Commit
·
ee4839e
1
Parent(s):
4968a0e
Training in progress, epoch 1
Browse files- .gitignore +1 -0
- config.json +58 -0
- pytorch_model.bin +3 -0
- runs/May22_20-29-57_a100vm/events.out.tfevents.1653253495.a100vm.1239.0 +3 -0
- runs/May23_21-51-11_a100vm/1653342901.8660035/events.out.tfevents.1653342901.a100vm.59062.1 +3 -0
- runs/May23_21-51-11_a100vm/events.out.tfevents.1653342851.a100vm.59062.0 +3 -0
- runs/May23_21-55-51_a100vm/1653342962.3782356/events.out.tfevents.1653342962.a100vm.62525.1 +3 -0
- runs/May23_21-55-51_a100vm/events.out.tfevents.1653342962.a100vm.62525.0 +3 -0
- runs/May23_21-56-18_a100vm/1653342984.0852232/events.out.tfevents.1653342984.a100vm.62525.3 +3 -0
- runs/May23_21-56-18_a100vm/events.out.tfevents.1653342984.a100vm.62525.2 +3 -0
- runs/May23_21-56-51_a100vm/1653343021.994845/events.out.tfevents.1653343021.a100vm.63396.1 +3 -0
- runs/May23_21-56-51_a100vm/events.out.tfevents.1653343021.a100vm.63396.0 +3 -0
- runs/May23_21-57-52_a100vm/1653343083.210807/events.out.tfevents.1653343083.a100vm.64215.1 +3 -0
- runs/May23_21-57-52_a100vm/events.out.tfevents.1653343083.a100vm.64215.0 +3 -0
- runs/May23_22-04-18_a100vm/1653343511.8982878/events.out.tfevents.1653343511.a100vm.1428.1 +3 -0
- runs/May23_22-04-18_a100vm/events.out.tfevents.1653343511.a100vm.1428.0 +3 -0
- runs/May23_22-05-35_a100vm/1653343540.4615602/events.out.tfevents.1653343540.a100vm.1428.3 +3 -0
- runs/May23_22-05-35_a100vm/events.out.tfevents.1653343540.a100vm.1428.2 +3 -0
- sentencepiece.bpe.model +3 -0
- special_tokens_map.json +1 -0
- tokenizer.json +0 -0
- tokenizer_config.json +1 -0
- training_args.bin +3 -0
.gitignore
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
checkpoint-*/
|
config.json
ADDED
@@ -0,0 +1,58 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "facebook/mbart-large-50-one-to-many-mmt",
|
3 |
+
"_num_labels": 3,
|
4 |
+
"activation_dropout": 0.0,
|
5 |
+
"activation_function": "relu",
|
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": true,
|
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": 200,
|
43 |
+
"max_position_embeddings": 1024,
|
44 |
+
"model_type": "mbart",
|
45 |
+
"normalize_before": true,
|
46 |
+
"normalize_embedding": true,
|
47 |
+
"num_beams": 5,
|
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": "4.13.0",
|
56 |
+
"use_cache": true,
|
57 |
+
"vocab_size": 250054
|
58 |
+
}
|
pytorch_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:bf5ff0b328cdceb94059f6f8bdd19e034be3e14f7733b537db86b93444587e1f
|
3 |
+
size 2444710009
|
runs/May22_20-29-57_a100vm/events.out.tfevents.1653253495.a100vm.1239.0
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:3562dd70b6dc0f9e4303767b7c5898d5f7329cd3ba97beac790cbc5b2b6d5268
|
3 |
+
size 294
|
runs/May23_21-51-11_a100vm/1653342901.8660035/events.out.tfevents.1653342901.a100vm.59062.1
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:0f1a943645bfdaeac9d0735dc17edbb4376791b734a23265f5b4bd669d9a2f3b
|
3 |
+
size 4940
|
runs/May23_21-51-11_a100vm/events.out.tfevents.1653342851.a100vm.59062.0
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:a2b9bfc88fa4b0983be9f730fe74599b8f5019e4acc734d75ade8213b429885e
|
3 |
+
size 4390
|
runs/May23_21-55-51_a100vm/1653342962.3782356/events.out.tfevents.1653342962.a100vm.62525.1
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:c601fae99b2ece8dc3f43c30f65415370984d506ebc9d83f58e59327d6bbadc1
|
3 |
+
size 4940
|
runs/May23_21-55-51_a100vm/events.out.tfevents.1653342962.a100vm.62525.0
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:0ad96322ebd596fc2077144cfa71e93d716743c3161908c4222e9f0bd9752827
|
3 |
+
size 4136
|
runs/May23_21-56-18_a100vm/1653342984.0852232/events.out.tfevents.1653342984.a100vm.62525.3
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:8eb74960626892fb6537fbb4287bb232150bb0370f0a8facb73eca9572bd3614
|
3 |
+
size 4940
|
runs/May23_21-56-18_a100vm/events.out.tfevents.1653342984.a100vm.62525.2
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:8e2a1e4b6f8e136c95af73ba0d8aef0cf87fdca32fff96ec5ca91f17efb283e6
|
3 |
+
size 4136
|
runs/May23_21-56-51_a100vm/1653343021.994845/events.out.tfevents.1653343021.a100vm.63396.1
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:1c5134851aa960e169144d4fe762065567913badd2199e182872d6aaf3c64c53
|
3 |
+
size 4940
|
runs/May23_21-56-51_a100vm/events.out.tfevents.1653343021.a100vm.63396.0
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:326a027336456e30d908db06bbf1cd73b2a89050267ebf19eb59548b4ad4f87e
|
3 |
+
size 4136
|
runs/May23_21-57-52_a100vm/1653343083.210807/events.out.tfevents.1653343083.a100vm.64215.1
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:047dfb9dc95b7e2f452ca29e852fec1363d96a3b56415ace8a6479ac3af16215
|
3 |
+
size 4940
|
runs/May23_21-57-52_a100vm/events.out.tfevents.1653343083.a100vm.64215.0
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:2c7f911d8e810c20dff0d4c1c36a4f408ae51b48f046f3e7be50db53708afa38
|
3 |
+
size 4136
|
runs/May23_22-04-18_a100vm/1653343511.8982878/events.out.tfevents.1653343511.a100vm.1428.1
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:e99dc04279a299ef3f13fd636627c8f7cade7ec83b0c714dd36ceb549858d012
|
3 |
+
size 4940
|
runs/May23_22-04-18_a100vm/events.out.tfevents.1653343511.a100vm.1428.0
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:512cab66115d2c24c7194501416733a054721db93c7d8ffd6302c93bd4b47412
|
3 |
+
size 4153
|
runs/May23_22-05-35_a100vm/1653343540.4615602/events.out.tfevents.1653343540.a100vm.1428.3
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:2592948deb0ccba85e950911e0597bde2d1e29950c884dfbe006b3004d15f74c
|
3 |
+
size 4940
|
runs/May23_22-05-35_a100vm/events.out.tfevents.1653343540.a100vm.1428.2
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:9059e4c56290be8ca273d1f4e4a9ab888dd73a08fac1c888e8a5d8327080b8c5
|
3 |
+
size 4625
|
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": {"content": "<mask>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true}, "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": "en_XX", "tgt_lang": "de_DE", "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>", "language_codes": "ML50", "special_tokens_map_file": "/home/suraj/projects/mbart-50/mbart-50/special_tokens_map.json", "name_or_path": "facebook/mbart-large-50-one-to-many-mmt", "model_max_length": 1024, "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"}
|
training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:df0a946e90712d096c439b1ab1939aed995ef5357865af6ed6ad1aa0b9f2c5df
|
3 |
+
size 3055
|