Hugging Face's logo

mekaneeky
/
dataset-tags-expanded-tokenizer

Model card Files
xet
Community
azawahry commited on
Commit
981a23b
·
1 Parent(s): 4a71f29

Initial commit

Browse files
Files changed (45) hide show
  1. checkpoint-150/added_tokens.json +44 -0
  2. checkpoint-150/config.json +61 -0
  3. checkpoint-150/generation_config.json +15 -0
  4. checkpoint-150/optimizer.pt +3 -0
  5. checkpoint-150/pytorch_model.bin +3 -0
  6. checkpoint-150/rng_state.pth +3 -0
  7. checkpoint-150/scaler.pt +3 -0
  8. checkpoint-150/scheduler.pt +3 -0
  9. checkpoint-150/source.spm +0 -0
  10. checkpoint-150/special_tokens_map.json +5 -0
  11. checkpoint-150/target.spm +0 -0
  12. checkpoint-150/tokenizer_config.json +12 -0
  13. checkpoint-150/trainer_state.json +30 -0
  14. checkpoint-150/training_args.bin +3 -0
  15. checkpoint-150/vocab.json +0 -0
  16. checkpoint-450/added_tokens.json +44 -0
  17. checkpoint-450/config.json +61 -0
  18. checkpoint-450/generation_config.json +15 -0
  19. checkpoint-450/optimizer.pt +3 -0
  20. checkpoint-450/pytorch_model.bin +3 -0
  21. checkpoint-450/rng_state.pth +3 -0
  22. checkpoint-450/scaler.pt +3 -0
  23. checkpoint-450/scheduler.pt +3 -0
  24. checkpoint-450/source.spm +0 -0
  25. checkpoint-450/special_tokens_map.json +5 -0
  26. checkpoint-450/target.spm +0 -0
  27. checkpoint-450/tokenizer_config.json +12 -0
  28. checkpoint-450/trainer_state.json +58 -0
  29. checkpoint-450/training_args.bin +3 -0
  30. checkpoint-450/vocab.json +0 -0
  31. checkpoint-600/added_tokens.json +44 -0
  32. checkpoint-600/config.json +61 -0
  33. checkpoint-600/generation_config.json +15 -0
  34. checkpoint-600/optimizer.pt +3 -0
  35. checkpoint-600/pytorch_model.bin +3 -0
  36. checkpoint-600/rng_state.pth +3 -0
  37. checkpoint-600/scaler.pt +3 -0
  38. checkpoint-600/scheduler.pt +3 -0
  39. checkpoint-600/source.spm +0 -0
  40. checkpoint-600/special_tokens_map.json +5 -0
  41. checkpoint-600/target.spm +0 -0
  42. checkpoint-600/tokenizer_config.json +12 -0
  43. checkpoint-600/trainer_state.json +78 -0
  44. checkpoint-600/training_args.bin +3 -0
  45. checkpoint-600/vocab.json +0 -0
checkpoint-150/added_tokens.json ADDED
@@ -0,0 +1,44 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "|ach_aug_bt|": 64199,
3
+ "|ach_aug_ggl|": 64197,
4
+ "|ach_aug_hq|": 64198,
5
+ "|ach_bt|": 64196,
6
+ "|ach_ggl|": 64194,
7
+ "|ach_hq|": 64195,
8
+ "|ach_ood|": 64193,
9
+ "|lgg_aug_bt|": 64191,
10
+ "|lgg_aug_ggl|": 64190,
11
+ "|lgg_aug_hq|": 64192,
12
+ "|lgg_bt|": 64188,
13
+ "|lgg_ggl|": 64187,
14
+ "|lgg_hq|": 64189,
15
+ "|lgg_ood|": 64186,
16
+ "|lug_aug_bt|": 64206,
17
+ "|lug_aug_ggl|": 64204,
18
+ "|lug_aug_hq|": 64205,
19
+ "|lug_bt|": 64203,
20
+ "|lug_ggl|": 64201,
21
+ "|lug_hq|": 64202,
22
+ "|lug_ood|": 64200,
23
+ "|luo_aug_bt|": 64213,
24
+ "|luo_aug_ggl|": 64211,
25
+ "|luo_aug_hq|": 64212,
26
+ "|luo_bt|": 64210,
27
+ "|luo_ggl|": 64208,
28
+ "|luo_hq|": 64209,
29
+ "|luo_ood|": 64207,
30
+ "|nyn_aug_bt|": 64178,
31
+ "|nyn_aug_ggl|": 64176,
32
+ "|nyn_aug_hq|": 64177,
33
+ "|nyn_bt|": 64175,
34
+ "|nyn_ggl|": 64173,
35
+ "|nyn_hq|": 64174,
36
+ "|nyn_ood|": 64172,
37
+ "|teo_aug_bt|": 64185,
38
+ "|teo_aug_ggl|": 64183,
39
+ "|teo_aug_hq|": 64184,
40
+ "|teo_bt|": 64182,
41
+ "|teo_ggl|": 64180,
42
+ "|teo_hq|": 64181,
43
+ "|teo_ood|": 64179
44
+ }
checkpoint-150/config.json ADDED
@@ -0,0 +1,61 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "/home/azawahry_sunbird_ai/models/m2e/dataset-tags-expanded-tokenizer/checkpoint-1050",
3
+ "activation_dropout": 0.0,
4
+ "activation_function": "swish",
5
+ "add_bias_logits": false,
6
+ "add_final_layer_norm": false,
7
+ "architectures": [
8
+ "MarianMTModel"
9
+ ],
10
+ "attention_dropout": 0.0,
11
+ "bad_words_ids": [
12
+ [
13
+ 64171
14
+ ]
15
+ ],
16
+ "bos_token_id": 0,
17
+ "classif_dropout": 0.0,
18
+ "classifier_dropout": 0.0,
19
+ "d_model": 512,
20
+ "decoder_attention_heads": 8,
21
+ "decoder_ffn_dim": 2048,
22
+ "decoder_layerdrop": 0.0,
23
+ "decoder_layers": 6,
24
+ "decoder_start_token_id": 64171,
25
+ "decoder_vocab_size": 64214,
26
+ "dropout": 0.1,
27
+ "encoder_attention_heads": 8,
28
+ "encoder_ffn_dim": 2048,
29
+ "encoder_layerdrop": 0.0,
30
+ "encoder_layers": 6,
31
+ "eos_token_id": 0,
32
+ "extra_pos_embeddings": 64172,
33
+ "forced_eos_token_id": 0,
34
+ "id2label": {
35
+ "0": "LABEL_0",
36
+ "1": "LABEL_1",
37
+ "2": "LABEL_2"
38
+ },
39
+ "init_std": 0.02,
40
+ "is_encoder_decoder": true,
41
+ "label2id": {
42
+ "LABEL_0": 0,
43
+ "LABEL_1": 1,
44
+ "LABEL_2": 2
45
+ },
46
+ "max_length": 512,
47
+ "max_position_embeddings": 512,
48
+ "model_type": "marian",
49
+ "normalize_before": false,
50
+ "normalize_embedding": false,
51
+ "num_beams": 6,
52
+ "num_hidden_layers": 6,
53
+ "pad_token_id": 64171,
54
+ "scale_embedding": true,
55
+ "share_encoder_decoder_embeddings": true,
56
+ "static_position_embeddings": true,
57
+ "torch_dtype": "float32",
58
+ "transformers_version": "4.27.3",
59
+ "use_cache": true,
60
+ "vocab_size": 64214
61
+ }
checkpoint-150/generation_config.json ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bad_words_ids": [
3
+ [
4
+ 64171
5
+ ]
6
+ ],
7
+ "bos_token_id": 0,
8
+ "decoder_start_token_id": 64171,
9
+ "eos_token_id": 0,
10
+ "forced_eos_token_id": 0,
11
+ "max_length": 512,
12
+ "num_beams": 6,
13
+ "pad_token_id": 64171,
14
+ "transformers_version": "4.27.3"
15
+ }
checkpoint-150/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3dae2189b093b86ba289af4f005a508ae538bb31354719e3db9116d35f245e32
3
+ size 616276741
checkpoint-150/pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2270bc13e810caca72693d79f82753223c8de32941ed9dc649902821d62a39ab
3
+ size 308407621
checkpoint-150/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:42a80f6998d2d4ab9eb7566bcf48c742bb00e92d4722e6ad00fe328123bdb3e9
3
+ size 14575
checkpoint-150/scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c51c8702e03bc24d0566e4b37e893acfcfe0af87dcc3883f22471d07682d7ac5
3
+ size 557
checkpoint-150/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f7c7bf40a66ebd048fdd4239a501cedcc9711d3fdb1b1d9a7f4d677727d2347a
3
+ size 627
checkpoint-150/source.spm ADDED
Binary file (707 kB). View file
 
checkpoint-150/special_tokens_map.json ADDED
@@ -0,0 +1,5 @@
 
 
 
 
 
 
1
+ {
2
+ "eos_token": "</s>",
3
+ "pad_token": "<pad>",
4
+ "unk_token": "<unk>"
5
+ }
checkpoint-150/target.spm ADDED
Binary file (791 kB). View file
 
checkpoint-150/tokenizer_config.json ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "eos_token": "</s>",
3
+ "model_max_length": 512,
4
+ "pad_token": "<pad>",
5
+ "separate_vocabs": false,
6
+ "source_lang": "mul",
7
+ "sp_model_kwargs": {},
8
+ "special_tokens_map_file": null,
9
+ "target_lang": "eng",
10
+ "tokenizer_class": "MarianTokenizer",
11
+ "unk_token": "<unk>"
12
+ }
checkpoint-150/trainer_state.json ADDED
@@ -0,0 +1,30 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 2.9462931156158447,
3
+ "best_model_checkpoint": "/home/azawahry_sunbird_ai/models/m2e/dataset-tags-expanded-tokenizer/checkpoint-150",
4
+ "epoch": 0.2562481849086902,
5
+ "global_step": 150,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.26,
12
+ "eval_BLEU_ach": 21.4505,
13
+ "eval_BLEU_lgg": 24.4794,
14
+ "eval_BLEU_lug": 35.1488,
15
+ "eval_BLEU_mean": 26.3376,
16
+ "eval_BLEU_nyn": 25.5547,
17
+ "eval_BLEU_teo": 25.0545,
18
+ "eval_loss": 2.9462931156158447,
19
+ "eval_runtime": 59.9936,
20
+ "eval_samples_per_second": 41.671,
21
+ "eval_steps_per_second": 0.833,
22
+ "step": 150
23
+ }
24
+ ],
25
+ "max_steps": 1755,
26
+ "num_train_epochs": 3,
27
+ "total_flos": 2.16451520888832e+16,
28
+ "trial_name": null,
29
+ "trial_params": null
30
+ }
checkpoint-150/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e46a42e4fff4eb7e4702f865fd1022499e6ad9fc3cf4d3f71aef8c990651b558
3
+ size 3707
checkpoint-150/vocab.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-450/added_tokens.json ADDED
@@ -0,0 +1,44 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "|ach_aug_bt|": 64199,
3
+ "|ach_aug_ggl|": 64197,
4
+ "|ach_aug_hq|": 64198,
5
+ "|ach_bt|": 64196,
6
+ "|ach_ggl|": 64194,
7
+ "|ach_hq|": 64195,
8
+ "|ach_ood|": 64193,
9
+ "|lgg_aug_bt|": 64191,
10
+ "|lgg_aug_ggl|": 64190,
11
+ "|lgg_aug_hq|": 64192,
12
+ "|lgg_bt|": 64188,
13
+ "|lgg_ggl|": 64187,
14
+ "|lgg_hq|": 64189,
15
+ "|lgg_ood|": 64186,
16
+ "|lug_aug_bt|": 64206,
17
+ "|lug_aug_ggl|": 64204,
18
+ "|lug_aug_hq|": 64205,
19
+ "|lug_bt|": 64203,
20
+ "|lug_ggl|": 64201,
21
+ "|lug_hq|": 64202,
22
+ "|lug_ood|": 64200,
23
+ "|luo_aug_bt|": 64213,
24
+ "|luo_aug_ggl|": 64211,
25
+ "|luo_aug_hq|": 64212,
26
+ "|luo_bt|": 64210,
27
+ "|luo_ggl|": 64208,
28
+ "|luo_hq|": 64209,
29
+ "|luo_ood|": 64207,
30
+ "|nyn_aug_bt|": 64178,
31
+ "|nyn_aug_ggl|": 64176,
32
+ "|nyn_aug_hq|": 64177,
33
+ "|nyn_bt|": 64175,
34
+ "|nyn_ggl|": 64173,
35
+ "|nyn_hq|": 64174,
36
+ "|nyn_ood|": 64172,
37
+ "|teo_aug_bt|": 64185,
38
+ "|teo_aug_ggl|": 64183,
39
+ "|teo_aug_hq|": 64184,
40
+ "|teo_bt|": 64182,
41
+ "|teo_ggl|": 64180,
42
+ "|teo_hq|": 64181,
43
+ "|teo_ood|": 64179
44
+ }
checkpoint-450/config.json ADDED
@@ -0,0 +1,61 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "/home/azawahry_sunbird_ai/models/m2e/dataset-tags-expanded-tokenizer/checkpoint-1050",
3
+ "activation_dropout": 0.0,
4
+ "activation_function": "swish",
5
+ "add_bias_logits": false,
6
+ "add_final_layer_norm": false,
7
+ "architectures": [
8
+ "MarianMTModel"
9
+ ],
10
+ "attention_dropout": 0.0,
11
+ "bad_words_ids": [
12
+ [
13
+ 64171
14
+ ]
15
+ ],
16
+ "bos_token_id": 0,
17
+ "classif_dropout": 0.0,
18
+ "classifier_dropout": 0.0,
19
+ "d_model": 512,
20
+ "decoder_attention_heads": 8,
21
+ "decoder_ffn_dim": 2048,
22
+ "decoder_layerdrop": 0.0,
23
+ "decoder_layers": 6,
24
+ "decoder_start_token_id": 64171,
25
+ "decoder_vocab_size": 64214,
26
+ "dropout": 0.1,
27
+ "encoder_attention_heads": 8,
28
+ "encoder_ffn_dim": 2048,
29
+ "encoder_layerdrop": 0.0,
30
+ "encoder_layers": 6,
31
+ "eos_token_id": 0,
32
+ "extra_pos_embeddings": 64172,
33
+ "forced_eos_token_id": 0,
34
+ "id2label": {
35
+ "0": "LABEL_0",
36
+ "1": "LABEL_1",
37
+ "2": "LABEL_2"
38
+ },
39
+ "init_std": 0.02,
40
+ "is_encoder_decoder": true,
41
+ "label2id": {
42
+ "LABEL_0": 0,
43
+ "LABEL_1": 1,
44
+ "LABEL_2": 2
45
+ },
46
+ "max_length": 512,
47
+ "max_position_embeddings": 512,
48
+ "model_type": "marian",
49
+ "normalize_before": false,
50
+ "normalize_embedding": false,
51
+ "num_beams": 6,
52
+ "num_hidden_layers": 6,
53
+ "pad_token_id": 64171,
54
+ "scale_embedding": true,
55
+ "share_encoder_decoder_embeddings": true,
56
+ "static_position_embeddings": true,
57
+ "torch_dtype": "float32",
58
+ "transformers_version": "4.27.3",
59
+ "use_cache": true,
60
+ "vocab_size": 64214
61
+ }
checkpoint-450/generation_config.json ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bad_words_ids": [
3
+ [
4
+ 64171
5
+ ]
6
+ ],
7
+ "bos_token_id": 0,
8
+ "decoder_start_token_id": 64171,
9
+ "eos_token_id": 0,
10
+ "forced_eos_token_id": 0,
11
+ "max_length": 512,
12
+ "num_beams": 6,
13
+ "pad_token_id": 64171,
14
+ "transformers_version": "4.27.3"
15
+ }
checkpoint-450/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0b18e3590002b877b6e3670c8f8c7ef25095ceb720573dd2db865e5e771cd1d6
3
+ size 616276997
checkpoint-450/pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8f98f0032593bbf0363c8517501389a11c533da912a79454abcd692f771f0c8e
3
+ size 308407621
checkpoint-450/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0f357c535c5b43fc8733e6460972e270983e96db90cb51bf9b46ee6d2b7c4965
3
+ size 14639
checkpoint-450/scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:206389f090abcc3222507a1886e7890301684ad43da4a05adfa7cac210314c85
3
+ size 557
checkpoint-450/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1cf1d48671f6e11f1aa6e8904b2ccf053a76b8f1676467007e8acdff715fc054
3
+ size 627
checkpoint-450/source.spm ADDED
Binary file (707 kB). View file
 
checkpoint-450/special_tokens_map.json ADDED
@@ -0,0 +1,5 @@
 
 
 
 
 
 
1
+ {
2
+ "eos_token": "</s>",
3
+ "pad_token": "<pad>",
4
+ "unk_token": "<unk>"
5
+ }
checkpoint-450/target.spm ADDED
Binary file (791 kB). View file
 
checkpoint-450/tokenizer_config.json ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "eos_token": "</s>",
3
+ "model_max_length": 512,
4
+ "pad_token": "<pad>",
5
+ "separate_vocabs": false,
6
+ "source_lang": "mul",
7
+ "sp_model_kwargs": {},
8
+ "special_tokens_map_file": null,
9
+ "target_lang": "eng",
10
+ "tokenizer_class": "MarianTokenizer",
11
+ "unk_token": "<unk>"
12
+ }
checkpoint-450/trainer_state.json ADDED
@@ -0,0 +1,58 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 2.9462931156158447,
3
+ "best_model_checkpoint": "/home/azawahry_sunbird_ai/models/m2e/dataset-tags-expanded-tokenizer/checkpoint-150",
4
+ "epoch": 0.7687445547260707,
5
+ "global_step": 450,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.26,
12
+ "eval_BLEU_ach": 21.4505,
13
+ "eval_BLEU_lgg": 24.4794,
14
+ "eval_BLEU_lug": 35.1488,
15
+ "eval_BLEU_mean": 26.3376,
16
+ "eval_BLEU_nyn": 25.5547,
17
+ "eval_BLEU_teo": 25.0545,
18
+ "eval_loss": 2.9462931156158447,
19
+ "eval_runtime": 59.9936,
20
+ "eval_samples_per_second": 41.671,
21
+ "eval_steps_per_second": 0.833,
22
+ "step": 150
23
+ },
24
+ {
25
+ "epoch": 0.51,
26
+ "eval_BLEU_ach": 22.0556,
27
+ "eval_BLEU_lgg": 23.8502,
28
+ "eval_BLEU_lug": 34.6592,
29
+ "eval_BLEU_mean": 26.1806,
30
+ "eval_BLEU_nyn": 25.075,
31
+ "eval_BLEU_teo": 25.2631,
32
+ "eval_loss": 2.9494848251342773,
33
+ "eval_runtime": 58.1523,
34
+ "eval_samples_per_second": 42.991,
35
+ "eval_steps_per_second": 0.86,
36
+ "step": 300
37
+ },
38
+ {
39
+ "epoch": 0.77,
40
+ "eval_BLEU_ach": 21.6851,
41
+ "eval_BLEU_lgg": 24.2555,
42
+ "eval_BLEU_lug": 34.7523,
43
+ "eval_BLEU_mean": 26.1102,
44
+ "eval_BLEU_nyn": 24.5712,
45
+ "eval_BLEU_teo": 25.2871,
46
+ "eval_loss": 2.9563159942626953,
47
+ "eval_runtime": 57.8967,
48
+ "eval_samples_per_second": 43.18,
49
+ "eval_steps_per_second": 0.864,
50
+ "step": 450
51
+ }
52
+ ],
53
+ "max_steps": 1755,
54
+ "num_train_epochs": 3,
55
+ "total_flos": 6.48118331670528e+16,
56
+ "trial_name": null,
57
+ "trial_params": null
58
+ }
checkpoint-450/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e46a42e4fff4eb7e4702f865fd1022499e6ad9fc3cf4d3f71aef8c990651b558
3
+ size 3707
checkpoint-450/vocab.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-600/added_tokens.json ADDED
@@ -0,0 +1,44 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "|ach_aug_bt|": 64199,
3
+ "|ach_aug_ggl|": 64197,
4
+ "|ach_aug_hq|": 64198,
5
+ "|ach_bt|": 64196,
6
+ "|ach_ggl|": 64194,
7
+ "|ach_hq|": 64195,
8
+ "|ach_ood|": 64193,
9
+ "|lgg_aug_bt|": 64191,
10
+ "|lgg_aug_ggl|": 64190,
11
+ "|lgg_aug_hq|": 64192,
12
+ "|lgg_bt|": 64188,
13
+ "|lgg_ggl|": 64187,
14
+ "|lgg_hq|": 64189,
15
+ "|lgg_ood|": 64186,
16
+ "|lug_aug_bt|": 64206,
17
+ "|lug_aug_ggl|": 64204,
18
+ "|lug_aug_hq|": 64205,
19
+ "|lug_bt|": 64203,
20
+ "|lug_ggl|": 64201,
21
+ "|lug_hq|": 64202,
22
+ "|lug_ood|": 64200,
23
+ "|luo_aug_bt|": 64213,
24
+ "|luo_aug_ggl|": 64211,
25
+ "|luo_aug_hq|": 64212,
26
+ "|luo_bt|": 64210,
27
+ "|luo_ggl|": 64208,
28
+ "|luo_hq|": 64209,
29
+ "|luo_ood|": 64207,
30
+ "|nyn_aug_bt|": 64178,
31
+ "|nyn_aug_ggl|": 64176,
32
+ "|nyn_aug_hq|": 64177,
33
+ "|nyn_bt|": 64175,
34
+ "|nyn_ggl|": 64173,
35
+ "|nyn_hq|": 64174,
36
+ "|nyn_ood|": 64172,
37
+ "|teo_aug_bt|": 64185,
38
+ "|teo_aug_ggl|": 64183,
39
+ "|teo_aug_hq|": 64184,
40
+ "|teo_bt|": 64182,
41
+ "|teo_ggl|": 64180,
42
+ "|teo_hq|": 64181,
43
+ "|teo_ood|": 64179
44
+ }
checkpoint-600/config.json ADDED
@@ -0,0 +1,61 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "/home/azawahry_sunbird_ai/models/m2e/dataset-tags-expanded-tokenizer/checkpoint-1050",
3
+ "activation_dropout": 0.0,
4
+ "activation_function": "swish",
5
+ "add_bias_logits": false,
6
+ "add_final_layer_norm": false,
7
+ "architectures": [
8
+ "MarianMTModel"
9
+ ],
10
+ "attention_dropout": 0.0,
11
+ "bad_words_ids": [
12
+ [
13
+ 64171
14
+ ]
15
+ ],
16
+ "bos_token_id": 0,
17
+ "classif_dropout": 0.0,
18
+ "classifier_dropout": 0.0,
19
+ "d_model": 512,
20
+ "decoder_attention_heads": 8,
21
+ "decoder_ffn_dim": 2048,
22
+ "decoder_layerdrop": 0.0,
23
+ "decoder_layers": 6,
24
+ "decoder_start_token_id": 64171,
25
+ "decoder_vocab_size": 64214,
26
+ "dropout": 0.1,
27
+ "encoder_attention_heads": 8,
28
+ "encoder_ffn_dim": 2048,
29
+ "encoder_layerdrop": 0.0,
30
+ "encoder_layers": 6,
31
+ "eos_token_id": 0,
32
+ "extra_pos_embeddings": 64172,
33
+ "forced_eos_token_id": 0,
34
+ "id2label": {
35
+ "0": "LABEL_0",
36
+ "1": "LABEL_1",
37
+ "2": "LABEL_2"
38
+ },
39
+ "init_std": 0.02,
40
+ "is_encoder_decoder": true,
41
+ "label2id": {
42
+ "LABEL_0": 0,
43
+ "LABEL_1": 1,
44
+ "LABEL_2": 2
45
+ },
46
+ "max_length": 512,
47
+ "max_position_embeddings": 512,
48
+ "model_type": "marian",
49
+ "normalize_before": false,
50
+ "normalize_embedding": false,
51
+ "num_beams": 6,
52
+ "num_hidden_layers": 6,
53
+ "pad_token_id": 64171,
54
+ "scale_embedding": true,
55
+ "share_encoder_decoder_embeddings": true,
56
+ "static_position_embeddings": true,
57
+ "torch_dtype": "float32",
58
+ "transformers_version": "4.27.3",
59
+ "use_cache": true,
60
+ "vocab_size": 64214
61
+ }
checkpoint-600/generation_config.json ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bad_words_ids": [
3
+ [
4
+ 64171
5
+ ]
6
+ ],
7
+ "bos_token_id": 0,
8
+ "decoder_start_token_id": 64171,
9
+ "eos_token_id": 0,
10
+ "forced_eos_token_id": 0,
11
+ "max_length": 512,
12
+ "num_beams": 6,
13
+ "pad_token_id": 64171,
14
+ "transformers_version": "4.27.3"
15
+ }
checkpoint-600/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2a0e4755674f056eacf97d924962a2e5d841acdb773975e2a4bedd4f5bf6d75b
3
+ size 616276997
checkpoint-600/pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3eea04476d9697fa71962473976b3416e5248309674ea553768c0f389a3abe77
3
+ size 308407621
checkpoint-600/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:47ac1c0f8b136e6cdcb221a075d1f58aff1fff607dc25f74f47fb8a74d814335
3
+ size 14575
checkpoint-600/scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5dc9eacfeb00bd0bfeb98934a2309be01be65b288e0d747bbfc423b32679169f
3
+ size 557
checkpoint-600/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d2ae1c671816119b4d45adcd15ae17d1daa9fa8260c3798bcd3486935e995c78
3
+ size 627
checkpoint-600/source.spm ADDED
Binary file (707 kB). View file
 
checkpoint-600/special_tokens_map.json ADDED
@@ -0,0 +1,5 @@
 
 
 
 
 
 
1
+ {
2
+ "eos_token": "</s>",
3
+ "pad_token": "<pad>",
4
+ "unk_token": "<unk>"
5
+ }
checkpoint-600/target.spm ADDED
Binary file (791 kB). View file
 
checkpoint-600/tokenizer_config.json ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "eos_token": "</s>",
3
+ "model_max_length": 512,
4
+ "pad_token": "<pad>",
5
+ "separate_vocabs": false,
6
+ "source_lang": "mul",
7
+ "sp_model_kwargs": {},
8
+ "special_tokens_map_file": null,
9
+ "target_lang": "eng",
10
+ "tokenizer_class": "MarianTokenizer",
11
+ "unk_token": "<unk>"
12
+ }
checkpoint-600/trainer_state.json ADDED
@@ -0,0 +1,78 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 2.9462931156158447,
3
+ "best_model_checkpoint": "/home/azawahry_sunbird_ai/models/m2e/dataset-tags-expanded-tokenizer/checkpoint-150",
4
+ "epoch": 1.024992739634761,
5
+ "global_step": 600,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.26,
12
+ "eval_BLEU_ach": 21.4505,
13
+ "eval_BLEU_lgg": 24.4794,
14
+ "eval_BLEU_lug": 35.1488,
15
+ "eval_BLEU_mean": 26.3376,
16
+ "eval_BLEU_nyn": 25.5547,
17
+ "eval_BLEU_teo": 25.0545,
18
+ "eval_loss": 2.9462931156158447,
19
+ "eval_runtime": 59.9936,
20
+ "eval_samples_per_second": 41.671,
21
+ "eval_steps_per_second": 0.833,
22
+ "step": 150
23
+ },
24
+ {
25
+ "epoch": 0.51,
26
+ "eval_BLEU_ach": 22.0556,
27
+ "eval_BLEU_lgg": 23.8502,
28
+ "eval_BLEU_lug": 34.6592,
29
+ "eval_BLEU_mean": 26.1806,
30
+ "eval_BLEU_nyn": 25.075,
31
+ "eval_BLEU_teo": 25.2631,
32
+ "eval_loss": 2.9494848251342773,
33
+ "eval_runtime": 58.1523,
34
+ "eval_samples_per_second": 42.991,
35
+ "eval_steps_per_second": 0.86,
36
+ "step": 300
37
+ },
38
+ {
39
+ "epoch": 0.77,
40
+ "eval_BLEU_ach": 21.6851,
41
+ "eval_BLEU_lgg": 24.2555,
42
+ "eval_BLEU_lug": 34.7523,
43
+ "eval_BLEU_mean": 26.1102,
44
+ "eval_BLEU_nyn": 24.5712,
45
+ "eval_BLEU_teo": 25.2871,
46
+ "eval_loss": 2.9563159942626953,
47
+ "eval_runtime": 57.8967,
48
+ "eval_samples_per_second": 43.18,
49
+ "eval_steps_per_second": 0.864,
50
+ "step": 450
51
+ },
52
+ {
53
+ "epoch": 0.85,
54
+ "learning_rate": 7.150997150997152e-05,
55
+ "loss": 2.4257,
56
+ "step": 500
57
+ },
58
+ {
59
+ "epoch": 1.02,
60
+ "eval_BLEU_ach": 21.8237,
61
+ "eval_BLEU_lgg": 23.3647,
62
+ "eval_BLEU_lug": 35.1063,
63
+ "eval_BLEU_mean": 25.832,
64
+ "eval_BLEU_nyn": 24.8711,
65
+ "eval_BLEU_teo": 23.9942,
66
+ "eval_loss": 2.9721972942352295,
67
+ "eval_runtime": 57.6727,
68
+ "eval_samples_per_second": 43.348,
69
+ "eval_steps_per_second": 0.867,
70
+ "step": 600
71
+ }
72
+ ],
73
+ "max_steps": 1755,
74
+ "num_train_epochs": 3,
75
+ "total_flos": 8.641348729778995e+16,
76
+ "trial_name": null,
77
+ "trial_params": null
78
+ }
checkpoint-600/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e46a42e4fff4eb7e4702f865fd1022499e6ad9fc3cf4d3f71aef8c990651b558
3
+ size 3707
checkpoint-600/vocab.json ADDED
The diff for this file is too large to render. See raw diff