Mix Trained marian model over 9 low resource languages of ALT dataset + 18K senetences from en-ja opus dataset.

last-checkpoint/config.json

@@ -1,44 +0,0 @@
-{
-  "_name_or_path": "Helsinki-NLP/opus-tatoeba-en-ja",
-  "activation_dropout": 0.0,
-  "activation_function": "swish",
-  "architectures": [
-    "MarianMTModel"
-  ],
-  "attention_dropout": 0.0,
-  "bad_words_ids": [
-    [
-      65000
-    ]
-  ],
-  "bos_token_id": 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": 65000,
-  "dropout": 0.1,
-  "encoder_attention_heads": 8,
-  "encoder_ffn_dim": 2048,
-  "encoder_layerdrop": 0.0,
-  "encoder_layers": 6,
-  "eos_token_id": 0,
-  "forced_eos_token_id": 0,
-  "init_std": 0.02,
-  "is_encoder_decoder": true,
-  "max_length": 512,
-  "max_position_embeddings": 512,
-  "model_type": "marian",
-  "normalize_embedding": false,
-  "num_beams": 6,
-  "num_hidden_layers": 6,
-  "pad_token_id": 65000,
-  "scale_embedding": true,
-  "static_position_embeddings": true,
-  "torch_dtype": "float32",
-  "transformers_version": "4.12.5",
-  "use_cache": true,
-  "vocab_size": 97154
-}

last-checkpoint/optimizer.pt

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:eddb74787696dede4919ec271555a6350a568c321174a7c15dc3e1d67bee90d9
-size 751197645

last-checkpoint/pytorch_model.bin

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:ab9255443ab93767d06e31e700d16c826a1b9c4bd217525b669b2851a871e8fd
-size 376008389

last-checkpoint/rng_state.pth

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:955f2254b1e2583a0c202f4167da7de0b528b48ae4ee34006c7519a67f4f2166
-size 14567

last-checkpoint/scaler.pt

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:35de3af080a175fc9162ccb44a37ba550925fde9c9502c74bafce2c813e66388
-size 559

last-checkpoint/scheduler.pt

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:265e019a5d08011062ca46680adb46fea92faeb6e1ff9bcaf722eb0c685c75c6
-size 623

last-checkpoint/special_tokens_map.json

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

last-checkpoint/tokenizer_config.json

@@ -1 +0,0 @@
-{"source_lang": "eng", "target_lang": "jpn", "unk_token": "<unk>", "eos_token": "</s>", "pad_token": "<pad>", "model_max_length": 512, "sp_model_kwargs": {}, "special_tokens_map_file": null, "tokenizer_file": null, "name_or_path": "Helsinki-NLP/opus-tatoeba-en-ja", "tokenizer_class": "MarianTokenizer"}

last-checkpoint/trainer_state.json

@@ -1,276 +0,0 @@
-{
-  "best_metric": null,
-  "best_model_checkpoint": null,
-  "epoch": 3.5727045373347623,
-  "global_step": 20000,
-  "is_hyper_param_search": false,
-  "is_local_process_zero": true,
-  "is_world_process_zero": true,
-  "log_history": [
-    {
-      "epoch": 0.09,
-      "learning_rate": 9.825589706933525e-05,
-      "loss": 2.1183,
-      "step": 500
-    },
-    {
-      "epoch": 0.18,
-      "learning_rate": 9.646890636168693e-05,
-      "loss": 1.8604,
-      "step": 1000
-    },
-    {
-      "epoch": 0.27,
-      "learning_rate": 9.46819156540386e-05,
-      "loss": 1.7038,
-      "step": 1500
-    },
-    {
-      "epoch": 0.36,
-      "learning_rate": 9.289492494639028e-05,
-      "loss": 1.5787,
-      "step": 2000
-    },
-    {
-      "epoch": 0.45,
-      "learning_rate": 9.110793423874197e-05,
-      "loss": 1.4618,
-      "step": 2500
-    },
-    {
-      "epoch": 0.54,
-      "learning_rate": 8.932094353109363e-05,
-      "loss": 1.3833,
-      "step": 3000
-    },
-    {
-      "epoch": 0.63,
-      "learning_rate": 8.753395282344532e-05,
-      "loss": 1.3247,
-      "step": 3500
-    },
-    {
-      "epoch": 0.71,
-      "learning_rate": 8.5746962115797e-05,
-      "loss": 1.2699,
-      "step": 4000
-    },
-    {
-      "epoch": 0.8,
-      "learning_rate": 8.395997140814869e-05,
-      "loss": 1.2196,
-      "step": 4500
-    },
-    {
-      "epoch": 0.89,
-      "learning_rate": 8.217298070050035e-05,
-      "loss": 1.1743,
-      "step": 5000
-    },
-    {
-      "epoch": 0.98,
-      "learning_rate": 8.038598999285204e-05,
-      "loss": 1.1334,
-      "step": 5500
-    },
-    {
-      "epoch": 1.07,
-      "learning_rate": 7.859899928520372e-05,
-      "loss": 1.0688,
-      "step": 6000
-    },
-    {
-      "epoch": 1.16,
-      "learning_rate": 7.681200857755541e-05,
-      "loss": 1.0209,
-      "step": 6500
-    },
-    {
-      "epoch": 1.25,
-      "learning_rate": 7.502501786990707e-05,
-      "loss": 1.0123,
-      "step": 7000
-    },
-    {
-      "epoch": 1.34,
-      "learning_rate": 7.323802716225876e-05,
-      "loss": 0.9964,
-      "step": 7500
-    },
-    {
-      "epoch": 1.43,
-      "learning_rate": 7.145461043602574e-05,
-      "loss": 0.9671,
-      "step": 8000
-    },
-    {
-      "epoch": 1.52,
-      "learning_rate": 6.966761972837742e-05,
-      "loss": 0.9496,
-      "step": 8500
-    },
-    {
-      "epoch": 1.61,
-      "learning_rate": 6.788062902072909e-05,
-      "loss": 0.9354,
-      "step": 9000
-    },
-    {
-      "epoch": 1.7,
-      "learning_rate": 6.609363831308077e-05,
-      "loss": 0.9291,
-      "step": 9500
-    },
-    {
-      "epoch": 1.79,
-      "learning_rate": 6.430664760543246e-05,
-      "loss": 0.9039,
-      "step": 10000
-    },
-    {
-      "epoch": 1.79,
-      "eval_bleu": 4.2041,
-      "eval_gen_len": 43.7816,
-      "eval_loss": 0.9678672552108765,
-      "eval_runtime": 1428.8589,
-      "eval_samples_per_second": 6.25,
-      "eval_steps_per_second": 0.196,
-      "step": 10000
-    },
-    {
-      "epoch": 1.88,
-      "learning_rate": 6.251965689778412e-05,
-      "loss": 0.9,
-      "step": 10500
-    },
-    {
-      "epoch": 1.96,
-      "learning_rate": 6.073266619013581e-05,
-      "loss": 0.8792,
-      "step": 11000
-    },
-    {
-      "epoch": 2.05,
-      "learning_rate": 5.894924946390279e-05,
-      "loss": 0.8345,
-      "step": 11500
-    },
-    {
-      "epoch": 2.14,
-      "learning_rate": 5.716583273766977e-05,
-      "loss": 0.8162,
-      "step": 12000
-    },
-    {
-      "epoch": 2.23,
-      "learning_rate": 5.537884203002145e-05,
-      "loss": 0.8095,
-      "step": 12500
-    },
-    {
-      "epoch": 2.32,
-      "learning_rate": 5.359185132237312e-05,
-      "loss": 0.7957,
-      "step": 13000
-    },
-    {
-      "epoch": 2.41,
-      "learning_rate": 5.18048606147248e-05,
-      "loss": 0.7941,
-      "step": 13500
-    },
-    {
-      "epoch": 2.5,
-      "learning_rate": 5.0017869907076487e-05,
-      "loss": 0.7864,
-      "step": 14000
-    },
-    {
-      "epoch": 2.59,
-      "learning_rate": 4.823087919942816e-05,
-      "loss": 0.7805,
-      "step": 14500
-    },
-    {
-      "epoch": 2.68,
-      "learning_rate": 4.6443888491779846e-05,
-      "loss": 0.7744,
-      "step": 15000
-    },
-    {
-      "epoch": 2.77,
-      "learning_rate": 4.465689778413152e-05,
-      "loss": 0.7757,
-      "step": 15500
-    },
-    {
-      "epoch": 2.86,
-      "learning_rate": 4.2869907076483206e-05,
-      "loss": 0.7671,
-      "step": 16000
-    },
-    {
-      "epoch": 2.95,
-      "learning_rate": 4.108291636883488e-05,
-      "loss": 0.7618,
-      "step": 16500
-    },
-    {
-      "epoch": 3.04,
-      "learning_rate": 3.929592566118656e-05,
-      "loss": 0.7375,
-      "step": 17000
-    },
-    {
-      "epoch": 3.13,
-      "learning_rate": 3.750893495353824e-05,
-      "loss": 0.6979,
-      "step": 17500
-    },
-    {
-      "epoch": 3.22,
-      "learning_rate": 3.5729092208720514e-05,
-      "loss": 0.7052,
-      "step": 18000
-    },
-    {
-      "epoch": 3.3,
-      "learning_rate": 3.39421015010722e-05,
-      "loss": 0.6974,
-      "step": 18500
-    },
-    {
-      "epoch": 3.39,
-      "learning_rate": 3.2155110793423873e-05,
-      "loss": 0.6921,
-      "step": 19000
-    },
-    {
-      "epoch": 3.48,
-      "learning_rate": 3.0368120085775553e-05,
-      "loss": 0.696,
-      "step": 19500
-    },
-    {
-      "epoch": 3.57,
-      "learning_rate": 2.8581129378127237e-05,
-      "loss": 0.6971,
-      "step": 20000
-    },
-    {
-      "epoch": 3.57,
-      "eval_bleu": 5.3321,
-      "eval_gen_len": 44.1972,
-      "eval_loss": 0.8275578618049622,
-      "eval_runtime": 1432.7377,
-      "eval_samples_per_second": 6.233,
-      "eval_steps_per_second": 0.195,
-      "step": 20000
-    }
-  ],
-  "max_steps": 27990,
-  "num_train_epochs": 5,
-  "total_flos": 4.33848223531008e+16,
-  "trial_name": null,
-  "trial_params": null
-}

last-checkpoint/training_args.bin

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:788f94a24f9e0e59617584da9ecf7e726c3cbc0e93e03fa9a8b7ffebbb2ba793
-size 2991

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ab9255443ab93767d06e31e700d16c826a1b9c4bd217525b669b2851a871e8fd
+oid sha256:d751ab90d05d0d52fffef4a7a32ef85d75248f9e1c6e91b1b232e10d268cfea5
 size 376008389

runs/Feb08_19-24-11_fadcca9c27d0/events.out.tfevents.1644348258.fadcca9c27d0.35.2

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a6d8d0dbfec2c2425f8dbde5ffbd5b91ba063ed34d1bef26590e5f7a9537720b
-size 10759
+oid sha256:cd005f18ed18f94e7a52cc7365ddce803ae7292578769405b657f0d653b6f9ef
+size 13519