stage best base simple tfidf wce model

checkpoint-60000/config.json

@@ -0,0 +1,62 @@
+{
+  "_name_or_path": "Helsinki-NLP/opus-mt-en-fr",
+  "_num_labels": 3,
+  "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": [
+    [
+      59513
+    ]
+  ],
+  "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": 59513,
+  "decoder_vocab_size": 59514,
+  "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,
+  "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": 512,
+  "max_position_embeddings": 512,
+  "model_type": "marian",
+  "normalize_before": false,
+  "normalize_embedding": false,
+  "num_beams": 4,
+  "num_hidden_layers": 6,
+  "pad_token_id": 59513,
+  "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": 59514
+}

checkpoint-60000/generation_config.json

@@ -0,0 +1,16 @@
+{
+  "bad_words_ids": [
+    [
+      59513
+    ]
+  ],
+  "bos_token_id": 0,
+  "decoder_start_token_id": 59513,
+  "eos_token_id": 0,
+  "forced_eos_token_id": 0,
+  "max_length": 512,
+  "num_beams": 4,
+  "pad_token_id": 59513,
+  "renormalize_logits": true,
+  "transformers_version": "4.28.0"
+}

checkpoint-60000/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2e59520691c117ea1bd877e1bfe01882afb6344ca21e55b98715bd485704a575
+size 597025797

checkpoint-60000/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8fc98e1224f3c436077f903fb1738893ae0cc176dff85bbd440b582504cf6967
+size 298763205

checkpoint-60000/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:425f4664411f9ce455fd3bd3524c36aefb9f3d55834e6f423d5ea5f4789b68ac
+size 14511

checkpoint-60000/scaler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6387ad04f732286d3be24c378e3195aff7d77378a7b0a5ad2c05e2d6bd20582e
+size 557

checkpoint-60000/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:16591d2163f8bc85d56572a600f474f5f9444cf73f0e3bfd3e50b0fda488247e
+size 627

checkpoint-60000/special_tokens_map.json

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

checkpoint-60000/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>"
+}

checkpoint-60000/trainer_state.json

@@ -0,0 +1,256 @@
+{
+  "best_metric": 42.5736,
+  "best_model_checkpoint": "opus_base_simple_tfidf_wce/checkpoint-60000",
+  "epoch": 2.9574132492113563,
+  "global_step": 60000,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.2,
+      "learning_rate": 1.9753918572555207e-05,
+      "loss": 0.2098,
+      "step": 4000
+    },
+    {
+      "epoch": 0.2,
+      "eval_bleu": 35.7996,
+      "eval_gen_len": 39.4171,
+      "eval_loss": 0.10897957533597946,
+      "eval_runtime": 150.8888,
+      "eval_samples_per_second": 6.912,
+      "eval_steps_per_second": 0.219,
+      "step": 4000
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1.950746746845426e-05,
+      "loss": 0.1762,
+      "step": 8000
+    },
+    {
+      "epoch": 0.39,
+      "eval_bleu": 36.8077,
+      "eval_gen_len": 39.0182,
+      "eval_loss": 0.10654434561729431,
+      "eval_runtime": 116.4461,
+      "eval_samples_per_second": 8.957,
+      "eval_steps_per_second": 0.283,
+      "step": 8000
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 1.9261016364353314e-05,
+      "loss": 0.1691,
+      "step": 12000
+    },
+    {
+      "epoch": 0.59,
+      "eval_bleu": 41.2732,
+      "eval_gen_len": 34.2013,
+      "eval_loss": 0.10531464219093323,
+      "eval_runtime": 133.3831,
+      "eval_samples_per_second": 7.82,
+      "eval_steps_per_second": 0.247,
+      "step": 12000
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 1.9014565260252367e-05,
+      "loss": 0.1635,
+      "step": 16000
+    },
+    {
+      "epoch": 0.79,
+      "eval_bleu": 41.6402,
+      "eval_gen_len": 34.6366,
+      "eval_loss": 0.10416054725646973,
+      "eval_runtime": 104.9246,
+      "eval_samples_per_second": 9.94,
+      "eval_steps_per_second": 0.315,
+      "step": 16000
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 1.876811415615142e-05,
+      "loss": 0.1611,
+      "step": 20000
+    },
+    {
+      "epoch": 0.99,
+      "eval_bleu": 41.7834,
+      "eval_gen_len": 35.3193,
+      "eval_loss": 0.10361454635858536,
+      "eval_runtime": 153.2125,
+      "eval_samples_per_second": 6.808,
+      "eval_steps_per_second": 0.215,
+      "step": 20000
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 1.8521786277602524e-05,
+      "loss": 0.1493,
+      "step": 24000
+    },
+    {
+      "epoch": 1.18,
+      "eval_bleu": 41.5067,
+      "eval_gen_len": 35.8773,
+      "eval_loss": 0.10355237871408463,
+      "eval_runtime": 92.4584,
+      "eval_samples_per_second": 11.281,
+      "eval_steps_per_second": 0.357,
+      "step": 24000
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 1.827545839905363e-05,
+      "loss": 0.1491,
+      "step": 28000
+    },
+    {
+      "epoch": 1.38,
+      "eval_bleu": 41.1362,
+      "eval_gen_len": 36.0518,
+      "eval_loss": 0.10270049422979355,
+      "eval_runtime": 90.7046,
+      "eval_samples_per_second": 11.499,
+      "eval_steps_per_second": 0.364,
+      "step": 28000
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 1.8029130520504733e-05,
+      "loss": 0.1486,
+      "step": 32000
+    },
+    {
+      "epoch": 1.58,
+      "eval_bleu": 42.0407,
+      "eval_gen_len": 34.8581,
+      "eval_loss": 0.10218680649995804,
+      "eval_runtime": 93.8654,
+      "eval_samples_per_second": 11.112,
+      "eval_steps_per_second": 0.352,
+      "step": 32000
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 1.778286425473186e-05,
+      "loss": 0.1472,
+      "step": 36000
+    },
+    {
+      "epoch": 1.77,
+      "eval_bleu": 40.0238,
+      "eval_gen_len": 37.8265,
+      "eval_loss": 0.10189016908407211,
+      "eval_runtime": 108.5416,
+      "eval_samples_per_second": 9.609,
+      "eval_steps_per_second": 0.304,
+      "step": 36000
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 1.7536474763406942e-05,
+      "loss": 0.1463,
+      "step": 40000
+    },
+    {
+      "epoch": 1.97,
+      "eval_bleu": 42.5557,
+      "eval_gen_len": 34.1285,
+      "eval_loss": 0.10188464820384979,
+      "eval_runtime": 71.1911,
+      "eval_samples_per_second": 14.651,
+      "eval_steps_per_second": 0.464,
+      "step": 40000
+    },
+    {
+      "epoch": 2.17,
+      "learning_rate": 1.7290146884858045e-05,
+      "loss": 0.1388,
+      "step": 44000
+    },
+    {
+      "epoch": 2.17,
+      "eval_bleu": 42.1819,
+      "eval_gen_len": 35.1783,
+      "eval_loss": 0.10227210074663162,
+      "eval_runtime": 93.0555,
+      "eval_samples_per_second": 11.208,
+      "eval_steps_per_second": 0.355,
+      "step": 44000
+    },
+    {
+      "epoch": 2.37,
+      "learning_rate": 1.704381900630915e-05,
+      "loss": 0.1378,
+      "step": 48000
+    },
+    {
+      "epoch": 2.37,
+      "eval_bleu": 42.5524,
+      "eval_gen_len": 35.2148,
+      "eval_loss": 0.10202094167470932,
+      "eval_runtime": 104.8789,
+      "eval_samples_per_second": 9.945,
+      "eval_steps_per_second": 0.315,
+      "step": 48000
+    },
+    {
+      "epoch": 2.56,
+      "learning_rate": 1.679755274053628e-05,
+      "loss": 0.1367,
+      "step": 52000
+    },
+    {
+      "epoch": 2.56,
+      "eval_bleu": 42.2979,
+      "eval_gen_len": 34.6817,
+      "eval_loss": 0.10162690281867981,
+      "eval_runtime": 90.8884,
+      "eval_samples_per_second": 11.476,
+      "eval_steps_per_second": 0.363,
+      "step": 52000
+    },
+    {
+      "epoch": 2.76,
+      "learning_rate": 1.6551163249211358e-05,
+      "loss": 0.137,
+      "step": 56000
+    },
+    {
+      "epoch": 2.76,
+      "eval_bleu": 42.4162,
+      "eval_gen_len": 35.652,
+      "eval_loss": 0.10157214105129242,
+      "eval_runtime": 132.7314,
+      "eval_samples_per_second": 7.858,
+      "eval_steps_per_second": 0.249,
+      "step": 56000
+    },
+    {
+      "epoch": 2.96,
+      "learning_rate": 1.6304773757886436e-05,
+      "loss": 0.1373,
+      "step": 60000
+    },
+    {
+      "epoch": 2.96,
+      "eval_bleu": 42.5736,
+      "eval_gen_len": 34.418,
+      "eval_loss": 0.10088738054037094,
+      "eval_runtime": 113.8031,
+      "eval_samples_per_second": 9.165,
+      "eval_steps_per_second": 0.29,
+      "step": 60000
+    }
+  ],
+  "max_steps": 324608,
+  "num_train_epochs": 16,
+  "total_flos": 1.0348852591617638e+17,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-60000/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a905ba110b0d9116123e168c70df7fb6514427860ac3c19aa00cc6bd4069f0c4
+size 3771

config.json

@@ -0,0 +1,62 @@
+{
+  "_name_or_path": "Helsinki-NLP/opus-mt-en-fr",
+  "_num_labels": 3,
+  "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": [
+    [
+      59513
+    ]
+  ],
+  "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": 59513,
+  "decoder_vocab_size": 59514,
+  "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,
+  "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": 512,
+  "max_position_embeddings": 512,
+  "model_type": "marian",
+  "normalize_before": false,
+  "normalize_embedding": false,
+  "num_beams": 4,
+  "num_hidden_layers": 6,
+  "pad_token_id": 59513,
+  "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": 59514
+}

generation_config.json

@@ -0,0 +1,16 @@
+{
+  "bad_words_ids": [
+    [
+      59513
+    ]
+  ],
+  "bos_token_id": 0,
+  "decoder_start_token_id": 59513,
+  "eos_token_id": 0,
+  "forced_eos_token_id": 0,
+  "max_length": 512,
+  "num_beams": 4,
+  "pad_token_id": 59513,
+  "renormalize_logits": true,
+  "transformers_version": "4.28.0"
+}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8fc98e1224f3c436077f903fb1738893ae0cc176dff85bbd440b582504cf6967
+size 298763205

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>"
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a905ba110b0d9116123e168c70df7fb6514427860ac3c19aa00cc6bd4069f0c4
+size 3771