Training in progress, epoch 22
Browse filesThis view is limited to 50 files because it contains too many changes.
See raw diff
- checkpoint-1330/added_tokens.json +1 -0
- checkpoint-1330/bpe.codes +0 -0
- checkpoint-1330/config.json +30 -0
- checkpoint-1330/optimizer.pt +3 -0
- checkpoint-1330/pytorch_model.bin +3 -0
- checkpoint-1330/rng_state.pth +3 -0
- checkpoint-1330/scaler.pt +3 -0
- checkpoint-1330/scheduler.pt +3 -0
- checkpoint-1330/special_tokens_map.json +1 -0
- checkpoint-1330/tokenizer_config.json +1 -0
- checkpoint-1330/trainer_state.json +320 -0
- checkpoint-1330/training_args.bin +3 -0
- checkpoint-1330/vocab.txt +0 -0
- checkpoint-1400/added_tokens.json +1 -0
- checkpoint-1400/bpe.codes +0 -0
- checkpoint-1400/config.json +30 -0
- checkpoint-1400/optimizer.pt +3 -0
- checkpoint-1400/pytorch_model.bin +3 -0
- checkpoint-1400/rng_state.pth +3 -0
- checkpoint-1400/scaler.pt +3 -0
- checkpoint-1400/scheduler.pt +3 -0
- checkpoint-1400/special_tokens_map.json +1 -0
- checkpoint-1400/tokenizer_config.json +1 -0
- checkpoint-1400/trainer_state.json +336 -0
- checkpoint-1400/training_args.bin +3 -0
- checkpoint-1400/vocab.txt +0 -0
- checkpoint-1470/added_tokens.json +1 -0
- checkpoint-1470/bpe.codes +0 -0
- checkpoint-1470/config.json +30 -0
- checkpoint-1470/optimizer.pt +3 -0
- checkpoint-1470/pytorch_model.bin +3 -0
- checkpoint-1470/rng_state.pth +3 -0
- checkpoint-1470/scaler.pt +3 -0
- checkpoint-1470/scheduler.pt +3 -0
- checkpoint-1470/special_tokens_map.json +1 -0
- checkpoint-1470/tokenizer_config.json +1 -0
- checkpoint-1470/trainer_state.json +352 -0
- checkpoint-1470/training_args.bin +3 -0
- checkpoint-1470/vocab.txt +0 -0
- checkpoint-1540/added_tokens.json +1 -0
- checkpoint-1540/bpe.codes +0 -0
- checkpoint-1540/config.json +30 -0
- checkpoint-1540/optimizer.pt +3 -0
- checkpoint-1540/pytorch_model.bin +3 -0
- checkpoint-1540/rng_state.pth +3 -0
- checkpoint-1540/scaler.pt +3 -0
- checkpoint-1540/scheduler.pt +3 -0
- checkpoint-1540/special_tokens_map.json +1 -0
- checkpoint-1540/tokenizer_config.json +1 -0
- checkpoint-1540/trainer_state.json +368 -0
checkpoint-1330/added_tokens.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"<mask>": 64000}
|
checkpoint-1330/bpe.codes
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-1330/config.json
ADDED
|
@@ -0,0 +1,30 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "vinai/bertweet-covid19-base-cased",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"RobertaForSequenceClassification"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.1,
|
| 7 |
+
"bos_token_id": 0,
|
| 8 |
+
"classifier_dropout": null,
|
| 9 |
+
"eos_token_id": 2,
|
| 10 |
+
"gradient_checkpointing": false,
|
| 11 |
+
"hidden_act": "gelu",
|
| 12 |
+
"hidden_dropout_prob": 0.1,
|
| 13 |
+
"hidden_size": 768,
|
| 14 |
+
"initializer_range": 0.02,
|
| 15 |
+
"intermediate_size": 3072,
|
| 16 |
+
"layer_norm_eps": 1e-05,
|
| 17 |
+
"max_position_embeddings": 130,
|
| 18 |
+
"model_type": "roberta",
|
| 19 |
+
"num_attention_heads": 12,
|
| 20 |
+
"num_hidden_layers": 12,
|
| 21 |
+
"pad_token_id": 1,
|
| 22 |
+
"position_embedding_type": "absolute",
|
| 23 |
+
"problem_type": "multi_label_classification",
|
| 24 |
+
"tokenizer_class": "BertweetTokenizer",
|
| 25 |
+
"torch_dtype": "float32",
|
| 26 |
+
"transformers_version": "4.17.0",
|
| 27 |
+
"type_vocab_size": 1,
|
| 28 |
+
"use_cache": true,
|
| 29 |
+
"vocab_size": 64001
|
| 30 |
+
}
|
checkpoint-1330/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4d79fbfe759c3583a43abee91f65a2977718528bca5783c9c20a70b84555c1fd
|
| 3 |
+
size 1079327581
|
checkpoint-1330/pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:44deb2d58abf435d327415d352a0bfd94423225874db95e908814055e2b2d7c6
|
| 3 |
+
size 539689325
|
checkpoint-1330/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ccafcd94cd6d254de33fe8808d3c3a34351e682a026ad95a77bc150452387801
|
| 3 |
+
size 14503
|
checkpoint-1330/scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8dc3946004caddb6adba34c071a7d9792f32df7ba8c5b63f145906d49992e840
|
| 3 |
+
size 559
|
checkpoint-1330/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:af96afa518cd53faa03c73d98543e74de04e0b23d408c7d82c044c4b83da10f6
|
| 3 |
+
size 623
|
checkpoint-1330/special_tokens_map.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"bos_token": "<s>", "eos_token": "</s>", "unk_token": "<unk>", "sep_token": "</s>", "pad_token": "<pad>", "cls_token": "<s>", "mask_token": "<mask>"}
|
checkpoint-1330/tokenizer_config.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"normalization": true, "bos_token": "<s>", "eos_token": "</s>", "sep_token": "</s>", "cls_token": "<s>", "unk_token": "<unk>", "pad_token": "<pad>", "mask_token": "<mask>", "special_tokens_map_file": null, "name_or_path": "vinai/bertweet-covid19-base-cased", "tokenizer_class": "BertweetTokenizer"}
|
checkpoint-1330/trainer_state.json
ADDED
|
@@ -0,0 +1,320 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.7711201859949647,
|
| 3 |
+
"best_model_checkpoint": "outputs/bertweet/checkpoint-1260",
|
| 4 |
+
"epoch": 18.99290780141844,
|
| 5 |
+
"global_step": 1330,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.99,
|
| 12 |
+
"learning_rate": 9.666666666666666e-07,
|
| 13 |
+
"loss": 0.6925,
|
| 14 |
+
"step": 70
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.99,
|
| 18 |
+
"eval_accuracy": 0.573,
|
| 19 |
+
"eval_f1": 0.3642720915448188,
|
| 20 |
+
"eval_loss": 0.6826930046081543,
|
| 21 |
+
"eval_runtime": 2.9877,
|
| 22 |
+
"eval_samples_per_second": 334.703,
|
| 23 |
+
"eval_steps_per_second": 41.838,
|
| 24 |
+
"step": 70
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 1.99,
|
| 28 |
+
"learning_rate": 9.333333333333333e-07,
|
| 29 |
+
"loss": 0.6823,
|
| 30 |
+
"step": 140
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 1.99,
|
| 34 |
+
"eval_accuracy": 0.573,
|
| 35 |
+
"eval_f1": 0.3642720915448188,
|
| 36 |
+
"eval_loss": 0.6736124753952026,
|
| 37 |
+
"eval_runtime": 3.0497,
|
| 38 |
+
"eval_samples_per_second": 327.901,
|
| 39 |
+
"eval_steps_per_second": 40.988,
|
| 40 |
+
"step": 140
|
| 41 |
+
},
|
| 42 |
+
{
|
| 43 |
+
"epoch": 2.99,
|
| 44 |
+
"learning_rate": 9e-07,
|
| 45 |
+
"loss": 0.6713,
|
| 46 |
+
"step": 210
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 2.99,
|
| 50 |
+
"eval_accuracy": 0.587,
|
| 51 |
+
"eval_f1": 0.39928932873127865,
|
| 52 |
+
"eval_loss": 0.656767725944519,
|
| 53 |
+
"eval_runtime": 3.0012,
|
| 54 |
+
"eval_samples_per_second": 333.2,
|
| 55 |
+
"eval_steps_per_second": 41.65,
|
| 56 |
+
"step": 210
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 3.99,
|
| 60 |
+
"learning_rate": 8.666666666666667e-07,
|
| 61 |
+
"loss": 0.6468,
|
| 62 |
+
"step": 280
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 3.99,
|
| 66 |
+
"eval_accuracy": 0.7,
|
| 67 |
+
"eval_f1": 0.6707623826267894,
|
| 68 |
+
"eval_loss": 0.6210379600524902,
|
| 69 |
+
"eval_runtime": 3.0442,
|
| 70 |
+
"eval_samples_per_second": 328.492,
|
| 71 |
+
"eval_steps_per_second": 41.062,
|
| 72 |
+
"step": 280
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 4.99,
|
| 76 |
+
"learning_rate": 8.333333333333333e-07,
|
| 77 |
+
"loss": 0.6047,
|
| 78 |
+
"step": 350
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"epoch": 4.99,
|
| 82 |
+
"eval_accuracy": 0.732,
|
| 83 |
+
"eval_f1": 0.7285955019767969,
|
| 84 |
+
"eval_loss": 0.5784569382667542,
|
| 85 |
+
"eval_runtime": 3.0658,
|
| 86 |
+
"eval_samples_per_second": 326.18,
|
| 87 |
+
"eval_steps_per_second": 40.773,
|
| 88 |
+
"step": 350
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 5.99,
|
| 92 |
+
"learning_rate": 8e-07,
|
| 93 |
+
"loss": 0.5648,
|
| 94 |
+
"step": 420
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 5.99,
|
| 98 |
+
"eval_accuracy": 0.733,
|
| 99 |
+
"eval_f1": 0.7318671386608421,
|
| 100 |
+
"eval_loss": 0.5537055730819702,
|
| 101 |
+
"eval_runtime": 3.061,
|
| 102 |
+
"eval_samples_per_second": 326.692,
|
| 103 |
+
"eval_steps_per_second": 40.836,
|
| 104 |
+
"step": 420
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 6.99,
|
| 108 |
+
"learning_rate": 7.666666666666667e-07,
|
| 109 |
+
"loss": 0.536,
|
| 110 |
+
"step": 490
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 6.99,
|
| 114 |
+
"eval_accuracy": 0.739,
|
| 115 |
+
"eval_f1": 0.7381492469031885,
|
| 116 |
+
"eval_loss": 0.5406165719032288,
|
| 117 |
+
"eval_runtime": 3.3555,
|
| 118 |
+
"eval_samples_per_second": 298.016,
|
| 119 |
+
"eval_steps_per_second": 37.252,
|
| 120 |
+
"step": 490
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 7.99,
|
| 124 |
+
"learning_rate": 7.333333333333332e-07,
|
| 125 |
+
"loss": 0.5175,
|
| 126 |
+
"step": 560
|
| 127 |
+
},
|
| 128 |
+
{
|
| 129 |
+
"epoch": 7.99,
|
| 130 |
+
"eval_accuracy": 0.744,
|
| 131 |
+
"eval_f1": 0.7430750702529104,
|
| 132 |
+
"eval_loss": 0.5307690501213074,
|
| 133 |
+
"eval_runtime": 3.1364,
|
| 134 |
+
"eval_samples_per_second": 318.842,
|
| 135 |
+
"eval_steps_per_second": 39.855,
|
| 136 |
+
"step": 560
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 8.99,
|
| 140 |
+
"learning_rate": 7e-07,
|
| 141 |
+
"loss": 0.5018,
|
| 142 |
+
"step": 630
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 8.99,
|
| 146 |
+
"eval_accuracy": 0.751,
|
| 147 |
+
"eval_f1": 0.7504007121097755,
|
| 148 |
+
"eval_loss": 0.5235322117805481,
|
| 149 |
+
"eval_runtime": 2.9996,
|
| 150 |
+
"eval_samples_per_second": 333.377,
|
| 151 |
+
"eval_steps_per_second": 41.672,
|
| 152 |
+
"step": 630
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 9.99,
|
| 156 |
+
"learning_rate": 6.666666666666666e-07,
|
| 157 |
+
"loss": 0.4874,
|
| 158 |
+
"step": 700
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 9.99,
|
| 162 |
+
"eval_accuracy": 0.749,
|
| 163 |
+
"eval_f1": 0.7478681802611924,
|
| 164 |
+
"eval_loss": 0.5145103931427002,
|
| 165 |
+
"eval_runtime": 3.1795,
|
| 166 |
+
"eval_samples_per_second": 314.515,
|
| 167 |
+
"eval_steps_per_second": 39.314,
|
| 168 |
+
"step": 700
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 10.99,
|
| 172 |
+
"learning_rate": 6.333333333333332e-07,
|
| 173 |
+
"loss": 0.4749,
|
| 174 |
+
"step": 770
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 10.99,
|
| 178 |
+
"eval_accuracy": 0.754,
|
| 179 |
+
"eval_f1": 0.7533330124657072,
|
| 180 |
+
"eval_loss": 0.510367751121521,
|
| 181 |
+
"eval_runtime": 3.0347,
|
| 182 |
+
"eval_samples_per_second": 329.518,
|
| 183 |
+
"eval_steps_per_second": 41.19,
|
| 184 |
+
"step": 770
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 11.99,
|
| 188 |
+
"learning_rate": 6e-07,
|
| 189 |
+
"loss": 0.4666,
|
| 190 |
+
"step": 840
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 11.99,
|
| 194 |
+
"eval_accuracy": 0.761,
|
| 195 |
+
"eval_f1": 0.760515042961998,
|
| 196 |
+
"eval_loss": 0.5052358508110046,
|
| 197 |
+
"eval_runtime": 3.0155,
|
| 198 |
+
"eval_samples_per_second": 331.615,
|
| 199 |
+
"eval_steps_per_second": 41.452,
|
| 200 |
+
"step": 840
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 12.99,
|
| 204 |
+
"learning_rate": 5.666666666666666e-07,
|
| 205 |
+
"loss": 0.456,
|
| 206 |
+
"step": 910
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 12.99,
|
| 210 |
+
"eval_accuracy": 0.761,
|
| 211 |
+
"eval_f1": 0.7604247798965316,
|
| 212 |
+
"eval_loss": 0.5017005205154419,
|
| 213 |
+
"eval_runtime": 2.9976,
|
| 214 |
+
"eval_samples_per_second": 333.599,
|
| 215 |
+
"eval_steps_per_second": 41.7,
|
| 216 |
+
"step": 910
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 13.99,
|
| 220 |
+
"learning_rate": 5.333333333333333e-07,
|
| 221 |
+
"loss": 0.4489,
|
| 222 |
+
"step": 980
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"epoch": 13.99,
|
| 226 |
+
"eval_accuracy": 0.764,
|
| 227 |
+
"eval_f1": 0.7634995650797087,
|
| 228 |
+
"eval_loss": 0.49855634570121765,
|
| 229 |
+
"eval_runtime": 3.0658,
|
| 230 |
+
"eval_samples_per_second": 326.176,
|
| 231 |
+
"eval_steps_per_second": 40.772,
|
| 232 |
+
"step": 980
|
| 233 |
+
},
|
| 234 |
+
{
|
| 235 |
+
"epoch": 14.99,
|
| 236 |
+
"learning_rate": 5e-07,
|
| 237 |
+
"loss": 0.4375,
|
| 238 |
+
"step": 1050
|
| 239 |
+
},
|
| 240 |
+
{
|
| 241 |
+
"epoch": 14.99,
|
| 242 |
+
"eval_accuracy": 0.764,
|
| 243 |
+
"eval_f1": 0.7624798711755234,
|
| 244 |
+
"eval_loss": 0.49321362376213074,
|
| 245 |
+
"eval_runtime": 3.1634,
|
| 246 |
+
"eval_samples_per_second": 316.111,
|
| 247 |
+
"eval_steps_per_second": 39.514,
|
| 248 |
+
"step": 1050
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 15.99,
|
| 252 |
+
"learning_rate": 4.6666666666666666e-07,
|
| 253 |
+
"loss": 0.4319,
|
| 254 |
+
"step": 1120
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 15.99,
|
| 258 |
+
"eval_accuracy": 0.762,
|
| 259 |
+
"eval_f1": 0.7608280574816602,
|
| 260 |
+
"eval_loss": 0.4916852116584778,
|
| 261 |
+
"eval_runtime": 3.0371,
|
| 262 |
+
"eval_samples_per_second": 329.257,
|
| 263 |
+
"eval_steps_per_second": 41.157,
|
| 264 |
+
"step": 1120
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 16.99,
|
| 268 |
+
"learning_rate": 4.3333333333333335e-07,
|
| 269 |
+
"loss": 0.427,
|
| 270 |
+
"step": 1190
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 16.99,
|
| 274 |
+
"eval_accuracy": 0.77,
|
| 275 |
+
"eval_f1": 0.7693273585776123,
|
| 276 |
+
"eval_loss": 0.4917564392089844,
|
| 277 |
+
"eval_runtime": 3.011,
|
| 278 |
+
"eval_samples_per_second": 332.114,
|
| 279 |
+
"eval_steps_per_second": 41.514,
|
| 280 |
+
"step": 1190
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"epoch": 17.99,
|
| 284 |
+
"learning_rate": 4e-07,
|
| 285 |
+
"loss": 0.4226,
|
| 286 |
+
"step": 1260
|
| 287 |
+
},
|
| 288 |
+
{
|
| 289 |
+
"epoch": 17.99,
|
| 290 |
+
"eval_accuracy": 0.772,
|
| 291 |
+
"eval_f1": 0.7711201859949647,
|
| 292 |
+
"eval_loss": 0.48886075615882874,
|
| 293 |
+
"eval_runtime": 2.9854,
|
| 294 |
+
"eval_samples_per_second": 334.968,
|
| 295 |
+
"eval_steps_per_second": 41.871,
|
| 296 |
+
"step": 1260
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 18.99,
|
| 300 |
+
"learning_rate": 3.666666666666666e-07,
|
| 301 |
+
"loss": 0.4167,
|
| 302 |
+
"step": 1330
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 18.99,
|
| 306 |
+
"eval_accuracy": 0.769,
|
| 307 |
+
"eval_f1": 0.7681372386650728,
|
| 308 |
+
"eval_loss": 0.48737427592277527,
|
| 309 |
+
"eval_runtime": 3.4249,
|
| 310 |
+
"eval_samples_per_second": 291.979,
|
| 311 |
+
"eval_steps_per_second": 36.497,
|
| 312 |
+
"step": 1330
|
| 313 |
+
}
|
| 314 |
+
],
|
| 315 |
+
"max_steps": 2100,
|
| 316 |
+
"num_train_epochs": 30,
|
| 317 |
+
"total_flos": 1.687199642496e+16,
|
| 318 |
+
"trial_name": null,
|
| 319 |
+
"trial_params": null
|
| 320 |
+
}
|
checkpoint-1330/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8be60ed58549fe1b3185c4d116e2ed7593eafbe162cc3300999b64c649675d04
|
| 3 |
+
size 3119
|
checkpoint-1330/vocab.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-1400/added_tokens.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"<mask>": 64000}
|
checkpoint-1400/bpe.codes
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-1400/config.json
ADDED
|
@@ -0,0 +1,30 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "vinai/bertweet-covid19-base-cased",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"RobertaForSequenceClassification"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.1,
|
| 7 |
+
"bos_token_id": 0,
|
| 8 |
+
"classifier_dropout": null,
|
| 9 |
+
"eos_token_id": 2,
|
| 10 |
+
"gradient_checkpointing": false,
|
| 11 |
+
"hidden_act": "gelu",
|
| 12 |
+
"hidden_dropout_prob": 0.1,
|
| 13 |
+
"hidden_size": 768,
|
| 14 |
+
"initializer_range": 0.02,
|
| 15 |
+
"intermediate_size": 3072,
|
| 16 |
+
"layer_norm_eps": 1e-05,
|
| 17 |
+
"max_position_embeddings": 130,
|
| 18 |
+
"model_type": "roberta",
|
| 19 |
+
"num_attention_heads": 12,
|
| 20 |
+
"num_hidden_layers": 12,
|
| 21 |
+
"pad_token_id": 1,
|
| 22 |
+
"position_embedding_type": "absolute",
|
| 23 |
+
"problem_type": "multi_label_classification",
|
| 24 |
+
"tokenizer_class": "BertweetTokenizer",
|
| 25 |
+
"torch_dtype": "float32",
|
| 26 |
+
"transformers_version": "4.17.0",
|
| 27 |
+
"type_vocab_size": 1,
|
| 28 |
+
"use_cache": true,
|
| 29 |
+
"vocab_size": 64001
|
| 30 |
+
}
|
checkpoint-1400/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:95cc0637412b7f437977ec372c70565c41d32d4adc0e845bdf1d6f47198d66a3
|
| 3 |
+
size 1079327581
|
checkpoint-1400/pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b6ba4a253da0877e7e7d26236c2dab142771fd554b1d86ee21d92b3dd4c92c7a
|
| 3 |
+
size 539689325
|
checkpoint-1400/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0f9ce8b30aa5cf25571abe72f151e001ed799e950e69987801ab1d0a9c21e293
|
| 3 |
+
size 14503
|
checkpoint-1400/scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:26339048a0b88d77fc1a5e9abedb9d0653c3cd4b707b2f8b4a992e2612d53b7f
|
| 3 |
+
size 559
|
checkpoint-1400/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a53f072a9a487fcaa99cc1848987d79f9aeb930a3b2cb329cc2f069c68ecfe48
|
| 3 |
+
size 623
|
checkpoint-1400/special_tokens_map.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"bos_token": "<s>", "eos_token": "</s>", "unk_token": "<unk>", "sep_token": "</s>", "pad_token": "<pad>", "cls_token": "<s>", "mask_token": "<mask>"}
|
checkpoint-1400/tokenizer_config.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"normalization": true, "bos_token": "<s>", "eos_token": "</s>", "sep_token": "</s>", "cls_token": "<s>", "unk_token": "<unk>", "pad_token": "<pad>", "mask_token": "<mask>", "special_tokens_map_file": null, "name_or_path": "vinai/bertweet-covid19-base-cased", "tokenizer_class": "BertweetTokenizer"}
|
checkpoint-1400/trainer_state.json
ADDED
|
@@ -0,0 +1,336 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.7711201859949647,
|
| 3 |
+
"best_model_checkpoint": "outputs/bertweet/checkpoint-1260",
|
| 4 |
+
"epoch": 19.99290780141844,
|
| 5 |
+
"global_step": 1400,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.99,
|
| 12 |
+
"learning_rate": 9.666666666666666e-07,
|
| 13 |
+
"loss": 0.6925,
|
| 14 |
+
"step": 70
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.99,
|
| 18 |
+
"eval_accuracy": 0.573,
|
| 19 |
+
"eval_f1": 0.3642720915448188,
|
| 20 |
+
"eval_loss": 0.6826930046081543,
|
| 21 |
+
"eval_runtime": 2.9877,
|
| 22 |
+
"eval_samples_per_second": 334.703,
|
| 23 |
+
"eval_steps_per_second": 41.838,
|
| 24 |
+
"step": 70
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 1.99,
|
| 28 |
+
"learning_rate": 9.333333333333333e-07,
|
| 29 |
+
"loss": 0.6823,
|
| 30 |
+
"step": 140
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 1.99,
|
| 34 |
+
"eval_accuracy": 0.573,
|
| 35 |
+
"eval_f1": 0.3642720915448188,
|
| 36 |
+
"eval_loss": 0.6736124753952026,
|
| 37 |
+
"eval_runtime": 3.0497,
|
| 38 |
+
"eval_samples_per_second": 327.901,
|
| 39 |
+
"eval_steps_per_second": 40.988,
|
| 40 |
+
"step": 140
|
| 41 |
+
},
|
| 42 |
+
{
|
| 43 |
+
"epoch": 2.99,
|
| 44 |
+
"learning_rate": 9e-07,
|
| 45 |
+
"loss": 0.6713,
|
| 46 |
+
"step": 210
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 2.99,
|
| 50 |
+
"eval_accuracy": 0.587,
|
| 51 |
+
"eval_f1": 0.39928932873127865,
|
| 52 |
+
"eval_loss": 0.656767725944519,
|
| 53 |
+
"eval_runtime": 3.0012,
|
| 54 |
+
"eval_samples_per_second": 333.2,
|
| 55 |
+
"eval_steps_per_second": 41.65,
|
| 56 |
+
"step": 210
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 3.99,
|
| 60 |
+
"learning_rate": 8.666666666666667e-07,
|
| 61 |
+
"loss": 0.6468,
|
| 62 |
+
"step": 280
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 3.99,
|
| 66 |
+
"eval_accuracy": 0.7,
|
| 67 |
+
"eval_f1": 0.6707623826267894,
|
| 68 |
+
"eval_loss": 0.6210379600524902,
|
| 69 |
+
"eval_runtime": 3.0442,
|
| 70 |
+
"eval_samples_per_second": 328.492,
|
| 71 |
+
"eval_steps_per_second": 41.062,
|
| 72 |
+
"step": 280
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 4.99,
|
| 76 |
+
"learning_rate": 8.333333333333333e-07,
|
| 77 |
+
"loss": 0.6047,
|
| 78 |
+
"step": 350
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"epoch": 4.99,
|
| 82 |
+
"eval_accuracy": 0.732,
|
| 83 |
+
"eval_f1": 0.7285955019767969,
|
| 84 |
+
"eval_loss": 0.5784569382667542,
|
| 85 |
+
"eval_runtime": 3.0658,
|
| 86 |
+
"eval_samples_per_second": 326.18,
|
| 87 |
+
"eval_steps_per_second": 40.773,
|
| 88 |
+
"step": 350
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 5.99,
|
| 92 |
+
"learning_rate": 8e-07,
|
| 93 |
+
"loss": 0.5648,
|
| 94 |
+
"step": 420
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 5.99,
|
| 98 |
+
"eval_accuracy": 0.733,
|
| 99 |
+
"eval_f1": 0.7318671386608421,
|
| 100 |
+
"eval_loss": 0.5537055730819702,
|
| 101 |
+
"eval_runtime": 3.061,
|
| 102 |
+
"eval_samples_per_second": 326.692,
|
| 103 |
+
"eval_steps_per_second": 40.836,
|
| 104 |
+
"step": 420
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 6.99,
|
| 108 |
+
"learning_rate": 7.666666666666667e-07,
|
| 109 |
+
"loss": 0.536,
|
| 110 |
+
"step": 490
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 6.99,
|
| 114 |
+
"eval_accuracy": 0.739,
|
| 115 |
+
"eval_f1": 0.7381492469031885,
|
| 116 |
+
"eval_loss": 0.5406165719032288,
|
| 117 |
+
"eval_runtime": 3.3555,
|
| 118 |
+
"eval_samples_per_second": 298.016,
|
| 119 |
+
"eval_steps_per_second": 37.252,
|
| 120 |
+
"step": 490
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 7.99,
|
| 124 |
+
"learning_rate": 7.333333333333332e-07,
|
| 125 |
+
"loss": 0.5175,
|
| 126 |
+
"step": 560
|
| 127 |
+
},
|
| 128 |
+
{
|
| 129 |
+
"epoch": 7.99,
|
| 130 |
+
"eval_accuracy": 0.744,
|
| 131 |
+
"eval_f1": 0.7430750702529104,
|
| 132 |
+
"eval_loss": 0.5307690501213074,
|
| 133 |
+
"eval_runtime": 3.1364,
|
| 134 |
+
"eval_samples_per_second": 318.842,
|
| 135 |
+
"eval_steps_per_second": 39.855,
|
| 136 |
+
"step": 560
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 8.99,
|
| 140 |
+
"learning_rate": 7e-07,
|
| 141 |
+
"loss": 0.5018,
|
| 142 |
+
"step": 630
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 8.99,
|
| 146 |
+
"eval_accuracy": 0.751,
|
| 147 |
+
"eval_f1": 0.7504007121097755,
|
| 148 |
+
"eval_loss": 0.5235322117805481,
|
| 149 |
+
"eval_runtime": 2.9996,
|
| 150 |
+
"eval_samples_per_second": 333.377,
|
| 151 |
+
"eval_steps_per_second": 41.672,
|
| 152 |
+
"step": 630
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 9.99,
|
| 156 |
+
"learning_rate": 6.666666666666666e-07,
|
| 157 |
+
"loss": 0.4874,
|
| 158 |
+
"step": 700
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 9.99,
|
| 162 |
+
"eval_accuracy": 0.749,
|
| 163 |
+
"eval_f1": 0.7478681802611924,
|
| 164 |
+
"eval_loss": 0.5145103931427002,
|
| 165 |
+
"eval_runtime": 3.1795,
|
| 166 |
+
"eval_samples_per_second": 314.515,
|
| 167 |
+
"eval_steps_per_second": 39.314,
|
| 168 |
+
"step": 700
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 10.99,
|
| 172 |
+
"learning_rate": 6.333333333333332e-07,
|
| 173 |
+
"loss": 0.4749,
|
| 174 |
+
"step": 770
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 10.99,
|
| 178 |
+
"eval_accuracy": 0.754,
|
| 179 |
+
"eval_f1": 0.7533330124657072,
|
| 180 |
+
"eval_loss": 0.510367751121521,
|
| 181 |
+
"eval_runtime": 3.0347,
|
| 182 |
+
"eval_samples_per_second": 329.518,
|
| 183 |
+
"eval_steps_per_second": 41.19,
|
| 184 |
+
"step": 770
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 11.99,
|
| 188 |
+
"learning_rate": 6e-07,
|
| 189 |
+
"loss": 0.4666,
|
| 190 |
+
"step": 840
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 11.99,
|
| 194 |
+
"eval_accuracy": 0.761,
|
| 195 |
+
"eval_f1": 0.760515042961998,
|
| 196 |
+
"eval_loss": 0.5052358508110046,
|
| 197 |
+
"eval_runtime": 3.0155,
|
| 198 |
+
"eval_samples_per_second": 331.615,
|
| 199 |
+
"eval_steps_per_second": 41.452,
|
| 200 |
+
"step": 840
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 12.99,
|
| 204 |
+
"learning_rate": 5.666666666666666e-07,
|
| 205 |
+
"loss": 0.456,
|
| 206 |
+
"step": 910
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 12.99,
|
| 210 |
+
"eval_accuracy": 0.761,
|
| 211 |
+
"eval_f1": 0.7604247798965316,
|
| 212 |
+
"eval_loss": 0.5017005205154419,
|
| 213 |
+
"eval_runtime": 2.9976,
|
| 214 |
+
"eval_samples_per_second": 333.599,
|
| 215 |
+
"eval_steps_per_second": 41.7,
|
| 216 |
+
"step": 910
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 13.99,
|
| 220 |
+
"learning_rate": 5.333333333333333e-07,
|
| 221 |
+
"loss": 0.4489,
|
| 222 |
+
"step": 980
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"epoch": 13.99,
|
| 226 |
+
"eval_accuracy": 0.764,
|
| 227 |
+
"eval_f1": 0.7634995650797087,
|
| 228 |
+
"eval_loss": 0.49855634570121765,
|
| 229 |
+
"eval_runtime": 3.0658,
|
| 230 |
+
"eval_samples_per_second": 326.176,
|
| 231 |
+
"eval_steps_per_second": 40.772,
|
| 232 |
+
"step": 980
|
| 233 |
+
},
|
| 234 |
+
{
|
| 235 |
+
"epoch": 14.99,
|
| 236 |
+
"learning_rate": 5e-07,
|
| 237 |
+
"loss": 0.4375,
|
| 238 |
+
"step": 1050
|
| 239 |
+
},
|
| 240 |
+
{
|
| 241 |
+
"epoch": 14.99,
|
| 242 |
+
"eval_accuracy": 0.764,
|
| 243 |
+
"eval_f1": 0.7624798711755234,
|
| 244 |
+
"eval_loss": 0.49321362376213074,
|
| 245 |
+
"eval_runtime": 3.1634,
|
| 246 |
+
"eval_samples_per_second": 316.111,
|
| 247 |
+
"eval_steps_per_second": 39.514,
|
| 248 |
+
"step": 1050
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 15.99,
|
| 252 |
+
"learning_rate": 4.6666666666666666e-07,
|
| 253 |
+
"loss": 0.4319,
|
| 254 |
+
"step": 1120
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 15.99,
|
| 258 |
+
"eval_accuracy": 0.762,
|
| 259 |
+
"eval_f1": 0.7608280574816602,
|
| 260 |
+
"eval_loss": 0.4916852116584778,
|
| 261 |
+
"eval_runtime": 3.0371,
|
| 262 |
+
"eval_samples_per_second": 329.257,
|
| 263 |
+
"eval_steps_per_second": 41.157,
|
| 264 |
+
"step": 1120
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 16.99,
|
| 268 |
+
"learning_rate": 4.3333333333333335e-07,
|
| 269 |
+
"loss": 0.427,
|
| 270 |
+
"step": 1190
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 16.99,
|
| 274 |
+
"eval_accuracy": 0.77,
|
| 275 |
+
"eval_f1": 0.7693273585776123,
|
| 276 |
+
"eval_loss": 0.4917564392089844,
|
| 277 |
+
"eval_runtime": 3.011,
|
| 278 |
+
"eval_samples_per_second": 332.114,
|
| 279 |
+
"eval_steps_per_second": 41.514,
|
| 280 |
+
"step": 1190
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"epoch": 17.99,
|
| 284 |
+
"learning_rate": 4e-07,
|
| 285 |
+
"loss": 0.4226,
|
| 286 |
+
"step": 1260
|
| 287 |
+
},
|
| 288 |
+
{
|
| 289 |
+
"epoch": 17.99,
|
| 290 |
+
"eval_accuracy": 0.772,
|
| 291 |
+
"eval_f1": 0.7711201859949647,
|
| 292 |
+
"eval_loss": 0.48886075615882874,
|
| 293 |
+
"eval_runtime": 2.9854,
|
| 294 |
+
"eval_samples_per_second": 334.968,
|
| 295 |
+
"eval_steps_per_second": 41.871,
|
| 296 |
+
"step": 1260
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 18.99,
|
| 300 |
+
"learning_rate": 3.666666666666666e-07,
|
| 301 |
+
"loss": 0.4167,
|
| 302 |
+
"step": 1330
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 18.99,
|
| 306 |
+
"eval_accuracy": 0.769,
|
| 307 |
+
"eval_f1": 0.7681372386650728,
|
| 308 |
+
"eval_loss": 0.48737427592277527,
|
| 309 |
+
"eval_runtime": 3.4249,
|
| 310 |
+
"eval_samples_per_second": 291.979,
|
| 311 |
+
"eval_steps_per_second": 36.497,
|
| 312 |
+
"step": 1330
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 19.99,
|
| 316 |
+
"learning_rate": 3.333333333333333e-07,
|
| 317 |
+
"loss": 0.4127,
|
| 318 |
+
"step": 1400
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 19.99,
|
| 322 |
+
"eval_accuracy": 0.768,
|
| 323 |
+
"eval_f1": 0.7672701592193119,
|
| 324 |
+
"eval_loss": 0.4868090748786926,
|
| 325 |
+
"eval_runtime": 3.2921,
|
| 326 |
+
"eval_samples_per_second": 303.76,
|
| 327 |
+
"eval_steps_per_second": 37.97,
|
| 328 |
+
"step": 1400
|
| 329 |
+
}
|
| 330 |
+
],
|
| 331 |
+
"max_steps": 2100,
|
| 332 |
+
"num_train_epochs": 30,
|
| 333 |
+
"total_flos": 1.77599962368e+16,
|
| 334 |
+
"trial_name": null,
|
| 335 |
+
"trial_params": null
|
| 336 |
+
}
|
checkpoint-1400/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8be60ed58549fe1b3185c4d116e2ed7593eafbe162cc3300999b64c649675d04
|
| 3 |
+
size 3119
|
checkpoint-1400/vocab.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-1470/added_tokens.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"<mask>": 64000}
|
checkpoint-1470/bpe.codes
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-1470/config.json
ADDED
|
@@ -0,0 +1,30 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "vinai/bertweet-covid19-base-cased",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"RobertaForSequenceClassification"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.1,
|
| 7 |
+
"bos_token_id": 0,
|
| 8 |
+
"classifier_dropout": null,
|
| 9 |
+
"eos_token_id": 2,
|
| 10 |
+
"gradient_checkpointing": false,
|
| 11 |
+
"hidden_act": "gelu",
|
| 12 |
+
"hidden_dropout_prob": 0.1,
|
| 13 |
+
"hidden_size": 768,
|
| 14 |
+
"initializer_range": 0.02,
|
| 15 |
+
"intermediate_size": 3072,
|
| 16 |
+
"layer_norm_eps": 1e-05,
|
| 17 |
+
"max_position_embeddings": 130,
|
| 18 |
+
"model_type": "roberta",
|
| 19 |
+
"num_attention_heads": 12,
|
| 20 |
+
"num_hidden_layers": 12,
|
| 21 |
+
"pad_token_id": 1,
|
| 22 |
+
"position_embedding_type": "absolute",
|
| 23 |
+
"problem_type": "multi_label_classification",
|
| 24 |
+
"tokenizer_class": "BertweetTokenizer",
|
| 25 |
+
"torch_dtype": "float32",
|
| 26 |
+
"transformers_version": "4.17.0",
|
| 27 |
+
"type_vocab_size": 1,
|
| 28 |
+
"use_cache": true,
|
| 29 |
+
"vocab_size": 64001
|
| 30 |
+
}
|
checkpoint-1470/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:36ffc173471e4bdba58c7db01980fa65e27f21d4188606254634e6c70ffa1836
|
| 3 |
+
size 1079327581
|
checkpoint-1470/pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0d1b682289991d82e8e89325fd39281dba62d73060ec6334d90aadc592d2e9fd
|
| 3 |
+
size 539689325
|
checkpoint-1470/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2496cb4dda83e285c649376bc99e31389ad4e997dd89381fec04c2299a7cd73f
|
| 3 |
+
size 14503
|
checkpoint-1470/scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ba7cf413f9784564be7d3db98289e6744519b3ab56fad8705f9eacbcd9349abe
|
| 3 |
+
size 559
|
checkpoint-1470/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2f82c3f571656da611259d9571d9e85f05f9f98bf1a3953875df732d6af21c8f
|
| 3 |
+
size 623
|
checkpoint-1470/special_tokens_map.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"bos_token": "<s>", "eos_token": "</s>", "unk_token": "<unk>", "sep_token": "</s>", "pad_token": "<pad>", "cls_token": "<s>", "mask_token": "<mask>"}
|
checkpoint-1470/tokenizer_config.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"normalization": true, "bos_token": "<s>", "eos_token": "</s>", "sep_token": "</s>", "cls_token": "<s>", "unk_token": "<unk>", "pad_token": "<pad>", "mask_token": "<mask>", "special_tokens_map_file": null, "name_or_path": "vinai/bertweet-covid19-base-cased", "tokenizer_class": "BertweetTokenizer"}
|
checkpoint-1470/trainer_state.json
ADDED
|
@@ -0,0 +1,352 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.7730704967547073,
|
| 3 |
+
"best_model_checkpoint": "outputs/bertweet/checkpoint-1470",
|
| 4 |
+
"epoch": 20.99290780141844,
|
| 5 |
+
"global_step": 1470,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.99,
|
| 12 |
+
"learning_rate": 9.666666666666666e-07,
|
| 13 |
+
"loss": 0.6925,
|
| 14 |
+
"step": 70
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.99,
|
| 18 |
+
"eval_accuracy": 0.573,
|
| 19 |
+
"eval_f1": 0.3642720915448188,
|
| 20 |
+
"eval_loss": 0.6826930046081543,
|
| 21 |
+
"eval_runtime": 2.9877,
|
| 22 |
+
"eval_samples_per_second": 334.703,
|
| 23 |
+
"eval_steps_per_second": 41.838,
|
| 24 |
+
"step": 70
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 1.99,
|
| 28 |
+
"learning_rate": 9.333333333333333e-07,
|
| 29 |
+
"loss": 0.6823,
|
| 30 |
+
"step": 140
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 1.99,
|
| 34 |
+
"eval_accuracy": 0.573,
|
| 35 |
+
"eval_f1": 0.3642720915448188,
|
| 36 |
+
"eval_loss": 0.6736124753952026,
|
| 37 |
+
"eval_runtime": 3.0497,
|
| 38 |
+
"eval_samples_per_second": 327.901,
|
| 39 |
+
"eval_steps_per_second": 40.988,
|
| 40 |
+
"step": 140
|
| 41 |
+
},
|
| 42 |
+
{
|
| 43 |
+
"epoch": 2.99,
|
| 44 |
+
"learning_rate": 9e-07,
|
| 45 |
+
"loss": 0.6713,
|
| 46 |
+
"step": 210
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 2.99,
|
| 50 |
+
"eval_accuracy": 0.587,
|
| 51 |
+
"eval_f1": 0.39928932873127865,
|
| 52 |
+
"eval_loss": 0.656767725944519,
|
| 53 |
+
"eval_runtime": 3.0012,
|
| 54 |
+
"eval_samples_per_second": 333.2,
|
| 55 |
+
"eval_steps_per_second": 41.65,
|
| 56 |
+
"step": 210
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 3.99,
|
| 60 |
+
"learning_rate": 8.666666666666667e-07,
|
| 61 |
+
"loss": 0.6468,
|
| 62 |
+
"step": 280
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 3.99,
|
| 66 |
+
"eval_accuracy": 0.7,
|
| 67 |
+
"eval_f1": 0.6707623826267894,
|
| 68 |
+
"eval_loss": 0.6210379600524902,
|
| 69 |
+
"eval_runtime": 3.0442,
|
| 70 |
+
"eval_samples_per_second": 328.492,
|
| 71 |
+
"eval_steps_per_second": 41.062,
|
| 72 |
+
"step": 280
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 4.99,
|
| 76 |
+
"learning_rate": 8.333333333333333e-07,
|
| 77 |
+
"loss": 0.6047,
|
| 78 |
+
"step": 350
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"epoch": 4.99,
|
| 82 |
+
"eval_accuracy": 0.732,
|
| 83 |
+
"eval_f1": 0.7285955019767969,
|
| 84 |
+
"eval_loss": 0.5784569382667542,
|
| 85 |
+
"eval_runtime": 3.0658,
|
| 86 |
+
"eval_samples_per_second": 326.18,
|
| 87 |
+
"eval_steps_per_second": 40.773,
|
| 88 |
+
"step": 350
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 5.99,
|
| 92 |
+
"learning_rate": 8e-07,
|
| 93 |
+
"loss": 0.5648,
|
| 94 |
+
"step": 420
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 5.99,
|
| 98 |
+
"eval_accuracy": 0.733,
|
| 99 |
+
"eval_f1": 0.7318671386608421,
|
| 100 |
+
"eval_loss": 0.5537055730819702,
|
| 101 |
+
"eval_runtime": 3.061,
|
| 102 |
+
"eval_samples_per_second": 326.692,
|
| 103 |
+
"eval_steps_per_second": 40.836,
|
| 104 |
+
"step": 420
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 6.99,
|
| 108 |
+
"learning_rate": 7.666666666666667e-07,
|
| 109 |
+
"loss": 0.536,
|
| 110 |
+
"step": 490
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 6.99,
|
| 114 |
+
"eval_accuracy": 0.739,
|
| 115 |
+
"eval_f1": 0.7381492469031885,
|
| 116 |
+
"eval_loss": 0.5406165719032288,
|
| 117 |
+
"eval_runtime": 3.3555,
|
| 118 |
+
"eval_samples_per_second": 298.016,
|
| 119 |
+
"eval_steps_per_second": 37.252,
|
| 120 |
+
"step": 490
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 7.99,
|
| 124 |
+
"learning_rate": 7.333333333333332e-07,
|
| 125 |
+
"loss": 0.5175,
|
| 126 |
+
"step": 560
|
| 127 |
+
},
|
| 128 |
+
{
|
| 129 |
+
"epoch": 7.99,
|
| 130 |
+
"eval_accuracy": 0.744,
|
| 131 |
+
"eval_f1": 0.7430750702529104,
|
| 132 |
+
"eval_loss": 0.5307690501213074,
|
| 133 |
+
"eval_runtime": 3.1364,
|
| 134 |
+
"eval_samples_per_second": 318.842,
|
| 135 |
+
"eval_steps_per_second": 39.855,
|
| 136 |
+
"step": 560
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 8.99,
|
| 140 |
+
"learning_rate": 7e-07,
|
| 141 |
+
"loss": 0.5018,
|
| 142 |
+
"step": 630
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 8.99,
|
| 146 |
+
"eval_accuracy": 0.751,
|
| 147 |
+
"eval_f1": 0.7504007121097755,
|
| 148 |
+
"eval_loss": 0.5235322117805481,
|
| 149 |
+
"eval_runtime": 2.9996,
|
| 150 |
+
"eval_samples_per_second": 333.377,
|
| 151 |
+
"eval_steps_per_second": 41.672,
|
| 152 |
+
"step": 630
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 9.99,
|
| 156 |
+
"learning_rate": 6.666666666666666e-07,
|
| 157 |
+
"loss": 0.4874,
|
| 158 |
+
"step": 700
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 9.99,
|
| 162 |
+
"eval_accuracy": 0.749,
|
| 163 |
+
"eval_f1": 0.7478681802611924,
|
| 164 |
+
"eval_loss": 0.5145103931427002,
|
| 165 |
+
"eval_runtime": 3.1795,
|
| 166 |
+
"eval_samples_per_second": 314.515,
|
| 167 |
+
"eval_steps_per_second": 39.314,
|
| 168 |
+
"step": 700
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 10.99,
|
| 172 |
+
"learning_rate": 6.333333333333332e-07,
|
| 173 |
+
"loss": 0.4749,
|
| 174 |
+
"step": 770
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 10.99,
|
| 178 |
+
"eval_accuracy": 0.754,
|
| 179 |
+
"eval_f1": 0.7533330124657072,
|
| 180 |
+
"eval_loss": 0.510367751121521,
|
| 181 |
+
"eval_runtime": 3.0347,
|
| 182 |
+
"eval_samples_per_second": 329.518,
|
| 183 |
+
"eval_steps_per_second": 41.19,
|
| 184 |
+
"step": 770
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 11.99,
|
| 188 |
+
"learning_rate": 6e-07,
|
| 189 |
+
"loss": 0.4666,
|
| 190 |
+
"step": 840
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 11.99,
|
| 194 |
+
"eval_accuracy": 0.761,
|
| 195 |
+
"eval_f1": 0.760515042961998,
|
| 196 |
+
"eval_loss": 0.5052358508110046,
|
| 197 |
+
"eval_runtime": 3.0155,
|
| 198 |
+
"eval_samples_per_second": 331.615,
|
| 199 |
+
"eval_steps_per_second": 41.452,
|
| 200 |
+
"step": 840
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 12.99,
|
| 204 |
+
"learning_rate": 5.666666666666666e-07,
|
| 205 |
+
"loss": 0.456,
|
| 206 |
+
"step": 910
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 12.99,
|
| 210 |
+
"eval_accuracy": 0.761,
|
| 211 |
+
"eval_f1": 0.7604247798965316,
|
| 212 |
+
"eval_loss": 0.5017005205154419,
|
| 213 |
+
"eval_runtime": 2.9976,
|
| 214 |
+
"eval_samples_per_second": 333.599,
|
| 215 |
+
"eval_steps_per_second": 41.7,
|
| 216 |
+
"step": 910
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 13.99,
|
| 220 |
+
"learning_rate": 5.333333333333333e-07,
|
| 221 |
+
"loss": 0.4489,
|
| 222 |
+
"step": 980
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"epoch": 13.99,
|
| 226 |
+
"eval_accuracy": 0.764,
|
| 227 |
+
"eval_f1": 0.7634995650797087,
|
| 228 |
+
"eval_loss": 0.49855634570121765,
|
| 229 |
+
"eval_runtime": 3.0658,
|
| 230 |
+
"eval_samples_per_second": 326.176,
|
| 231 |
+
"eval_steps_per_second": 40.772,
|
| 232 |
+
"step": 980
|
| 233 |
+
},
|
| 234 |
+
{
|
| 235 |
+
"epoch": 14.99,
|
| 236 |
+
"learning_rate": 5e-07,
|
| 237 |
+
"loss": 0.4375,
|
| 238 |
+
"step": 1050
|
| 239 |
+
},
|
| 240 |
+
{
|
| 241 |
+
"epoch": 14.99,
|
| 242 |
+
"eval_accuracy": 0.764,
|
| 243 |
+
"eval_f1": 0.7624798711755234,
|
| 244 |
+
"eval_loss": 0.49321362376213074,
|
| 245 |
+
"eval_runtime": 3.1634,
|
| 246 |
+
"eval_samples_per_second": 316.111,
|
| 247 |
+
"eval_steps_per_second": 39.514,
|
| 248 |
+
"step": 1050
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 15.99,
|
| 252 |
+
"learning_rate": 4.6666666666666666e-07,
|
| 253 |
+
"loss": 0.4319,
|
| 254 |
+
"step": 1120
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 15.99,
|
| 258 |
+
"eval_accuracy": 0.762,
|
| 259 |
+
"eval_f1": 0.7608280574816602,
|
| 260 |
+
"eval_loss": 0.4916852116584778,
|
| 261 |
+
"eval_runtime": 3.0371,
|
| 262 |
+
"eval_samples_per_second": 329.257,
|
| 263 |
+
"eval_steps_per_second": 41.157,
|
| 264 |
+
"step": 1120
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 16.99,
|
| 268 |
+
"learning_rate": 4.3333333333333335e-07,
|
| 269 |
+
"loss": 0.427,
|
| 270 |
+
"step": 1190
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 16.99,
|
| 274 |
+
"eval_accuracy": 0.77,
|
| 275 |
+
"eval_f1": 0.7693273585776123,
|
| 276 |
+
"eval_loss": 0.4917564392089844,
|
| 277 |
+
"eval_runtime": 3.011,
|
| 278 |
+
"eval_samples_per_second": 332.114,
|
| 279 |
+
"eval_steps_per_second": 41.514,
|
| 280 |
+
"step": 1190
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"epoch": 17.99,
|
| 284 |
+
"learning_rate": 4e-07,
|
| 285 |
+
"loss": 0.4226,
|
| 286 |
+
"step": 1260
|
| 287 |
+
},
|
| 288 |
+
{
|
| 289 |
+
"epoch": 17.99,
|
| 290 |
+
"eval_accuracy": 0.772,
|
| 291 |
+
"eval_f1": 0.7711201859949647,
|
| 292 |
+
"eval_loss": 0.48886075615882874,
|
| 293 |
+
"eval_runtime": 2.9854,
|
| 294 |
+
"eval_samples_per_second": 334.968,
|
| 295 |
+
"eval_steps_per_second": 41.871,
|
| 296 |
+
"step": 1260
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 18.99,
|
| 300 |
+
"learning_rate": 3.666666666666666e-07,
|
| 301 |
+
"loss": 0.4167,
|
| 302 |
+
"step": 1330
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 18.99,
|
| 306 |
+
"eval_accuracy": 0.769,
|
| 307 |
+
"eval_f1": 0.7681372386650728,
|
| 308 |
+
"eval_loss": 0.48737427592277527,
|
| 309 |
+
"eval_runtime": 3.4249,
|
| 310 |
+
"eval_samples_per_second": 291.979,
|
| 311 |
+
"eval_steps_per_second": 36.497,
|
| 312 |
+
"step": 1330
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 19.99,
|
| 316 |
+
"learning_rate": 3.333333333333333e-07,
|
| 317 |
+
"loss": 0.4127,
|
| 318 |
+
"step": 1400
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 19.99,
|
| 322 |
+
"eval_accuracy": 0.768,
|
| 323 |
+
"eval_f1": 0.7672701592193119,
|
| 324 |
+
"eval_loss": 0.4868090748786926,
|
| 325 |
+
"eval_runtime": 3.2921,
|
| 326 |
+
"eval_samples_per_second": 303.76,
|
| 327 |
+
"eval_steps_per_second": 37.97,
|
| 328 |
+
"step": 1400
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"epoch": 20.99,
|
| 332 |
+
"learning_rate": 3e-07,
|
| 333 |
+
"loss": 0.4095,
|
| 334 |
+
"step": 1470
|
| 335 |
+
},
|
| 336 |
+
{
|
| 337 |
+
"epoch": 20.99,
|
| 338 |
+
"eval_accuracy": 0.774,
|
| 339 |
+
"eval_f1": 0.7730704967547073,
|
| 340 |
+
"eval_loss": 0.4835589826107025,
|
| 341 |
+
"eval_runtime": 3.0995,
|
| 342 |
+
"eval_samples_per_second": 322.633,
|
| 343 |
+
"eval_steps_per_second": 40.329,
|
| 344 |
+
"step": 1470
|
| 345 |
+
}
|
| 346 |
+
],
|
| 347 |
+
"max_steps": 2100,
|
| 348 |
+
"num_train_epochs": 30,
|
| 349 |
+
"total_flos": 1.864799604864e+16,
|
| 350 |
+
"trial_name": null,
|
| 351 |
+
"trial_params": null
|
| 352 |
+
}
|
checkpoint-1470/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8be60ed58549fe1b3185c4d116e2ed7593eafbe162cc3300999b64c649675d04
|
| 3 |
+
size 3119
|
checkpoint-1470/vocab.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-1540/added_tokens.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"<mask>": 64000}
|
checkpoint-1540/bpe.codes
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-1540/config.json
ADDED
|
@@ -0,0 +1,30 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "vinai/bertweet-covid19-base-cased",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"RobertaForSequenceClassification"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.1,
|
| 7 |
+
"bos_token_id": 0,
|
| 8 |
+
"classifier_dropout": null,
|
| 9 |
+
"eos_token_id": 2,
|
| 10 |
+
"gradient_checkpointing": false,
|
| 11 |
+
"hidden_act": "gelu",
|
| 12 |
+
"hidden_dropout_prob": 0.1,
|
| 13 |
+
"hidden_size": 768,
|
| 14 |
+
"initializer_range": 0.02,
|
| 15 |
+
"intermediate_size": 3072,
|
| 16 |
+
"layer_norm_eps": 1e-05,
|
| 17 |
+
"max_position_embeddings": 130,
|
| 18 |
+
"model_type": "roberta",
|
| 19 |
+
"num_attention_heads": 12,
|
| 20 |
+
"num_hidden_layers": 12,
|
| 21 |
+
"pad_token_id": 1,
|
| 22 |
+
"position_embedding_type": "absolute",
|
| 23 |
+
"problem_type": "multi_label_classification",
|
| 24 |
+
"tokenizer_class": "BertweetTokenizer",
|
| 25 |
+
"torch_dtype": "float32",
|
| 26 |
+
"transformers_version": "4.17.0",
|
| 27 |
+
"type_vocab_size": 1,
|
| 28 |
+
"use_cache": true,
|
| 29 |
+
"vocab_size": 64001
|
| 30 |
+
}
|
checkpoint-1540/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:989be34e0dcc176b627283388df2d03dc7e212a55ad576026dc0b4d9adb41966
|
| 3 |
+
size 1079327581
|
checkpoint-1540/pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5520bccced16e1e0956381c54026c8f357a21a74cb0e08e379646452f7b06e7b
|
| 3 |
+
size 539689325
|
checkpoint-1540/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:afb47569cd26fe2e64395f3d330d4478529e57882ba5cdaf4b5d52e6ec08a336
|
| 3 |
+
size 14503
|
checkpoint-1540/scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:70e910e56af9cd0b07792d3c4dc0c36b4bdd190e5b8ca42eb40efb9f4a791081
|
| 3 |
+
size 559
|
checkpoint-1540/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:05ac3f94efc19a82d292b01c5f7f2be57d5bdf788911ae3884f3fa65532cae59
|
| 3 |
+
size 623
|
checkpoint-1540/special_tokens_map.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"bos_token": "<s>", "eos_token": "</s>", "unk_token": "<unk>", "sep_token": "</s>", "pad_token": "<pad>", "cls_token": "<s>", "mask_token": "<mask>"}
|
checkpoint-1540/tokenizer_config.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"normalization": true, "bos_token": "<s>", "eos_token": "</s>", "sep_token": "</s>", "cls_token": "<s>", "unk_token": "<unk>", "pad_token": "<pad>", "mask_token": "<mask>", "special_tokens_map_file": null, "name_or_path": "vinai/bertweet-covid19-base-cased", "tokenizer_class": "BertweetTokenizer"}
|
checkpoint-1540/trainer_state.json
ADDED
|
@@ -0,0 +1,368 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.7730704967547073,
|
| 3 |
+
"best_model_checkpoint": "outputs/bertweet/checkpoint-1470",
|
| 4 |
+
"epoch": 21.99290780141844,
|
| 5 |
+
"global_step": 1540,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.99,
|
| 12 |
+
"learning_rate": 9.666666666666666e-07,
|
| 13 |
+
"loss": 0.6925,
|
| 14 |
+
"step": 70
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.99,
|
| 18 |
+
"eval_accuracy": 0.573,
|
| 19 |
+
"eval_f1": 0.3642720915448188,
|
| 20 |
+
"eval_loss": 0.6826930046081543,
|
| 21 |
+
"eval_runtime": 2.9877,
|
| 22 |
+
"eval_samples_per_second": 334.703,
|
| 23 |
+
"eval_steps_per_second": 41.838,
|
| 24 |
+
"step": 70
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 1.99,
|
| 28 |
+
"learning_rate": 9.333333333333333e-07,
|
| 29 |
+
"loss": 0.6823,
|
| 30 |
+
"step": 140
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 1.99,
|
| 34 |
+
"eval_accuracy": 0.573,
|
| 35 |
+
"eval_f1": 0.3642720915448188,
|
| 36 |
+
"eval_loss": 0.6736124753952026,
|
| 37 |
+
"eval_runtime": 3.0497,
|
| 38 |
+
"eval_samples_per_second": 327.901,
|
| 39 |
+
"eval_steps_per_second": 40.988,
|
| 40 |
+
"step": 140
|
| 41 |
+
},
|
| 42 |
+
{
|
| 43 |
+
"epoch": 2.99,
|
| 44 |
+
"learning_rate": 9e-07,
|
| 45 |
+
"loss": 0.6713,
|
| 46 |
+
"step": 210
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 2.99,
|
| 50 |
+
"eval_accuracy": 0.587,
|
| 51 |
+
"eval_f1": 0.39928932873127865,
|
| 52 |
+
"eval_loss": 0.656767725944519,
|
| 53 |
+
"eval_runtime": 3.0012,
|
| 54 |
+
"eval_samples_per_second": 333.2,
|
| 55 |
+
"eval_steps_per_second": 41.65,
|
| 56 |
+
"step": 210
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 3.99,
|
| 60 |
+
"learning_rate": 8.666666666666667e-07,
|
| 61 |
+
"loss": 0.6468,
|
| 62 |
+
"step": 280
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 3.99,
|
| 66 |
+
"eval_accuracy": 0.7,
|
| 67 |
+
"eval_f1": 0.6707623826267894,
|
| 68 |
+
"eval_loss": 0.6210379600524902,
|
| 69 |
+
"eval_runtime": 3.0442,
|
| 70 |
+
"eval_samples_per_second": 328.492,
|
| 71 |
+
"eval_steps_per_second": 41.062,
|
| 72 |
+
"step": 280
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 4.99,
|
| 76 |
+
"learning_rate": 8.333333333333333e-07,
|
| 77 |
+
"loss": 0.6047,
|
| 78 |
+
"step": 350
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"epoch": 4.99,
|
| 82 |
+
"eval_accuracy": 0.732,
|
| 83 |
+
"eval_f1": 0.7285955019767969,
|
| 84 |
+
"eval_loss": 0.5784569382667542,
|
| 85 |
+
"eval_runtime": 3.0658,
|
| 86 |
+
"eval_samples_per_second": 326.18,
|
| 87 |
+
"eval_steps_per_second": 40.773,
|
| 88 |
+
"step": 350
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 5.99,
|
| 92 |
+
"learning_rate": 8e-07,
|
| 93 |
+
"loss": 0.5648,
|
| 94 |
+
"step": 420
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 5.99,
|
| 98 |
+
"eval_accuracy": 0.733,
|
| 99 |
+
"eval_f1": 0.7318671386608421,
|
| 100 |
+
"eval_loss": 0.5537055730819702,
|
| 101 |
+
"eval_runtime": 3.061,
|
| 102 |
+
"eval_samples_per_second": 326.692,
|
| 103 |
+
"eval_steps_per_second": 40.836,
|
| 104 |
+
"step": 420
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 6.99,
|
| 108 |
+
"learning_rate": 7.666666666666667e-07,
|
| 109 |
+
"loss": 0.536,
|
| 110 |
+
"step": 490
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 6.99,
|
| 114 |
+
"eval_accuracy": 0.739,
|
| 115 |
+
"eval_f1": 0.7381492469031885,
|
| 116 |
+
"eval_loss": 0.5406165719032288,
|
| 117 |
+
"eval_runtime": 3.3555,
|
| 118 |
+
"eval_samples_per_second": 298.016,
|
| 119 |
+
"eval_steps_per_second": 37.252,
|
| 120 |
+
"step": 490
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 7.99,
|
| 124 |
+
"learning_rate": 7.333333333333332e-07,
|
| 125 |
+
"loss": 0.5175,
|
| 126 |
+
"step": 560
|
| 127 |
+
},
|
| 128 |
+
{
|
| 129 |
+
"epoch": 7.99,
|
| 130 |
+
"eval_accuracy": 0.744,
|
| 131 |
+
"eval_f1": 0.7430750702529104,
|
| 132 |
+
"eval_loss": 0.5307690501213074,
|
| 133 |
+
"eval_runtime": 3.1364,
|
| 134 |
+
"eval_samples_per_second": 318.842,
|
| 135 |
+
"eval_steps_per_second": 39.855,
|
| 136 |
+
"step": 560
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 8.99,
|
| 140 |
+
"learning_rate": 7e-07,
|
| 141 |
+
"loss": 0.5018,
|
| 142 |
+
"step": 630
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 8.99,
|
| 146 |
+
"eval_accuracy": 0.751,
|
| 147 |
+
"eval_f1": 0.7504007121097755,
|
| 148 |
+
"eval_loss": 0.5235322117805481,
|
| 149 |
+
"eval_runtime": 2.9996,
|
| 150 |
+
"eval_samples_per_second": 333.377,
|
| 151 |
+
"eval_steps_per_second": 41.672,
|
| 152 |
+
"step": 630
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 9.99,
|
| 156 |
+
"learning_rate": 6.666666666666666e-07,
|
| 157 |
+
"loss": 0.4874,
|
| 158 |
+
"step": 700
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 9.99,
|
| 162 |
+
"eval_accuracy": 0.749,
|
| 163 |
+
"eval_f1": 0.7478681802611924,
|
| 164 |
+
"eval_loss": 0.5145103931427002,
|
| 165 |
+
"eval_runtime": 3.1795,
|
| 166 |
+
"eval_samples_per_second": 314.515,
|
| 167 |
+
"eval_steps_per_second": 39.314,
|
| 168 |
+
"step": 700
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 10.99,
|
| 172 |
+
"learning_rate": 6.333333333333332e-07,
|
| 173 |
+
"loss": 0.4749,
|
| 174 |
+
"step": 770
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 10.99,
|
| 178 |
+
"eval_accuracy": 0.754,
|
| 179 |
+
"eval_f1": 0.7533330124657072,
|
| 180 |
+
"eval_loss": 0.510367751121521,
|
| 181 |
+
"eval_runtime": 3.0347,
|
| 182 |
+
"eval_samples_per_second": 329.518,
|
| 183 |
+
"eval_steps_per_second": 41.19,
|
| 184 |
+
"step": 770
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 11.99,
|
| 188 |
+
"learning_rate": 6e-07,
|
| 189 |
+
"loss": 0.4666,
|
| 190 |
+
"step": 840
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 11.99,
|
| 194 |
+
"eval_accuracy": 0.761,
|
| 195 |
+
"eval_f1": 0.760515042961998,
|
| 196 |
+
"eval_loss": 0.5052358508110046,
|
| 197 |
+
"eval_runtime": 3.0155,
|
| 198 |
+
"eval_samples_per_second": 331.615,
|
| 199 |
+
"eval_steps_per_second": 41.452,
|
| 200 |
+
"step": 840
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 12.99,
|
| 204 |
+
"learning_rate": 5.666666666666666e-07,
|
| 205 |
+
"loss": 0.456,
|
| 206 |
+
"step": 910
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 12.99,
|
| 210 |
+
"eval_accuracy": 0.761,
|
| 211 |
+
"eval_f1": 0.7604247798965316,
|
| 212 |
+
"eval_loss": 0.5017005205154419,
|
| 213 |
+
"eval_runtime": 2.9976,
|
| 214 |
+
"eval_samples_per_second": 333.599,
|
| 215 |
+
"eval_steps_per_second": 41.7,
|
| 216 |
+
"step": 910
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 13.99,
|
| 220 |
+
"learning_rate": 5.333333333333333e-07,
|
| 221 |
+
"loss": 0.4489,
|
| 222 |
+
"step": 980
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"epoch": 13.99,
|
| 226 |
+
"eval_accuracy": 0.764,
|
| 227 |
+
"eval_f1": 0.7634995650797087,
|
| 228 |
+
"eval_loss": 0.49855634570121765,
|
| 229 |
+
"eval_runtime": 3.0658,
|
| 230 |
+
"eval_samples_per_second": 326.176,
|
| 231 |
+
"eval_steps_per_second": 40.772,
|
| 232 |
+
"step": 980
|
| 233 |
+
},
|
| 234 |
+
{
|
| 235 |
+
"epoch": 14.99,
|
| 236 |
+
"learning_rate": 5e-07,
|
| 237 |
+
"loss": 0.4375,
|
| 238 |
+
"step": 1050
|
| 239 |
+
},
|
| 240 |
+
{
|
| 241 |
+
"epoch": 14.99,
|
| 242 |
+
"eval_accuracy": 0.764,
|
| 243 |
+
"eval_f1": 0.7624798711755234,
|
| 244 |
+
"eval_loss": 0.49321362376213074,
|
| 245 |
+
"eval_runtime": 3.1634,
|
| 246 |
+
"eval_samples_per_second": 316.111,
|
| 247 |
+
"eval_steps_per_second": 39.514,
|
| 248 |
+
"step": 1050
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 15.99,
|
| 252 |
+
"learning_rate": 4.6666666666666666e-07,
|
| 253 |
+
"loss": 0.4319,
|
| 254 |
+
"step": 1120
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 15.99,
|
| 258 |
+
"eval_accuracy": 0.762,
|
| 259 |
+
"eval_f1": 0.7608280574816602,
|
| 260 |
+
"eval_loss": 0.4916852116584778,
|
| 261 |
+
"eval_runtime": 3.0371,
|
| 262 |
+
"eval_samples_per_second": 329.257,
|
| 263 |
+
"eval_steps_per_second": 41.157,
|
| 264 |
+
"step": 1120
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 16.99,
|
| 268 |
+
"learning_rate": 4.3333333333333335e-07,
|
| 269 |
+
"loss": 0.427,
|
| 270 |
+
"step": 1190
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 16.99,
|
| 274 |
+
"eval_accuracy": 0.77,
|
| 275 |
+
"eval_f1": 0.7693273585776123,
|
| 276 |
+
"eval_loss": 0.4917564392089844,
|
| 277 |
+
"eval_runtime": 3.011,
|
| 278 |
+
"eval_samples_per_second": 332.114,
|
| 279 |
+
"eval_steps_per_second": 41.514,
|
| 280 |
+
"step": 1190
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"epoch": 17.99,
|
| 284 |
+
"learning_rate": 4e-07,
|
| 285 |
+
"loss": 0.4226,
|
| 286 |
+
"step": 1260
|
| 287 |
+
},
|
| 288 |
+
{
|
| 289 |
+
"epoch": 17.99,
|
| 290 |
+
"eval_accuracy": 0.772,
|
| 291 |
+
"eval_f1": 0.7711201859949647,
|
| 292 |
+
"eval_loss": 0.48886075615882874,
|
| 293 |
+
"eval_runtime": 2.9854,
|
| 294 |
+
"eval_samples_per_second": 334.968,
|
| 295 |
+
"eval_steps_per_second": 41.871,
|
| 296 |
+
"step": 1260
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 18.99,
|
| 300 |
+
"learning_rate": 3.666666666666666e-07,
|
| 301 |
+
"loss": 0.4167,
|
| 302 |
+
"step": 1330
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 18.99,
|
| 306 |
+
"eval_accuracy": 0.769,
|
| 307 |
+
"eval_f1": 0.7681372386650728,
|
| 308 |
+
"eval_loss": 0.48737427592277527,
|
| 309 |
+
"eval_runtime": 3.4249,
|
| 310 |
+
"eval_samples_per_second": 291.979,
|
| 311 |
+
"eval_steps_per_second": 36.497,
|
| 312 |
+
"step": 1330
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 19.99,
|
| 316 |
+
"learning_rate": 3.333333333333333e-07,
|
| 317 |
+
"loss": 0.4127,
|
| 318 |
+
"step": 1400
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 19.99,
|
| 322 |
+
"eval_accuracy": 0.768,
|
| 323 |
+
"eval_f1": 0.7672701592193119,
|
| 324 |
+
"eval_loss": 0.4868090748786926,
|
| 325 |
+
"eval_runtime": 3.2921,
|
| 326 |
+
"eval_samples_per_second": 303.76,
|
| 327 |
+
"eval_steps_per_second": 37.97,
|
| 328 |
+
"step": 1400
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"epoch": 20.99,
|
| 332 |
+
"learning_rate": 3e-07,
|
| 333 |
+
"loss": 0.4095,
|
| 334 |
+
"step": 1470
|
| 335 |
+
},
|
| 336 |
+
{
|
| 337 |
+
"epoch": 20.99,
|
| 338 |
+
"eval_accuracy": 0.774,
|
| 339 |
+
"eval_f1": 0.7730704967547073,
|
| 340 |
+
"eval_loss": 0.4835589826107025,
|
| 341 |
+
"eval_runtime": 3.0995,
|
| 342 |
+
"eval_samples_per_second": 322.633,
|
| 343 |
+
"eval_steps_per_second": 40.329,
|
| 344 |
+
"step": 1470
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 21.99,
|
| 348 |
+
"learning_rate": 2.6666666666666667e-07,
|
| 349 |
+
"loss": 0.4066,
|
| 350 |
+
"step": 1540
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 21.99,
|
| 354 |
+
"eval_accuracy": 0.77,
|
| 355 |
+
"eval_f1": 0.7689937367171399,
|
| 356 |
+
"eval_loss": 0.4828569293022156,
|
| 357 |
+
"eval_runtime": 3.0514,
|
| 358 |
+
"eval_samples_per_second": 327.723,
|
| 359 |
+
"eval_steps_per_second": 40.965,
|
| 360 |
+
"step": 1540
|
| 361 |
+
}
|
| 362 |
+
],
|
| 363 |
+
"max_steps": 2100,
|
| 364 |
+
"num_train_epochs": 30,
|
| 365 |
+
"total_flos": 1.953599586048e+16,
|
| 366 |
+
"trial_name": null,
|
| 367 |
+
"trial_params": null
|
| 368 |
+
}
|