stage best unsampled glossary simple adapt wce model

config.json

@@ -0,0 +1,46 @@
+{
+  "_name_or_path": "Helsinki-NLP/opus-mt-tc-big-en-fr",
+  "activation_dropout": 0.0,
+  "activation_function": "relu",
+  "architectures": [
+    "MarianMTModel"
+  ],
+  "attention_dropout": 0.0,
+  "bad_words_ids": [
+    [
+      53016
+    ]
+  ],
+  "bos_token_id": 0,
+  "classifier_dropout": 0.0,
+  "d_model": 1024,
+  "decoder_attention_heads": 16,
+  "decoder_ffn_dim": 4096,
+  "decoder_layerdrop": 0.0,
+  "decoder_layers": 6,
+  "decoder_start_token_id": 53016,
+  "decoder_vocab_size": 53017,
+  "dropout": 0.1,
+  "encoder_attention_heads": 16,
+  "encoder_ffn_dim": 4096,
+  "encoder_layerdrop": 0.0,
+  "encoder_layers": 6,
+  "eos_token_id": 43311,
+  "forced_eos_token_id": 43311,
+  "init_std": 0.02,
+  "is_encoder_decoder": true,
+  "max_length": 512,
+  "max_position_embeddings": 1024,
+  "model_type": "marian",
+  "normalize_embedding": false,
+  "num_beams": 4,
+  "num_hidden_layers": 6,
+  "pad_token_id": 53016,
+  "scale_embedding": true,
+  "share_encoder_decoder_embeddings": true,
+  "static_position_embeddings": true,
+  "torch_dtype": "float32",
+  "transformers_version": "4.28.0",
+  "use_cache": true,
+  "vocab_size": 53017
+}

generation_config.json

@@ -0,0 +1,15 @@
+{
+  "bad_words_ids": [
+    [
+      53016
+    ]
+  ],
+  "bos_token_id": 0,
+  "decoder_start_token_id": 53016,
+  "eos_token_id": 43311,
+  "forced_eos_token_id": 43311,
+  "max_length": 512,
+  "num_beams": 4,
+  "pad_token_id": 53016,
+  "transformers_version": "4.28.0"
+}

optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8a48b2aae37d5a73aec4d4b2589b64eeb546d915de6174e2995774308539791e
+size 1845323269

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:05af2e0d14ab4e4aaf567f6e8cc1c20129b38ef54698b6858f0f1d8ded93288f
+size 922885701

rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cbfba05a1fccae790a539c4ff8ee13ef1f633735f45b066ff11e812f3e37065e
+size 14575

scaler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8553d83270175bc39173b92f7b41aea01d380c65e36dd70550db2a9fc3983b41
+size 557

scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f3af67343aa0f8b4109961c1ed1ad806d5b136e6b0592737a57873f77d6c9558
+size 627

special_tokens_map.json

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

tokenizer_config.json

@@ -0,0 +1,12 @@
+{
+  "clean_up_tokenization_spaces": true,
+  "eos_token": "</s>",
+  "model_max_length": 512,
+  "pad_token": "<pad>",
+  "separate_vocabs": false,
+  "source_lang": "en",
+  "sp_model_kwargs": {},
+  "target_lang": "fr",
+  "tokenizer_class": "MarianTokenizer",
+  "unk_token": "<unk>"
+}

trainer_state.json

@@ -0,0 +1,176 @@
+{
+  "best_metric": 43.6441,
+  "best_model_checkpoint": "opus_big_adapt_wce_gloss_unsampled_bands_7_ubweight_1.5/checkpoint-80000",
+  "epoch": 1.9716574245224892,
+  "global_step": 80000,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.2,
+      "learning_rate": 1.9753635243376465e-05,
+      "loss": 0.1779,
+      "step": 8000
+    },
+    {
+      "epoch": 0.2,
+      "eval_bleu": 42.3466,
+      "eval_gen_len": 34.1314,
+      "eval_loss": 0.09819881618022919,
+      "eval_runtime": 147.9956,
+      "eval_samples_per_second": 7.048,
+      "eval_steps_per_second": 0.446,
+      "step": 8000
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1.9507270486752928e-05,
+      "loss": 0.1471,
+      "step": 16000
+    },
+    {
+      "epoch": 0.39,
+      "eval_bleu": 42.6127,
+      "eval_gen_len": 34.3682,
+      "eval_loss": 0.09763215482234955,
+      "eval_runtime": 92.3945,
+      "eval_samples_per_second": 11.289,
+      "eval_steps_per_second": 0.714,
+      "step": 16000
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 1.9260844115834877e-05,
+      "loss": 0.143,
+      "step": 24000
+    },
+    {
+      "epoch": 0.59,
+      "eval_bleu": 42.3437,
+      "eval_gen_len": 33.6031,
+      "eval_loss": 0.09766430407762527,
+      "eval_runtime": 84.1375,
+      "eval_samples_per_second": 12.396,
+      "eval_steps_per_second": 0.784,
+      "step": 24000
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 1.901447935921134e-05,
+      "loss": 0.1391,
+      "step": 32000
+    },
+    {
+      "epoch": 0.79,
+      "eval_bleu": 42.6785,
+      "eval_gen_len": 33.7076,
+      "eval_loss": 0.0967756137251854,
+      "eval_runtime": 80.1108,
+      "eval_samples_per_second": 13.019,
+      "eval_steps_per_second": 0.824,
+      "step": 32000
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 1.8768083795440544e-05,
+      "loss": 0.1377,
+      "step": 40000
+    },
+    {
+      "epoch": 0.99,
+      "eval_bleu": 42.593,
+      "eval_gen_len": 34.2493,
+      "eval_loss": 0.09624858945608139,
+      "eval_runtime": 89.2577,
+      "eval_samples_per_second": 11.685,
+      "eval_steps_per_second": 0.739,
+      "step": 40000
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 1.8521749845964266e-05,
+      "loss": 0.1226,
+      "step": 48000
+    },
+    {
+      "epoch": 1.18,
+      "eval_bleu": 43.0072,
+      "eval_gen_len": 33.9262,
+      "eval_loss": 0.09690112620592117,
+      "eval_runtime": 81.2861,
+      "eval_samples_per_second": 12.831,
+      "eval_steps_per_second": 0.812,
+      "step": 48000
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 1.827535428219347e-05,
+      "loss": 0.1229,
+      "step": 56000
+    },
+    {
+      "epoch": 1.38,
+      "eval_bleu": 42.0603,
+      "eval_gen_len": 34.1515,
+      "eval_loss": 0.09700236469507217,
+      "eval_runtime": 106.3162,
+      "eval_samples_per_second": 9.81,
+      "eval_steps_per_second": 0.621,
+      "step": 56000
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 1.8028989525569933e-05,
+      "loss": 0.1235,
+      "step": 64000
+    },
+    {
+      "epoch": 1.58,
+      "eval_bleu": 43.1273,
+      "eval_gen_len": 34.5503,
+      "eval_loss": 0.09687265008687973,
+      "eval_runtime": 104.2899,
+      "eval_samples_per_second": 10.001,
+      "eval_steps_per_second": 0.633,
+      "step": 64000
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 1.7782655576093655e-05,
+      "loss": 0.123,
+      "step": 72000
+    },
+    {
+      "epoch": 1.77,
+      "eval_bleu": 42.8922,
+      "eval_gen_len": 34.3643,
+      "eval_loss": 0.09673970192670822,
+      "eval_runtime": 85.0152,
+      "eval_samples_per_second": 12.268,
+      "eval_steps_per_second": 0.776,
+      "step": 72000
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 1.753626001232286e-05,
+      "loss": 0.1227,
+      "step": 80000
+    },
+    {
+      "epoch": 1.97,
+      "eval_bleu": 43.6441,
+      "eval_gen_len": 33.999,
+      "eval_loss": 0.09602336585521698,
+      "eval_runtime": 92.3918,
+      "eval_samples_per_second": 11.289,
+      "eval_steps_per_second": 0.714,
+      "step": 80000
+    }
+  ],
+  "max_steps": 649200,
+  "num_train_epochs": 16,
+  "total_flos": 2.5844916958632346e+17,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0eb9e6cd285688a1599292d4bb1836ecc103b3d560996907eb799a94f98469b5
+size 3835