Initial commit

many-eng-mBART.zip

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:11624a6648d7825fe5a5ecacae1386e0f72dc206ba9edd0611452a7517bae090
+size 2272087269

many-eng-mBART/config.json

@@ -0,0 +1,63 @@
+{
+  "_name_or_path": "facebook/mbart-large-50-many-to-one-mmt",
+  "_num_labels": 3,
+  "activation_dropout": 0.0,
+  "activation_function": "relu",
+  "add_bias_logits": false,
+  "add_final_layer_norm": true,
+  "architectures": [
+    "MBartForConditionalGeneration"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 0,
+  "classif_dropout": 0.0,
+  "classifier_dropout": 0.0,
+  "d_model": 1024,
+  "decoder_attention_heads": 16,
+  "decoder_ffn_dim": 4096,
+  "decoder_layerdrop": 0.0,
+  "decoder_layers": 12,
+  "decoder_start_token_id": 2,
+  "dropout": 0.1,
+  "encoder_attention_heads": 16,
+  "encoder_ffn_dim": 4096,
+  "encoder_layerdrop": 0.0,
+  "encoder_layers": 12,
+  "eos_token_id": 2,
+  "forced_bos_token_id": 250004,
+  "forced_eos_token_id": 2,
+  "gradient_checkpointing": false,
+  "id2label": {
+    "0": "LABEL_0",
+    "1": "LABEL_1",
+    "2": "LABEL_2"
+  },
+  "init_std": 0.02,
+  "is_encoder_decoder": true,
+  "label2id": {
+    "LABEL_0": 0,
+    "LABEL_1": 1,
+    "LABEL_2": 2
+  },
+  "max_length": 200,
+  "max_position_embeddings": 1024,
+  "model_type": "mbart",
+  "normalize_before": true,
+  "normalize_embedding": true,
+  "num_beams": 5,
+  "num_hidden_layers": 12,
+  "output_past": true,
+  "pad_token_id": 1,
+  "scale_embedding": true,
+  "static_position_embeddings": false,
+  "task_specific_params": {
+    "translation_en_to_ro": {
+      "decoder_start_token_id": 250020
+    }
+  },
+  "tokenizer_class": "MBart50Tokenizer",
+  "torch_dtype": "float32",
+  "transformers_version": "4.27.4",
+  "use_cache": true,
+  "vocab_size": 250054
+}

many-eng-mBART/generation_config.json

@@ -0,0 +1,11 @@
+{
+  "bos_token_id": 0,
+  "decoder_start_token_id": 2,
+  "eos_token_id": 2,
+  "forced_bos_token_id": 250004,
+  "forced_eos_token_id": 2,
+  "max_length": 200,
+  "num_beams": 5,
+  "pad_token_id": 1,
+  "transformers_version": "4.27.4"
+}

many-eng-mBART/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:46e22995f5874e7e7cfe9e2694987824149a3b241d4a1507cd7562eb94d4df41
+size 2444694045

many-eng-mBART/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9187c1decfd14dc82482484b6be1866e0c64b0a7043e882d7d8cd60abd1c2d48
+size 14575

many-eng-mBART/scaler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cd6850eab3abc1e94035c14fc8e445e12f0627e346bb34450b997690a1c11cea
+size 557

many-eng-mBART/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c0d563bc42c2a289c1fb610196fe20b04a37a40982fa364fa1a23dd3f43368c4
+size 627

many-eng-mBART/sentencepiece.bpe.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cfc8146abe2a0488e9e2a0c56de7952f7c11ab059eca145a0a727afce0db2865
+size 5069051

many-eng-mBART/special_tokens_map.json

@@ -0,0 +1,63 @@
+{
+  "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"
+  ],
+  "bos_token": "<s>",
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "mask_token": "<mask>",
+  "pad_token": "<pad>",
+  "sep_token": "</s>",
+  "unk_token": "<unk>"
+}

many-eng-mBART/tokenizer_config.json

@@ -0,0 +1,78 @@
+{
+  "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"
+  ],
+  "bos_token": "<s>",
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "language_codes": "ML50",
+  "mask_token": {
+    "__type": "AddedToken",
+    "content": "<mask>",
+    "lstrip": true,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<pad>",
+  "sep_token": "</s>",
+  "sp_model_kwargs": {},
+  "special_tokens_map_file": "special_tokens_map.json",
+  "src_lang": null,
+  "tgt_lang": null,
+  "tokenizer_class": "MBart50Tokenizer",
+  "tokenizer_file": null,
+  "unk_token": "<unk>"
+}

many-eng-mBART/trainer_state.json

@@ -0,0 +1,260 @@
+{
+  "best_metric": 2.5316762924194336,
+  "best_model_checkpoint": "/content/gdrive/MyDrive/Translation/marianmt-many-eng-tagged-mbart/checkpoint-600",
+  "epoch": 1.6465887765603244,
+  "global_step": 680,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.1,
+      "eval_BLEU_ach": 7.374,
+      "eval_BLEU_lgg": 2.6457,
+      "eval_BLEU_lug": 14.2348,
+      "eval_BLEU_mean": 7.7324,
+      "eval_BLEU_nyn": 9.4935,
+      "eval_BLEU_teo": 4.9141,
+      "eval_loss": 3.311436653137207,
+      "eval_runtime": 174.8119,
+      "eval_samples_per_second": 14.301,
+      "eval_steps_per_second": 0.572,
+      "step": 40
+    },
+    {
+      "epoch": 0.19,
+      "eval_BLEU_ach": 17.4994,
+      "eval_BLEU_lgg": 13.7961,
+      "eval_BLEU_lug": 28.3699,
+      "eval_BLEU_mean": 19.0648,
+      "eval_BLEU_nyn": 19.5416,
+      "eval_BLEU_teo": 16.117,
+      "eval_loss": 2.849926233291626,
+      "eval_runtime": 143.9679,
+      "eval_samples_per_second": 17.365,
+      "eval_steps_per_second": 0.695,
+      "step": 80
+    },
+    {
+      "epoch": 0.29,
+      "eval_BLEU_ach": 20.4475,
+      "eval_BLEU_lgg": 18.8818,
+      "eval_BLEU_lug": 31.465,
+      "eval_BLEU_mean": 22.8451,
+      "eval_BLEU_nyn": 22.6147,
+      "eval_BLEU_teo": 20.8166,
+      "eval_loss": 2.7193143367767334,
+      "eval_runtime": 151.2716,
+      "eval_samples_per_second": 16.527,
+      "eval_steps_per_second": 0.661,
+      "step": 120
+    },
+    {
+      "epoch": 0.39,
+      "eval_BLEU_ach": 21.1792,
+      "eval_BLEU_lgg": 22.6167,
+      "eval_BLEU_lug": 33.6773,
+      "eval_BLEU_mean": 25.1052,
+      "eval_BLEU_nyn": 24.9998,
+      "eval_BLEU_teo": 23.0531,
+      "eval_loss": 2.6529085636138916,
+      "eval_runtime": 151.3691,
+      "eval_samples_per_second": 16.516,
+      "eval_steps_per_second": 0.661,
+      "step": 160
+    },
+    {
+      "epoch": 0.48,
+      "eval_BLEU_ach": 21.8405,
+      "eval_BLEU_lgg": 24.3787,
+      "eval_BLEU_lug": 34.771,
+      "eval_BLEU_mean": 26.5409,
+      "eval_BLEU_nyn": 26.1483,
+      "eval_BLEU_teo": 25.5659,
+      "eval_loss": 2.6228654384613037,
+      "eval_runtime": 145.7246,
+      "eval_samples_per_second": 17.156,
+      "eval_steps_per_second": 0.686,
+      "step": 200
+    },
+    {
+      "epoch": 0.58,
+      "eval_BLEU_ach": 23.8728,
+      "eval_BLEU_lgg": 26.4161,
+      "eval_BLEU_lug": 35.6208,
+      "eval_BLEU_mean": 27.7489,
+      "eval_BLEU_nyn": 26.7958,
+      "eval_BLEU_teo": 26.0391,
+      "eval_loss": 2.5896105766296387,
+      "eval_runtime": 145.5739,
+      "eval_samples_per_second": 17.173,
+      "eval_steps_per_second": 0.687,
+      "step": 240
+    },
+    {
+      "epoch": 0.68,
+      "eval_BLEU_ach": 24.1946,
+      "eval_BLEU_lgg": 26.7135,
+      "eval_BLEU_lug": 36.3971,
+      "eval_BLEU_mean": 28.5216,
+      "eval_BLEU_nyn": 27.9368,
+      "eval_BLEU_teo": 27.3661,
+      "eval_loss": 2.5668087005615234,
+      "eval_runtime": 139.7436,
+      "eval_samples_per_second": 17.89,
+      "eval_steps_per_second": 0.716,
+      "step": 280
+    },
+    {
+      "epoch": 0.77,
+      "eval_BLEU_ach": 25.5654,
+      "eval_BLEU_lgg": 26.7393,
+      "eval_BLEU_lug": 36.9048,
+      "eval_BLEU_mean": 29.2994,
+      "eval_BLEU_nyn": 28.5134,
+      "eval_BLEU_teo": 28.7739,
+      "eval_loss": 2.5546505451202393,
+      "eval_runtime": 140.7306,
+      "eval_samples_per_second": 17.764,
+      "eval_steps_per_second": 0.711,
+      "step": 320
+    },
+    {
+      "epoch": 0.87,
+      "eval_BLEU_ach": 25.4308,
+      "eval_BLEU_lgg": 28.1125,
+      "eval_BLEU_lug": 36.9446,
+      "eval_BLEU_mean": 29.4833,
+      "eval_BLEU_nyn": 28.1238,
+      "eval_BLEU_teo": 28.8046,
+      "eval_loss": 2.5436601638793945,
+      "eval_runtime": 143.9611,
+      "eval_samples_per_second": 17.366,
+      "eval_steps_per_second": 0.695,
+      "step": 360
+    },
+    {
+      "epoch": 0.97,
+      "eval_BLEU_ach": 26.7467,
+      "eval_BLEU_lgg": 28.6318,
+      "eval_BLEU_lug": 36.9987,
+      "eval_BLEU_mean": 30.0818,
+      "eval_BLEU_nyn": 29.1607,
+      "eval_BLEU_teo": 28.871,
+      "eval_loss": 2.538221836090088,
+      "eval_runtime": 144.2328,
+      "eval_samples_per_second": 17.333,
+      "eval_steps_per_second": 0.693,
+      "step": 400
+    },
+    {
+      "epoch": 1.07,
+      "eval_BLEU_ach": 25.3994,
+      "eval_BLEU_lgg": 28.9482,
+      "eval_BLEU_lug": 36.3868,
+      "eval_BLEU_mean": 29.5297,
+      "eval_BLEU_nyn": 27.99,
+      "eval_BLEU_teo": 28.9242,
+      "eval_loss": 2.546099901199341,
+      "eval_runtime": 142.9851,
+      "eval_samples_per_second": 17.484,
+      "eval_steps_per_second": 0.699,
+      "step": 440
+    },
+    {
+      "epoch": 1.16,
+      "eval_BLEU_ach": 26.368,
+      "eval_BLEU_lgg": 28.3352,
+      "eval_BLEU_lug": 37.4236,
+      "eval_BLEU_mean": 29.7199,
+      "eval_BLEU_nyn": 28.3613,
+      "eval_BLEU_teo": 28.1114,
+      "eval_loss": 2.545441150665283,
+      "eval_runtime": 139.7722,
+      "eval_samples_per_second": 17.886,
+      "eval_steps_per_second": 0.715,
+      "step": 480
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 0.0002757281553398058,
+      "loss": 2.7247,
+      "step": 500
+    },
+    {
+      "epoch": 1.26,
+      "eval_BLEU_ach": 25.8616,
+      "eval_BLEU_lgg": 30.7302,
+      "eval_BLEU_lug": 38.6363,
+      "eval_BLEU_mean": 30.8524,
+      "eval_BLEU_nyn": 29.9434,
+      "eval_BLEU_teo": 29.0903,
+      "eval_loss": 2.533596992492676,
+      "eval_runtime": 137.9606,
+      "eval_samples_per_second": 18.121,
+      "eval_steps_per_second": 0.725,
+      "step": 520
+    },
+    {
+      "epoch": 1.36,
+      "eval_BLEU_ach": 26.0392,
+      "eval_BLEU_lgg": 30.1594,
+      "eval_BLEU_lug": 37.9999,
+      "eval_BLEU_mean": 30.3508,
+      "eval_BLEU_nyn": 27.9994,
+      "eval_BLEU_teo": 29.5562,
+      "eval_loss": 2.5437333583831787,
+      "eval_runtime": 138.8107,
+      "eval_samples_per_second": 18.01,
+      "eval_steps_per_second": 0.72,
+      "step": 560
+    },
+    {
+      "epoch": 1.45,
+      "eval_BLEU_ach": 26.2529,
+      "eval_BLEU_lgg": 29.8407,
+      "eval_BLEU_lug": 38.411,
+      "eval_BLEU_mean": 30.6186,
+      "eval_BLEU_nyn": 28.8491,
+      "eval_BLEU_teo": 29.7392,
+      "eval_loss": 2.5316762924194336,
+      "eval_runtime": 137.1551,
+      "eval_samples_per_second": 18.228,
+      "eval_steps_per_second": 0.729,
+      "step": 600
+    },
+    {
+      "epoch": 1.55,
+      "eval_BLEU_ach": 26.2604,
+      "eval_BLEU_lgg": 29.9693,
+      "eval_BLEU_lug": 38.5596,
+      "eval_BLEU_mean": 30.9198,
+      "eval_BLEU_nyn": 29.4182,
+      "eval_BLEU_teo": 30.3917,
+      "eval_loss": 2.5321297645568848,
+      "eval_runtime": 139.1502,
+      "eval_samples_per_second": 17.966,
+      "eval_steps_per_second": 0.719,
+      "step": 640
+    },
+    {
+      "epoch": 1.65,
+      "eval_BLEU_ach": 26.291,
+      "eval_BLEU_lgg": 28.9904,
+      "eval_BLEU_lug": 39.0724,
+      "eval_BLEU_mean": 30.8232,
+      "eval_BLEU_nyn": 28.9724,
+      "eval_BLEU_teo": 30.7897,
+      "eval_loss": 2.532381296157837,
+      "eval_runtime": 138.8494,
+      "eval_samples_per_second": 18.005,
+      "eval_steps_per_second": 0.72,
+      "step": 680
+    }
+  ],
+  "max_steps": 6180,
+  "num_train_epochs": 15,
+  "total_flos": 3.946090637857751e+17,
+  "trial_name": null,
+  "trial_params": null
+}

many-eng-mBART/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d307267bdf0f70e748efd0e15b3395ed29d02bb4cb326450e1d5ab37e4350151
+size 3771