Initial model commit

config.json

@@ -0,0 +1,61 @@
+{
+  "_name_or_path": "Helsinki-NLP/opus-mt-en-mul",
+  "activation_dropout": 0.0,
+  "activation_function": "swish",
+  "add_bias_logits": false,
+  "add_final_layer_norm": false,
+  "architectures": [
+    "MarianMTModel"
+  ],
+  "attention_dropout": 0.0,
+  "bad_words_ids": [
+    [
+      64109
+    ]
+  ],
+  "bos_token_id": 0,
+  "classif_dropout": 0.0,
+  "classifier_dropout": 0.0,
+  "d_model": 512,
+  "decoder_attention_heads": 8,
+  "decoder_ffn_dim": 2048,
+  "decoder_layerdrop": 0.0,
+  "decoder_layers": 6,
+  "decoder_start_token_id": 64109,
+  "decoder_vocab_size": 64110,
+  "dropout": 0.1,
+  "encoder_attention_heads": 8,
+  "encoder_ffn_dim": 2048,
+  "encoder_layerdrop": 0.0,
+  "encoder_layers": 6,
+  "eos_token_id": 0,
+  "extra_pos_embeddings": 64110,
+  "forced_eos_token_id": 0,
+  "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": 512,
+  "max_position_embeddings": 512,
+  "model_type": "marian",
+  "normalize_before": false,
+  "normalize_embedding": false,
+  "num_beams": 4,
+  "num_hidden_layers": 6,
+  "pad_token_id": 64109,
+  "scale_embedding": true,
+  "share_encoder_decoder_embeddings": true,
+  "static_position_embeddings": true,
+  "torch_dtype": "float32",
+  "transformers_version": "4.26.1",
+  "use_cache": true,
+  "vocab_size": 64110
+}

generation_config.json

@@ -0,0 +1,16 @@
+{
+  "_from_model_config": true,
+  "bad_words_ids": [
+    [
+      64109
+    ]
+  ],
+  "bos_token_id": 0,
+  "decoder_start_token_id": 64109,
+  "eos_token_id": 0,
+  "forced_eos_token_id": 0,
+  "max_length": 512,
+  "num_beams": 4,
+  "pad_token_id": 64109,
+  "transformers_version": "4.26.1"
+}

optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:644bd2d51850f8ebe962807c167a9aeb0dd89f165e25262b3eb2a31c3989c33e
+size 615851013

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:746409eb0b7e482f3b854c89692bc91306e3ee4a32bd65882e6fc6d73d0343d9
+size 308194181

rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:651a53e801c5a2790308a62468a078ec69803120b315687e518d76de24c461f2
+size 14575

scaler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:68cff80b680ddf6e7abbef98b5f336b97f9b5963e2209307f639383870e8cc71
+size 557

scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:91c9b267cde425c39aa53b1bd1888694f560f02c2c05398ea8df564fc7429dee
+size 627

special_tokens_map.json

@@ -0,0 +1,5 @@
+{
+  "eos_token": "</s>",
+  "pad_token": "<pad>",
+  "unk_token": "<unk>"
+}

tokenizer_config.json

@@ -0,0 +1,13 @@
+{
+  "eos_token": "</s>",
+  "model_max_length": 512,
+  "name_or_path": "Helsinki-NLP/opus-mt-en-mul",
+  "pad_token": "<pad>",
+  "separate_vocabs": false,
+  "source_lang": "eng",
+  "sp_model_kwargs": {},
+  "special_tokens_map_file": null,
+  "target_lang": "mul",
+  "tokenizer_class": "MarianTokenizer",
+  "unk_token": "<unk>"
+}

trainer_state.json

@@ -0,0 +1,308 @@
+{
+  "best_metric": 2.5794920921325684,
+  "best_model_checkpoint": "/kaggle/working/best/checkpoint-950",
+  "epoch": 2.6095654892153797,
+  "global_step": 1000,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.13,
+      "eval_BLEU_ach": 4.4065,
+      "eval_BLEU_lgg": 1.7478,
+      "eval_BLEU_lug": 19.3906,
+      "eval_BLEU_mean": 6.3725,
+      "eval_BLEU_nyn": 5.6412,
+      "eval_BLEU_teo": 0.6765,
+      "eval_loss": 3.7197346687316895,
+      "eval_runtime": 81.4138,
+      "eval_samples_per_second": 30.707,
+      "eval_steps_per_second": 0.614,
+      "step": 50
+    },
+    {
+      "epoch": 0.26,
+      "eval_BLEU_ach": 10.4487,
+      "eval_BLEU_lgg": 6.9863,
+      "eval_BLEU_lug": 22.1636,
+      "eval_BLEU_mean": 11.2797,
+      "eval_BLEU_nyn": 9.4768,
+      "eval_BLEU_teo": 7.3231,
+      "eval_loss": 3.190028429031372,
+      "eval_runtime": 80.5044,
+      "eval_samples_per_second": 31.054,
+      "eval_steps_per_second": 0.621,
+      "step": 100
+    },
+    {
+      "epoch": 0.39,
+      "eval_BLEU_ach": 12.7721,
+      "eval_BLEU_lgg": 10.1002,
+      "eval_BLEU_lug": 23.7283,
+      "eval_BLEU_mean": 13.7176,
+      "eval_BLEU_nyn": 11.3816,
+      "eval_BLEU_teo": 10.606,
+      "eval_loss": 2.9923548698425293,
+      "eval_runtime": 80.2426,
+      "eval_samples_per_second": 31.156,
+      "eval_steps_per_second": 0.623,
+      "step": 150
+    },
+    {
+      "epoch": 0.52,
+      "eval_BLEU_ach": 14.146,
+      "eval_BLEU_lgg": 13.022,
+      "eval_BLEU_lug": 24.6207,
+      "eval_BLEU_mean": 15.4215,
+      "eval_BLEU_nyn": 11.8514,
+      "eval_BLEU_teo": 13.4677,
+      "eval_loss": 2.884697437286377,
+      "eval_runtime": 81.1433,
+      "eval_samples_per_second": 30.81,
+      "eval_steps_per_second": 0.616,
+      "step": 200
+    },
+    {
+      "epoch": 0.65,
+      "eval_BLEU_ach": 14.4905,
+      "eval_BLEU_lgg": 13.8605,
+      "eval_BLEU_lug": 25.8326,
+      "eval_BLEU_mean": 15.9588,
+      "eval_BLEU_nyn": 11.7846,
+      "eval_BLEU_teo": 13.8258,
+      "eval_loss": 2.819976806640625,
+      "eval_runtime": 82.5563,
+      "eval_samples_per_second": 30.282,
+      "eval_steps_per_second": 0.606,
+      "step": 250
+    },
+    {
+      "epoch": 0.78,
+      "eval_BLEU_ach": 15.9859,
+      "eval_BLEU_lgg": 15.2204,
+      "eval_BLEU_lug": 25.4128,
+      "eval_BLEU_mean": 17.0546,
+      "eval_BLEU_nyn": 13.1463,
+      "eval_BLEU_teo": 15.5075,
+      "eval_loss": 2.7662155628204346,
+      "eval_runtime": 80.9944,
+      "eval_samples_per_second": 30.866,
+      "eval_steps_per_second": 0.617,
+      "step": 300
+    },
+    {
+      "epoch": 0.91,
+      "eval_BLEU_ach": 16.6088,
+      "eval_BLEU_lgg": 15.9575,
+      "eval_BLEU_lug": 25.2816,
+      "eval_BLEU_mean": 17.3754,
+      "eval_BLEU_nyn": 12.738,
+      "eval_BLEU_teo": 16.2911,
+      "eval_loss": 2.72863507270813,
+      "eval_runtime": 80.9213,
+      "eval_samples_per_second": 30.894,
+      "eval_steps_per_second": 0.618,
+      "step": 350
+    },
+    {
+      "epoch": 1.04,
+      "eval_BLEU_ach": 17.0217,
+      "eval_BLEU_lgg": 16.0095,
+      "eval_BLEU_lug": 25.841,
+      "eval_BLEU_mean": 17.8794,
+      "eval_BLEU_nyn": 13.4439,
+      "eval_BLEU_teo": 17.0808,
+      "eval_loss": 2.703911781311035,
+      "eval_runtime": 82.5765,
+      "eval_samples_per_second": 30.275,
+      "eval_steps_per_second": 0.605,
+      "step": 400
+    },
+    {
+      "epoch": 1.17,
+      "eval_BLEU_ach": 16.8919,
+      "eval_BLEU_lgg": 17.5364,
+      "eval_BLEU_lug": 25.8493,
+      "eval_BLEU_mean": 18.3704,
+      "eval_BLEU_nyn": 13.432,
+      "eval_BLEU_teo": 18.1425,
+      "eval_loss": 2.6822402477264404,
+      "eval_runtime": 82.4142,
+      "eval_samples_per_second": 30.335,
+      "eval_steps_per_second": 0.607,
+      "step": 450
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 0.00043472584856396867,
+      "loss": 2.7655,
+      "step": 500
+    },
+    {
+      "epoch": 1.3,
+      "eval_BLEU_ach": 16.8644,
+      "eval_BLEU_lgg": 17.662,
+      "eval_BLEU_lug": 26.7519,
+      "eval_BLEU_mean": 18.8597,
+      "eval_BLEU_nyn": 14.0338,
+      "eval_BLEU_teo": 18.9866,
+      "eval_loss": 2.6606709957122803,
+      "eval_runtime": 82.2093,
+      "eval_samples_per_second": 30.41,
+      "eval_steps_per_second": 0.608,
+      "step": 500
+    },
+    {
+      "epoch": 1.44,
+      "eval_BLEU_ach": 17.7792,
+      "eval_BLEU_lgg": 17.2092,
+      "eval_BLEU_lug": 26.9857,
+      "eval_BLEU_mean": 18.9292,
+      "eval_BLEU_nyn": 13.7654,
+      "eval_BLEU_teo": 18.9065,
+      "eval_loss": 2.6529688835144043,
+      "eval_runtime": 81.4731,
+      "eval_samples_per_second": 30.685,
+      "eval_steps_per_second": 0.614,
+      "step": 550
+    },
+    {
+      "epoch": 1.57,
+      "eval_BLEU_ach": 17.5146,
+      "eval_BLEU_lgg": 17.6765,
+      "eval_BLEU_lug": 26.2161,
+      "eval_BLEU_mean": 18.7786,
+      "eval_BLEU_nyn": 13.9067,
+      "eval_BLEU_teo": 18.5792,
+      "eval_loss": 2.640359401702881,
+      "eval_runtime": 82.7573,
+      "eval_samples_per_second": 30.209,
+      "eval_steps_per_second": 0.604,
+      "step": 600
+    },
+    {
+      "epoch": 1.7,
+      "eval_BLEU_ach": 18.0286,
+      "eval_BLEU_lgg": 17.2142,
+      "eval_BLEU_lug": 27.1338,
+      "eval_BLEU_mean": 19.1169,
+      "eval_BLEU_nyn": 14.3399,
+      "eval_BLEU_teo": 18.8677,
+      "eval_loss": 2.6239874362945557,
+      "eval_runtime": 83.0768,
+      "eval_samples_per_second": 30.093,
+      "eval_steps_per_second": 0.602,
+      "step": 650
+    },
+    {
+      "epoch": 1.83,
+      "eval_BLEU_ach": 18.2373,
+      "eval_BLEU_lgg": 17.8534,
+      "eval_BLEU_lug": 27.2152,
+      "eval_BLEU_mean": 19.2291,
+      "eval_BLEU_nyn": 14.9249,
+      "eval_BLEU_teo": 17.9148,
+      "eval_loss": 2.6153488159179688,
+      "eval_runtime": 83.4372,
+      "eval_samples_per_second": 29.963,
+      "eval_steps_per_second": 0.599,
+      "step": 700
+    },
+    {
+      "epoch": 1.96,
+      "eval_BLEU_ach": 18.1915,
+      "eval_BLEU_lgg": 18.7514,
+      "eval_BLEU_lug": 27.1244,
+      "eval_BLEU_mean": 19.5482,
+      "eval_BLEU_nyn": 14.923,
+      "eval_BLEU_teo": 18.7509,
+      "eval_loss": 2.605884075164795,
+      "eval_runtime": 83.7259,
+      "eval_samples_per_second": 29.859,
+      "eval_steps_per_second": 0.597,
+      "step": 750
+    },
+    {
+      "epoch": 2.09,
+      "eval_BLEU_ach": 18.1001,
+      "eval_BLEU_lgg": 18.7958,
+      "eval_BLEU_lug": 27.0829,
+      "eval_BLEU_mean": 19.6395,
+      "eval_BLEU_nyn": 14.6849,
+      "eval_BLEU_teo": 19.534,
+      "eval_loss": 2.5994479656219482,
+      "eval_runtime": 81.9571,
+      "eval_samples_per_second": 30.504,
+      "eval_steps_per_second": 0.61,
+      "step": 800
+    },
+    {
+      "epoch": 2.22,
+      "eval_BLEU_ach": 18.4623,
+      "eval_BLEU_lgg": 18.923,
+      "eval_BLEU_lug": 27.8801,
+      "eval_BLEU_mean": 20.1857,
+      "eval_BLEU_nyn": 15.4508,
+      "eval_BLEU_teo": 20.2125,
+      "eval_loss": 2.6000475883483887,
+      "eval_runtime": 82.634,
+      "eval_samples_per_second": 30.254,
+      "eval_steps_per_second": 0.605,
+      "step": 850
+    },
+    {
+      "epoch": 2.35,
+      "eval_BLEU_ach": 17.984,
+      "eval_BLEU_lgg": 19.1317,
+      "eval_BLEU_lug": 27.4247,
+      "eval_BLEU_mean": 19.9146,
+      "eval_BLEU_nyn": 15.4422,
+      "eval_BLEU_teo": 19.5903,
+      "eval_loss": 2.595747470855713,
+      "eval_runtime": 82.2257,
+      "eval_samples_per_second": 30.404,
+      "eval_steps_per_second": 0.608,
+      "step": 900
+    },
+    {
+      "epoch": 2.48,
+      "eval_BLEU_ach": 18.2931,
+      "eval_BLEU_lgg": 19.5551,
+      "eval_BLEU_lug": 26.6605,
+      "eval_BLEU_mean": 19.8366,
+      "eval_BLEU_nyn": 14.4395,
+      "eval_BLEU_teo": 20.2348,
+      "eval_loss": 2.5794920921325684,
+      "eval_runtime": 83.6122,
+      "eval_samples_per_second": 29.9,
+      "eval_steps_per_second": 0.598,
+      "step": 950
+    },
+    {
+      "epoch": 2.61,
+      "learning_rate": 0.0003694516971279374,
+      "loss": 2.2915,
+      "step": 1000
+    },
+    {
+      "epoch": 2.61,
+      "eval_BLEU_ach": 19.06,
+      "eval_BLEU_lgg": 19.0945,
+      "eval_BLEU_lug": 27.6451,
+      "eval_BLEU_mean": 20.1345,
+      "eval_BLEU_nyn": 15.1057,
+      "eval_BLEU_teo": 19.7672,
+      "eval_loss": 2.580702781677246,
+      "eval_runtime": 84.0777,
+      "eval_samples_per_second": 29.734,
+      "eval_steps_per_second": 0.595,
+      "step": 1000
+    }
+  ],
+  "max_steps": 3830,
+  "num_train_epochs": 10,
+  "total_flos": 7.668611525246976e+16,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:479ed7995aeb584398016ef5f3b9cc182e0a4fc4b175c860fada41bec51fed6c
+size 3579