37
+ "n/a": 0,
38
+ "natural sustainability": 5,
39
+ "participation and engagement": 13,
40
+ "policy dialogues": 7,
41
+ "renewable energy": 4,
42
+ "research and innovation": 12,
43
+ "studying abroad": 6,
44
+ "youth and the world": 8
45
+ },
46
+ "max_position_embeddings": 512,
47
+ "model_type": "distilbert",
48
+ "n_heads": 12,
49
+ "n_layers": 6,
50
+ "pad_token_id": 0,
51
+ "qa_dropout": 0.1,
52
+ "seq_classif_dropout": 0.2,
53
+ "sinusoidal_pos_embds": false,
54
+ "tie_weights_": true,
55
+ "torch_dtype": "float32",
56
+ "transformers_version": "4.17.0",
57
+ "vocab_size": 30522
58
+ }
checkpoint-3000/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:98471503a23771c92da9f2abd65ae2e6497a83995190e0a9366078a7ac69638c
3
+ size 535779937
checkpoint-3000/pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4ec61b8275bc495c29cdc8c60b01b3edd75fb2912b7719f96fac16e343d9b9f3
3
+ size 267900465
checkpoint-3000/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7e823145ceff848d3304a03dbfda599b201f8c385e9c3bd781564b37210ac22e
3
+ size 14503
checkpoint-3000/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:41625199ddc3548f3fb8e3d4a786a1e52abf1e78aa1125676a9cb9dff1785538
3
+ size 623
checkpoint-3000/special_tokens_map.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]"}