Training in progress, step 500
Browse files- .gitattributes +3 -0
- config.json +58 -0
- generation_config.json +12 -0
- model.safetensors +3 -0
- phase1_test/checkpoint-1/config.json +58 -0
- phase1_test/checkpoint-1/generation_config.json +12 -0
- phase1_test/checkpoint-1/model.safetensors +3 -0
- phase1_test/checkpoint-1/optimizer.pt +3 -0
- phase1_test/checkpoint-1/rng_state.pth +3 -0
- phase1_test/checkpoint-1/scaler.pt +3 -0
- phase1_test/checkpoint-1/scheduler.pt +3 -0
- phase1_test/checkpoint-1/sentencepiece.bpe.model +3 -0
- phase1_test/checkpoint-1/special_tokens_map.json +69 -0
- phase1_test/checkpoint-1/tokenizer.json +3 -0
- phase1_test/checkpoint-1/tokenizer_config.json +528 -0
- phase1_test/checkpoint-1/trainer_state.json +41 -0
- phase1_test/checkpoint-1/training_args.bin +3 -0
- phase1_test/config.json +58 -0
- phase1_test/generation_config.json +11 -0
- phase1_test/model.safetensors +3 -0
- phase1_test/sentencepiece.bpe.model +3 -0
- phase1_test/special_tokens_map.json +69 -0
- phase1_test/test_results.json +190 -0
- phase1_test/tokenizer.json +3 -0
- phase1_test/tokenizer_config.json +528 -0
- phase1_test/training_args.bin +3 -0
- sentencepiece.bpe.model +3 -0
- special_tokens_map.json +69 -0
- tokenizer.json +3 -0
- tokenizer_config.json +528 -0
- training_args.bin +3 -0
.gitattributes
CHANGED
@@ -33,3 +33,6 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
|
33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
|
|
33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
36 |
+
phase1_test/checkpoint-1/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
37 |
+
phase1_test/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
38 |
+
tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
config.json
ADDED
@@ -0,0 +1,58 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "facebook/mbart-large-50",
|
3 |
+
"_num_labels": 3,
|
4 |
+
"activation_dropout": 0.0,
|
5 |
+
"activation_function": "gelu",
|
6 |
+
"add_bias_logits": false,
|
7 |
+
"add_final_layer_norm": true,
|
8 |
+
"architectures": [
|
9 |
+
"MBartForConditionalGeneration"
|
10 |
+
],
|
11 |
+
"attention_dropout": 0.0,
|
12 |
+
"bos_token_id": 0,
|
13 |
+
"classif_dropout": 0.0,
|
14 |
+
"classifier_dropout": 0.0,
|
15 |
+
"d_model": 1024,
|
16 |
+
"decoder_attention_heads": 16,
|
17 |
+
"decoder_ffn_dim": 4096,
|
18 |
+
"decoder_layerdrop": 0.0,
|
19 |
+
"decoder_layers": 12,
|
20 |
+
"decoder_start_token_id": 2,
|
21 |
+
"dropout": 0.1,
|
22 |
+
"early_stopping": null,
|
23 |
+
"encoder_attention_heads": 16,
|
24 |
+
"encoder_ffn_dim": 4096,
|
25 |
+
"encoder_layerdrop": 0.0,
|
26 |
+
"encoder_layers": 12,
|
27 |
+
"eos_token_id": 2,
|
28 |
+
"forced_eos_token_id": 2,
|
29 |
+
"gradient_checkpointing": false,
|
30 |
+
"id2label": {
|
31 |
+
"0": "LABEL_0",
|
32 |
+
"1": "LABEL_1",
|
33 |
+
"2": "LABEL_2"
|
34 |
+
},
|
35 |
+
"init_std": 0.02,
|
36 |
+
"is_encoder_decoder": true,
|
37 |
+
"label2id": {
|
38 |
+
"LABEL_0": 0,
|
39 |
+
"LABEL_1": 1,
|
40 |
+
"LABEL_2": 2
|
41 |
+
},
|
42 |
+
"max_length": null,
|
43 |
+
"max_position_embeddings": 1024,
|
44 |
+
"model_type": "mbart",
|
45 |
+
"normalize_before": true,
|
46 |
+
"normalize_embedding": true,
|
47 |
+
"num_beams": null,
|
48 |
+
"num_hidden_layers": 12,
|
49 |
+
"output_past": true,
|
50 |
+
"pad_token_id": 1,
|
51 |
+
"scale_embedding": true,
|
52 |
+
"static_position_embeddings": false,
|
53 |
+
"tokenizer_class": "MBart50Tokenizer",
|
54 |
+
"torch_dtype": "float32",
|
55 |
+
"transformers_version": "4.49.0",
|
56 |
+
"use_cache": true,
|
57 |
+
"vocab_size": 250054
|
58 |
+
}
|
generation_config.json
ADDED
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_from_model_config": true,
|
3 |
+
"bos_token_id": 0,
|
4 |
+
"decoder_start_token_id": 2,
|
5 |
+
"early_stopping": true,
|
6 |
+
"eos_token_id": 2,
|
7 |
+
"forced_eos_token_id": 2,
|
8 |
+
"max_length": 200,
|
9 |
+
"num_beams": 5,
|
10 |
+
"pad_token_id": 1,
|
11 |
+
"transformers_version": "4.49.0"
|
12 |
+
}
|
model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:fbd62e13cf6e658725dd6c29a2ee76b4c6263cdffc2ca228042c31d2ba643c20
|
3 |
+
size 2444578688
|
phase1_test/checkpoint-1/config.json
ADDED
@@ -0,0 +1,58 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "facebook/mbart-large-50",
|
3 |
+
"_num_labels": 3,
|
4 |
+
"activation_dropout": 0.0,
|
5 |
+
"activation_function": "gelu",
|
6 |
+
"add_bias_logits": false,
|
7 |
+
"add_final_layer_norm": true,
|
8 |
+
"architectures": [
|
9 |
+
"MBartForConditionalGeneration"
|
10 |
+
],
|
11 |
+
"attention_dropout": 0.0,
|
12 |
+
"bos_token_id": 0,
|
13 |
+
"classif_dropout": 0.0,
|
14 |
+
"classifier_dropout": 0.0,
|
15 |
+
"d_model": 1024,
|
16 |
+
"decoder_attention_heads": 16,
|
17 |
+
"decoder_ffn_dim": 4096,
|
18 |
+
"decoder_layerdrop": 0.0,
|
19 |
+
"decoder_layers": 12,
|
20 |
+
"decoder_start_token_id": 2,
|
21 |
+
"dropout": 0.1,
|
22 |
+
"early_stopping": null,
|
23 |
+
"encoder_attention_heads": 16,
|
24 |
+
"encoder_ffn_dim": 4096,
|
25 |
+
"encoder_layerdrop": 0.0,
|
26 |
+
"encoder_layers": 12,
|
27 |
+
"eos_token_id": 2,
|
28 |
+
"forced_eos_token_id": 2,
|
29 |
+
"gradient_checkpointing": false,
|
30 |
+
"id2label": {
|
31 |
+
"0": "LABEL_0",
|
32 |
+
"1": "LABEL_1",
|
33 |
+
"2": "LABEL_2"
|
34 |
+
},
|
35 |
+
"init_std": 0.02,
|
36 |
+
"is_encoder_decoder": true,
|
37 |
+
"label2id": {
|
38 |
+
"LABEL_0": 0,
|
39 |
+
"LABEL_1": 1,
|
40 |
+
"LABEL_2": 2
|
41 |
+
},
|
42 |
+
"max_length": null,
|
43 |
+
"max_position_embeddings": 1024,
|
44 |
+
"model_type": "mbart",
|
45 |
+
"normalize_before": true,
|
46 |
+
"normalize_embedding": true,
|
47 |
+
"num_beams": null,
|
48 |
+
"num_hidden_layers": 12,
|
49 |
+
"output_past": true,
|
50 |
+
"pad_token_id": 1,
|
51 |
+
"scale_embedding": true,
|
52 |
+
"static_position_embeddings": false,
|
53 |
+
"tokenizer_class": "MBart50Tokenizer",
|
54 |
+
"torch_dtype": "float32",
|
55 |
+
"transformers_version": "4.49.0",
|
56 |
+
"use_cache": true,
|
57 |
+
"vocab_size": 250054
|
58 |
+
}
|
phase1_test/checkpoint-1/generation_config.json
ADDED
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_from_model_config": true,
|
3 |
+
"bos_token_id": 0,
|
4 |
+
"decoder_start_token_id": 2,
|
5 |
+
"early_stopping": true,
|
6 |
+
"eos_token_id": 2,
|
7 |
+
"forced_eos_token_id": 2,
|
8 |
+
"max_length": 200,
|
9 |
+
"num_beams": 5,
|
10 |
+
"pad_token_id": 1,
|
11 |
+
"transformers_version": "4.49.0"
|
12 |
+
}
|
phase1_test/checkpoint-1/model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:02d71e6372d8d882bb9da77dee837ddbc40d6f65f57ec48febec02f34b0b494d
|
3 |
+
size 2444578688
|
phase1_test/checkpoint-1/optimizer.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:d3cfbafd47563e2a98a3f4498c247fba80489bdf72d690de94ae139a9f33af49
|
3 |
+
size 2536
|
phase1_test/checkpoint-1/rng_state.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:29086b6f5c59a8b951f71ddb74b34c9ca2dd50ea6554745ecf2848a6a770812c
|
3 |
+
size 14244
|
phase1_test/checkpoint-1/scaler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:bd88c1400734381e4f63d3ace0ea9a85534c65976e4e34bd6f4e74b7cb7c880b
|
3 |
+
size 988
|
phase1_test/checkpoint-1/scheduler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:8977ed50352cd3f78d91c962273e8d3a3891256fa18db85e55de46f9f672d88b
|
3 |
+
size 1064
|
phase1_test/checkpoint-1/sentencepiece.bpe.model
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:cfc8146abe2a0488e9e2a0c56de7952f7c11ab059eca145a0a727afce0db2865
|
3 |
+
size 5069051
|
phase1_test/checkpoint-1/special_tokens_map.json
ADDED
@@ -0,0 +1,69 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"additional_special_tokens": [
|
3 |
+
"ar_AR",
|
4 |
+
"cs_CZ",
|
5 |
+
"de_DE",
|
6 |
+
"en_XX",
|
7 |
+
"es_XX",
|
8 |
+
"et_EE",
|
9 |
+
"fi_FI",
|
10 |
+
"fr_XX",
|
11 |
+
"gu_IN",
|
12 |
+
"hi_IN",
|
13 |
+
"it_IT",
|
14 |
+
"ja_XX",
|
15 |
+
"kk_KZ",
|
16 |
+
"ko_KR",
|
17 |
+
"lt_LT",
|
18 |
+
"lv_LV",
|
19 |
+
"my_MM",
|
20 |
+
"ne_NP",
|
21 |
+
"nl_XX",
|
22 |
+
"ro_RO",
|
23 |
+
"ru_RU",
|
24 |
+
"si_LK",
|
25 |
+
"tr_TR",
|
26 |
+
"vi_VN",
|
27 |
+
"zh_CN",
|
28 |
+
"af_ZA",
|
29 |
+
"az_AZ",
|
30 |
+
"bn_IN",
|
31 |
+
"fa_IR",
|
32 |
+
"he_IL",
|
33 |
+
"hr_HR",
|
34 |
+
"id_ID",
|
35 |
+
"ka_GE",
|
36 |
+
"km_KH",
|
37 |
+
"mk_MK",
|
38 |
+
"ml_IN",
|
39 |
+
"mn_MN",
|
40 |
+
"mr_IN",
|
41 |
+
"pl_PL",
|
42 |
+
"ps_AF",
|
43 |
+
"pt_XX",
|
44 |
+
"sv_SE",
|
45 |
+
"sw_KE",
|
46 |
+
"ta_IN",
|
47 |
+
"te_IN",
|
48 |
+
"th_TH",
|
49 |
+
"tl_XX",
|
50 |
+
"uk_UA",
|
51 |
+
"ur_PK",
|
52 |
+
"xh_ZA",
|
53 |
+
"gl_ES",
|
54 |
+
"sl_SI"
|
55 |
+
],
|
56 |
+
"bos_token": "<s>",
|
57 |
+
"cls_token": "<s>",
|
58 |
+
"eos_token": "</s>",
|
59 |
+
"mask_token": {
|
60 |
+
"content": "<mask>",
|
61 |
+
"lstrip": true,
|
62 |
+
"normalized": true,
|
63 |
+
"rstrip": false,
|
64 |
+
"single_word": false
|
65 |
+
},
|
66 |
+
"pad_token": "<pad>",
|
67 |
+
"sep_token": "</s>",
|
68 |
+
"unk_token": "<unk>"
|
69 |
+
}
|
phase1_test/checkpoint-1/tokenizer.json
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:500bc815ab1b3a039f7de9cf48b4c97a58e503054d23c7a53a2e59f8d6666629
|
3 |
+
size 17110184
|
phase1_test/checkpoint-1/tokenizer_config.json
ADDED
@@ -0,0 +1,528 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"added_tokens_decoder": {
|
3 |
+
"0": {
|
4 |
+
"content": "<s>",
|
5 |
+
"lstrip": false,
|
6 |
+
"normalized": false,
|
7 |
+
"rstrip": false,
|
8 |
+
"single_word": false,
|
9 |
+
"special": true
|
10 |
+
},
|
11 |
+
"1": {
|
12 |
+
"content": "<pad>",
|
13 |
+
"lstrip": false,
|
14 |
+
"normalized": false,
|
15 |
+
"rstrip": false,
|
16 |
+
"single_word": false,
|
17 |
+
"special": true
|
18 |
+
},
|
19 |
+
"2": {
|
20 |
+
"content": "</s>",
|
21 |
+
"lstrip": false,
|
22 |
+
"normalized": false,
|
23 |
+
"rstrip": false,
|
24 |
+
"single_word": false,
|
25 |
+
"special": true
|
26 |
+
},
|
27 |
+
"3": {
|
28 |
+
"content": "<unk>",
|
29 |
+
"lstrip": false,
|
30 |
+
"normalized": false,
|
31 |
+
"rstrip": false,
|
32 |
+
"single_word": false,
|
33 |
+
"special": true
|
34 |
+
},
|
35 |
+
"250001": {
|
36 |
+
"content": "ar_AR",
|
37 |
+
"lstrip": false,
|
38 |
+
"normalized": false,
|
39 |
+
"rstrip": false,
|
40 |
+
"single_word": false,
|
41 |
+
"special": true
|
42 |
+
},
|
43 |
+
"250002": {
|
44 |
+
"content": "cs_CZ",
|
45 |
+
"lstrip": false,
|
46 |
+
"normalized": false,
|
47 |
+
"rstrip": false,
|
48 |
+
"single_word": false,
|
49 |
+
"special": true
|
50 |
+
},
|
51 |
+
"250003": {
|
52 |
+
"content": "de_DE",
|
53 |
+
"lstrip": false,
|
54 |
+
"normalized": false,
|
55 |
+
"rstrip": false,
|
56 |
+
"single_word": false,
|
57 |
+
"special": true
|
58 |
+
},
|
59 |
+
"250004": {
|
60 |
+
"content": "en_XX",
|
61 |
+
"lstrip": false,
|
62 |
+
"normalized": false,
|
63 |
+
"rstrip": false,
|
64 |
+
"single_word": false,
|
65 |
+
"special": true
|
66 |
+
},
|
67 |
+
"250005": {
|
68 |
+
"content": "es_XX",
|
69 |
+
"lstrip": false,
|
70 |
+
"normalized": false,
|
71 |
+
"rstrip": false,
|
72 |
+
"single_word": false,
|
73 |
+
"special": true
|
74 |
+
},
|
75 |
+
"250006": {
|
76 |
+
"content": "et_EE",
|
77 |
+
"lstrip": false,
|
78 |
+
"normalized": false,
|
79 |
+
"rstrip": false,
|
80 |
+
"single_word": false,
|
81 |
+
"special": true
|
82 |
+
},
|
83 |
+
"250007": {
|
84 |
+
"content": "fi_FI",
|
85 |
+
"lstrip": false,
|
86 |
+
"normalized": false,
|
87 |
+
"rstrip": false,
|
88 |
+
"single_word": false,
|
89 |
+
"special": true
|
90 |
+
},
|
91 |
+
"250008": {
|
92 |
+
"content": "fr_XX",
|
93 |
+
"lstrip": false,
|
94 |
+
"normalized": false,
|
95 |
+
"rstrip": false,
|
96 |
+
"single_word": false,
|
97 |
+
"special": true
|
98 |
+
},
|
99 |
+
"250009": {
|
100 |
+
"content": "gu_IN",
|
101 |
+
"lstrip": false,
|
102 |
+
"normalized": false,
|
103 |
+
"rstrip": false,
|
104 |
+
"single_word": false,
|
105 |
+
"special": true
|
106 |
+
},
|
107 |
+
"250010": {
|
108 |
+
"content": "hi_IN",
|
109 |
+
"lstrip": false,
|
110 |
+
"normalized": false,
|
111 |
+
"rstrip": false,
|
112 |
+
"single_word": false,
|
113 |
+
"special": true
|
114 |
+
},
|
115 |
+
"250011": {
|
116 |
+
"content": "it_IT",
|
117 |
+
"lstrip": false,
|
118 |
+
"normalized": false,
|
119 |
+
"rstrip": false,
|
120 |
+
"single_word": false,
|
121 |
+
"special": true
|
122 |
+
},
|
123 |
+
"250012": {
|
124 |
+
"content": "ja_XX",
|
125 |
+
"lstrip": false,
|
126 |
+
"normalized": false,
|
127 |
+
"rstrip": false,
|
128 |
+
"single_word": false,
|
129 |
+
"special": true
|
130 |
+
},
|
131 |
+
"250013": {
|
132 |
+
"content": "kk_KZ",
|
133 |
+
"lstrip": false,
|
134 |
+
"normalized": false,
|
135 |
+
"rstrip": false,
|
136 |
+
"single_word": false,
|
137 |
+
"special": true
|
138 |
+
},
|
139 |
+
"250014": {
|
140 |
+
"content": "ko_KR",
|
141 |
+
"lstrip": false,
|
142 |
+
"normalized": false,
|
143 |
+
"rstrip": false,
|
144 |
+
"single_word": false,
|
145 |
+
"special": true
|
146 |
+
},
|
147 |
+
"250015": {
|
148 |
+
"content": "lt_LT",
|
149 |
+
"lstrip": false,
|
150 |
+
"normalized": false,
|
151 |
+
"rstrip": false,
|
152 |
+
"single_word": false,
|
153 |
+
"special": true
|
154 |
+
},
|
155 |
+
"250016": {
|
156 |
+
"content": "lv_LV",
|
157 |
+
"lstrip": false,
|
158 |
+
"normalized": false,
|
159 |
+
"rstrip": false,
|
160 |
+
"single_word": false,
|
161 |
+
"special": true
|
162 |
+
},
|
163 |
+
"250017": {
|
164 |
+
"content": "my_MM",
|
165 |
+
"lstrip": false,
|
166 |
+
"normalized": false,
|
167 |
+
"rstrip": false,
|
168 |
+
"single_word": false,
|
169 |
+
"special": true
|
170 |
+
},
|
171 |
+
"250018": {
|
172 |
+
"content": "ne_NP",
|
173 |
+
"lstrip": false,
|
174 |
+
"normalized": false,
|
175 |
+
"rstrip": false,
|
176 |
+
"single_word": false,
|
177 |
+
"special": true
|
178 |
+
},
|
179 |
+
"250019": {
|
180 |
+
"content": "nl_XX",
|
181 |
+
"lstrip": false,
|
182 |
+
"normalized": false,
|
183 |
+
"rstrip": false,
|
184 |
+
"single_word": false,
|
185 |
+
"special": true
|
186 |
+
},
|
187 |
+
"250020": {
|
188 |
+
"content": "ro_RO",
|
189 |
+
"lstrip": false,
|
190 |
+
"normalized": false,
|
191 |
+
"rstrip": false,
|
192 |
+
"single_word": false,
|
193 |
+
"special": true
|
194 |
+
},
|
195 |
+
"250021": {
|
196 |
+
"content": "ru_RU",
|
197 |
+
"lstrip": false,
|
198 |
+
"normalized": false,
|
199 |
+
"rstrip": false,
|
200 |
+
"single_word": false,
|
201 |
+
"special": true
|
202 |
+
},
|
203 |
+
"250022": {
|
204 |
+
"content": "si_LK",
|
205 |
+
"lstrip": false,
|
206 |
+
"normalized": false,
|
207 |
+
"rstrip": false,
|
208 |
+
"single_word": false,
|
209 |
+
"special": true
|
210 |
+
},
|
211 |
+
"250023": {
|
212 |
+
"content": "tr_TR",
|
213 |
+
"lstrip": false,
|
214 |
+
"normalized": false,
|
215 |
+
"rstrip": false,
|
216 |
+
"single_word": false,
|
217 |
+
"special": true
|
218 |
+
},
|
219 |
+
"250024": {
|
220 |
+
"content": "vi_VN",
|
221 |
+
"lstrip": false,
|
222 |
+
"normalized": false,
|
223 |
+
"rstrip": false,
|
224 |
+
"single_word": false,
|
225 |
+
"special": true
|
226 |
+
},
|
227 |
+
"250025": {
|
228 |
+
"content": "zh_CN",
|
229 |
+
"lstrip": false,
|
230 |
+
"normalized": false,
|
231 |
+
"rstrip": false,
|
232 |
+
"single_word": false,
|
233 |
+
"special": true
|
234 |
+
},
|
235 |
+
"250026": {
|
236 |
+
"content": "af_ZA",
|
237 |
+
"lstrip": false,
|
238 |
+
"normalized": false,
|
239 |
+
"rstrip": false,
|
240 |
+
"single_word": false,
|
241 |
+
"special": true
|
242 |
+
},
|
243 |
+
"250027": {
|
244 |
+
"content": "az_AZ",
|
245 |
+
"lstrip": false,
|
246 |
+
"normalized": false,
|
247 |
+
"rstrip": false,
|
248 |
+
"single_word": false,
|
249 |
+
"special": true
|
250 |
+
},
|
251 |
+
"250028": {
|
252 |
+
"content": "bn_IN",
|
253 |
+
"lstrip": false,
|
254 |
+
"normalized": false,
|
255 |
+
"rstrip": false,
|
256 |
+
"single_word": false,
|
257 |
+
"special": true
|
258 |
+
},
|
259 |
+
"250029": {
|
260 |
+
"content": "fa_IR",
|
261 |
+
"lstrip": false,
|
262 |
+
"normalized": false,
|
263 |
+
"rstrip": false,
|
264 |
+
"single_word": false,
|
265 |
+
"special": true
|
266 |
+
},
|
267 |
+
"250030": {
|
268 |
+
"content": "he_IL",
|
269 |
+
"lstrip": false,
|
270 |
+
"normalized": false,
|
271 |
+
"rstrip": false,
|
272 |
+
"single_word": false,
|
273 |
+
"special": true
|
274 |
+
},
|
275 |
+
"250031": {
|
276 |
+
"content": "hr_HR",
|
277 |
+
"lstrip": false,
|
278 |
+
"normalized": false,
|
279 |
+
"rstrip": false,
|
280 |
+
"single_word": false,
|
281 |
+
"special": true
|
282 |
+
},
|
283 |
+
"250032": {
|
284 |
+
"content": "id_ID",
|
285 |
+
"lstrip": false,
|
286 |
+
"normalized": false,
|
287 |
+
"rstrip": false,
|
288 |
+
"single_word": false,
|
289 |
+
"special": true
|
290 |
+
},
|
291 |
+
"250033": {
|
292 |
+
"content": "ka_GE",
|
293 |
+
"lstrip": false,
|
294 |
+
"normalized": false,
|
295 |
+
"rstrip": false,
|
296 |
+
"single_word": false,
|
297 |
+
"special": true
|
298 |
+
},
|
299 |
+
"250034": {
|
300 |
+
"content": "km_KH",
|
301 |
+
"lstrip": false,
|
302 |
+
"normalized": false,
|
303 |
+
"rstrip": false,
|
304 |
+
"single_word": false,
|
305 |
+
"special": true
|
306 |
+
},
|
307 |
+
"250035": {
|
308 |
+
"content": "mk_MK",
|
309 |
+
"lstrip": false,
|
310 |
+
"normalized": false,
|
311 |
+
"rstrip": false,
|
312 |
+
"single_word": false,
|
313 |
+
"special": true
|
314 |
+
},
|
315 |
+
"250036": {
|
316 |
+
"content": "ml_IN",
|
317 |
+
"lstrip": false,
|
318 |
+
"normalized": false,
|
319 |
+
"rstrip": false,
|
320 |
+
"single_word": false,
|
321 |
+
"special": true
|
322 |
+
},
|
323 |
+
"250037": {
|
324 |
+
"content": "mn_MN",
|
325 |
+
"lstrip": false,
|
326 |
+
"normalized": false,
|
327 |
+
"rstrip": false,
|
328 |
+
"single_word": false,
|
329 |
+
"special": true
|
330 |
+
},
|
331 |
+
"250038": {
|
332 |
+
"content": "mr_IN",
|
333 |
+
"lstrip": false,
|
334 |
+
"normalized": false,
|
335 |
+
"rstrip": false,
|
336 |
+
"single_word": false,
|
337 |
+
"special": true
|
338 |
+
},
|
339 |
+
"250039": {
|
340 |
+
"content": "pl_PL",
|
341 |
+
"lstrip": false,
|
342 |
+
"normalized": false,
|
343 |
+
"rstrip": false,
|
344 |
+
"single_word": false,
|
345 |
+
"special": true
|
346 |
+
},
|
347 |
+
"250040": {
|
348 |
+
"content": "ps_AF",
|
349 |
+
"lstrip": false,
|
350 |
+
"normalized": false,
|
351 |
+
"rstrip": false,
|
352 |
+
"single_word": false,
|
353 |
+
"special": true
|
354 |
+
},
|
355 |
+
"250041": {
|
356 |
+
"content": "pt_XX",
|
357 |
+
"lstrip": false,
|
358 |
+
"normalized": false,
|
359 |
+
"rstrip": false,
|
360 |
+
"single_word": false,
|
361 |
+
"special": true
|
362 |
+
},
|
363 |
+
"250042": {
|
364 |
+
"content": "sv_SE",
|
365 |
+
"lstrip": false,
|
366 |
+
"normalized": false,
|
367 |
+
"rstrip": false,
|
368 |
+
"single_word": false,
|
369 |
+
"special": true
|
370 |
+
},
|
371 |
+
"250043": {
|
372 |
+
"content": "sw_KE",
|
373 |
+
"lstrip": false,
|
374 |
+
"normalized": false,
|
375 |
+
"rstrip": false,
|
376 |
+
"single_word": false,
|
377 |
+
"special": true
|
378 |
+
},
|
379 |
+
"250044": {
|
380 |
+
"content": "ta_IN",
|
381 |
+
"lstrip": false,
|
382 |
+
"normalized": false,
|
383 |
+
"rstrip": false,
|
384 |
+
"single_word": false,
|
385 |
+
"special": true
|
386 |
+
},
|
387 |
+
"250045": {
|
388 |
+
"content": "te_IN",
|
389 |
+
"lstrip": false,
|
390 |
+
"normalized": false,
|
391 |
+
"rstrip": false,
|
392 |
+
"single_word": false,
|
393 |
+
"special": true
|
394 |
+
},
|
395 |
+
"250046": {
|
396 |
+
"content": "th_TH",
|
397 |
+
"lstrip": false,
|
398 |
+
"normalized": false,
|
399 |
+
"rstrip": false,
|
400 |
+
"single_word": false,
|
401 |
+
"special": true
|
402 |
+
},
|
403 |
+
"250047": {
|
404 |
+
"content": "tl_XX",
|
405 |
+
"lstrip": false,
|
406 |
+
"normalized": false,
|
407 |
+
"rstrip": false,
|
408 |
+
"single_word": false,
|
409 |
+
"special": true
|
410 |
+
},
|
411 |
+
"250048": {
|
412 |
+
"content": "uk_UA",
|
413 |
+
"lstrip": false,
|
414 |
+
"normalized": false,
|
415 |
+
"rstrip": false,
|
416 |
+
"single_word": false,
|
417 |
+
"special": true
|
418 |
+
},
|
419 |
+
"250049": {
|
420 |
+
"content": "ur_PK",
|
421 |
+
"lstrip": false,
|
422 |
+
"normalized": false,
|
423 |
+
"rstrip": false,
|
424 |
+
"single_word": false,
|
425 |
+
"special": true
|
426 |
+
},
|
427 |
+
"250050": {
|
428 |
+
"content": "xh_ZA",
|
429 |
+
"lstrip": false,
|
430 |
+
"normalized": false,
|
431 |
+
"rstrip": false,
|
432 |
+
"single_word": false,
|
433 |
+
"special": true
|
434 |
+
},
|
435 |
+
"250051": {
|
436 |
+
"content": "gl_ES",
|
437 |
+
"lstrip": false,
|
438 |
+
"normalized": false,
|
439 |
+
"rstrip": false,
|
440 |
+
"single_word": false,
|
441 |
+
"special": true
|
442 |
+
},
|
443 |
+
"250052": {
|
444 |
+
"content": "sl_SI",
|
445 |
+
"lstrip": false,
|
446 |
+
"normalized": false,
|
447 |
+
"rstrip": false,
|
448 |
+
"single_word": false,
|
449 |
+
"special": true
|
450 |
+
},
|
451 |
+
"250053": {
|
452 |
+
"content": "<mask>",
|
453 |
+
"lstrip": true,
|
454 |
+
"normalized": true,
|
455 |
+
"rstrip": false,
|
456 |
+
"single_word": false,
|
457 |
+
"special": true
|
458 |
+
}
|
459 |
+
},
|
460 |
+
"additional_special_tokens": [
|
461 |
+
"ar_AR",
|
462 |
+
"cs_CZ",
|
463 |
+
"de_DE",
|
464 |
+
"en_XX",
|
465 |
+
"es_XX",
|
466 |
+
"et_EE",
|
467 |
+
"fi_FI",
|
468 |
+
"fr_XX",
|
469 |
+
"gu_IN",
|
470 |
+
"hi_IN",
|
471 |
+
"it_IT",
|
472 |
+
"ja_XX",
|
473 |
+
"kk_KZ",
|
474 |
+
"ko_KR",
|
475 |
+
"lt_LT",
|
476 |
+
"lv_LV",
|
477 |
+
"my_MM",
|
478 |
+
"ne_NP",
|
479 |
+
"nl_XX",
|
480 |
+
"ro_RO",
|
481 |
+
"ru_RU",
|
482 |
+
"si_LK",
|
483 |
+
"tr_TR",
|
484 |
+
"vi_VN",
|
485 |
+
"zh_CN",
|
486 |
+
"af_ZA",
|
487 |
+
"az_AZ",
|
488 |
+
"bn_IN",
|
489 |
+
"fa_IR",
|
490 |
+
"he_IL",
|
491 |
+
"hr_HR",
|
492 |
+
"id_ID",
|
493 |
+
"ka_GE",
|
494 |
+
"km_KH",
|
495 |
+
"mk_MK",
|
496 |
+
"ml_IN",
|
497 |
+
"mn_MN",
|
498 |
+
"mr_IN",
|
499 |
+
"pl_PL",
|
500 |
+
"ps_AF",
|
501 |
+
"pt_XX",
|
502 |
+
"sv_SE",
|
503 |
+
"sw_KE",
|
504 |
+
"ta_IN",
|
505 |
+
"te_IN",
|
506 |
+
"th_TH",
|
507 |
+
"tl_XX",
|
508 |
+
"uk_UA",
|
509 |
+
"ur_PK",
|
510 |
+
"xh_ZA",
|
511 |
+
"gl_ES",
|
512 |
+
"sl_SI"
|
513 |
+
],
|
514 |
+
"bos_token": "<s>",
|
515 |
+
"clean_up_tokenization_spaces": false,
|
516 |
+
"cls_token": "<s>",
|
517 |
+
"eos_token": "</s>",
|
518 |
+
"extra_special_tokens": {},
|
519 |
+
"mask_token": "<mask>",
|
520 |
+
"model_max_length": 1024,
|
521 |
+
"pad_token": "<pad>",
|
522 |
+
"sep_token": "</s>",
|
523 |
+
"sp_model_kwargs": {},
|
524 |
+
"src_lang": "ur_PK",
|
525 |
+
"tgt_lang": "ur_PK",
|
526 |
+
"tokenizer_class": "MBart50Tokenizer",
|
527 |
+
"unk_token": "<unk>"
|
528 |
+
}
|
phase1_test/checkpoint-1/trainer_state.json
ADDED
@@ -0,0 +1,41 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 1.0,
|
5 |
+
"eval_steps": 100,
|
6 |
+
"global_step": 1,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [],
|
11 |
+
"logging_steps": 100,
|
12 |
+
"max_steps": 1,
|
13 |
+
"num_input_tokens_seen": 0,
|
14 |
+
"num_train_epochs": 1,
|
15 |
+
"save_steps": 100,
|
16 |
+
"stateful_callbacks": {
|
17 |
+
"EarlyStoppingCallback": {
|
18 |
+
"args": {
|
19 |
+
"early_stopping_patience": 2,
|
20 |
+
"early_stopping_threshold": 0.001
|
21 |
+
},
|
22 |
+
"attributes": {
|
23 |
+
"early_stopping_patience_counter": 0
|
24 |
+
}
|
25 |
+
},
|
26 |
+
"TrainerControl": {
|
27 |
+
"args": {
|
28 |
+
"should_epoch_stop": false,
|
29 |
+
"should_evaluate": false,
|
30 |
+
"should_log": false,
|
31 |
+
"should_save": true,
|
32 |
+
"should_training_stop": true
|
33 |
+
},
|
34 |
+
"attributes": {}
|
35 |
+
}
|
36 |
+
},
|
37 |
+
"total_flos": 13544561049600.0,
|
38 |
+
"train_batch_size": 32,
|
39 |
+
"trial_name": null,
|
40 |
+
"trial_params": null
|
41 |
+
}
|
phase1_test/checkpoint-1/training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:609cf31c34e6e192184094c345a1f99375fa32c74a07075bd3a9d0dedea23412
|
3 |
+
size 5432
|
phase1_test/config.json
ADDED
@@ -0,0 +1,58 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "facebook/mbart-large-50",
|
3 |
+
"_num_labels": 3,
|
4 |
+
"activation_dropout": 0.0,
|
5 |
+
"activation_function": "gelu",
|
6 |
+
"add_bias_logits": false,
|
7 |
+
"add_final_layer_norm": true,
|
8 |
+
"architectures": [
|
9 |
+
"MBartForConditionalGeneration"
|
10 |
+
],
|
11 |
+
"attention_dropout": 0.0,
|
12 |
+
"bos_token_id": 0,
|
13 |
+
"classif_dropout": 0.0,
|
14 |
+
"classifier_dropout": 0.0,
|
15 |
+
"d_model": 1024,
|
16 |
+
"decoder_attention_heads": 16,
|
17 |
+
"decoder_ffn_dim": 4096,
|
18 |
+
"decoder_layerdrop": 0.0,
|
19 |
+
"decoder_layers": 12,
|
20 |
+
"decoder_start_token_id": 2,
|
21 |
+
"dropout": 0.1,
|
22 |
+
"early_stopping": null,
|
23 |
+
"encoder_attention_heads": 16,
|
24 |
+
"encoder_ffn_dim": 4096,
|
25 |
+
"encoder_layerdrop": 0.0,
|
26 |
+
"encoder_layers": 12,
|
27 |
+
"eos_token_id": 2,
|
28 |
+
"forced_eos_token_id": 2,
|
29 |
+
"gradient_checkpointing": false,
|
30 |
+
"id2label": {
|
31 |
+
"0": "LABEL_0",
|
32 |
+
"1": "LABEL_1",
|
33 |
+
"2": "LABEL_2"
|
34 |
+
},
|
35 |
+
"init_std": 0.02,
|
36 |
+
"is_encoder_decoder": true,
|
37 |
+
"label2id": {
|
38 |
+
"LABEL_0": 0,
|
39 |
+
"LABEL_1": 1,
|
40 |
+
"LABEL_2": 2
|
41 |
+
},
|
42 |
+
"max_length": null,
|
43 |
+
"max_position_embeddings": 1024,
|
44 |
+
"model_type": "mbart",
|
45 |
+
"normalize_before": true,
|
46 |
+
"normalize_embedding": true,
|
47 |
+
"num_beams": null,
|
48 |
+
"num_hidden_layers": 12,
|
49 |
+
"output_past": true,
|
50 |
+
"pad_token_id": 1,
|
51 |
+
"scale_embedding": true,
|
52 |
+
"static_position_embeddings": false,
|
53 |
+
"tokenizer_class": "MBart50Tokenizer",
|
54 |
+
"torch_dtype": "float32",
|
55 |
+
"transformers_version": "4.49.0",
|
56 |
+
"use_cache": true,
|
57 |
+
"vocab_size": 250054
|
58 |
+
}
|
phase1_test/generation_config.json
ADDED
@@ -0,0 +1,11 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"bos_token_id": 0,
|
3 |
+
"decoder_start_token_id": 250049,
|
4 |
+
"early_stopping": true,
|
5 |
+
"forced_eos_token_id": 2,
|
6 |
+
"max_length": 64,
|
7 |
+
"no_repeat_ngram_size": 3,
|
8 |
+
"num_beams": 5,
|
9 |
+
"pad_token_id": 1,
|
10 |
+
"transformers_version": "4.49.0"
|
11 |
+
}
|
phase1_test/model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:02d71e6372d8d882bb9da77dee837ddbc40d6f65f57ec48febec02f34b0b494d
|
3 |
+
size 2444578688
|
phase1_test/sentencepiece.bpe.model
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:cfc8146abe2a0488e9e2a0c56de7952f7c11ab059eca145a0a727afce0db2865
|
3 |
+
size 5069051
|
phase1_test/special_tokens_map.json
ADDED
@@ -0,0 +1,69 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"additional_special_tokens": [
|
3 |
+
"ar_AR",
|
4 |
+
"cs_CZ",
|
5 |
+
"de_DE",
|
6 |
+
"en_XX",
|
7 |
+
"es_XX",
|
8 |
+
"et_EE",
|
9 |
+
"fi_FI",
|
10 |
+
"fr_XX",
|
11 |
+
"gu_IN",
|
12 |
+
"hi_IN",
|
13 |
+
"it_IT",
|
14 |
+
"ja_XX",
|
15 |
+
"kk_KZ",
|
16 |
+
"ko_KR",
|
17 |
+
"lt_LT",
|
18 |
+
"lv_LV",
|
19 |
+
"my_MM",
|
20 |
+
"ne_NP",
|
21 |
+
"nl_XX",
|
22 |
+
"ro_RO",
|
23 |
+
"ru_RU",
|
24 |
+
"si_LK",
|
25 |
+
"tr_TR",
|
26 |
+
"vi_VN",
|
27 |
+
"zh_CN",
|
28 |
+
"af_ZA",
|
29 |
+
"az_AZ",
|
30 |
+
"bn_IN",
|
31 |
+
"fa_IR",
|
32 |
+
"he_IL",
|
33 |
+
"hr_HR",
|
34 |
+
"id_ID",
|
35 |
+
"ka_GE",
|
36 |
+
"km_KH",
|
37 |
+
"mk_MK",
|
38 |
+
"ml_IN",
|
39 |
+
"mn_MN",
|
40 |
+
"mr_IN",
|
41 |
+
"pl_PL",
|
42 |
+
"ps_AF",
|
43 |
+
"pt_XX",
|
44 |
+
"sv_SE",
|
45 |
+
"sw_KE",
|
46 |
+
"ta_IN",
|
47 |
+
"te_IN",
|
48 |
+
"th_TH",
|
49 |
+
"tl_XX",
|
50 |
+
"uk_UA",
|
51 |
+
"ur_PK",
|
52 |
+
"xh_ZA",
|
53 |
+
"gl_ES",
|
54 |
+
"sl_SI"
|
55 |
+
],
|
56 |
+
"bos_token": "<s>",
|
57 |
+
"cls_token": "<s>",
|
58 |
+
"eos_token": "</s>",
|
59 |
+
"mask_token": {
|
60 |
+
"content": "<mask>",
|
61 |
+
"lstrip": true,
|
62 |
+
"normalized": true,
|
63 |
+
"rstrip": false,
|
64 |
+
"single_word": false
|
65 |
+
},
|
66 |
+
"pad_token": "<pad>",
|
67 |
+
"sep_token": "</s>",
|
68 |
+
"unk_token": "<unk>"
|
69 |
+
}
|
phase1_test/test_results.json
ADDED
@@ -0,0 +1,190 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"metrics": {
|
3 |
+
"bleu": 0.0,
|
4 |
+
"wer": 1.0466666666666666,
|
5 |
+
"cer": 0.9398562179165824,
|
6 |
+
"exact_match_rate": 0.0
|
7 |
+
},
|
8 |
+
"examples": [
|
9 |
+
{
|
10 |
+
"source": "لاوہرراد پوائنٹ ابارلازہررری 05 ستمب2015ء انعگلشک اونٹیییک رکٹ سعید ایجلکمیئا ڈراونا خاب ںںن",
|
11 |
+
"target": "لاہوراردو پوائنٹ اخبار تازہ ترین 05 ستمبر2015ء انگلش کاونٹی کرکٹ سعید اجمل کیلئے ڈراونا خواب بن",
|
12 |
+
"prediction": "Posted in: Uncategorized _ Tagged: ,,,,,,,,,,,,,,,,,,,,,",
|
13 |
+
"exact_match": false,
|
14 |
+
"bleu": 0.0,
|
15 |
+
"wer": 1.0,
|
16 |
+
"cer": 0.9473684210526315
|
17 |
+
},
|
18 |
+
{
|
19 |
+
"source": "سچنطٹبڈلرنے 19اووور بھارہ وکےےے ساروگنگولی نے 214ون ڈے یچز میں 20سسسغنریا ںسکورکی تھیں",
|
20 |
+
"target": "سچن ٹبڈولکرنے 197اور بھارت کے ساروگنگولی نے 214ون ڈے میچز میں 20سنچریاں سکورکی تھیں",
|
21 |
+
"prediction": ". . . . . . . . .",
|
22 |
+
"exact_match": false,
|
23 |
+
"bleu": 0.0,
|
24 |
+
"wer": 1.0,
|
25 |
+
"cer": 0.9036144578313253
|
26 |
+
},
|
27 |
+
{
|
28 |
+
"source": "لکاست کوگلگتب لتسیییاننن میںڈسرگڈسز کے اااغازززپر مبارکب دیت ہیں ور ڈیجیکٹل زکاستان کےم قاصد شے رصول",
|
29 |
+
"target": "پاکستان کو گلگت بلتستان میں سروسز کے اغازپر مبارکباد دیتے ہیں اور ڈیجیٹل پاکستان کے مقاصد کے حصول",
|
30 |
+
"prediction": ". . . . . . . . . .",
|
31 |
+
"exact_match": false,
|
32 |
+
"bleu": 0.0,
|
33 |
+
"wer": 1.0,
|
34 |
+
"cer": 0.9072164948453608
|
35 |
+
},
|
36 |
+
{
|
37 |
+
"source": "بکسیر دییا غغغتچھا",
|
38 |
+
"target": "بکھیر دیا تھا",
|
39 |
+
"prediction": ". . . . . . . . . .",
|
40 |
+
"exact_match": false,
|
41 |
+
"bleu": 0.0,
|
42 |
+
"wer": 3.3333333333333335,
|
43 |
+
"cer": 1.3076923076923077
|
44 |
+
},
|
45 |
+
{
|
46 |
+
"source": "اءونٹ غی ںکھیلنے پتاحیاہ پابنعی عائد کردی صیلاتکے بمطابق فیفا کپفیڈریشن کپ 2005 کرے دورانذمممروہ",
|
47 |
+
"target": "ایونٹ میں کھیلنے پر تاحیات پابندی عائد کردی تفصیلات کے مطابق فیفا کنفیڈریشن کپ 2005 کے دوران مذکورہ",
|
48 |
+
"prediction": ",,,,,,,,,,,,,,,,,,,,,",
|
49 |
+
"exact_match": false,
|
50 |
+
"bleu": 0.0,
|
51 |
+
"wer": 1.0,
|
52 |
+
"cer": 1.0
|
53 |
+
},
|
54 |
+
{
|
55 |
+
"source": "ورلڈ کپچ کے میچغں مممیں ٹیم کی کءءءرکردگی جوبی اعدریقہ سے اچھی تیھ میڈڈڈیا سے یت چگ تکپتےگہوذے ظہیررر عععباس",
|
56 |
+
"target": "ورلڈ کپ کے میچوں میں ٹیم کی کارکردگی جنوبی افریقہ سے اچھی تھی میڈیا سے بات چیت کرتے ہوئے ظہیر عباس",
|
57 |
+
"prediction": ". . . . . . . . . . .",
|
58 |
+
"exact_match": false,
|
59 |
+
"bleu": 0.0,
|
60 |
+
"wer": 1.0,
|
61 |
+
"cer": 0.8979591836734694
|
62 |
+
},
|
63 |
+
{
|
64 |
+
"source": "بدلا مشکککل ہے بوتے نے کککا مصصصدپٹصصصم کا حوصلہ نڑدا نےہ کتپان سسسرفراز بحمد نے کہا کہچ نجوبیڑافریقققی",
|
65 |
+
"target": "بدلنا مشکل ہے بولتے رہنے کا مقصد ٹیم کا حوصلہ بڑھانا ہے کپتان سرفراز احمد نے کہا کہ جنوبی افریقی",
|
66 |
+
"prediction": ".:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.",
|
67 |
+
"exact_match": false,
|
68 |
+
"bleu": 0.0,
|
69 |
+
"wer": 1.0,
|
70 |
+
"cer": 1.0
|
71 |
+
},
|
72 |
+
{
|
73 |
+
"source": "اسلام ادادائی این پچپاکستان اروا سکا ٹلللیخڈ کیزکرکٹ ٹیموں کے درمصیانچدو ٹی ٹوئئئٹذیمایچوں پر مشژل",
|
74 |
+
"target": "اسلام ابادائی این پیپاکستان اورا سکاٹ لینڈ کی کرکٹ ٹیموں کے درمیان دو ٹی ٹوئنٹی میچوں پر مشتمل",
|
75 |
+
"prediction": "Posted in: Uncategorized _ Tagged: ,,,,,,,,,,,,,,,,,,,,,,,,,",
|
76 |
+
"exact_match": false,
|
77 |
+
"bleu": 0.0,
|
78 |
+
"wer": 1.0,
|
79 |
+
"cer": 0.9468085106382979
|
80 |
+
},
|
81 |
+
{
|
82 |
+
"source": "ھبء گشررریکککروں گے گس میں نہ صرف قءومی کرکٹٹٹ ٹیم اور کپںںںان کی ورلڈ زپ مےں پفررمنس کا جائزہ لیا جائے گا",
|
83 |
+
"target": "بھی شریک ہوں گے جس میں نہ صرف قومی کرکٹ ٹیم اور کپتان کی ورلڈ کپ میں پرفارمنس کا جائزہ لیا جائے گا",
|
84 |
+
"prediction": "Posted in: Uncategorized _ Tagged: ,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,",
|
85 |
+
"exact_match": false,
|
86 |
+
"bleu": 0.0,
|
87 |
+
"wer": 1.0,
|
88 |
+
"cer": 0.9489795918367347
|
89 |
+
},
|
90 |
+
{
|
91 |
+
"source": "کی سفارشات وزارت ئم یٹی ءو بشھیجی تھیییں ضیااان مءیں اااس کا حوالہطدیتے وئاک ہا گیییا وزززغتکے پساننننے متعد",
|
92 |
+
"target": "کی سفارشات وزارت ئی ٹی کو بھیجی تھیں بیان میں اس کا حوالہ دیتے ہوئے کہا گیا وزارت کے سامنے متعدد",
|
93 |
+
"prediction": ". . . . . . . . . . . .",
|
94 |
+
"exact_match": false,
|
95 |
+
"bleu": 0.0,
|
96 |
+
"wer": 1.0,
|
97 |
+
"cer": 0.8854166666666666
|
98 |
+
},
|
99 |
+
{
|
100 |
+
"source": "بلیوز ے زاداعظمتٹرایف کے فائنننل میب مھیں لاوہرگسٹٹٹ کے خلایف وکٹقں رپ خ802 رن بنائے تھے دوسری جانب سے",
|
101 |
+
"target": "بلیوز نے قائداعظم ٹرافی کے فائنل میچ میں لاہور سٹی کے خلاف وکٹوں پر 802 رنز بنائے تھے دوسری جانب سے",
|
102 |
+
"prediction": ". . . . . . . . . . . . .",
|
103 |
+
"exact_match": false,
|
104 |
+
"bleu": 0.0,
|
105 |
+
"wer": 1.0,
|
106 |
+
"cer": 0.8787878787878788
|
107 |
+
},
|
108 |
+
{
|
109 |
+
"source": "کر چرے تھے داران گتگمو اس وقت سب لواوں ک دک الگاج بٹہ پایتک دم ہمیشپڈ گر گئے جسسس کے بعدد اہنیں",
|
110 |
+
"target": "کر رہے تھے دوران گفتگو اس وقت سب لوگوں کو دھچکا لگا جب وہ ایک دم زمین پر گر گئے جس کے بعد انہیں",
|
111 |
+
"prediction": ". . . . . . . . . . .",
|
112 |
+
"exact_match": false,
|
113 |
+
"bleu": 0.0,
|
114 |
+
"wer": 1.0,
|
115 |
+
"cer": 0.8947368421052632
|
116 |
+
},
|
117 |
+
{
|
118 |
+
"source": "کامیابی ک جانب جللس یمرے بڑھتے ہئے قصدم دککک ھکر کی ستھگی فنکارغایئغ پریششش اہیں میری کامیابی کی",
|
119 |
+
"target": "کامیابی کی جانب مسلسل میرے بڑھتے ہوئے قدم دیکھ کر کئی ساتھی فنکارائیں پریشان ہیں میری کامیابی کی",
|
120 |
+
"prediction": ". . . . . . . . . . .",
|
121 |
+
"exact_match": false,
|
122 |
+
"bleu": 0.0,
|
123 |
+
"wer": 1.0,
|
124 |
+
"cer": 0.8958333333333334
|
125 |
+
},
|
126 |
+
{
|
127 |
+
"source": "ہاتے ہیں ہک ضہہہو ورپڈاٹی ٹوئنٹی واررر نیشیا ک جیییسےےےاہم یاوٹنسسسسے قبل ٹف ہو جایپںاکستا پر لیگلمممیںںں",
|
128 |
+
"target": "چاہتے ہیں کہ وہ ورلڈ ٹی ٹوئنٹی اور ایشیا کپ جیسے اہم ایونٹس سے قبل فٹ ہو جائیںپاکستان سپر لیگ میں",
|
129 |
+
"prediction": ". . . . . . . . . . .",
|
130 |
+
"exact_match": false,
|
131 |
+
"bleu": 0.0,
|
132 |
+
"wer": 1.0,
|
133 |
+
"cer": 0.8969072164948454
|
134 |
+
},
|
135 |
+
{
|
136 |
+
"source": "چونہک وہ وبرڈک ے ضاہطہژ اخالق میکں ثامل ہیں ج کفی وجہ سےقان پر ٹٹٹاضح کرددتیاہےذکجہ بورڈ کسیقبھجی کھلھاڑی",
|
137 |
+
"target": "چونکہ وہ بورڈ کے ضابطہ اخلاق میں شامل ہیں جس کی وجہ سے ان پر واضح کر دیا ہے کہ بورڈ کسی بھی کھلاڑی",
|
138 |
+
"prediction": ",,,,,,,,,,,,,,,,,,,,,,,,",
|
139 |
+
"exact_match": false,
|
140 |
+
"bleu": 0.0,
|
141 |
+
"wer": 1.0,
|
142 |
+
"cer": 1.0
|
143 |
+
},
|
144 |
+
{
|
145 |
+
"source": "کککس ے کمظ نیں ببمہ اجک ے دوشر میں کوچزک ااا تعزھی یےاتفہ قہوا بہت ضرروییی ہےےے ناوءکاااخن ویتھبگیما زمیں",
|
146 |
+
"target": "کسی سے کم نہیں بلکہ اج کے دور میں کوچز کا تعلیم یافتہ ہونا بہت ضروری ہے اورکامن ویلتھ گیمز میں",
|
147 |
+
"prediction": ". . . . . . . . . . .",
|
148 |
+
"exact_match": false,
|
149 |
+
"bleu": 0.0,
|
150 |
+
"wer": 1.0,
|
151 |
+
"cer": 0.8936170212765957
|
152 |
+
},
|
153 |
+
{
|
154 |
+
"source": "پن یباذذذئ ٹمکینک لی کووو ودتراہ ہتٹحرک کیااا ہےیلر ہے ک ہاجملا ںیو ںپاکستانننپی غھلا یہہہں جنتتتکا",
|
155 |
+
"target": "اپنی بائی مکینک لیب کو دوبارہ متحرک کیا ہےخیال رہے کہ اجمل ساتویں پاکستانی کھلاڑی ہیں جن کا",
|
156 |
+
"prediction": ". . . . . . . . .",
|
157 |
+
"exact_match": false,
|
158 |
+
"bleu": 0.0,
|
159 |
+
"wer": 1.0,
|
160 |
+
"cer": 0.9120879120879121
|
161 |
+
},
|
162 |
+
{
|
163 |
+
"source": "سرژمااایییہ کاری مڈا تزبردست کمی ٹہے یتنامب ھامممرت بنگلہ دلش اور غسرٹی لنکات ک یم پاکستٹن سےےے دگنی سے زیداہ",
|
164 |
+
"target": "سرمایہ کاری میں زبردست کمی ہے ویتنام بھارت بنگلہ دیش اور سری لنکا تک میں پاکستان سے دگنی سے زیادہ",
|
165 |
+
"prediction": ". . . . . . . . . . .",
|
166 |
+
"exact_match": false,
|
167 |
+
"bleu": 0.0,
|
168 |
+
"wer": 1.0,
|
169 |
+
"cer": 0.8969072164948454
|
170 |
+
},
|
171 |
+
{
|
172 |
+
"source": "فففانڈڈڈیشنمممک ی جابنننب سبے دراموی جانے والی2017ماااڈڈڈل کی لینڈ رکوزکےل اے وزززاصت تجارترنپ این او فی جی",
|
173 |
+
"target": "فانڈیشن کی جانب سے درامدکی جانے والی 2017ماڈل کی لینڈ کروزر کے لیے وزارت تجارت نے این او سی جاری",
|
174 |
+
"prediction": ". . . . . . . . . . .",
|
175 |
+
"exact_match": false,
|
176 |
+
"bleu": 0.0,
|
177 |
+
"wer": 1.0,
|
178 |
+
"cer": 0.8958333333333334
|
179 |
+
},
|
180 |
+
{
|
181 |
+
"source": "لاااشہو رسپورٹپس رپورٹرعاپحححچیںاکککیشین بچیہینڈ بال چیمپنئ ھشپ ح2015ءءءمیں اپپپکساتتتن ٹیمے اپےمدوسرےپ مین یں",
|
182 |
+
"target": "لاہور سپورٹس رپورٹر پانچویں ایشین بیچ ہینڈ بال چیمپئن شپ 2015ءمیں پاکستان ٹیم اپنے دوسرے میچ میں",
|
183 |
+
"prediction": ".::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::",
|
184 |
+
"exact_match": false,
|
185 |
+
"bleu": 0.0,
|
186 |
+
"wer": 1.0,
|
187 |
+
"cer": 1.0
|
188 |
+
}
|
189 |
+
]
|
190 |
+
}
|
phase1_test/tokenizer.json
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:500bc815ab1b3a039f7de9cf48b4c97a58e503054d23c7a53a2e59f8d6666629
|
3 |
+
size 17110184
|
phase1_test/tokenizer_config.json
ADDED
@@ -0,0 +1,528 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"added_tokens_decoder": {
|
3 |
+
"0": {
|
4 |
+
"content": "<s>",
|
5 |
+
"lstrip": false,
|
6 |
+
"normalized": false,
|
7 |
+
"rstrip": false,
|
8 |
+
"single_word": false,
|
9 |
+
"special": true
|
10 |
+
},
|
11 |
+
"1": {
|
12 |
+
"content": "<pad>",
|
13 |
+
"lstrip": false,
|
14 |
+
"normalized": false,
|
15 |
+
"rstrip": false,
|
16 |
+
"single_word": false,
|
17 |
+
"special": true
|
18 |
+
},
|
19 |
+
"2": {
|
20 |
+
"content": "</s>",
|
21 |
+
"lstrip": false,
|
22 |
+
"normalized": false,
|
23 |
+
"rstrip": false,
|
24 |
+
"single_word": false,
|
25 |
+
"special": true
|
26 |
+
},
|
27 |
+
"3": {
|
28 |
+
"content": "<unk>",
|
29 |
+
"lstrip": false,
|
30 |
+
"normalized": false,
|
31 |
+
"rstrip": false,
|
32 |
+
"single_word": false,
|
33 |
+
"special": true
|
34 |
+
},
|
35 |
+
"250001": {
|
36 |
+
"content": "ar_AR",
|
37 |
+
"lstrip": false,
|
38 |
+
"normalized": false,
|
39 |
+
"rstrip": false,
|
40 |
+
"single_word": false,
|
41 |
+
"special": true
|
42 |
+
},
|
43 |
+
"250002": {
|
44 |
+
"content": "cs_CZ",
|
45 |
+
"lstrip": false,
|
46 |
+
"normalized": false,
|
47 |
+
"rstrip": false,
|
48 |
+
"single_word": false,
|
49 |
+
"special": true
|
50 |
+
},
|
51 |
+
"250003": {
|
52 |
+
"content": "de_DE",
|
53 |
+
"lstrip": false,
|
54 |
+
"normalized": false,
|
55 |
+
"rstrip": false,
|
56 |
+
"single_word": false,
|
57 |
+
"special": true
|
58 |
+
},
|
59 |
+
"250004": {
|
60 |
+
"content": "en_XX",
|
61 |
+
"lstrip": false,
|
62 |
+
"normalized": false,
|
63 |
+
"rstrip": false,
|
64 |
+
"single_word": false,
|
65 |
+
"special": true
|
66 |
+
},
|
67 |
+
"250005": {
|
68 |
+
"content": "es_XX",
|
69 |
+
"lstrip": false,
|
70 |
+
"normalized": false,
|
71 |
+
"rstrip": false,
|
72 |
+
"single_word": false,
|
73 |
+
"special": true
|
74 |
+
},
|
75 |
+
"250006": {
|
76 |
+
"content": "et_EE",
|
77 |
+
"lstrip": false,
|
78 |
+
"normalized": false,
|
79 |
+
"rstrip": false,
|
80 |
+
"single_word": false,
|
81 |
+
"special": true
|
82 |
+
},
|
83 |
+
"250007": {
|
84 |
+
"content": "fi_FI",
|
85 |
+
"lstrip": false,
|
86 |
+
"normalized": false,
|
87 |
+
"rstrip": false,
|
88 |
+
"single_word": false,
|
89 |
+
"special": true
|
90 |
+
},
|
91 |
+
"250008": {
|
92 |
+
"content": "fr_XX",
|
93 |
+
"lstrip": false,
|
94 |
+
"normalized": false,
|
95 |
+
"rstrip": false,
|
96 |
+
"single_word": false,
|
97 |
+
"special": true
|
98 |
+
},
|
99 |
+
"250009": {
|
100 |
+
"content": "gu_IN",
|
101 |
+
"lstrip": false,
|
102 |
+
"normalized": false,
|
103 |
+
"rstrip": false,
|
104 |
+
"single_word": false,
|
105 |
+
"special": true
|
106 |
+
},
|
107 |
+
"250010": {
|
108 |
+
"content": "hi_IN",
|
109 |
+
"lstrip": false,
|
110 |
+
"normalized": false,
|
111 |
+
"rstrip": false,
|
112 |
+
"single_word": false,
|
113 |
+
"special": true
|
114 |
+
},
|
115 |
+
"250011": {
|
116 |
+
"content": "it_IT",
|
117 |
+
"lstrip": false,
|
118 |
+
"normalized": false,
|
119 |
+
"rstrip": false,
|
120 |
+
"single_word": false,
|
121 |
+
"special": true
|
122 |
+
},
|
123 |
+
"250012": {
|
124 |
+
"content": "ja_XX",
|
125 |
+
"lstrip": false,
|
126 |
+
"normalized": false,
|
127 |
+
"rstrip": false,
|
128 |
+
"single_word": false,
|
129 |
+
"special": true
|
130 |
+
},
|
131 |
+
"250013": {
|
132 |
+
"content": "kk_KZ",
|
133 |
+
"lstrip": false,
|
134 |
+
"normalized": false,
|
135 |
+
"rstrip": false,
|
136 |
+
"single_word": false,
|
137 |
+
"special": true
|
138 |
+
},
|
139 |
+
"250014": {
|
140 |
+
"content": "ko_KR",
|
141 |
+
"lstrip": false,
|
142 |
+
"normalized": false,
|
143 |
+
"rstrip": false,
|
144 |
+
"single_word": false,
|
145 |
+
"special": true
|
146 |
+
},
|
147 |
+
"250015": {
|
148 |
+
"content": "lt_LT",
|
149 |
+
"lstrip": false,
|
150 |
+
"normalized": false,
|
151 |
+
"rstrip": false,
|
152 |
+
"single_word": false,
|
153 |
+
"special": true
|
154 |
+
},
|
155 |
+
"250016": {
|
156 |
+
"content": "lv_LV",
|
157 |
+
"lstrip": false,
|
158 |
+
"normalized": false,
|
159 |
+
"rstrip": false,
|
160 |
+
"single_word": false,
|
161 |
+
"special": true
|
162 |
+
},
|
163 |
+
"250017": {
|
164 |
+
"content": "my_MM",
|
165 |
+
"lstrip": false,
|
166 |
+
"normalized": false,
|
167 |
+
"rstrip": false,
|
168 |
+
"single_word": false,
|
169 |
+
"special": true
|
170 |
+
},
|
171 |
+
"250018": {
|
172 |
+
"content": "ne_NP",
|
173 |
+
"lstrip": false,
|
174 |
+
"normalized": false,
|
175 |
+
"rstrip": false,
|
176 |
+
"single_word": false,
|
177 |
+
"special": true
|
178 |
+
},
|
179 |
+
"250019": {
|
180 |
+
"content": "nl_XX",
|
181 |
+
"lstrip": false,
|
182 |
+
"normalized": false,
|
183 |
+
"rstrip": false,
|
184 |
+
"single_word": false,
|
185 |
+
"special": true
|
186 |
+
},
|
187 |
+
"250020": {
|
188 |
+
"content": "ro_RO",
|
189 |
+
"lstrip": false,
|
190 |
+
"normalized": false,
|
191 |
+
"rstrip": false,
|
192 |
+
"single_word": false,
|
193 |
+
"special": true
|
194 |
+
},
|
195 |
+
"250021": {
|
196 |
+
"content": "ru_RU",
|
197 |
+
"lstrip": false,
|
198 |
+
"normalized": false,
|
199 |
+
"rstrip": false,
|
200 |
+
"single_word": false,
|
201 |
+
"special": true
|
202 |
+
},
|
203 |
+
"250022": {
|
204 |
+
"content": "si_LK",
|
205 |
+
"lstrip": false,
|
206 |
+
"normalized": false,
|
207 |
+
"rstrip": false,
|
208 |
+
"single_word": false,
|
209 |
+
"special": true
|
210 |
+
},
|
211 |
+
"250023": {
|
212 |
+
"content": "tr_TR",
|
213 |
+
"lstrip": false,
|
214 |
+
"normalized": false,
|
215 |
+
"rstrip": false,
|
216 |
+
"single_word": false,
|
217 |
+
"special": true
|
218 |
+
},
|
219 |
+
"250024": {
|
220 |
+
"content": "vi_VN",
|
221 |
+
"lstrip": false,
|
222 |
+
"normalized": false,
|
223 |
+
"rstrip": false,
|
224 |
+
"single_word": false,
|
225 |
+
"special": true
|
226 |
+
},
|
227 |
+
"250025": {
|
228 |
+
"content": "zh_CN",
|
229 |
+
"lstrip": false,
|
230 |
+
"normalized": false,
|
231 |
+
"rstrip": false,
|
232 |
+
"single_word": false,
|
233 |
+
"special": true
|
234 |
+
},
|
235 |
+
"250026": {
|
236 |
+
"content": "af_ZA",
|
237 |
+
"lstrip": false,
|
238 |
+
"normalized": false,
|
239 |
+
"rstrip": false,
|
240 |
+
"single_word": false,
|
241 |
+
"special": true
|
242 |
+
},
|
243 |
+
"250027": {
|
244 |
+
"content": "az_AZ",
|
245 |
+
"lstrip": false,
|
246 |
+
"normalized": false,
|
247 |
+
"rstrip": false,
|
248 |
+
"single_word": false,
|
249 |
+
"special": true
|
250 |
+
},
|
251 |
+
"250028": {
|
252 |
+
"content": "bn_IN",
|
253 |
+
"lstrip": false,
|
254 |
+
"normalized": false,
|
255 |
+
"rstrip": false,
|
256 |
+
"single_word": false,
|
257 |
+
"special": true
|
258 |
+
},
|
259 |
+
"250029": {
|
260 |
+
"content": "fa_IR",
|
261 |
+
"lstrip": false,
|
262 |
+
"normalized": false,
|
263 |
+
"rstrip": false,
|
264 |
+
"single_word": false,
|
265 |
+
"special": true
|
266 |
+
},
|
267 |
+
"250030": {
|
268 |
+
"content": "he_IL",
|
269 |
+
"lstrip": false,
|
270 |
+
"normalized": false,
|
271 |
+
"rstrip": false,
|
272 |
+
"single_word": false,
|
273 |
+
"special": true
|
274 |
+
},
|
275 |
+
"250031": {
|
276 |
+
"content": "hr_HR",
|
277 |
+
"lstrip": false,
|
278 |
+
"normalized": false,
|
279 |
+
"rstrip": false,
|
280 |
+
"single_word": false,
|
281 |
+
"special": true
|
282 |
+
},
|
283 |
+
"250032": {
|
284 |
+
"content": "id_ID",
|
285 |
+
"lstrip": false,
|
286 |
+
"normalized": false,
|
287 |
+
"rstrip": false,
|
288 |
+
"single_word": false,
|
289 |
+
"special": true
|
290 |
+
},
|
291 |
+
"250033": {
|
292 |
+
"content": "ka_GE",
|
293 |
+
"lstrip": false,
|
294 |
+
"normalized": false,
|
295 |
+
"rstrip": false,
|
296 |
+
"single_word": false,
|
297 |
+
"special": true
|
298 |
+
},
|
299 |
+
"250034": {
|
300 |
+
"content": "km_KH",
|
301 |
+
"lstrip": false,
|
302 |
+
"normalized": false,
|
303 |
+
"rstrip": false,
|
304 |
+
"single_word": false,
|
305 |
+
"special": true
|
306 |
+
},
|
307 |
+
"250035": {
|
308 |
+
"content": "mk_MK",
|
309 |
+
"lstrip": false,
|
310 |
+
"normalized": false,
|
311 |
+
"rstrip": false,
|
312 |
+
"single_word": false,
|
313 |
+
"special": true
|
314 |
+
},
|
315 |
+
"250036": {
|
316 |
+
"content": "ml_IN",
|
317 |
+
"lstrip": false,
|
318 |
+
"normalized": false,
|
319 |
+
"rstrip": false,
|
320 |
+
"single_word": false,
|
321 |
+
"special": true
|
322 |
+
},
|
323 |
+
"250037": {
|
324 |
+
"content": "mn_MN",
|
325 |
+
"lstrip": false,
|
326 |
+
"normalized": false,
|
327 |
+
"rstrip": false,
|
328 |
+
"single_word": false,
|
329 |
+
"special": true
|
330 |
+
},
|
331 |
+
"250038": {
|
332 |
+
"content": "mr_IN",
|
333 |
+
"lstrip": false,
|
334 |
+
"normalized": false,
|
335 |
+
"rstrip": false,
|
336 |
+
"single_word": false,
|
337 |
+
"special": true
|
338 |
+
},
|
339 |
+
"250039": {
|
340 |
+
"content": "pl_PL",
|
341 |
+
"lstrip": false,
|
342 |
+
"normalized": false,
|
343 |
+
"rstrip": false,
|
344 |
+
"single_word": false,
|
345 |
+
"special": true
|
346 |
+
},
|
347 |
+
"250040": {
|
348 |
+
"content": "ps_AF",
|
349 |
+
"lstrip": false,
|
350 |
+
"normalized": false,
|
351 |
+
"rstrip": false,
|
352 |
+
"single_word": false,
|
353 |
+
"special": true
|
354 |
+
},
|
355 |
+
"250041": {
|
356 |
+
"content": "pt_XX",
|
357 |
+
"lstrip": false,
|
358 |
+
"normalized": false,
|
359 |
+
"rstrip": false,
|
360 |
+
"single_word": false,
|
361 |
+
"special": true
|
362 |
+
},
|
363 |
+
"250042": {
|
364 |
+
"content": "sv_SE",
|
365 |
+
"lstrip": false,
|
366 |
+
"normalized": false,
|
367 |
+
"rstrip": false,
|
368 |
+
"single_word": false,
|
369 |
+
"special": true
|
370 |
+
},
|
371 |
+
"250043": {
|
372 |
+
"content": "sw_KE",
|
373 |
+
"lstrip": false,
|
374 |
+
"normalized": false,
|
375 |
+
"rstrip": false,
|
376 |
+
"single_word": false,
|
377 |
+
"special": true
|
378 |
+
},
|
379 |
+
"250044": {
|
380 |
+
"content": "ta_IN",
|
381 |
+
"lstrip": false,
|
382 |
+
"normalized": false,
|
383 |
+
"rstrip": false,
|
384 |
+
"single_word": false,
|
385 |
+
"special": true
|
386 |
+
},
|
387 |
+
"250045": {
|
388 |
+
"content": "te_IN",
|
389 |
+
"lstrip": false,
|
390 |
+
"normalized": false,
|
391 |
+
"rstrip": false,
|
392 |
+
"single_word": false,
|
393 |
+
"special": true
|
394 |
+
},
|
395 |
+
"250046": {
|
396 |
+
"content": "th_TH",
|
397 |
+
"lstrip": false,
|
398 |
+
"normalized": false,
|
399 |
+
"rstrip": false,
|
400 |
+
"single_word": false,
|
401 |
+
"special": true
|
402 |
+
},
|
403 |
+
"250047": {
|
404 |
+
"content": "tl_XX",
|
405 |
+
"lstrip": false,
|
406 |
+
"normalized": false,
|
407 |
+
"rstrip": false,
|
408 |
+
"single_word": false,
|
409 |
+
"special": true
|
410 |
+
},
|
411 |
+
"250048": {
|
412 |
+
"content": "uk_UA",
|
413 |
+
"lstrip": false,
|
414 |
+
"normalized": false,
|
415 |
+
"rstrip": false,
|
416 |
+
"single_word": false,
|
417 |
+
"special": true
|
418 |
+
},
|
419 |
+
"250049": {
|
420 |
+
"content": "ur_PK",
|
421 |
+
"lstrip": false,
|
422 |
+
"normalized": false,
|
423 |
+
"rstrip": false,
|
424 |
+
"single_word": false,
|
425 |
+
"special": true
|
426 |
+
},
|
427 |
+
"250050": {
|
428 |
+
"content": "xh_ZA",
|
429 |
+
"lstrip": false,
|
430 |
+
"normalized": false,
|
431 |
+
"rstrip": false,
|
432 |
+
"single_word": false,
|
433 |
+
"special": true
|
434 |
+
},
|
435 |
+
"250051": {
|
436 |
+
"content": "gl_ES",
|
437 |
+
"lstrip": false,
|
438 |
+
"normalized": false,
|
439 |
+
"rstrip": false,
|
440 |
+
"single_word": false,
|
441 |
+
"special": true
|
442 |
+
},
|
443 |
+
"250052": {
|
444 |
+
"content": "sl_SI",
|
445 |
+
"lstrip": false,
|
446 |
+
"normalized": false,
|
447 |
+
"rstrip": false,
|
448 |
+
"single_word": false,
|
449 |
+
"special": true
|
450 |
+
},
|
451 |
+
"250053": {
|
452 |
+
"content": "<mask>",
|
453 |
+
"lstrip": true,
|
454 |
+
"normalized": true,
|
455 |
+
"rstrip": false,
|
456 |
+
"single_word": false,
|
457 |
+
"special": true
|
458 |
+
}
|
459 |
+
},
|
460 |
+
"additional_special_tokens": [
|
461 |
+
"ar_AR",
|
462 |
+
"cs_CZ",
|
463 |
+
"de_DE",
|
464 |
+
"en_XX",
|
465 |
+
"es_XX",
|
466 |
+
"et_EE",
|
467 |
+
"fi_FI",
|
468 |
+
"fr_XX",
|
469 |
+
"gu_IN",
|
470 |
+
"hi_IN",
|
471 |
+
"it_IT",
|
472 |
+
"ja_XX",
|
473 |
+
"kk_KZ",
|
474 |
+
"ko_KR",
|
475 |
+
"lt_LT",
|
476 |
+
"lv_LV",
|
477 |
+
"my_MM",
|
478 |
+
"ne_NP",
|
479 |
+
"nl_XX",
|
480 |
+
"ro_RO",
|
481 |
+
"ru_RU",
|
482 |
+
"si_LK",
|
483 |
+
"tr_TR",
|
484 |
+
"vi_VN",
|
485 |
+
"zh_CN",
|
486 |
+
"af_ZA",
|
487 |
+
"az_AZ",
|
488 |
+
"bn_IN",
|
489 |
+
"fa_IR",
|
490 |
+
"he_IL",
|
491 |
+
"hr_HR",
|
492 |
+
"id_ID",
|
493 |
+
"ka_GE",
|
494 |
+
"km_KH",
|
495 |
+
"mk_MK",
|
496 |
+
"ml_IN",
|
497 |
+
"mn_MN",
|
498 |
+
"mr_IN",
|
499 |
+
"pl_PL",
|
500 |
+
"ps_AF",
|
501 |
+
"pt_XX",
|
502 |
+
"sv_SE",
|
503 |
+
"sw_KE",
|
504 |
+
"ta_IN",
|
505 |
+
"te_IN",
|
506 |
+
"th_TH",
|
507 |
+
"tl_XX",
|
508 |
+
"uk_UA",
|
509 |
+
"ur_PK",
|
510 |
+
"xh_ZA",
|
511 |
+
"gl_ES",
|
512 |
+
"sl_SI"
|
513 |
+
],
|
514 |
+
"bos_token": "<s>",
|
515 |
+
"clean_up_tokenization_spaces": false,
|
516 |
+
"cls_token": "<s>",
|
517 |
+
"eos_token": "</s>",
|
518 |
+
"extra_special_tokens": {},
|
519 |
+
"mask_token": "<mask>",
|
520 |
+
"model_max_length": 1024,
|
521 |
+
"pad_token": "<pad>",
|
522 |
+
"sep_token": "</s>",
|
523 |
+
"sp_model_kwargs": {},
|
524 |
+
"src_lang": "ur_PK",
|
525 |
+
"tgt_lang": "ur_PK",
|
526 |
+
"tokenizer_class": "MBart50Tokenizer",
|
527 |
+
"unk_token": "<unk>"
|
528 |
+
}
|
phase1_test/training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:609cf31c34e6e192184094c345a1f99375fa32c74a07075bd3a9d0dedea23412
|
3 |
+
size 5432
|
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 +1,69 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"additional_special_tokens": [
|
3 |
+
"ar_AR",
|
4 |
+
"cs_CZ",
|
5 |
+
"de_DE",
|
6 |
+
"en_XX",
|
7 |
+
"es_XX",
|
8 |
+
"et_EE",
|
9 |
+
"fi_FI",
|
10 |
+
"fr_XX",
|
11 |
+
"gu_IN",
|
12 |
+
"hi_IN",
|
13 |
+
"it_IT",
|
14 |
+
"ja_XX",
|
15 |
+
"kk_KZ",
|
16 |
+
"ko_KR",
|
17 |
+
"lt_LT",
|
18 |
+
"lv_LV",
|
19 |
+
"my_MM",
|
20 |
+
"ne_NP",
|
21 |
+
"nl_XX",
|
22 |
+
"ro_RO",
|
23 |
+
"ru_RU",
|
24 |
+
"si_LK",
|
25 |
+
"tr_TR",
|
26 |
+
"vi_VN",
|
27 |
+
"zh_CN",
|
28 |
+
"af_ZA",
|
29 |
+
"az_AZ",
|
30 |
+
"bn_IN",
|
31 |
+
"fa_IR",
|
32 |
+
"he_IL",
|
33 |
+
"hr_HR",
|
34 |
+
"id_ID",
|
35 |
+
"ka_GE",
|
36 |
+
"km_KH",
|
37 |
+
"mk_MK",
|
38 |
+
"ml_IN",
|
39 |
+
"mn_MN",
|
40 |
+
"mr_IN",
|
41 |
+
"pl_PL",
|
42 |
+
"ps_AF",
|
43 |
+
"pt_XX",
|
44 |
+
"sv_SE",
|
45 |
+
"sw_KE",
|
46 |
+
"ta_IN",
|
47 |
+
"te_IN",
|
48 |
+
"th_TH",
|
49 |
+
"tl_XX",
|
50 |
+
"uk_UA",
|
51 |
+
"ur_PK",
|
52 |
+
"xh_ZA",
|
53 |
+
"gl_ES",
|
54 |
+
"sl_SI"
|
55 |
+
],
|
56 |
+
"bos_token": "<s>",
|
57 |
+
"cls_token": "<s>",
|
58 |
+
"eos_token": "</s>",
|
59 |
+
"mask_token": {
|
60 |
+
"content": "<mask>",
|
61 |
+
"lstrip": true,
|
62 |
+
"normalized": true,
|
63 |
+
"rstrip": false,
|
64 |
+
"single_word": false
|
65 |
+
},
|
66 |
+
"pad_token": "<pad>",
|
67 |
+
"sep_token": "</s>",
|
68 |
+
"unk_token": "<unk>"
|
69 |
+
}
|
tokenizer.json
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:500bc815ab1b3a039f7de9cf48b4c97a58e503054d23c7a53a2e59f8d6666629
|
3 |
+
size 17110184
|
tokenizer_config.json
ADDED
@@ -0,0 +1,528 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"added_tokens_decoder": {
|
3 |
+
"0": {
|
4 |
+
"content": "<s>",
|
5 |
+
"lstrip": false,
|
6 |
+
"normalized": false,
|
7 |
+
"rstrip": false,
|
8 |
+
"single_word": false,
|
9 |
+
"special": true
|
10 |
+
},
|
11 |
+
"1": {
|
12 |
+
"content": "<pad>",
|
13 |
+
"lstrip": false,
|
14 |
+
"normalized": false,
|
15 |
+
"rstrip": false,
|
16 |
+
"single_word": false,
|
17 |
+
"special": true
|
18 |
+
},
|
19 |
+
"2": {
|
20 |
+
"content": "</s>",
|
21 |
+
"lstrip": false,
|
22 |
+
"normalized": false,
|
23 |
+
"rstrip": false,
|
24 |
+
"single_word": false,
|
25 |
+
"special": true
|
26 |
+
},
|
27 |
+
"3": {
|
28 |
+
"content": "<unk>",
|
29 |
+
"lstrip": false,
|
30 |
+
"normalized": false,
|
31 |
+
"rstrip": false,
|
32 |
+
"single_word": false,
|
33 |
+
"special": true
|
34 |
+
},
|
35 |
+
"250001": {
|
36 |
+
"content": "ar_AR",
|
37 |
+
"lstrip": false,
|
38 |
+
"normalized": false,
|
39 |
+
"rstrip": false,
|
40 |
+
"single_word": false,
|
41 |
+
"special": true
|
42 |
+
},
|
43 |
+
"250002": {
|
44 |
+
"content": "cs_CZ",
|
45 |
+
"lstrip": false,
|
46 |
+
"normalized": false,
|
47 |
+
"rstrip": false,
|
48 |
+
"single_word": false,
|
49 |
+
"special": true
|
50 |
+
},
|
51 |
+
"250003": {
|
52 |
+
"content": "de_DE",
|
53 |
+
"lstrip": false,
|
54 |
+
"normalized": false,
|
55 |
+
"rstrip": false,
|
56 |
+
"single_word": false,
|
57 |
+
"special": true
|
58 |
+
},
|
59 |
+
"250004": {
|
60 |
+
"content": "en_XX",
|
61 |
+
"lstrip": false,
|
62 |
+
"normalized": false,
|
63 |
+
"rstrip": false,
|
64 |
+
"single_word": false,
|
65 |
+
"special": true
|
66 |
+
},
|
67 |
+
"250005": {
|
68 |
+
"content": "es_XX",
|
69 |
+
"lstrip": false,
|
70 |
+
"normalized": false,
|
71 |
+
"rstrip": false,
|
72 |
+
"single_word": false,
|
73 |
+
"special": true
|
74 |
+
},
|
75 |
+
"250006": {
|
76 |
+
"content": "et_EE",
|
77 |
+
"lstrip": false,
|
78 |
+
"normalized": false,
|
79 |
+
"rstrip": false,
|
80 |
+
"single_word": false,
|
81 |
+
"special": true
|
82 |
+
},
|
83 |
+
"250007": {
|
84 |
+
"content": "fi_FI",
|
85 |
+
"lstrip": false,
|
86 |
+
"normalized": false,
|
87 |
+
"rstrip": false,
|
88 |
+
"single_word": false,
|
89 |
+
"special": true
|
90 |
+
},
|
91 |
+
"250008": {
|
92 |
+
"content": "fr_XX",
|
93 |
+
"lstrip": false,
|
94 |
+
"normalized": false,
|
95 |
+
"rstrip": false,
|
96 |
+
"single_word": false,
|
97 |
+
"special": true
|
98 |
+
},
|
99 |
+
"250009": {
|
100 |
+
"content": "gu_IN",
|
101 |
+
"lstrip": false,
|
102 |
+
"normalized": false,
|
103 |
+
"rstrip": false,
|
104 |
+
"single_word": false,
|
105 |
+
"special": true
|
106 |
+
},
|
107 |
+
"250010": {
|
108 |
+
"content": "hi_IN",
|
109 |
+
"lstrip": false,
|
110 |
+
"normalized": false,
|
111 |
+
"rstrip": false,
|
112 |
+
"single_word": false,
|
113 |
+
"special": true
|
114 |
+
},
|
115 |
+
"250011": {
|
116 |
+
"content": "it_IT",
|
117 |
+
"lstrip": false,
|
118 |
+
"normalized": false,
|
119 |
+
"rstrip": false,
|
120 |
+
"single_word": false,
|
121 |
+
"special": true
|
122 |
+
},
|
123 |
+
"250012": {
|
124 |
+
"content": "ja_XX",
|
125 |
+
"lstrip": false,
|
126 |
+
"normalized": false,
|
127 |
+
"rstrip": false,
|
128 |
+
"single_word": false,
|
129 |
+
"special": true
|
130 |
+
},
|
131 |
+
"250013": {
|
132 |
+
"content": "kk_KZ",
|
133 |
+
"lstrip": false,
|
134 |
+
"normalized": false,
|
135 |
+
"rstrip": false,
|
136 |
+
"single_word": false,
|
137 |
+
"special": true
|
138 |
+
},
|
139 |
+
"250014": {
|
140 |
+
"content": "ko_KR",
|
141 |
+
"lstrip": false,
|
142 |
+
"normalized": false,
|
143 |
+
"rstrip": false,
|
144 |
+
"single_word": false,
|
145 |
+
"special": true
|
146 |
+
},
|
147 |
+
"250015": {
|
148 |
+
"content": "lt_LT",
|
149 |
+
"lstrip": false,
|
150 |
+
"normalized": false,
|
151 |
+
"rstrip": false,
|
152 |
+
"single_word": false,
|
153 |
+
"special": true
|
154 |
+
},
|
155 |
+
"250016": {
|
156 |
+
"content": "lv_LV",
|
157 |
+
"lstrip": false,
|
158 |
+
"normalized": false,
|
159 |
+
"rstrip": false,
|
160 |
+
"single_word": false,
|
161 |
+
"special": true
|
162 |
+
},
|
163 |
+
"250017": {
|
164 |
+
"content": "my_MM",
|
165 |
+
"lstrip": false,
|
166 |
+
"normalized": false,
|
167 |
+
"rstrip": false,
|
168 |
+
"single_word": false,
|
169 |
+
"special": true
|
170 |
+
},
|
171 |
+
"250018": {
|
172 |
+
"content": "ne_NP",
|
173 |
+
"lstrip": false,
|
174 |
+
"normalized": false,
|
175 |
+
"rstrip": false,
|
176 |
+
"single_word": false,
|
177 |
+
"special": true
|
178 |
+
},
|
179 |
+
"250019": {
|
180 |
+
"content": "nl_XX",
|
181 |
+
"lstrip": false,
|
182 |
+
"normalized": false,
|
183 |
+
"rstrip": false,
|
184 |
+
"single_word": false,
|
185 |
+
"special": true
|
186 |
+
},
|
187 |
+
"250020": {
|
188 |
+
"content": "ro_RO",
|
189 |
+
"lstrip": false,
|
190 |
+
"normalized": false,
|
191 |
+
"rstrip": false,
|
192 |
+
"single_word": false,
|
193 |
+
"special": true
|
194 |
+
},
|
195 |
+
"250021": {
|
196 |
+
"content": "ru_RU",
|
197 |
+
"lstrip": false,
|
198 |
+
"normalized": false,
|
199 |
+
"rstrip": false,
|
200 |
+
"single_word": false,
|
201 |
+
"special": true
|
202 |
+
},
|
203 |
+
"250022": {
|
204 |
+
"content": "si_LK",
|
205 |
+
"lstrip": false,
|
206 |
+
"normalized": false,
|
207 |
+
"rstrip": false,
|
208 |
+
"single_word": false,
|
209 |
+
"special": true
|
210 |
+
},
|
211 |
+
"250023": {
|
212 |
+
"content": "tr_TR",
|
213 |
+
"lstrip": false,
|
214 |
+
"normalized": false,
|
215 |
+
"rstrip": false,
|
216 |
+
"single_word": false,
|
217 |
+
"special": true
|
218 |
+
},
|
219 |
+
"250024": {
|
220 |
+
"content": "vi_VN",
|
221 |
+
"lstrip": false,
|
222 |
+
"normalized": false,
|
223 |
+
"rstrip": false,
|
224 |
+
"single_word": false,
|
225 |
+
"special": true
|
226 |
+
},
|
227 |
+
"250025": {
|
228 |
+
"content": "zh_CN",
|
229 |
+
"lstrip": false,
|
230 |
+
"normalized": false,
|
231 |
+
"rstrip": false,
|
232 |
+
"single_word": false,
|
233 |
+
"special": true
|
234 |
+
},
|
235 |
+
"250026": {
|
236 |
+
"content": "af_ZA",
|
237 |
+
"lstrip": false,
|
238 |
+
"normalized": false,
|
239 |
+
"rstrip": false,
|
240 |
+
"single_word": false,
|
241 |
+
"special": true
|
242 |
+
},
|
243 |
+
"250027": {
|
244 |
+
"content": "az_AZ",
|
245 |
+
"lstrip": false,
|
246 |
+
"normalized": false,
|
247 |
+
"rstrip": false,
|
248 |
+
"single_word": false,
|
249 |
+
"special": true
|
250 |
+
},
|
251 |
+
"250028": {
|
252 |
+
"content": "bn_IN",
|
253 |
+
"lstrip": false,
|
254 |
+
"normalized": false,
|
255 |
+
"rstrip": false,
|
256 |
+
"single_word": false,
|
257 |
+
"special": true
|
258 |
+
},
|
259 |
+
"250029": {
|
260 |
+
"content": "fa_IR",
|
261 |
+
"lstrip": false,
|
262 |
+
"normalized": false,
|
263 |
+
"rstrip": false,
|
264 |
+
"single_word": false,
|
265 |
+
"special": true
|
266 |
+
},
|
267 |
+
"250030": {
|
268 |
+
"content": "he_IL",
|
269 |
+
"lstrip": false,
|
270 |
+
"normalized": false,
|
271 |
+
"rstrip": false,
|
272 |
+
"single_word": false,
|
273 |
+
"special": true
|
274 |
+
},
|
275 |
+
"250031": {
|
276 |
+
"content": "hr_HR",
|
277 |
+
"lstrip": false,
|
278 |
+
"normalized": false,
|
279 |
+
"rstrip": false,
|
280 |
+
"single_word": false,
|
281 |
+
"special": true
|
282 |
+
},
|
283 |
+
"250032": {
|
284 |
+
"content": "id_ID",
|
285 |
+
"lstrip": false,
|
286 |
+
"normalized": false,
|
287 |
+
"rstrip": false,
|
288 |
+
"single_word": false,
|
289 |
+
"special": true
|
290 |
+
},
|
291 |
+
"250033": {
|
292 |
+
"content": "ka_GE",
|
293 |
+
"lstrip": false,
|
294 |
+
"normalized": false,
|
295 |
+
"rstrip": false,
|
296 |
+
"single_word": false,
|
297 |
+
"special": true
|
298 |
+
},
|
299 |
+
"250034": {
|
300 |
+
"content": "km_KH",
|
301 |
+
"lstrip": false,
|
302 |
+
"normalized": false,
|
303 |
+
"rstrip": false,
|
304 |
+
"single_word": false,
|
305 |
+
"special": true
|
306 |
+
},
|
307 |
+
"250035": {
|
308 |
+
"content": "mk_MK",
|
309 |
+
"lstrip": false,
|
310 |
+
"normalized": false,
|
311 |
+
"rstrip": false,
|
312 |
+
"single_word": false,
|
313 |
+
"special": true
|
314 |
+
},
|
315 |
+
"250036": {
|
316 |
+
"content": "ml_IN",
|
317 |
+
"lstrip": false,
|
318 |
+
"normalized": false,
|
319 |
+
"rstrip": false,
|
320 |
+
"single_word": false,
|
321 |
+
"special": true
|
322 |
+
},
|
323 |
+
"250037": {
|
324 |
+
"content": "mn_MN",
|
325 |
+
"lstrip": false,
|
326 |
+
"normalized": false,
|
327 |
+
"rstrip": false,
|
328 |
+
"single_word": false,
|
329 |
+
"special": true
|
330 |
+
},
|
331 |
+
"250038": {
|
332 |
+
"content": "mr_IN",
|
333 |
+
"lstrip": false,
|
334 |
+
"normalized": false,
|
335 |
+
"rstrip": false,
|
336 |
+
"single_word": false,
|
337 |
+
"special": true
|
338 |
+
},
|
339 |
+
"250039": {
|
340 |
+
"content": "pl_PL",
|
341 |
+
"lstrip": false,
|
342 |
+
"normalized": false,
|
343 |
+
"rstrip": false,
|
344 |
+
"single_word": false,
|
345 |
+
"special": true
|
346 |
+
},
|
347 |
+
"250040": {
|
348 |
+
"content": "ps_AF",
|
349 |
+
"lstrip": false,
|
350 |
+
"normalized": false,
|
351 |
+
"rstrip": false,
|
352 |
+
"single_word": false,
|
353 |
+
"special": true
|
354 |
+
},
|
355 |
+
"250041": {
|
356 |
+
"content": "pt_XX",
|
357 |
+
"lstrip": false,
|
358 |
+
"normalized": false,
|
359 |
+
"rstrip": false,
|
360 |
+
"single_word": false,
|
361 |
+
"special": true
|
362 |
+
},
|
363 |
+
"250042": {
|
364 |
+
"content": "sv_SE",
|
365 |
+
"lstrip": false,
|
366 |
+
"normalized": false,
|
367 |
+
"rstrip": false,
|
368 |
+
"single_word": false,
|
369 |
+
"special": true
|
370 |
+
},
|
371 |
+
"250043": {
|
372 |
+
"content": "sw_KE",
|
373 |
+
"lstrip": false,
|
374 |
+
"normalized": false,
|
375 |
+
"rstrip": false,
|
376 |
+
"single_word": false,
|
377 |
+
"special": true
|
378 |
+
},
|
379 |
+
"250044": {
|
380 |
+
"content": "ta_IN",
|
381 |
+
"lstrip": false,
|
382 |
+
"normalized": false,
|
383 |
+
"rstrip": false,
|
384 |
+
"single_word": false,
|
385 |
+
"special": true
|
386 |
+
},
|
387 |
+
"250045": {
|
388 |
+
"content": "te_IN",
|
389 |
+
"lstrip": false,
|
390 |
+
"normalized": false,
|
391 |
+
"rstrip": false,
|
392 |
+
"single_word": false,
|
393 |
+
"special": true
|
394 |
+
},
|
395 |
+
"250046": {
|
396 |
+
"content": "th_TH",
|
397 |
+
"lstrip": false,
|
398 |
+
"normalized": false,
|
399 |
+
"rstrip": false,
|
400 |
+
"single_word": false,
|
401 |
+
"special": true
|
402 |
+
},
|
403 |
+
"250047": {
|
404 |
+
"content": "tl_XX",
|
405 |
+
"lstrip": false,
|
406 |
+
"normalized": false,
|
407 |
+
"rstrip": false,
|
408 |
+
"single_word": false,
|
409 |
+
"special": true
|
410 |
+
},
|
411 |
+
"250048": {
|
412 |
+
"content": "uk_UA",
|
413 |
+
"lstrip": false,
|
414 |
+
"normalized": false,
|
415 |
+
"rstrip": false,
|
416 |
+
"single_word": false,
|
417 |
+
"special": true
|
418 |
+
},
|
419 |
+
"250049": {
|
420 |
+
"content": "ur_PK",
|
421 |
+
"lstrip": false,
|
422 |
+
"normalized": false,
|
423 |
+
"rstrip": false,
|
424 |
+
"single_word": false,
|
425 |
+
"special": true
|
426 |
+
},
|
427 |
+
"250050": {
|
428 |
+
"content": "xh_ZA",
|
429 |
+
"lstrip": false,
|
430 |
+
"normalized": false,
|
431 |
+
"rstrip": false,
|
432 |
+
"single_word": false,
|
433 |
+
"special": true
|
434 |
+
},
|
435 |
+
"250051": {
|
436 |
+
"content": "gl_ES",
|
437 |
+
"lstrip": false,
|
438 |
+
"normalized": false,
|
439 |
+
"rstrip": false,
|
440 |
+
"single_word": false,
|
441 |
+
"special": true
|
442 |
+
},
|
443 |
+
"250052": {
|
444 |
+
"content": "sl_SI",
|
445 |
+
"lstrip": false,
|
446 |
+
"normalized": false,
|
447 |
+
"rstrip": false,
|
448 |
+
"single_word": false,
|
449 |
+
"special": true
|
450 |
+
},
|
451 |
+
"250053": {
|
452 |
+
"content": "<mask>",
|
453 |
+
"lstrip": true,
|
454 |
+
"normalized": true,
|
455 |
+
"rstrip": false,
|
456 |
+
"single_word": false,
|
457 |
+
"special": true
|
458 |
+
}
|
459 |
+
},
|
460 |
+
"additional_special_tokens": [
|
461 |
+
"ar_AR",
|
462 |
+
"cs_CZ",
|
463 |
+
"de_DE",
|
464 |
+
"en_XX",
|
465 |
+
"es_XX",
|
466 |
+
"et_EE",
|
467 |
+
"fi_FI",
|
468 |
+
"fr_XX",
|
469 |
+
"gu_IN",
|
470 |
+
"hi_IN",
|
471 |
+
"it_IT",
|
472 |
+
"ja_XX",
|
473 |
+
"kk_KZ",
|
474 |
+
"ko_KR",
|
475 |
+
"lt_LT",
|
476 |
+
"lv_LV",
|
477 |
+
"my_MM",
|
478 |
+
"ne_NP",
|
479 |
+
"nl_XX",
|
480 |
+
"ro_RO",
|
481 |
+
"ru_RU",
|
482 |
+
"si_LK",
|
483 |
+
"tr_TR",
|
484 |
+
"vi_VN",
|
485 |
+
"zh_CN",
|
486 |
+
"af_ZA",
|
487 |
+
"az_AZ",
|
488 |
+
"bn_IN",
|
489 |
+
"fa_IR",
|
490 |
+
"he_IL",
|
491 |
+
"hr_HR",
|
492 |
+
"id_ID",
|
493 |
+
"ka_GE",
|
494 |
+
"km_KH",
|
495 |
+
"mk_MK",
|
496 |
+
"ml_IN",
|
497 |
+
"mn_MN",
|
498 |
+
"mr_IN",
|
499 |
+
"pl_PL",
|
500 |
+
"ps_AF",
|
501 |
+
"pt_XX",
|
502 |
+
"sv_SE",
|
503 |
+
"sw_KE",
|
504 |
+
"ta_IN",
|
505 |
+
"te_IN",
|
506 |
+
"th_TH",
|
507 |
+
"tl_XX",
|
508 |
+
"uk_UA",
|
509 |
+
"ur_PK",
|
510 |
+
"xh_ZA",
|
511 |
+
"gl_ES",
|
512 |
+
"sl_SI"
|
513 |
+
],
|
514 |
+
"bos_token": "<s>",
|
515 |
+
"clean_up_tokenization_spaces": false,
|
516 |
+
"cls_token": "<s>",
|
517 |
+
"eos_token": "</s>",
|
518 |
+
"extra_special_tokens": {},
|
519 |
+
"mask_token": "<mask>",
|
520 |
+
"model_max_length": 1024,
|
521 |
+
"pad_token": "<pad>",
|
522 |
+
"sep_token": "</s>",
|
523 |
+
"sp_model_kwargs": {},
|
524 |
+
"src_lang": "ur_PK",
|
525 |
+
"tgt_lang": "ur_PK",
|
526 |
+
"tokenizer_class": "MBart50Tokenizer",
|
527 |
+
"unk_token": "<unk>"
|
528 |
+
}
|
training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:59857a579e21b10bd94022946f8aae7a25c8dcb51c7dd8e8e5b517aad10cf7f8
|
3 |
+
size 5496
|