stage best big simple unsampled tfidf model

checkpoint-80000/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
+}

checkpoint-80000/generation_config.json

@@ -0,0 +1,16 @@
+{
+  "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,
+  "renormalize_logits": true,
+  "transformers_version": "4.28.0"
+}

checkpoint-80000/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:12544e28c4106146163dbb38b028ee691c497971cd89e8aace536a0ae8213e4a
+size 1845323269

checkpoint-80000/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:495127f14b870fe57639ecd59beacae76655b6fbccf4f3e62f317b48f3a9174a
+size 922885701

checkpoint-80000/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b6a94ecd596cea8258e793cba5af3dd1e89c782a8dacde9b57f6ecb51c176f9b
+size 14575

checkpoint-80000/scaler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:935f47817eaaae7db111675237853f41b8b8a68532f35b103b7b16d7bdc728f4
+size 557

checkpoint-80000/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d841877dae97f3484921bd550ae3afec1339fe3ea6e45d722c6eee90ccd95ea9
+size 627

checkpoint-80000/special_tokens_map.json

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

checkpoint-80000/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-80000/trainer_state.json

@@ -0,0 +1,176 @@
+{
+  "best_metric": 43.3778,
+  "best_model_checkpoint": "opus_big_simple_tfidf_wce_unsampled/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": 0.1925,
+      "step": 8000
+    },
+    {
+      "epoch": 0.2,
+      "eval_bleu": 42.1412,
+      "eval_gen_len": 33.8792,
+      "eval_loss": 0.09833964705467224,
+      "eval_runtime": 207.8182,
+      "eval_samples_per_second": 5.019,
+      "eval_steps_per_second": 0.318,
+      "step": 8000
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1.9507270486752928e-05,
+      "loss": 0.161,
+      "step": 16000
+    },
+    {
+      "epoch": 0.39,
+      "eval_bleu": 42.3996,
+      "eval_gen_len": 34.0614,
+      "eval_loss": 0.09759480506181717,
+      "eval_runtime": 125.4626,
+      "eval_samples_per_second": 8.313,
+      "eval_steps_per_second": 0.526,
+      "step": 16000
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 1.9260874922982132e-05,
+      "loss": 0.1564,
+      "step": 24000
+    },
+    {
+      "epoch": 0.59,
+      "eval_bleu": 42.5797,
+      "eval_gen_len": 33.9444,
+      "eval_loss": 0.0980168879032135,
+      "eval_runtime": 161.031,
+      "eval_samples_per_second": 6.477,
+      "eval_steps_per_second": 0.41,
+      "step": 24000
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 1.901447935921134e-05,
+      "loss": 0.1521,
+      "step": 32000
+    },
+    {
+      "epoch": 0.79,
+      "eval_bleu": 42.7821,
+      "eval_gen_len": 34.1016,
+      "eval_loss": 0.0970257818698883,
+      "eval_runtime": 117.7754,
+      "eval_samples_per_second": 8.856,
+      "eval_steps_per_second": 0.56,
+      "step": 32000
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 1.876817621688232e-05,
+      "loss": 0.1504,
+      "step": 40000
+    },
+    {
+      "epoch": 0.99,
+      "eval_bleu": 42.584,
+      "eval_gen_len": 33.5369,
+      "eval_loss": 0.09632710367441177,
+      "eval_runtime": 98.5843,
+      "eval_samples_per_second": 10.58,
+      "eval_steps_per_second": 0.669,
+      "step": 40000
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 1.852178065311152e-05,
+      "loss": 0.134,
+      "step": 48000
+    },
+    {
+      "epoch": 1.18,
+      "eval_bleu": 42.828,
+      "eval_gen_len": 34.0058,
+      "eval_loss": 0.09739922732114792,
+      "eval_runtime": 114.6861,
+      "eval_samples_per_second": 9.094,
+      "eval_steps_per_second": 0.575,
+      "step": 48000
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 1.8275415896487988e-05,
+      "loss": 0.1343,
+      "step": 56000
+    },
+    {
+      "epoch": 1.38,
+      "eval_bleu": 42.4346,
+      "eval_gen_len": 33.8926,
+      "eval_loss": 0.097435861825943,
+      "eval_runtime": 236.3987,
+      "eval_samples_per_second": 4.412,
+      "eval_steps_per_second": 0.279,
+      "step": 56000
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 1.8029020332717192e-05,
+      "loss": 0.1349,
+      "step": 64000
+    },
+    {
+      "epoch": 1.58,
+      "eval_bleu": 43.1683,
+      "eval_gen_len": 34.4803,
+      "eval_loss": 0.09670189768075943,
+      "eval_runtime": 133.1801,
+      "eval_samples_per_second": 7.832,
+      "eval_steps_per_second": 0.496,
+      "step": 64000
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 1.7782655576093655e-05,
+      "loss": 0.1343,
+      "step": 72000
+    },
+    {
+      "epoch": 1.77,
+      "eval_bleu": 42.0682,
+      "eval_gen_len": 33.9291,
+      "eval_loss": 0.09708751738071442,
+      "eval_runtime": 130.5758,
+      "eval_samples_per_second": 7.988,
+      "eval_steps_per_second": 0.505,
+      "step": 72000
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 1.7536321626617377e-05,
+      "loss": 0.134,
+      "step": 80000
+    },
+    {
+      "epoch": 1.97,
+      "eval_bleu": 43.3778,
+      "eval_gen_len": 33.7421,
+      "eval_loss": 0.09596329927444458,
+      "eval_runtime": 106.3126,
+      "eval_samples_per_second": 9.811,
+      "eval_steps_per_second": 0.621,
+      "step": 80000
+    }
+  ],
+  "max_steps": 649200,
+  "num_train_epochs": 16,
+  "total_flos": 2.5844916958632346e+17,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-80000/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b183cb667854fd7b42444b9137058b3df13b3220a47d3bf3ca869b006cddd5b6
+size 3771

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 +1,16 @@
+{
+  "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,
+  "renormalize_logits": true,
+  "transformers_version": "4.28.0"
+}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:495127f14b870fe57639ecd59beacae76655b6fbccf4f3e62f317b48f3a9174a
+size 922885701

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:b183cb667854fd7b42444b9137058b3df13b3220a47d3bf3ca869b006cddd5b6
+size 3771