Upload folder using huggingface_hub
Browse files- checkpoint-1610/config.json +61 -0
- checkpoint-1610/generation_config.json +16 -0
- checkpoint-1610/model.safetensors +3 -0
- checkpoint-1610/optimizer.pt +3 -0
- checkpoint-1610/rng_state.pth +3 -0
- checkpoint-1610/scheduler.pt +3 -0
- checkpoint-1610/source.spm +0 -0
- checkpoint-1610/special_tokens_map.json +5 -0
- checkpoint-1610/target.spm +0 -0
- checkpoint-1610/tokenizer_config.json +39 -0
- checkpoint-1610/trainer_state.json +161 -0
- checkpoint-1610/training_args.bin +3 -0
- checkpoint-1610/vocab.json +0 -0
- checkpoint-1840/config.json +61 -0
- checkpoint-1840/generation_config.json +16 -0
- checkpoint-1840/model.safetensors +3 -0
- checkpoint-1840/optimizer.pt +3 -0
- checkpoint-1840/rng_state.pth +3 -0
- checkpoint-1840/scheduler.pt +3 -0
- checkpoint-1840/source.spm +0 -0
- checkpoint-1840/special_tokens_map.json +5 -0
- checkpoint-1840/target.spm +0 -0
- checkpoint-1840/tokenizer_config.json +39 -0
- checkpoint-1840/trainer_state.json +178 -0
- checkpoint-1840/training_args.bin +3 -0
- checkpoint-1840/vocab.json +0 -0
- checkpoint-2070/config.json +61 -0
- checkpoint-2070/generation_config.json +16 -0
- checkpoint-2070/model.safetensors +3 -0
- checkpoint-2070/optimizer.pt +3 -0
- checkpoint-2070/rng_state.pth +3 -0
- checkpoint-2070/scheduler.pt +3 -0
- checkpoint-2070/source.spm +0 -0
- checkpoint-2070/special_tokens_map.json +5 -0
- checkpoint-2070/target.spm +0 -0
- checkpoint-2070/tokenizer_config.json +39 -0
- checkpoint-2070/trainer_state.json +195 -0
- checkpoint-2070/training_args.bin +3 -0
- checkpoint-2070/vocab.json +0 -0
- test_metrics.json +1 -0
checkpoint-1610/config.json
ADDED
|
@@ -0,0 +1,61 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "Helsinki-NLP/opus-mt-en-ar",
|
| 3 |
+
"activation_dropout": 0.0,
|
| 4 |
+
"activation_function": "swish",
|
| 5 |
+
"add_bias_logits": false,
|
| 6 |
+
"add_final_layer_norm": false,
|
| 7 |
+
"architectures": [
|
| 8 |
+
"MarianMTModel"
|
| 9 |
+
],
|
| 10 |
+
"attention_dropout": 0.0,
|
| 11 |
+
"bad_words_ids": [
|
| 12 |
+
[
|
| 13 |
+
62801
|
| 14 |
+
]
|
| 15 |
+
],
|
| 16 |
+
"bos_token_id": 0,
|
| 17 |
+
"classif_dropout": 0.0,
|
| 18 |
+
"classifier_dropout": 0.0,
|
| 19 |
+
"d_model": 512,
|
| 20 |
+
"decoder_attention_heads": 8,
|
| 21 |
+
"decoder_ffn_dim": 2048,
|
| 22 |
+
"decoder_layerdrop": 0.0,
|
| 23 |
+
"decoder_layers": 6,
|
| 24 |
+
"decoder_start_token_id": 62801,
|
| 25 |
+
"decoder_vocab_size": 62802,
|
| 26 |
+
"dropout": 0.1,
|
| 27 |
+
"encoder_attention_heads": 8,
|
| 28 |
+
"encoder_ffn_dim": 2048,
|
| 29 |
+
"encoder_layerdrop": 0.0,
|
| 30 |
+
"encoder_layers": 6,
|
| 31 |
+
"eos_token_id": 0,
|
| 32 |
+
"extra_pos_embeddings": 62802,
|
| 33 |
+
"forced_eos_token_id": 0,
|
| 34 |
+
"id2label": {
|
| 35 |
+
"0": "LABEL_0",
|
| 36 |
+
"1": "LABEL_1",
|
| 37 |
+
"2": "LABEL_2"
|
| 38 |
+
},
|
| 39 |
+
"init_std": 0.02,
|
| 40 |
+
"is_encoder_decoder": true,
|
| 41 |
+
"label2id": {
|
| 42 |
+
"LABEL_0": 0,
|
| 43 |
+
"LABEL_1": 1,
|
| 44 |
+
"LABEL_2": 2
|
| 45 |
+
},
|
| 46 |
+
"max_length": 512,
|
| 47 |
+
"max_position_embeddings": 512,
|
| 48 |
+
"model_type": "marian",
|
| 49 |
+
"normalize_before": false,
|
| 50 |
+
"normalize_embedding": false,
|
| 51 |
+
"num_beams": 4,
|
| 52 |
+
"num_hidden_layers": 6,
|
| 53 |
+
"pad_token_id": 62801,
|
| 54 |
+
"scale_embedding": true,
|
| 55 |
+
"share_encoder_decoder_embeddings": true,
|
| 56 |
+
"static_position_embeddings": true,
|
| 57 |
+
"torch_dtype": "float32",
|
| 58 |
+
"transformers_version": "4.42.4",
|
| 59 |
+
"use_cache": true,
|
| 60 |
+
"vocab_size": 62802
|
| 61 |
+
}
|
checkpoint-1610/generation_config.json
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bad_words_ids": [
|
| 3 |
+
[
|
| 4 |
+
62801
|
| 5 |
+
]
|
| 6 |
+
],
|
| 7 |
+
"bos_token_id": 0,
|
| 8 |
+
"decoder_start_token_id": 62801,
|
| 9 |
+
"eos_token_id": 0,
|
| 10 |
+
"forced_eos_token_id": 0,
|
| 11 |
+
"max_length": 512,
|
| 12 |
+
"num_beams": 4,
|
| 13 |
+
"pad_token_id": 62801,
|
| 14 |
+
"renormalize_logits": true,
|
| 15 |
+
"transformers_version": "4.42.4"
|
| 16 |
+
}
|
checkpoint-1610/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a99b0ef8c6daedf0454a3dd267ab56891cf738aeefabbaacac0846944621c1f7
|
| 3 |
+
size 305452744
|
checkpoint-1610/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ca022c729816f000bd5682f71db79dfae70a5c6b83619da050c104153885a26f
|
| 3 |
+
size 610554131
|
checkpoint-1610/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:be1831c2d877434e2ca71aa111ed2cd0d4dae185bb839542f315a1c9b77b1470
|
| 3 |
+
size 14503
|
checkpoint-1610/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2acb05a53fc1d87f27dded8dd2e8510a49ccdf0200c96a7fe23969533b3977f7
|
| 3 |
+
size 623
|
checkpoint-1610/source.spm
ADDED
|
Binary file (801 kB). View file
|
|
|
checkpoint-1610/special_tokens_map.json
ADDED
|
@@ -0,0 +1,5 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"eos_token": "</s>",
|
| 3 |
+
"pad_token": "<pad>",
|
| 4 |
+
"unk_token": "<unk>"
|
| 5 |
+
}
|
checkpoint-1610/target.spm
ADDED
|
Binary file (917 kB). View file
|
|
|
checkpoint-1610/tokenizer_config.json
ADDED
|
@@ -0,0 +1,39 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"added_tokens_decoder": {
|
| 3 |
+
"0": {
|
| 4 |
+
"content": "</s>",
|
| 5 |
+
"lstrip": false,
|
| 6 |
+
"normalized": false,
|
| 7 |
+
"rstrip": false,
|
| 8 |
+
"single_word": false,
|
| 9 |
+
"special": true
|
| 10 |
+
},
|
| 11 |
+
"1": {
|
| 12 |
+
"content": "<unk>",
|
| 13 |
+
"lstrip": false,
|
| 14 |
+
"normalized": false,
|
| 15 |
+
"rstrip": false,
|
| 16 |
+
"single_word": false,
|
| 17 |
+
"special": true
|
| 18 |
+
},
|
| 19 |
+
"62801": {
|
| 20 |
+
"content": "<pad>",
|
| 21 |
+
"lstrip": false,
|
| 22 |
+
"normalized": false,
|
| 23 |
+
"rstrip": false,
|
| 24 |
+
"single_word": false,
|
| 25 |
+
"special": true
|
| 26 |
+
}
|
| 27 |
+
},
|
| 28 |
+
"clean_up_tokenization_spaces": true,
|
| 29 |
+
"eos_token": "</s>",
|
| 30 |
+
"model_max_length": 512,
|
| 31 |
+
"pad_token": "<pad>",
|
| 32 |
+
"return_tensors": "pt",
|
| 33 |
+
"separate_vocabs": false,
|
| 34 |
+
"source_lang": "eng",
|
| 35 |
+
"sp_model_kwargs": {},
|
| 36 |
+
"target_lang": "ara",
|
| 37 |
+
"tokenizer_class": "MarianTokenizer",
|
| 38 |
+
"unk_token": "<unk>"
|
| 39 |
+
}
|
checkpoint-1610/trainer_state.json
ADDED
|
@@ -0,0 +1,161 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.03832539543509483,
|
| 3 |
+
"best_model_checkpoint": "text2gloss_ar/checkpoint-1380",
|
| 4 |
+
"epoch": 7.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 1610,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 1.0,
|
| 13 |
+
"grad_norm": 9.208166122436523,
|
| 14 |
+
"learning_rate": 1.933623188405797e-05,
|
| 15 |
+
"loss": 2.726,
|
| 16 |
+
"step": 230
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 1.0,
|
| 20 |
+
"eval_char_bleu": 42.04703354604912,
|
| 21 |
+
"eval_loss": 0.8206332921981812,
|
| 22 |
+
"eval_runtime": 31.4529,
|
| 23 |
+
"eval_samples_per_second": 41.236,
|
| 24 |
+
"eval_steps_per_second": 0.668,
|
| 25 |
+
"eval_word_bleu": 24.85614940942985,
|
| 26 |
+
"step": 230
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 2.0,
|
| 30 |
+
"grad_norm": 6.826977729797363,
|
| 31 |
+
"learning_rate": 1.8669565217391307e-05,
|
| 32 |
+
"loss": 0.6983,
|
| 33 |
+
"step": 460
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 2.0,
|
| 37 |
+
"eval_char_bleu": 74.73746769123755,
|
| 38 |
+
"eval_loss": 0.31659388542175293,
|
| 39 |
+
"eval_runtime": 30.2056,
|
| 40 |
+
"eval_samples_per_second": 42.939,
|
| 41 |
+
"eval_steps_per_second": 0.695,
|
| 42 |
+
"eval_word_bleu": 61.86433916154899,
|
| 43 |
+
"step": 460
|
| 44 |
+
},
|
| 45 |
+
{
|
| 46 |
+
"epoch": 3.0,
|
| 47 |
+
"grad_norm": 5.772645473480225,
|
| 48 |
+
"learning_rate": 1.800289855072464e-05,
|
| 49 |
+
"loss": 0.3167,
|
| 50 |
+
"step": 690
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 3.0,
|
| 54 |
+
"eval_char_bleu": 92.15392118666509,
|
| 55 |
+
"eval_loss": 0.12875397503376007,
|
| 56 |
+
"eval_runtime": 29.9006,
|
| 57 |
+
"eval_samples_per_second": 43.377,
|
| 58 |
+
"eval_steps_per_second": 0.702,
|
| 59 |
+
"eval_word_bleu": 85.4787040239448,
|
| 60 |
+
"step": 690
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 4.0,
|
| 64 |
+
"grad_norm": 5.245423793792725,
|
| 65 |
+
"learning_rate": 1.7336231884057974e-05,
|
| 66 |
+
"loss": 0.1599,
|
| 67 |
+
"step": 920
|
| 68 |
+
},
|
| 69 |
+
{
|
| 70 |
+
"epoch": 4.0,
|
| 71 |
+
"eval_char_bleu": 97.20198228076768,
|
| 72 |
+
"eval_loss": 0.06993339955806732,
|
| 73 |
+
"eval_runtime": 29.0824,
|
| 74 |
+
"eval_samples_per_second": 44.597,
|
| 75 |
+
"eval_steps_per_second": 0.722,
|
| 76 |
+
"eval_word_bleu": 92.92866546086414,
|
| 77 |
+
"step": 920
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 5.0,
|
| 81 |
+
"grad_norm": 4.459423542022705,
|
| 82 |
+
"learning_rate": 1.6669565217391305e-05,
|
| 83 |
+
"loss": 0.0971,
|
| 84 |
+
"step": 1150
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 5.0,
|
| 88 |
+
"eval_char_bleu": 97.69668982414296,
|
| 89 |
+
"eval_loss": 0.05042978748679161,
|
| 90 |
+
"eval_runtime": 29.4348,
|
| 91 |
+
"eval_samples_per_second": 44.064,
|
| 92 |
+
"eval_steps_per_second": 0.713,
|
| 93 |
+
"eval_word_bleu": 94.6364324777254,
|
| 94 |
+
"step": 1150
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 6.0,
|
| 98 |
+
"grad_norm": 2.5048043727874756,
|
| 99 |
+
"learning_rate": 1.600289855072464e-05,
|
| 100 |
+
"loss": 0.0626,
|
| 101 |
+
"step": 1380
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 6.0,
|
| 105 |
+
"eval_char_bleu": 98.60004486244999,
|
| 106 |
+
"eval_loss": 0.03832539543509483,
|
| 107 |
+
"eval_runtime": 29.4835,
|
| 108 |
+
"eval_samples_per_second": 43.991,
|
| 109 |
+
"eval_steps_per_second": 0.712,
|
| 110 |
+
"eval_word_bleu": 96.34412173641856,
|
| 111 |
+
"step": 1380
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 7.0,
|
| 115 |
+
"grad_norm": 2.593268871307373,
|
| 116 |
+
"learning_rate": 1.5336231884057972e-05,
|
| 117 |
+
"loss": 0.0507,
|
| 118 |
+
"step": 1610
|
| 119 |
+
},
|
| 120 |
+
{
|
| 121 |
+
"epoch": 7.0,
|
| 122 |
+
"eval_char_bleu": 98.50276122630902,
|
| 123 |
+
"eval_loss": 0.03961858153343201,
|
| 124 |
+
"eval_runtime": 30.4579,
|
| 125 |
+
"eval_samples_per_second": 42.583,
|
| 126 |
+
"eval_steps_per_second": 0.689,
|
| 127 |
+
"eval_word_bleu": 95.94399159765237,
|
| 128 |
+
"step": 1610
|
| 129 |
+
}
|
| 130 |
+
],
|
| 131 |
+
"logging_steps": 500,
|
| 132 |
+
"max_steps": 6900,
|
| 133 |
+
"num_input_tokens_seen": 0,
|
| 134 |
+
"num_train_epochs": 30,
|
| 135 |
+
"save_steps": 500,
|
| 136 |
+
"stateful_callbacks": {
|
| 137 |
+
"EarlyStoppingCallback": {
|
| 138 |
+
"args": {
|
| 139 |
+
"early_stopping_patience": 3,
|
| 140 |
+
"early_stopping_threshold": 0.01
|
| 141 |
+
},
|
| 142 |
+
"attributes": {
|
| 143 |
+
"early_stopping_patience_counter": 0
|
| 144 |
+
}
|
| 145 |
+
},
|
| 146 |
+
"TrainerControl": {
|
| 147 |
+
"args": {
|
| 148 |
+
"should_epoch_stop": false,
|
| 149 |
+
"should_evaluate": false,
|
| 150 |
+
"should_log": false,
|
| 151 |
+
"should_save": true,
|
| 152 |
+
"should_training_stop": false
|
| 153 |
+
},
|
| 154 |
+
"attributes": {}
|
| 155 |
+
}
|
| 156 |
+
},
|
| 157 |
+
"total_flos": 760319933349888.0,
|
| 158 |
+
"train_batch_size": 32,
|
| 159 |
+
"trial_name": null,
|
| 160 |
+
"trial_params": null
|
| 161 |
+
}
|
checkpoint-1610/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a34aef40641c75ceb07f5ee7c8c3d1673038476484d7d3f2997edc701fec171a
|
| 3 |
+
size 4783
|
checkpoint-1610/vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-1840/config.json
ADDED
|
@@ -0,0 +1,61 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "Helsinki-NLP/opus-mt-en-ar",
|
| 3 |
+
"activation_dropout": 0.0,
|
| 4 |
+
"activation_function": "swish",
|
| 5 |
+
"add_bias_logits": false,
|
| 6 |
+
"add_final_layer_norm": false,
|
| 7 |
+
"architectures": [
|
| 8 |
+
"MarianMTModel"
|
| 9 |
+
],
|
| 10 |
+
"attention_dropout": 0.0,
|
| 11 |
+
"bad_words_ids": [
|
| 12 |
+
[
|
| 13 |
+
62801
|
| 14 |
+
]
|
| 15 |
+
],
|
| 16 |
+
"bos_token_id": 0,
|
| 17 |
+
"classif_dropout": 0.0,
|
| 18 |
+
"classifier_dropout": 0.0,
|
| 19 |
+
"d_model": 512,
|
| 20 |
+
"decoder_attention_heads": 8,
|
| 21 |
+
"decoder_ffn_dim": 2048,
|
| 22 |
+
"decoder_layerdrop": 0.0,
|
| 23 |
+
"decoder_layers": 6,
|
| 24 |
+
"decoder_start_token_id": 62801,
|
| 25 |
+
"decoder_vocab_size": 62802,
|
| 26 |
+
"dropout": 0.1,
|
| 27 |
+
"encoder_attention_heads": 8,
|
| 28 |
+
"encoder_ffn_dim": 2048,
|
| 29 |
+
"encoder_layerdrop": 0.0,
|
| 30 |
+
"encoder_layers": 6,
|
| 31 |
+
"eos_token_id": 0,
|
| 32 |
+
"extra_pos_embeddings": 62802,
|
| 33 |
+
"forced_eos_token_id": 0,
|
| 34 |
+
"id2label": {
|
| 35 |
+
"0": "LABEL_0",
|
| 36 |
+
"1": "LABEL_1",
|
| 37 |
+
"2": "LABEL_2"
|
| 38 |
+
},
|
| 39 |
+
"init_std": 0.02,
|
| 40 |
+
"is_encoder_decoder": true,
|
| 41 |
+
"label2id": {
|
| 42 |
+
"LABEL_0": 0,
|
| 43 |
+
"LABEL_1": 1,
|
| 44 |
+
"LABEL_2": 2
|
| 45 |
+
},
|
| 46 |
+
"max_length": 512,
|
| 47 |
+
"max_position_embeddings": 512,
|
| 48 |
+
"model_type": "marian",
|
| 49 |
+
"normalize_before": false,
|
| 50 |
+
"normalize_embedding": false,
|
| 51 |
+
"num_beams": 4,
|
| 52 |
+
"num_hidden_layers": 6,
|
| 53 |
+
"pad_token_id": 62801,
|
| 54 |
+
"scale_embedding": true,
|
| 55 |
+
"share_encoder_decoder_embeddings": true,
|
| 56 |
+
"static_position_embeddings": true,
|
| 57 |
+
"torch_dtype": "float32",
|
| 58 |
+
"transformers_version": "4.42.4",
|
| 59 |
+
"use_cache": true,
|
| 60 |
+
"vocab_size": 62802
|
| 61 |
+
}
|
checkpoint-1840/generation_config.json
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bad_words_ids": [
|
| 3 |
+
[
|
| 4 |
+
62801
|
| 5 |
+
]
|
| 6 |
+
],
|
| 7 |
+
"bos_token_id": 0,
|
| 8 |
+
"decoder_start_token_id": 62801,
|
| 9 |
+
"eos_token_id": 0,
|
| 10 |
+
"forced_eos_token_id": 0,
|
| 11 |
+
"max_length": 512,
|
| 12 |
+
"num_beams": 4,
|
| 13 |
+
"pad_token_id": 62801,
|
| 14 |
+
"renormalize_logits": true,
|
| 15 |
+
"transformers_version": "4.42.4"
|
| 16 |
+
}
|
checkpoint-1840/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a8c2156b6a0e74966574eca738ad4d678bd632128e17ee635740721088c424ec
|
| 3 |
+
size 305452744
|
checkpoint-1840/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2259dda591a3813adc3ad1bce99419ee83ed7c326636215ff205e82fe415990d
|
| 3 |
+
size 610554131
|
checkpoint-1840/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3cbe311e94d9e3c283c82dbfd0ce147207e41af6a435c86141fc1de0ab7aaa29
|
| 3 |
+
size 14503
|
checkpoint-1840/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:675fbf6145bcb29a1206d6d2480b10212fcfd60c69f02f75740831a8974aa74a
|
| 3 |
+
size 623
|
checkpoint-1840/source.spm
ADDED
|
Binary file (801 kB). View file
|
|
|
checkpoint-1840/special_tokens_map.json
ADDED
|
@@ -0,0 +1,5 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"eos_token": "</s>",
|
| 3 |
+
"pad_token": "<pad>",
|
| 4 |
+
"unk_token": "<unk>"
|
| 5 |
+
}
|
checkpoint-1840/target.spm
ADDED
|
Binary file (917 kB). View file
|
|
|
checkpoint-1840/tokenizer_config.json
ADDED
|
@@ -0,0 +1,39 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"added_tokens_decoder": {
|
| 3 |
+
"0": {
|
| 4 |
+
"content": "</s>",
|
| 5 |
+
"lstrip": false,
|
| 6 |
+
"normalized": false,
|
| 7 |
+
"rstrip": false,
|
| 8 |
+
"single_word": false,
|
| 9 |
+
"special": true
|
| 10 |
+
},
|
| 11 |
+
"1": {
|
| 12 |
+
"content": "<unk>",
|
| 13 |
+
"lstrip": false,
|
| 14 |
+
"normalized": false,
|
| 15 |
+
"rstrip": false,
|
| 16 |
+
"single_word": false,
|
| 17 |
+
"special": true
|
| 18 |
+
},
|
| 19 |
+
"62801": {
|
| 20 |
+
"content": "<pad>",
|
| 21 |
+
"lstrip": false,
|
| 22 |
+
"normalized": false,
|
| 23 |
+
"rstrip": false,
|
| 24 |
+
"single_word": false,
|
| 25 |
+
"special": true
|
| 26 |
+
}
|
| 27 |
+
},
|
| 28 |
+
"clean_up_tokenization_spaces": true,
|
| 29 |
+
"eos_token": "</s>",
|
| 30 |
+
"model_max_length": 512,
|
| 31 |
+
"pad_token": "<pad>",
|
| 32 |
+
"return_tensors": "pt",
|
| 33 |
+
"separate_vocabs": false,
|
| 34 |
+
"source_lang": "eng",
|
| 35 |
+
"sp_model_kwargs": {},
|
| 36 |
+
"target_lang": "ara",
|
| 37 |
+
"tokenizer_class": "MarianTokenizer",
|
| 38 |
+
"unk_token": "<unk>"
|
| 39 |
+
}
|
checkpoint-1840/trainer_state.json
ADDED
|
@@ -0,0 +1,178 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.03635687381029129,
|
| 3 |
+
"best_model_checkpoint": "text2gloss_ar/checkpoint-1840",
|
| 4 |
+
"epoch": 8.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 1840,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 1.0,
|
| 13 |
+
"grad_norm": 9.208166122436523,
|
| 14 |
+
"learning_rate": 1.933623188405797e-05,
|
| 15 |
+
"loss": 2.726,
|
| 16 |
+
"step": 230
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 1.0,
|
| 20 |
+
"eval_char_bleu": 42.04703354604912,
|
| 21 |
+
"eval_loss": 0.8206332921981812,
|
| 22 |
+
"eval_runtime": 31.4529,
|
| 23 |
+
"eval_samples_per_second": 41.236,
|
| 24 |
+
"eval_steps_per_second": 0.668,
|
| 25 |
+
"eval_word_bleu": 24.85614940942985,
|
| 26 |
+
"step": 230
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 2.0,
|
| 30 |
+
"grad_norm": 6.826977729797363,
|
| 31 |
+
"learning_rate": 1.8669565217391307e-05,
|
| 32 |
+
"loss": 0.6983,
|
| 33 |
+
"step": 460
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 2.0,
|
| 37 |
+
"eval_char_bleu": 74.73746769123755,
|
| 38 |
+
"eval_loss": 0.31659388542175293,
|
| 39 |
+
"eval_runtime": 30.2056,
|
| 40 |
+
"eval_samples_per_second": 42.939,
|
| 41 |
+
"eval_steps_per_second": 0.695,
|
| 42 |
+
"eval_word_bleu": 61.86433916154899,
|
| 43 |
+
"step": 460
|
| 44 |
+
},
|
| 45 |
+
{
|
| 46 |
+
"epoch": 3.0,
|
| 47 |
+
"grad_norm": 5.772645473480225,
|
| 48 |
+
"learning_rate": 1.800289855072464e-05,
|
| 49 |
+
"loss": 0.3167,
|
| 50 |
+
"step": 690
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 3.0,
|
| 54 |
+
"eval_char_bleu": 92.15392118666509,
|
| 55 |
+
"eval_loss": 0.12875397503376007,
|
| 56 |
+
"eval_runtime": 29.9006,
|
| 57 |
+
"eval_samples_per_second": 43.377,
|
| 58 |
+
"eval_steps_per_second": 0.702,
|
| 59 |
+
"eval_word_bleu": 85.4787040239448,
|
| 60 |
+
"step": 690
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 4.0,
|
| 64 |
+
"grad_norm": 5.245423793792725,
|
| 65 |
+
"learning_rate": 1.7336231884057974e-05,
|
| 66 |
+
"loss": 0.1599,
|
| 67 |
+
"step": 920
|
| 68 |
+
},
|
| 69 |
+
{
|
| 70 |
+
"epoch": 4.0,
|
| 71 |
+
"eval_char_bleu": 97.20198228076768,
|
| 72 |
+
"eval_loss": 0.06993339955806732,
|
| 73 |
+
"eval_runtime": 29.0824,
|
| 74 |
+
"eval_samples_per_second": 44.597,
|
| 75 |
+
"eval_steps_per_second": 0.722,
|
| 76 |
+
"eval_word_bleu": 92.92866546086414,
|
| 77 |
+
"step": 920
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 5.0,
|
| 81 |
+
"grad_norm": 4.459423542022705,
|
| 82 |
+
"learning_rate": 1.6669565217391305e-05,
|
| 83 |
+
"loss": 0.0971,
|
| 84 |
+
"step": 1150
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 5.0,
|
| 88 |
+
"eval_char_bleu": 97.69668982414296,
|
| 89 |
+
"eval_loss": 0.05042978748679161,
|
| 90 |
+
"eval_runtime": 29.4348,
|
| 91 |
+
"eval_samples_per_second": 44.064,
|
| 92 |
+
"eval_steps_per_second": 0.713,
|
| 93 |
+
"eval_word_bleu": 94.6364324777254,
|
| 94 |
+
"step": 1150
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 6.0,
|
| 98 |
+
"grad_norm": 2.5048043727874756,
|
| 99 |
+
"learning_rate": 1.600289855072464e-05,
|
| 100 |
+
"loss": 0.0626,
|
| 101 |
+
"step": 1380
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 6.0,
|
| 105 |
+
"eval_char_bleu": 98.60004486244999,
|
| 106 |
+
"eval_loss": 0.03832539543509483,
|
| 107 |
+
"eval_runtime": 29.4835,
|
| 108 |
+
"eval_samples_per_second": 43.991,
|
| 109 |
+
"eval_steps_per_second": 0.712,
|
| 110 |
+
"eval_word_bleu": 96.34412173641856,
|
| 111 |
+
"step": 1380
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 7.0,
|
| 115 |
+
"grad_norm": 2.593268871307373,
|
| 116 |
+
"learning_rate": 1.5336231884057972e-05,
|
| 117 |
+
"loss": 0.0507,
|
| 118 |
+
"step": 1610
|
| 119 |
+
},
|
| 120 |
+
{
|
| 121 |
+
"epoch": 7.0,
|
| 122 |
+
"eval_char_bleu": 98.50276122630902,
|
| 123 |
+
"eval_loss": 0.03961858153343201,
|
| 124 |
+
"eval_runtime": 30.4579,
|
| 125 |
+
"eval_samples_per_second": 42.583,
|
| 126 |
+
"eval_steps_per_second": 0.689,
|
| 127 |
+
"eval_word_bleu": 95.94399159765237,
|
| 128 |
+
"step": 1610
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 8.0,
|
| 132 |
+
"grad_norm": 2.4942216873168945,
|
| 133 |
+
"learning_rate": 1.4669565217391306e-05,
|
| 134 |
+
"loss": 0.036,
|
| 135 |
+
"step": 1840
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 8.0,
|
| 139 |
+
"eval_char_bleu": 98.39569733809864,
|
| 140 |
+
"eval_loss": 0.03635687381029129,
|
| 141 |
+
"eval_runtime": 30.2971,
|
| 142 |
+
"eval_samples_per_second": 42.809,
|
| 143 |
+
"eval_steps_per_second": 0.693,
|
| 144 |
+
"eval_word_bleu": 96.0035686359632,
|
| 145 |
+
"step": 1840
|
| 146 |
+
}
|
| 147 |
+
],
|
| 148 |
+
"logging_steps": 500,
|
| 149 |
+
"max_steps": 6900,
|
| 150 |
+
"num_input_tokens_seen": 0,
|
| 151 |
+
"num_train_epochs": 30,
|
| 152 |
+
"save_steps": 500,
|
| 153 |
+
"stateful_callbacks": {
|
| 154 |
+
"EarlyStoppingCallback": {
|
| 155 |
+
"args": {
|
| 156 |
+
"early_stopping_patience": 3,
|
| 157 |
+
"early_stopping_threshold": 0.01
|
| 158 |
+
},
|
| 159 |
+
"attributes": {
|
| 160 |
+
"early_stopping_patience_counter": 0
|
| 161 |
+
}
|
| 162 |
+
},
|
| 163 |
+
"TrainerControl": {
|
| 164 |
+
"args": {
|
| 165 |
+
"should_epoch_stop": false,
|
| 166 |
+
"should_evaluate": false,
|
| 167 |
+
"should_log": false,
|
| 168 |
+
"should_save": true,
|
| 169 |
+
"should_training_stop": false
|
| 170 |
+
},
|
| 171 |
+
"attributes": {}
|
| 172 |
+
}
|
| 173 |
+
},
|
| 174 |
+
"total_flos": 869236968849408.0,
|
| 175 |
+
"train_batch_size": 32,
|
| 176 |
+
"trial_name": null,
|
| 177 |
+
"trial_params": null
|
| 178 |
+
}
|
checkpoint-1840/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a34aef40641c75ceb07f5ee7c8c3d1673038476484d7d3f2997edc701fec171a
|
| 3 |
+
size 4783
|
checkpoint-1840/vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-2070/config.json
ADDED
|
@@ -0,0 +1,61 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "Helsinki-NLP/opus-mt-en-ar",
|
| 3 |
+
"activation_dropout": 0.0,
|
| 4 |
+
"activation_function": "swish",
|
| 5 |
+
"add_bias_logits": false,
|
| 6 |
+
"add_final_layer_norm": false,
|
| 7 |
+
"architectures": [
|
| 8 |
+
"MarianMTModel"
|
| 9 |
+
],
|
| 10 |
+
"attention_dropout": 0.0,
|
| 11 |
+
"bad_words_ids": [
|
| 12 |
+
[
|
| 13 |
+
62801
|
| 14 |
+
]
|
| 15 |
+
],
|
| 16 |
+
"bos_token_id": 0,
|
| 17 |
+
"classif_dropout": 0.0,
|
| 18 |
+
"classifier_dropout": 0.0,
|
| 19 |
+
"d_model": 512,
|
| 20 |
+
"decoder_attention_heads": 8,
|
| 21 |
+
"decoder_ffn_dim": 2048,
|
| 22 |
+
"decoder_layerdrop": 0.0,
|
| 23 |
+
"decoder_layers": 6,
|
| 24 |
+
"decoder_start_token_id": 62801,
|
| 25 |
+
"decoder_vocab_size": 62802,
|
| 26 |
+
"dropout": 0.1,
|
| 27 |
+
"encoder_attention_heads": 8,
|
| 28 |
+
"encoder_ffn_dim": 2048,
|
| 29 |
+
"encoder_layerdrop": 0.0,
|
| 30 |
+
"encoder_layers": 6,
|
| 31 |
+
"eos_token_id": 0,
|
| 32 |
+
"extra_pos_embeddings": 62802,
|
| 33 |
+
"forced_eos_token_id": 0,
|
| 34 |
+
"id2label": {
|
| 35 |
+
"0": "LABEL_0",
|
| 36 |
+
"1": "LABEL_1",
|
| 37 |
+
"2": "LABEL_2"
|
| 38 |
+
},
|
| 39 |
+
"init_std": 0.02,
|
| 40 |
+
"is_encoder_decoder": true,
|
| 41 |
+
"label2id": {
|
| 42 |
+
"LABEL_0": 0,
|
| 43 |
+
"LABEL_1": 1,
|
| 44 |
+
"LABEL_2": 2
|
| 45 |
+
},
|
| 46 |
+
"max_length": 512,
|
| 47 |
+
"max_position_embeddings": 512,
|
| 48 |
+
"model_type": "marian",
|
| 49 |
+
"normalize_before": false,
|
| 50 |
+
"normalize_embedding": false,
|
| 51 |
+
"num_beams": 4,
|
| 52 |
+
"num_hidden_layers": 6,
|
| 53 |
+
"pad_token_id": 62801,
|
| 54 |
+
"scale_embedding": true,
|
| 55 |
+
"share_encoder_decoder_embeddings": true,
|
| 56 |
+
"static_position_embeddings": true,
|
| 57 |
+
"torch_dtype": "float32",
|
| 58 |
+
"transformers_version": "4.42.4",
|
| 59 |
+
"use_cache": true,
|
| 60 |
+
"vocab_size": 62802
|
| 61 |
+
}
|
checkpoint-2070/generation_config.json
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bad_words_ids": [
|
| 3 |
+
[
|
| 4 |
+
62801
|
| 5 |
+
]
|
| 6 |
+
],
|
| 7 |
+
"bos_token_id": 0,
|
| 8 |
+
"decoder_start_token_id": 62801,
|
| 9 |
+
"eos_token_id": 0,
|
| 10 |
+
"forced_eos_token_id": 0,
|
| 11 |
+
"max_length": 512,
|
| 12 |
+
"num_beams": 4,
|
| 13 |
+
"pad_token_id": 62801,
|
| 14 |
+
"renormalize_logits": true,
|
| 15 |
+
"transformers_version": "4.42.4"
|
| 16 |
+
}
|
checkpoint-2070/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7bfa91dfe245c06b42617223e922f3501645b8678fe3af6c60deb4a8117ae6f0
|
| 3 |
+
size 305452744
|
checkpoint-2070/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:78928934974ccebb2c100a886238db461e4def7bfaa590cd5bed738de43515ab
|
| 3 |
+
size 610554131
|
checkpoint-2070/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:df43f029a2fc1bb14a3582521ba694e61c93eadce6b8cc0cdbad2ff5763bdb31
|
| 3 |
+
size 14503
|
checkpoint-2070/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:edfd388feec0549579c96bd9e3714594160145d7a78ae750dea5c170bcad7332
|
| 3 |
+
size 623
|
checkpoint-2070/source.spm
ADDED
|
Binary file (801 kB). View file
|
|
|
checkpoint-2070/special_tokens_map.json
ADDED
|
@@ -0,0 +1,5 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"eos_token": "</s>",
|
| 3 |
+
"pad_token": "<pad>",
|
| 4 |
+
"unk_token": "<unk>"
|
| 5 |
+
}
|
checkpoint-2070/target.spm
ADDED
|
Binary file (917 kB). View file
|
|
|
checkpoint-2070/tokenizer_config.json
ADDED
|
@@ -0,0 +1,39 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"added_tokens_decoder": {
|
| 3 |
+
"0": {
|
| 4 |
+
"content": "</s>",
|
| 5 |
+
"lstrip": false,
|
| 6 |
+
"normalized": false,
|
| 7 |
+
"rstrip": false,
|
| 8 |
+
"single_word": false,
|
| 9 |
+
"special": true
|
| 10 |
+
},
|
| 11 |
+
"1": {
|
| 12 |
+
"content": "<unk>",
|
| 13 |
+
"lstrip": false,
|
| 14 |
+
"normalized": false,
|
| 15 |
+
"rstrip": false,
|
| 16 |
+
"single_word": false,
|
| 17 |
+
"special": true
|
| 18 |
+
},
|
| 19 |
+
"62801": {
|
| 20 |
+
"content": "<pad>",
|
| 21 |
+
"lstrip": false,
|
| 22 |
+
"normalized": false,
|
| 23 |
+
"rstrip": false,
|
| 24 |
+
"single_word": false,
|
| 25 |
+
"special": true
|
| 26 |
+
}
|
| 27 |
+
},
|
| 28 |
+
"clean_up_tokenization_spaces": true,
|
| 29 |
+
"eos_token": "</s>",
|
| 30 |
+
"model_max_length": 512,
|
| 31 |
+
"pad_token": "<pad>",
|
| 32 |
+
"return_tensors": "pt",
|
| 33 |
+
"separate_vocabs": false,
|
| 34 |
+
"source_lang": "eng",
|
| 35 |
+
"sp_model_kwargs": {},
|
| 36 |
+
"target_lang": "ara",
|
| 37 |
+
"tokenizer_class": "MarianTokenizer",
|
| 38 |
+
"unk_token": "<unk>"
|
| 39 |
+
}
|
checkpoint-2070/trainer_state.json
ADDED
|
@@ -0,0 +1,195 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.030597062781453133,
|
| 3 |
+
"best_model_checkpoint": "text2gloss_ar/checkpoint-2070",
|
| 4 |
+
"epoch": 9.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 2070,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 1.0,
|
| 13 |
+
"grad_norm": 9.208166122436523,
|
| 14 |
+
"learning_rate": 1.933623188405797e-05,
|
| 15 |
+
"loss": 2.726,
|
| 16 |
+
"step": 230
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 1.0,
|
| 20 |
+
"eval_char_bleu": 42.04703354604912,
|
| 21 |
+
"eval_loss": 0.8206332921981812,
|
| 22 |
+
"eval_runtime": 31.4529,
|
| 23 |
+
"eval_samples_per_second": 41.236,
|
| 24 |
+
"eval_steps_per_second": 0.668,
|
| 25 |
+
"eval_word_bleu": 24.85614940942985,
|
| 26 |
+
"step": 230
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 2.0,
|
| 30 |
+
"grad_norm": 6.826977729797363,
|
| 31 |
+
"learning_rate": 1.8669565217391307e-05,
|
| 32 |
+
"loss": 0.6983,
|
| 33 |
+
"step": 460
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 2.0,
|
| 37 |
+
"eval_char_bleu": 74.73746769123755,
|
| 38 |
+
"eval_loss": 0.31659388542175293,
|
| 39 |
+
"eval_runtime": 30.2056,
|
| 40 |
+
"eval_samples_per_second": 42.939,
|
| 41 |
+
"eval_steps_per_second": 0.695,
|
| 42 |
+
"eval_word_bleu": 61.86433916154899,
|
| 43 |
+
"step": 460
|
| 44 |
+
},
|
| 45 |
+
{
|
| 46 |
+
"epoch": 3.0,
|
| 47 |
+
"grad_norm": 5.772645473480225,
|
| 48 |
+
"learning_rate": 1.800289855072464e-05,
|
| 49 |
+
"loss": 0.3167,
|
| 50 |
+
"step": 690
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 3.0,
|
| 54 |
+
"eval_char_bleu": 92.15392118666509,
|
| 55 |
+
"eval_loss": 0.12875397503376007,
|
| 56 |
+
"eval_runtime": 29.9006,
|
| 57 |
+
"eval_samples_per_second": 43.377,
|
| 58 |
+
"eval_steps_per_second": 0.702,
|
| 59 |
+
"eval_word_bleu": 85.4787040239448,
|
| 60 |
+
"step": 690
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 4.0,
|
| 64 |
+
"grad_norm": 5.245423793792725,
|
| 65 |
+
"learning_rate": 1.7336231884057974e-05,
|
| 66 |
+
"loss": 0.1599,
|
| 67 |
+
"step": 920
|
| 68 |
+
},
|
| 69 |
+
{
|
| 70 |
+
"epoch": 4.0,
|
| 71 |
+
"eval_char_bleu": 97.20198228076768,
|
| 72 |
+
"eval_loss": 0.06993339955806732,
|
| 73 |
+
"eval_runtime": 29.0824,
|
| 74 |
+
"eval_samples_per_second": 44.597,
|
| 75 |
+
"eval_steps_per_second": 0.722,
|
| 76 |
+
"eval_word_bleu": 92.92866546086414,
|
| 77 |
+
"step": 920
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 5.0,
|
| 81 |
+
"grad_norm": 4.459423542022705,
|
| 82 |
+
"learning_rate": 1.6669565217391305e-05,
|
| 83 |
+
"loss": 0.0971,
|
| 84 |
+
"step": 1150
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 5.0,
|
| 88 |
+
"eval_char_bleu": 97.69668982414296,
|
| 89 |
+
"eval_loss": 0.05042978748679161,
|
| 90 |
+
"eval_runtime": 29.4348,
|
| 91 |
+
"eval_samples_per_second": 44.064,
|
| 92 |
+
"eval_steps_per_second": 0.713,
|
| 93 |
+
"eval_word_bleu": 94.6364324777254,
|
| 94 |
+
"step": 1150
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 6.0,
|
| 98 |
+
"grad_norm": 2.5048043727874756,
|
| 99 |
+
"learning_rate": 1.600289855072464e-05,
|
| 100 |
+
"loss": 0.0626,
|
| 101 |
+
"step": 1380
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 6.0,
|
| 105 |
+
"eval_char_bleu": 98.60004486244999,
|
| 106 |
+
"eval_loss": 0.03832539543509483,
|
| 107 |
+
"eval_runtime": 29.4835,
|
| 108 |
+
"eval_samples_per_second": 43.991,
|
| 109 |
+
"eval_steps_per_second": 0.712,
|
| 110 |
+
"eval_word_bleu": 96.34412173641856,
|
| 111 |
+
"step": 1380
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 7.0,
|
| 115 |
+
"grad_norm": 2.593268871307373,
|
| 116 |
+
"learning_rate": 1.5336231884057972e-05,
|
| 117 |
+
"loss": 0.0507,
|
| 118 |
+
"step": 1610
|
| 119 |
+
},
|
| 120 |
+
{
|
| 121 |
+
"epoch": 7.0,
|
| 122 |
+
"eval_char_bleu": 98.50276122630902,
|
| 123 |
+
"eval_loss": 0.03961858153343201,
|
| 124 |
+
"eval_runtime": 30.4579,
|
| 125 |
+
"eval_samples_per_second": 42.583,
|
| 126 |
+
"eval_steps_per_second": 0.689,
|
| 127 |
+
"eval_word_bleu": 95.94399159765237,
|
| 128 |
+
"step": 1610
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 8.0,
|
| 132 |
+
"grad_norm": 2.4942216873168945,
|
| 133 |
+
"learning_rate": 1.4669565217391306e-05,
|
| 134 |
+
"loss": 0.036,
|
| 135 |
+
"step": 1840
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 8.0,
|
| 139 |
+
"eval_char_bleu": 98.39569733809864,
|
| 140 |
+
"eval_loss": 0.03635687381029129,
|
| 141 |
+
"eval_runtime": 30.2971,
|
| 142 |
+
"eval_samples_per_second": 42.809,
|
| 143 |
+
"eval_steps_per_second": 0.693,
|
| 144 |
+
"eval_word_bleu": 96.0035686359632,
|
| 145 |
+
"step": 1840
|
| 146 |
+
},
|
| 147 |
+
{
|
| 148 |
+
"epoch": 9.0,
|
| 149 |
+
"grad_norm": 3.2667160034179688,
|
| 150 |
+
"learning_rate": 1.400289855072464e-05,
|
| 151 |
+
"loss": 0.0289,
|
| 152 |
+
"step": 2070
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 9.0,
|
| 156 |
+
"eval_char_bleu": 98.93912433138856,
|
| 157 |
+
"eval_loss": 0.030597062781453133,
|
| 158 |
+
"eval_runtime": 30.2718,
|
| 159 |
+
"eval_samples_per_second": 42.845,
|
| 160 |
+
"eval_steps_per_second": 0.694,
|
| 161 |
+
"eval_word_bleu": 97.08305126864244,
|
| 162 |
+
"step": 2070
|
| 163 |
+
}
|
| 164 |
+
],
|
| 165 |
+
"logging_steps": 500,
|
| 166 |
+
"max_steps": 6900,
|
| 167 |
+
"num_input_tokens_seen": 0,
|
| 168 |
+
"num_train_epochs": 30,
|
| 169 |
+
"save_steps": 500,
|
| 170 |
+
"stateful_callbacks": {
|
| 171 |
+
"EarlyStoppingCallback": {
|
| 172 |
+
"args": {
|
| 173 |
+
"early_stopping_patience": 3,
|
| 174 |
+
"early_stopping_threshold": 0.01
|
| 175 |
+
},
|
| 176 |
+
"attributes": {
|
| 177 |
+
"early_stopping_patience_counter": 0
|
| 178 |
+
}
|
| 179 |
+
},
|
| 180 |
+
"TrainerControl": {
|
| 181 |
+
"args": {
|
| 182 |
+
"should_epoch_stop": false,
|
| 183 |
+
"should_evaluate": false,
|
| 184 |
+
"should_log": false,
|
| 185 |
+
"should_save": true,
|
| 186 |
+
"should_training_stop": true
|
| 187 |
+
},
|
| 188 |
+
"attributes": {}
|
| 189 |
+
}
|
| 190 |
+
},
|
| 191 |
+
"total_flos": 978040391860224.0,
|
| 192 |
+
"train_batch_size": 32,
|
| 193 |
+
"trial_name": null,
|
| 194 |
+
"trial_params": null
|
| 195 |
+
}
|
checkpoint-2070/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a34aef40641c75ceb07f5ee7c8c3d1673038476484d7d3f2997edc701fec171a
|
| 3 |
+
size 4783
|
checkpoint-2070/vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
test_metrics.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"test_loss": 0.28931236267089844, "test_word_bleu": 76.91704413218343, "test_char_bleu": 86.30489121528109, "test_runtime": 1.1051, "test_samples_per_second": 41.626, "test_steps_per_second": 0.905}
|