Davlan commited on
Commit
e8528b2
1 Parent(s): 613bb48

add MT model

Browse files
all_results.json ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 10.0,
3
+ "predict_bleu": 8.7772,
4
+ "predict_gen_len": 86.2323,
5
+ "predict_loss": 2.777204990386963,
6
+ "predict_runtime": 1479.4496,
7
+ "predict_samples": 1558,
8
+ "predict_samples_per_second": 1.053,
9
+ "predict_steps_per_second": 0.105,
10
+ "train_loss": 1.0558881765379942,
11
+ "train_runtime": 3713.6591,
12
+ "train_samples": 6644,
13
+ "train_samples_per_second": 17.891,
14
+ "train_steps_per_second": 1.791
15
+ }
config.json ADDED
@@ -0,0 +1,63 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "/data/users/didelani/afribart/afrimbart",
3
+ "_num_labels": 3,
4
+ "activation_dropout": 0.0,
5
+ "activation_function": "gelu",
6
+ "adapters": {
7
+ "adapters": {},
8
+ "config_map": {},
9
+ "fusion_config_map": {},
10
+ "fusions": {}
11
+ },
12
+ "add_bias_logits": false,
13
+ "add_final_layer_norm": true,
14
+ "architectures": [
15
+ "MBartForConditionalGeneration"
16
+ ],
17
+ "attention_dropout": 0.0,
18
+ "bos_token_id": 0,
19
+ "classif_dropout": 0.0,
20
+ "classifier_dropout": 0.0,
21
+ "d_model": 1024,
22
+ "decoder_attention_heads": 16,
23
+ "decoder_ffn_dim": 4096,
24
+ "decoder_layerdrop": 0.0,
25
+ "decoder_layers": 12,
26
+ "decoder_start_token_id": 250020,
27
+ "dropout": 0.1,
28
+ "encoder_attention_heads": 16,
29
+ "encoder_ffn_dim": 4096,
30
+ "encoder_layerdrop": 0.0,
31
+ "encoder_layers": 12,
32
+ "eos_token_id": 2,
33
+ "forced_bos_token_id": 250043,
34
+ "forced_eos_token_id": 2,
35
+ "gradient_checkpointing": false,
36
+ "id2label": {
37
+ "0": "LABEL_0",
38
+ "1": "LABEL_1",
39
+ "2": "LABEL_2"
40
+ },
41
+ "init_std": 0.02,
42
+ "is_encoder_decoder": true,
43
+ "label2id": {
44
+ "LABEL_0": 0,
45
+ "LABEL_1": 1,
46
+ "LABEL_2": 2
47
+ },
48
+ "max_length": 1024,
49
+ "max_position_embeddings": 1024,
50
+ "model_type": "mbart",
51
+ "normalize_before": true,
52
+ "normalize_embedding": true,
53
+ "num_beams": 5,
54
+ "num_hidden_layers": 12,
55
+ "output_past": true,
56
+ "pad_token_id": 1,
57
+ "scale_embedding": true,
58
+ "static_position_embeddings": false,
59
+ "torch_dtype": "float32",
60
+ "transformers_version": "4.10.3",
61
+ "use_cache": true,
62
+ "vocab_size": 250054
63
+ }
generated_predictions.txt ADDED
The diff for this file is too large to render. See raw diff
 
predict_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "predict_bleu": 8.7772,
3
+ "predict_gen_len": 86.2323,
4
+ "predict_loss": 2.777204990386963,
5
+ "predict_runtime": 1479.4496,
6
+ "predict_samples": 1558,
7
+ "predict_samples_per_second": 1.053,
8
+ "predict_steps_per_second": 0.105
9
+ }
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2f9fcc96f22a2c1a2c3715fd2824c163ece893539bb8db186ae25be01899027e
3
+ size 2444723641
sentencepiece.bpe.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cfc8146abe2a0488e9e2a0c56de7952f7c11ab059eca145a0a727afce0db2865
3
+ size 5069051
special_tokens_map.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"bos_token": "<s>", "eos_token": "</s>", "unk_token": "<unk>", "sep_token": "</s>", "pad_token": "<pad>", "cls_token": "<s>", "mask_token": "<mask>", "additional_special_tokens": ["ar_AR", "cs_CZ", "de_DE", "en_XX", "es_XX", "et_EE", "fi_FI", "fr_XX", "gu_IN", "hi_IN", "it_IT", "ja_XX", "kk_KZ", "ko_KR", "lt_LT", "lv_LV", "my_MM", "ne_NP", "nl_XX", "ro_RO", "ru_RU", "si_LK", "tr_TR", "vi_VN", "zh_CN", "af_ZA", "az_AZ", "bn_IN", "fa_IR", "he_IL", "hr_HR", "id_ID", "ka_GE", "km_KH", "mk_MK", "ml_IN", "mn_MN", "mr_IN", "pl_PL", "ps_AF", "pt_XX", "sv_SE", "sw_KE", "ta_IN", "te_IN", "th_TH", "tl_XX", "uk_UA", "ur_PK", "xh_ZA", "gl_ES", "sl_SI"]}
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"src_lang": null, "tgt_lang": null, "eos_token": "</s>", "sep_token": "</s>", "cls_token": "<s>", "unk_token": "<unk>", "pad_token": "<pad>", "mask_token": {"content": "<mask>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "bos_token": "<s>", "model_max_length": 1024, "name_or_path": "/data/users/didelani/afribart/afrimbart", "special_tokens_map_file": "/home/suraj/projects/mbart-50/hf_models/mbart-50-large/special_tokens_map.json", "additional_special_tokens": ["ar_AR", "cs_CZ", "de_DE", "en_XX", "es_XX", "et_EE", "fi_FI", "fr_XX", "gu_IN", "hi_IN", "it_IT", "ja_XX", "kk_KZ", "ko_KR", "lt_LT", "lv_LV", "my_MM", "ne_NP", "nl_XX", "ro_RO", "ru_RU", "si_LK", "tr_TR", "vi_VN", "zh_CN", "af_ZA", "az_AZ", "bn_IN", "fa_IR", "he_IL", "hr_HR", "id_ID", "ka_GE", "km_KH", "mk_MK", "ml_IN", "mn_MN", "mr_IN", "pl_PL", "ps_AF", "pt_XX", "sv_SE", "sw_KE", "ta_IN", "te_IN", "th_TH", "tl_XX", "uk_UA", "ur_PK", "xh_ZA", "gl_ES", "sl_SI"], "sp_model_kwargs": {}, "tokenizer_class": "MBart50Tokenizer"}
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 10.0,
3
+ "train_loss": 1.0558881765379942,
4
+ "train_runtime": 3713.6591,
5
+ "train_samples": 6644,
6
+ "train_samples_per_second": 17.891,
7
+ "train_steps_per_second": 1.791
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,103 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 10.0,
5
+ "global_step": 6650,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.75,
12
+ "learning_rate": 4.62406015037594e-05,
13
+ "loss": 2.4571,
14
+ "step": 500
15
+ },
16
+ {
17
+ "epoch": 1.5,
18
+ "learning_rate": 4.24812030075188e-05,
19
+ "loss": 3.8854,
20
+ "step": 1000
21
+ },
22
+ {
23
+ "epoch": 2.26,
24
+ "learning_rate": 3.87218045112782e-05,
25
+ "loss": 1.9981,
26
+ "step": 1500
27
+ },
28
+ {
29
+ "epoch": 3.01,
30
+ "learning_rate": 3.49624060150376e-05,
31
+ "loss": 1.226,
32
+ "step": 2000
33
+ },
34
+ {
35
+ "epoch": 3.76,
36
+ "learning_rate": 3.120300751879699e-05,
37
+ "loss": 0.9436,
38
+ "step": 2500
39
+ },
40
+ {
41
+ "epoch": 4.51,
42
+ "learning_rate": 2.7443609022556393e-05,
43
+ "loss": 0.7817,
44
+ "step": 3000
45
+ },
46
+ {
47
+ "epoch": 5.26,
48
+ "learning_rate": 2.368421052631579e-05,
49
+ "loss": 0.6278,
50
+ "step": 3500
51
+ },
52
+ {
53
+ "epoch": 6.02,
54
+ "learning_rate": 1.9924812030075188e-05,
55
+ "loss": 0.5235,
56
+ "step": 4000
57
+ },
58
+ {
59
+ "epoch": 6.77,
60
+ "learning_rate": 1.6165413533834585e-05,
61
+ "loss": 0.3977,
62
+ "step": 4500
63
+ },
64
+ {
65
+ "epoch": 7.52,
66
+ "learning_rate": 1.2406015037593984e-05,
67
+ "loss": 0.3781,
68
+ "step": 5000
69
+ },
70
+ {
71
+ "epoch": 8.27,
72
+ "learning_rate": 8.646616541353383e-06,
73
+ "loss": 0.2964,
74
+ "step": 5500
75
+ },
76
+ {
77
+ "epoch": 9.02,
78
+ "learning_rate": 4.887218045112782e-06,
79
+ "loss": 0.2509,
80
+ "step": 6000
81
+ },
82
+ {
83
+ "epoch": 9.77,
84
+ "learning_rate": 1.1278195488721805e-06,
85
+ "loss": 0.2136,
86
+ "step": 6500
87
+ },
88
+ {
89
+ "epoch": 10.0,
90
+ "step": 6650,
91
+ "total_flos": 2.994539913333965e+16,
92
+ "train_loss": 1.0558881765379942,
93
+ "train_runtime": 3713.6591,
94
+ "train_samples_per_second": 17.891,
95
+ "train_steps_per_second": 1.791
96
+ }
97
+ ],
98
+ "max_steps": 6650,
99
+ "num_train_epochs": 10,
100
+ "total_flos": 2.994539913333965e+16,
101
+ "trial_name": null,
102
+ "trial_params": null
103
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fd79e134612951b1cb7cbd98443bc0e5fa81c697c7bcc81c6d118dbec8b5ea15
3
+ size 2799