End of training
Browse files- README.md +69 -0
- model.safetensors +3 -0
- preprocessor_config.json +9 -0
- release/.ipynb_checkpoints/preprocessor_config-checkpoint.json +9 -0
- release/.ipynb_checkpoints/trainer_state-checkpoint.json +783 -0
- release/model.safetensors +3 -0
- release/optimizer.pt +3 -0
- release/preprocessor_config.json +9 -0
- release/rng_state.pth +3 -0
- release/scheduler.pt +3 -0
- release/trainer_state.json +783 -0
- release/training_args.bin +3 -0
- trainer_state.json +67 -0
- training_args.bin +3 -0
README.md
ADDED
|
@@ -0,0 +1,69 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: cc-by-nc-4.0
|
| 3 |
+
base_model: m-a-p/MERT-v0
|
| 4 |
+
tags:
|
| 5 |
+
- generated_from_trainer
|
| 6 |
+
metrics:
|
| 7 |
+
- accuracy
|
| 8 |
+
model-index:
|
| 9 |
+
- name: mert_cmp_single
|
| 10 |
+
results: []
|
| 11 |
+
---
|
| 12 |
+
|
| 13 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 14 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 15 |
+
|
| 16 |
+
# mert_cmp_single
|
| 17 |
+
|
| 18 |
+
This model is a fine-tuned version of [m-a-p/MERT-v0](https://huggingface.co/m-a-p/MERT-v0) on an unknown dataset.
|
| 19 |
+
It achieves the following results on the evaluation set:
|
| 20 |
+
- Loss: 2.3228
|
| 21 |
+
- Accuracy: 0.0870
|
| 22 |
+
- Precision Micro: 0.0870
|
| 23 |
+
- Recall Micro: 0.0870
|
| 24 |
+
- F1 Micro: 0.0870
|
| 25 |
+
- Precision Macro: 0.0109
|
| 26 |
+
- Recall Macro: 0.125
|
| 27 |
+
- F1 Macro: 0.02
|
| 28 |
+
- Precision Weighted: 0.0076
|
| 29 |
+
- Recall Weighted: 0.0870
|
| 30 |
+
- F1 Weighted: 0.0139
|
| 31 |
+
|
| 32 |
+
## Model description
|
| 33 |
+
|
| 34 |
+
More information needed
|
| 35 |
+
|
| 36 |
+
## Intended uses & limitations
|
| 37 |
+
|
| 38 |
+
More information needed
|
| 39 |
+
|
| 40 |
+
## Training and evaluation data
|
| 41 |
+
|
| 42 |
+
More information needed
|
| 43 |
+
|
| 44 |
+
## Training procedure
|
| 45 |
+
|
| 46 |
+
### Training hyperparameters
|
| 47 |
+
|
| 48 |
+
The following hyperparameters were used during training:
|
| 49 |
+
- learning_rate: 0.005
|
| 50 |
+
- train_batch_size: 16
|
| 51 |
+
- eval_batch_size: 16
|
| 52 |
+
- seed: 42
|
| 53 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 54 |
+
- lr_scheduler_type: linear
|
| 55 |
+
- num_epochs: 1
|
| 56 |
+
|
| 57 |
+
### Training results
|
| 58 |
+
|
| 59 |
+
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision Micro | Recall Micro | F1 Micro | Precision Macro | Recall Macro | F1 Macro | Precision Weighted | Recall Weighted | F1 Weighted |
|
| 60 |
+
|:-------------:|:-----:|:----:|:---------------:|:--------:|:---------------:|:------------:|:--------:|:---------------:|:------------:|:--------:|:------------------:|:---------------:|:-----------:|
|
| 61 |
+
| 2.2433 | 1.0 | 4 | 2.3228 | 0.0870 | 0.0870 | 0.0870 | 0.0870 | 0.0109 | 0.125 | 0.02 | 0.0076 | 0.0870 | 0.0139 |
|
| 62 |
+
|
| 63 |
+
|
| 64 |
+
### Framework versions
|
| 65 |
+
|
| 66 |
+
- Transformers 4.44.0
|
| 67 |
+
- Pytorch 2.3.1+cu121
|
| 68 |
+
- Datasets 3.6.0
|
| 69 |
+
- Tokenizers 0.19.1
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3572d43e0b49b3db23d49aa98da1eb47ade01409d5bb802fd8b207fecc703a09
|
| 3 |
+
size 392738908
|
preprocessor_config.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"do_normalize": true,
|
| 3 |
+
"feature_extractor_type": "Wav2Vec2FeatureExtractor",
|
| 4 |
+
"feature_size": 1,
|
| 5 |
+
"padding_side": "right",
|
| 6 |
+
"padding_value": 0.0,
|
| 7 |
+
"return_attention_mask": true,
|
| 8 |
+
"sampling_rate": 16000
|
| 9 |
+
}
|
release/.ipynb_checkpoints/preprocessor_config-checkpoint.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"do_normalize": true,
|
| 3 |
+
"feature_extractor_type": "Wav2Vec2FeatureExtractor",
|
| 4 |
+
"feature_size": 1,
|
| 5 |
+
"padding_side": "right",
|
| 6 |
+
"padding_value": 0.0,
|
| 7 |
+
"return_attention_mask": true,
|
| 8 |
+
"sampling_rate": 16000
|
| 9 |
+
}
|
release/.ipynb_checkpoints/trainer_state-checkpoint.json
ADDED
|
@@ -0,0 +1,783 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.5166666666666667,
|
| 3 |
+
"best_model_checkpoint": "model/mert_cmp_single/checkpoint-5010",
|
| 4 |
+
"epoch": 30.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 5010,
|
| 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": 1.972764253616333,
|
| 14 |
+
"learning_rate": 0.004833333333333334,
|
| 15 |
+
"loss": 2.0548,
|
| 16 |
+
"step": 167
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 1.0,
|
| 20 |
+
"eval_accuracy": 0.3719298245614035,
|
| 21 |
+
"eval_f1_macro": 0.1752339726946509,
|
| 22 |
+
"eval_f1_micro": 0.3719298245614035,
|
| 23 |
+
"eval_f1_weighted": 0.3032743408676906,
|
| 24 |
+
"eval_loss": 1.8728982210159302,
|
| 25 |
+
"eval_precision_macro": 0.15349976094634282,
|
| 26 |
+
"eval_precision_micro": 0.3719298245614035,
|
| 27 |
+
"eval_precision_weighted": 0.2655915542913314,
|
| 28 |
+
"eval_recall_macro": 0.2199586762280261,
|
| 29 |
+
"eval_recall_micro": 0.3719298245614035,
|
| 30 |
+
"eval_recall_weighted": 0.3719298245614035,
|
| 31 |
+
"eval_runtime": 390.667,
|
| 32 |
+
"eval_samples_per_second": 2.918,
|
| 33 |
+
"eval_steps_per_second": 0.184,
|
| 34 |
+
"step": 167
|
| 35 |
+
},
|
| 36 |
+
{
|
| 37 |
+
"epoch": 2.0,
|
| 38 |
+
"grad_norm": 3.7620842456817627,
|
| 39 |
+
"learning_rate": 0.004666666666666667,
|
| 40 |
+
"loss": 1.8547,
|
| 41 |
+
"step": 334
|
| 42 |
+
},
|
| 43 |
+
{
|
| 44 |
+
"epoch": 2.0,
|
| 45 |
+
"eval_accuracy": 0.38596491228070173,
|
| 46 |
+
"eval_f1_macro": 0.20308576109052026,
|
| 47 |
+
"eval_f1_micro": 0.38596491228070173,
|
| 48 |
+
"eval_f1_weighted": 0.34326825024510915,
|
| 49 |
+
"eval_loss": 1.7156952619552612,
|
| 50 |
+
"eval_precision_macro": 0.21519653478856515,
|
| 51 |
+
"eval_precision_micro": 0.38596491228070173,
|
| 52 |
+
"eval_precision_weighted": 0.34210736396220265,
|
| 53 |
+
"eval_recall_macro": 0.22995235116630916,
|
| 54 |
+
"eval_recall_micro": 0.38596491228070173,
|
| 55 |
+
"eval_recall_weighted": 0.38596491228070173,
|
| 56 |
+
"eval_runtime": 388.1478,
|
| 57 |
+
"eval_samples_per_second": 2.937,
|
| 58 |
+
"eval_steps_per_second": 0.185,
|
| 59 |
+
"step": 334
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 3.0,
|
| 63 |
+
"grad_norm": 2.893453359603882,
|
| 64 |
+
"learning_rate": 0.0045000000000000005,
|
| 65 |
+
"loss": 1.7595,
|
| 66 |
+
"step": 501
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 3.0,
|
| 70 |
+
"eval_accuracy": 0.39035087719298245,
|
| 71 |
+
"eval_f1_macro": 0.27399951358210684,
|
| 72 |
+
"eval_f1_micro": 0.39035087719298245,
|
| 73 |
+
"eval_f1_weighted": 0.3326450313532406,
|
| 74 |
+
"eval_loss": 1.649847388267517,
|
| 75 |
+
"eval_precision_macro": 0.32755489540557126,
|
| 76 |
+
"eval_precision_micro": 0.39035087719298245,
|
| 77 |
+
"eval_precision_weighted": 0.3633738601389556,
|
| 78 |
+
"eval_recall_macro": 0.293769851069751,
|
| 79 |
+
"eval_recall_micro": 0.39035087719298245,
|
| 80 |
+
"eval_recall_weighted": 0.39035087719298245,
|
| 81 |
+
"eval_runtime": 387.7811,
|
| 82 |
+
"eval_samples_per_second": 2.94,
|
| 83 |
+
"eval_steps_per_second": 0.186,
|
| 84 |
+
"step": 501
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 4.0,
|
| 88 |
+
"grad_norm": 4.628668785095215,
|
| 89 |
+
"learning_rate": 0.004333333333333334,
|
| 90 |
+
"loss": 1.6863,
|
| 91 |
+
"step": 668
|
| 92 |
+
},
|
| 93 |
+
{
|
| 94 |
+
"epoch": 4.0,
|
| 95 |
+
"eval_accuracy": 0.4456140350877193,
|
| 96 |
+
"eval_f1_macro": 0.2949654893702652,
|
| 97 |
+
"eval_f1_micro": 0.4456140350877193,
|
| 98 |
+
"eval_f1_weighted": 0.4051374424710547,
|
| 99 |
+
"eval_loss": 1.4840811491012573,
|
| 100 |
+
"eval_precision_macro": 0.2723679270138862,
|
| 101 |
+
"eval_precision_micro": 0.4456140350877193,
|
| 102 |
+
"eval_precision_weighted": 0.3877325579722141,
|
| 103 |
+
"eval_recall_macro": 0.3506862934107516,
|
| 104 |
+
"eval_recall_micro": 0.4456140350877193,
|
| 105 |
+
"eval_recall_weighted": 0.4456140350877193,
|
| 106 |
+
"eval_runtime": 387.302,
|
| 107 |
+
"eval_samples_per_second": 2.943,
|
| 108 |
+
"eval_steps_per_second": 0.186,
|
| 109 |
+
"step": 668
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 5.0,
|
| 113 |
+
"grad_norm": 2.4614219665527344,
|
| 114 |
+
"learning_rate": 0.004166666666666667,
|
| 115 |
+
"loss": 1.6545,
|
| 116 |
+
"step": 835
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 5.0,
|
| 120 |
+
"eval_accuracy": 0.4710526315789474,
|
| 121 |
+
"eval_f1_macro": 0.32061989017679554,
|
| 122 |
+
"eval_f1_micro": 0.4710526315789474,
|
| 123 |
+
"eval_f1_weighted": 0.43804477640411854,
|
| 124 |
+
"eval_loss": 1.5168241262435913,
|
| 125 |
+
"eval_precision_macro": 0.3093839470299629,
|
| 126 |
+
"eval_precision_micro": 0.4710526315789474,
|
| 127 |
+
"eval_precision_weighted": 0.4152776413552219,
|
| 128 |
+
"eval_recall_macro": 0.3463593477939053,
|
| 129 |
+
"eval_recall_micro": 0.4710526315789474,
|
| 130 |
+
"eval_recall_weighted": 0.4710526315789474,
|
| 131 |
+
"eval_runtime": 388.1938,
|
| 132 |
+
"eval_samples_per_second": 2.937,
|
| 133 |
+
"eval_steps_per_second": 0.185,
|
| 134 |
+
"step": 835
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 6.0,
|
| 138 |
+
"grad_norm": 3.0677852630615234,
|
| 139 |
+
"learning_rate": 0.004,
|
| 140 |
+
"loss": 1.6326,
|
| 141 |
+
"step": 1002
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 6.0,
|
| 145 |
+
"eval_accuracy": 0.4324561403508772,
|
| 146 |
+
"eval_f1_macro": 0.3009858659534055,
|
| 147 |
+
"eval_f1_micro": 0.4324561403508772,
|
| 148 |
+
"eval_f1_weighted": 0.40163961306373386,
|
| 149 |
+
"eval_loss": 1.5056780576705933,
|
| 150 |
+
"eval_precision_macro": 0.3533036543942354,
|
| 151 |
+
"eval_precision_micro": 0.4324561403508772,
|
| 152 |
+
"eval_precision_weighted": 0.41979003159359185,
|
| 153 |
+
"eval_recall_macro": 0.33496368714252633,
|
| 154 |
+
"eval_recall_micro": 0.4324561403508772,
|
| 155 |
+
"eval_recall_weighted": 0.4324561403508772,
|
| 156 |
+
"eval_runtime": 389.6423,
|
| 157 |
+
"eval_samples_per_second": 2.926,
|
| 158 |
+
"eval_steps_per_second": 0.185,
|
| 159 |
+
"step": 1002
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 7.0,
|
| 163 |
+
"grad_norm": 4.243473529815674,
|
| 164 |
+
"learning_rate": 0.0038333333333333336,
|
| 165 |
+
"loss": 1.599,
|
| 166 |
+
"step": 1169
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"epoch": 7.0,
|
| 170 |
+
"eval_accuracy": 0.4789473684210526,
|
| 171 |
+
"eval_f1_macro": 0.3417367991830593,
|
| 172 |
+
"eval_f1_micro": 0.4789473684210526,
|
| 173 |
+
"eval_f1_weighted": 0.44204068485349196,
|
| 174 |
+
"eval_loss": 1.485596776008606,
|
| 175 |
+
"eval_precision_macro": 0.35341928050514015,
|
| 176 |
+
"eval_precision_micro": 0.4789473684210526,
|
| 177 |
+
"eval_precision_weighted": 0.43268574425886164,
|
| 178 |
+
"eval_recall_macro": 0.3618233001813015,
|
| 179 |
+
"eval_recall_micro": 0.4789473684210526,
|
| 180 |
+
"eval_recall_weighted": 0.4789473684210526,
|
| 181 |
+
"eval_runtime": 387.4622,
|
| 182 |
+
"eval_samples_per_second": 2.942,
|
| 183 |
+
"eval_steps_per_second": 0.186,
|
| 184 |
+
"step": 1169
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 8.0,
|
| 188 |
+
"grad_norm": 3.4915621280670166,
|
| 189 |
+
"learning_rate": 0.0036666666666666666,
|
| 190 |
+
"loss": 1.5571,
|
| 191 |
+
"step": 1336
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 8.0,
|
| 195 |
+
"eval_accuracy": 0.4649122807017544,
|
| 196 |
+
"eval_f1_macro": 0.33185435154616927,
|
| 197 |
+
"eval_f1_micro": 0.4649122807017544,
|
| 198 |
+
"eval_f1_weighted": 0.4375194270323329,
|
| 199 |
+
"eval_loss": 1.5075509548187256,
|
| 200 |
+
"eval_precision_macro": 0.402563382174571,
|
| 201 |
+
"eval_precision_micro": 0.4649122807017544,
|
| 202 |
+
"eval_precision_weighted": 0.4595468667733386,
|
| 203 |
+
"eval_recall_macro": 0.3376072239589585,
|
| 204 |
+
"eval_recall_micro": 0.4649122807017544,
|
| 205 |
+
"eval_recall_weighted": 0.4649122807017544,
|
| 206 |
+
"eval_runtime": 387.2913,
|
| 207 |
+
"eval_samples_per_second": 2.944,
|
| 208 |
+
"eval_steps_per_second": 0.186,
|
| 209 |
+
"step": 1336
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 9.0,
|
| 213 |
+
"grad_norm": 5.648871421813965,
|
| 214 |
+
"learning_rate": 0.0034999999999999996,
|
| 215 |
+
"loss": 1.5423,
|
| 216 |
+
"step": 1503
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 9.0,
|
| 220 |
+
"eval_accuracy": 0.4473684210526316,
|
| 221 |
+
"eval_f1_macro": 0.3633193692983963,
|
| 222 |
+
"eval_f1_micro": 0.4473684210526316,
|
| 223 |
+
"eval_f1_weighted": 0.43457480187919556,
|
| 224 |
+
"eval_loss": 1.5081905126571655,
|
| 225 |
+
"eval_precision_macro": 0.3975086633004074,
|
| 226 |
+
"eval_precision_micro": 0.4473684210526316,
|
| 227 |
+
"eval_precision_weighted": 0.453531478964179,
|
| 228 |
+
"eval_recall_macro": 0.35673163734767305,
|
| 229 |
+
"eval_recall_micro": 0.4473684210526316,
|
| 230 |
+
"eval_recall_weighted": 0.4473684210526316,
|
| 231 |
+
"eval_runtime": 387.3663,
|
| 232 |
+
"eval_samples_per_second": 2.943,
|
| 233 |
+
"eval_steps_per_second": 0.186,
|
| 234 |
+
"step": 1503
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 10.0,
|
| 238 |
+
"grad_norm": 3.1236555576324463,
|
| 239 |
+
"learning_rate": 0.003333333333333333,
|
| 240 |
+
"loss": 1.5356,
|
| 241 |
+
"step": 1670
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 10.0,
|
| 245 |
+
"eval_accuracy": 0.4649122807017544,
|
| 246 |
+
"eval_f1_macro": 0.32504503507830124,
|
| 247 |
+
"eval_f1_micro": 0.4649122807017544,
|
| 248 |
+
"eval_f1_weighted": 0.4232674688894728,
|
| 249 |
+
"eval_loss": 1.5755454301834106,
|
| 250 |
+
"eval_precision_macro": 0.361163616907195,
|
| 251 |
+
"eval_precision_micro": 0.4649122807017544,
|
| 252 |
+
"eval_precision_weighted": 0.4247075744687934,
|
| 253 |
+
"eval_recall_macro": 0.3384149762412627,
|
| 254 |
+
"eval_recall_micro": 0.4649122807017544,
|
| 255 |
+
"eval_recall_weighted": 0.4649122807017544,
|
| 256 |
+
"eval_runtime": 389.1667,
|
| 257 |
+
"eval_samples_per_second": 2.929,
|
| 258 |
+
"eval_steps_per_second": 0.185,
|
| 259 |
+
"step": 1670
|
| 260 |
+
},
|
| 261 |
+
{
|
| 262 |
+
"epoch": 11.0,
|
| 263 |
+
"grad_norm": 3.6445953845977783,
|
| 264 |
+
"learning_rate": 0.0031666666666666666,
|
| 265 |
+
"loss": 1.5174,
|
| 266 |
+
"step": 1837
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 11.0,
|
| 270 |
+
"eval_accuracy": 0.4666666666666667,
|
| 271 |
+
"eval_f1_macro": 0.35976985093947306,
|
| 272 |
+
"eval_f1_micro": 0.4666666666666667,
|
| 273 |
+
"eval_f1_weighted": 0.4458889953148088,
|
| 274 |
+
"eval_loss": 1.4875001907348633,
|
| 275 |
+
"eval_precision_macro": 0.4802628528097644,
|
| 276 |
+
"eval_precision_micro": 0.4666666666666667,
|
| 277 |
+
"eval_precision_weighted": 0.5174355391927419,
|
| 278 |
+
"eval_recall_macro": 0.3642094502216486,
|
| 279 |
+
"eval_recall_micro": 0.4666666666666667,
|
| 280 |
+
"eval_recall_weighted": 0.4666666666666667,
|
| 281 |
+
"eval_runtime": 388.4896,
|
| 282 |
+
"eval_samples_per_second": 2.934,
|
| 283 |
+
"eval_steps_per_second": 0.185,
|
| 284 |
+
"step": 1837
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 12.0,
|
| 288 |
+
"grad_norm": 3.5447826385498047,
|
| 289 |
+
"learning_rate": 0.003,
|
| 290 |
+
"loss": 1.4717,
|
| 291 |
+
"step": 2004
|
| 292 |
+
},
|
| 293 |
+
{
|
| 294 |
+
"epoch": 12.0,
|
| 295 |
+
"eval_accuracy": 0.4666666666666667,
|
| 296 |
+
"eval_f1_macro": 0.36601636096965706,
|
| 297 |
+
"eval_f1_micro": 0.4666666666666667,
|
| 298 |
+
"eval_f1_weighted": 0.44027602426120266,
|
| 299 |
+
"eval_loss": 1.4927704334259033,
|
| 300 |
+
"eval_precision_macro": 0.43193780369068935,
|
| 301 |
+
"eval_precision_micro": 0.4666666666666667,
|
| 302 |
+
"eval_precision_weighted": 0.47269213977673047,
|
| 303 |
+
"eval_recall_macro": 0.3798912460539077,
|
| 304 |
+
"eval_recall_micro": 0.4666666666666667,
|
| 305 |
+
"eval_recall_weighted": 0.4666666666666667,
|
| 306 |
+
"eval_runtime": 387.823,
|
| 307 |
+
"eval_samples_per_second": 2.939,
|
| 308 |
+
"eval_steps_per_second": 0.186,
|
| 309 |
+
"step": 2004
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 13.0,
|
| 313 |
+
"grad_norm": 4.328464508056641,
|
| 314 |
+
"learning_rate": 0.002833333333333333,
|
| 315 |
+
"loss": 1.4669,
|
| 316 |
+
"step": 2171
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 13.0,
|
| 320 |
+
"eval_accuracy": 0.4745614035087719,
|
| 321 |
+
"eval_f1_macro": 0.38859755562541454,
|
| 322 |
+
"eval_f1_micro": 0.4745614035087719,
|
| 323 |
+
"eval_f1_weighted": 0.46671787231935036,
|
| 324 |
+
"eval_loss": 1.4511743783950806,
|
| 325 |
+
"eval_precision_macro": 0.4015529868639453,
|
| 326 |
+
"eval_precision_micro": 0.4745614035087719,
|
| 327 |
+
"eval_precision_weighted": 0.4732841181218933,
|
| 328 |
+
"eval_recall_macro": 0.3975362907769957,
|
| 329 |
+
"eval_recall_micro": 0.4745614035087719,
|
| 330 |
+
"eval_recall_weighted": 0.4745614035087719,
|
| 331 |
+
"eval_runtime": 387.3991,
|
| 332 |
+
"eval_samples_per_second": 2.943,
|
| 333 |
+
"eval_steps_per_second": 0.186,
|
| 334 |
+
"step": 2171
|
| 335 |
+
},
|
| 336 |
+
{
|
| 337 |
+
"epoch": 14.0,
|
| 338 |
+
"grad_norm": 8.561325073242188,
|
| 339 |
+
"learning_rate": 0.0026666666666666666,
|
| 340 |
+
"loss": 1.4516,
|
| 341 |
+
"step": 2338
|
| 342 |
+
},
|
| 343 |
+
{
|
| 344 |
+
"epoch": 14.0,
|
| 345 |
+
"eval_accuracy": 0.47719298245614034,
|
| 346 |
+
"eval_f1_macro": 0.38778824262319195,
|
| 347 |
+
"eval_f1_micro": 0.47719298245614034,
|
| 348 |
+
"eval_f1_weighted": 0.4521071312833841,
|
| 349 |
+
"eval_loss": 1.5683324337005615,
|
| 350 |
+
"eval_precision_macro": 0.45015826319899144,
|
| 351 |
+
"eval_precision_micro": 0.47719298245614034,
|
| 352 |
+
"eval_precision_weighted": 0.48469205065002274,
|
| 353 |
+
"eval_recall_macro": 0.3807191719284666,
|
| 354 |
+
"eval_recall_micro": 0.47719298245614034,
|
| 355 |
+
"eval_recall_weighted": 0.47719298245614034,
|
| 356 |
+
"eval_runtime": 387.3112,
|
| 357 |
+
"eval_samples_per_second": 2.943,
|
| 358 |
+
"eval_steps_per_second": 0.186,
|
| 359 |
+
"step": 2338
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 15.0,
|
| 363 |
+
"grad_norm": 5.659043788909912,
|
| 364 |
+
"learning_rate": 0.0025,
|
| 365 |
+
"loss": 1.4468,
|
| 366 |
+
"step": 2505
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 15.0,
|
| 370 |
+
"eval_accuracy": 0.4649122807017544,
|
| 371 |
+
"eval_f1_macro": 0.3746786393581578,
|
| 372 |
+
"eval_f1_micro": 0.4649122807017544,
|
| 373 |
+
"eval_f1_weighted": 0.4490106368369649,
|
| 374 |
+
"eval_loss": 1.522507905960083,
|
| 375 |
+
"eval_precision_macro": 0.42440688015824424,
|
| 376 |
+
"eval_precision_micro": 0.4649122807017544,
|
| 377 |
+
"eval_precision_weighted": 0.48542408043393825,
|
| 378 |
+
"eval_recall_macro": 0.3792014977295045,
|
| 379 |
+
"eval_recall_micro": 0.4649122807017544,
|
| 380 |
+
"eval_recall_weighted": 0.4649122807017544,
|
| 381 |
+
"eval_runtime": 387.3309,
|
| 382 |
+
"eval_samples_per_second": 2.943,
|
| 383 |
+
"eval_steps_per_second": 0.186,
|
| 384 |
+
"step": 2505
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 16.0,
|
| 388 |
+
"grad_norm": 3.6838574409484863,
|
| 389 |
+
"learning_rate": 0.0023333333333333335,
|
| 390 |
+
"loss": 1.3935,
|
| 391 |
+
"step": 2672
|
| 392 |
+
},
|
| 393 |
+
{
|
| 394 |
+
"epoch": 16.0,
|
| 395 |
+
"eval_accuracy": 0.4850877192982456,
|
| 396 |
+
"eval_f1_macro": 0.41079129743155646,
|
| 397 |
+
"eval_f1_micro": 0.4850877192982456,
|
| 398 |
+
"eval_f1_weighted": 0.4818917262202846,
|
| 399 |
+
"eval_loss": 1.4960697889328003,
|
| 400 |
+
"eval_precision_macro": 0.42311229328764643,
|
| 401 |
+
"eval_precision_micro": 0.4850877192982456,
|
| 402 |
+
"eval_precision_weighted": 0.49189296461866266,
|
| 403 |
+
"eval_recall_macro": 0.4115979824480055,
|
| 404 |
+
"eval_recall_micro": 0.4850877192982456,
|
| 405 |
+
"eval_recall_weighted": 0.4850877192982456,
|
| 406 |
+
"eval_runtime": 386.9296,
|
| 407 |
+
"eval_samples_per_second": 2.946,
|
| 408 |
+
"eval_steps_per_second": 0.186,
|
| 409 |
+
"step": 2672
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 17.0,
|
| 413 |
+
"grad_norm": 6.394424915313721,
|
| 414 |
+
"learning_rate": 0.002166666666666667,
|
| 415 |
+
"loss": 1.4217,
|
| 416 |
+
"step": 2839
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 17.0,
|
| 420 |
+
"eval_accuracy": 0.48771929824561405,
|
| 421 |
+
"eval_f1_macro": 0.38735566453611514,
|
| 422 |
+
"eval_f1_micro": 0.48771929824561405,
|
| 423 |
+
"eval_f1_weighted": 0.4661251566654804,
|
| 424 |
+
"eval_loss": 1.524192214012146,
|
| 425 |
+
"eval_precision_macro": 0.4543356420217285,
|
| 426 |
+
"eval_precision_micro": 0.48771929824561405,
|
| 427 |
+
"eval_precision_weighted": 0.5054670595798253,
|
| 428 |
+
"eval_recall_macro": 0.4000549048543122,
|
| 429 |
+
"eval_recall_micro": 0.48771929824561405,
|
| 430 |
+
"eval_recall_weighted": 0.48771929824561405,
|
| 431 |
+
"eval_runtime": 387.3753,
|
| 432 |
+
"eval_samples_per_second": 2.943,
|
| 433 |
+
"eval_steps_per_second": 0.186,
|
| 434 |
+
"step": 2839
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 18.0,
|
| 438 |
+
"grad_norm": 5.048763751983643,
|
| 439 |
+
"learning_rate": 0.002,
|
| 440 |
+
"loss": 1.3831,
|
| 441 |
+
"step": 3006
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 18.0,
|
| 445 |
+
"eval_accuracy": 0.49298245614035086,
|
| 446 |
+
"eval_f1_macro": 0.3945057051668317,
|
| 447 |
+
"eval_f1_micro": 0.49298245614035086,
|
| 448 |
+
"eval_f1_weighted": 0.4768674757940579,
|
| 449 |
+
"eval_loss": 1.4723140001296997,
|
| 450 |
+
"eval_precision_macro": 0.45564765495926707,
|
| 451 |
+
"eval_precision_micro": 0.49298245614035086,
|
| 452 |
+
"eval_precision_weighted": 0.5182791073380009,
|
| 453 |
+
"eval_recall_macro": 0.4229398794166392,
|
| 454 |
+
"eval_recall_micro": 0.49298245614035086,
|
| 455 |
+
"eval_recall_weighted": 0.49298245614035086,
|
| 456 |
+
"eval_runtime": 387.6312,
|
| 457 |
+
"eval_samples_per_second": 2.941,
|
| 458 |
+
"eval_steps_per_second": 0.186,
|
| 459 |
+
"step": 3006
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 19.0,
|
| 463 |
+
"grad_norm": 8.706170082092285,
|
| 464 |
+
"learning_rate": 0.0018333333333333333,
|
| 465 |
+
"loss": 1.4249,
|
| 466 |
+
"step": 3173
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 19.0,
|
| 470 |
+
"eval_accuracy": 0.5026315789473684,
|
| 471 |
+
"eval_f1_macro": 0.3967450923381979,
|
| 472 |
+
"eval_f1_micro": 0.5026315789473684,
|
| 473 |
+
"eval_f1_weighted": 0.48467826189256524,
|
| 474 |
+
"eval_loss": 1.4886491298675537,
|
| 475 |
+
"eval_precision_macro": 0.4442558168809208,
|
| 476 |
+
"eval_precision_micro": 0.5026315789473684,
|
| 477 |
+
"eval_precision_weighted": 0.49883534500451954,
|
| 478 |
+
"eval_recall_macro": 0.39853885191451394,
|
| 479 |
+
"eval_recall_micro": 0.5026315789473684,
|
| 480 |
+
"eval_recall_weighted": 0.5026315789473684,
|
| 481 |
+
"eval_runtime": 387.008,
|
| 482 |
+
"eval_samples_per_second": 2.946,
|
| 483 |
+
"eval_steps_per_second": 0.186,
|
| 484 |
+
"step": 3173
|
| 485 |
+
},
|
| 486 |
+
{
|
| 487 |
+
"epoch": 20.0,
|
| 488 |
+
"grad_norm": 9.670186042785645,
|
| 489 |
+
"learning_rate": 0.0016666666666666666,
|
| 490 |
+
"loss": 1.3931,
|
| 491 |
+
"step": 3340
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 20.0,
|
| 495 |
+
"eval_accuracy": 0.49122807017543857,
|
| 496 |
+
"eval_f1_macro": 0.4000545127776559,
|
| 497 |
+
"eval_f1_micro": 0.49122807017543857,
|
| 498 |
+
"eval_f1_weighted": 0.47950452147445755,
|
| 499 |
+
"eval_loss": 1.5036280155181885,
|
| 500 |
+
"eval_precision_macro": 0.443839992334728,
|
| 501 |
+
"eval_precision_micro": 0.49122807017543857,
|
| 502 |
+
"eval_precision_weighted": 0.499565206319528,
|
| 503 |
+
"eval_recall_macro": 0.398944898086383,
|
| 504 |
+
"eval_recall_micro": 0.49122807017543857,
|
| 505 |
+
"eval_recall_weighted": 0.49122807017543857,
|
| 506 |
+
"eval_runtime": 387.8055,
|
| 507 |
+
"eval_samples_per_second": 2.94,
|
| 508 |
+
"eval_steps_per_second": 0.186,
|
| 509 |
+
"step": 3340
|
| 510 |
+
},
|
| 511 |
+
{
|
| 512 |
+
"epoch": 21.0,
|
| 513 |
+
"grad_norm": 3.482584238052368,
|
| 514 |
+
"learning_rate": 0.0015,
|
| 515 |
+
"loss": 1.3695,
|
| 516 |
+
"step": 3507
|
| 517 |
+
},
|
| 518 |
+
{
|
| 519 |
+
"epoch": 21.0,
|
| 520 |
+
"eval_accuracy": 0.4675438596491228,
|
| 521 |
+
"eval_f1_macro": 0.37912447794813103,
|
| 522 |
+
"eval_f1_micro": 0.4675438596491228,
|
| 523 |
+
"eval_f1_weighted": 0.4472561141761033,
|
| 524 |
+
"eval_loss": 1.5602822303771973,
|
| 525 |
+
"eval_precision_macro": 0.409817315097036,
|
| 526 |
+
"eval_precision_micro": 0.4675438596491228,
|
| 527 |
+
"eval_precision_weighted": 0.46947315014830876,
|
| 528 |
+
"eval_recall_macro": 0.39767343278642575,
|
| 529 |
+
"eval_recall_micro": 0.4675438596491228,
|
| 530 |
+
"eval_recall_weighted": 0.4675438596491228,
|
| 531 |
+
"eval_runtime": 387.5678,
|
| 532 |
+
"eval_samples_per_second": 2.941,
|
| 533 |
+
"eval_steps_per_second": 0.186,
|
| 534 |
+
"step": 3507
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 22.0,
|
| 538 |
+
"grad_norm": 5.20711088180542,
|
| 539 |
+
"learning_rate": 0.0013333333333333333,
|
| 540 |
+
"loss": 1.3424,
|
| 541 |
+
"step": 3674
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 22.0,
|
| 545 |
+
"eval_accuracy": 0.5026315789473684,
|
| 546 |
+
"eval_f1_macro": 0.41946040337720514,
|
| 547 |
+
"eval_f1_micro": 0.5026315789473684,
|
| 548 |
+
"eval_f1_weighted": 0.4942558483892827,
|
| 549 |
+
"eval_loss": 1.5268137454986572,
|
| 550 |
+
"eval_precision_macro": 0.45115966752712633,
|
| 551 |
+
"eval_precision_micro": 0.5026315789473684,
|
| 552 |
+
"eval_precision_weighted": 0.5051162872222296,
|
| 553 |
+
"eval_recall_macro": 0.41687400337731273,
|
| 554 |
+
"eval_recall_micro": 0.5026315789473684,
|
| 555 |
+
"eval_recall_weighted": 0.5026315789473684,
|
| 556 |
+
"eval_runtime": 389.0165,
|
| 557 |
+
"eval_samples_per_second": 2.93,
|
| 558 |
+
"eval_steps_per_second": 0.185,
|
| 559 |
+
"step": 3674
|
| 560 |
+
},
|
| 561 |
+
{
|
| 562 |
+
"epoch": 23.0,
|
| 563 |
+
"grad_norm": 3.0958826541900635,
|
| 564 |
+
"learning_rate": 0.0011666666666666668,
|
| 565 |
+
"loss": 1.3182,
|
| 566 |
+
"step": 3841
|
| 567 |
+
},
|
| 568 |
+
{
|
| 569 |
+
"epoch": 23.0,
|
| 570 |
+
"eval_accuracy": 0.5043859649122807,
|
| 571 |
+
"eval_f1_macro": 0.4250257135657705,
|
| 572 |
+
"eval_f1_micro": 0.5043859649122807,
|
| 573 |
+
"eval_f1_weighted": 0.49650300655947666,
|
| 574 |
+
"eval_loss": 1.5448613166809082,
|
| 575 |
+
"eval_precision_macro": 0.4553361061245826,
|
| 576 |
+
"eval_precision_micro": 0.5043859649122807,
|
| 577 |
+
"eval_precision_weighted": 0.5092558110886319,
|
| 578 |
+
"eval_recall_macro": 0.42373341360749667,
|
| 579 |
+
"eval_recall_micro": 0.5043859649122807,
|
| 580 |
+
"eval_recall_weighted": 0.5043859649122807,
|
| 581 |
+
"eval_runtime": 387.3958,
|
| 582 |
+
"eval_samples_per_second": 2.943,
|
| 583 |
+
"eval_steps_per_second": 0.186,
|
| 584 |
+
"step": 3841
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 24.0,
|
| 588 |
+
"grad_norm": 10.11511516571045,
|
| 589 |
+
"learning_rate": 0.001,
|
| 590 |
+
"loss": 1.3203,
|
| 591 |
+
"step": 4008
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 24.0,
|
| 595 |
+
"eval_accuracy": 0.5096491228070176,
|
| 596 |
+
"eval_f1_macro": 0.4226791843362771,
|
| 597 |
+
"eval_f1_micro": 0.5096491228070176,
|
| 598 |
+
"eval_f1_weighted": 0.49650965589318546,
|
| 599 |
+
"eval_loss": 1.5966026782989502,
|
| 600 |
+
"eval_precision_macro": 0.44893183685690363,
|
| 601 |
+
"eval_precision_micro": 0.5096491228070176,
|
| 602 |
+
"eval_precision_weighted": 0.5017696631956957,
|
| 603 |
+
"eval_recall_macro": 0.4174735986452103,
|
| 604 |
+
"eval_recall_micro": 0.5096491228070176,
|
| 605 |
+
"eval_recall_weighted": 0.5096491228070176,
|
| 606 |
+
"eval_runtime": 387.7731,
|
| 607 |
+
"eval_samples_per_second": 2.94,
|
| 608 |
+
"eval_steps_per_second": 0.186,
|
| 609 |
+
"step": 4008
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 25.0,
|
| 613 |
+
"grad_norm": 9.380126953125,
|
| 614 |
+
"learning_rate": 0.0008333333333333333,
|
| 615 |
+
"loss": 1.2721,
|
| 616 |
+
"step": 4175
|
| 617 |
+
},
|
| 618 |
+
{
|
| 619 |
+
"epoch": 25.0,
|
| 620 |
+
"eval_accuracy": 0.5070175438596491,
|
| 621 |
+
"eval_f1_macro": 0.42100594112908074,
|
| 622 |
+
"eval_f1_micro": 0.5070175438596491,
|
| 623 |
+
"eval_f1_weighted": 0.49391608845007934,
|
| 624 |
+
"eval_loss": 1.7380679845809937,
|
| 625 |
+
"eval_precision_macro": 0.45476177682378327,
|
| 626 |
+
"eval_precision_micro": 0.5070175438596491,
|
| 627 |
+
"eval_precision_weighted": 0.5038725473468296,
|
| 628 |
+
"eval_recall_macro": 0.41455121185805693,
|
| 629 |
+
"eval_recall_micro": 0.5070175438596491,
|
| 630 |
+
"eval_recall_weighted": 0.5070175438596491,
|
| 631 |
+
"eval_runtime": 387.0289,
|
| 632 |
+
"eval_samples_per_second": 2.946,
|
| 633 |
+
"eval_steps_per_second": 0.186,
|
| 634 |
+
"step": 4175
|
| 635 |
+
},
|
| 636 |
+
{
|
| 637 |
+
"epoch": 26.0,
|
| 638 |
+
"grad_norm": 9.300443649291992,
|
| 639 |
+
"learning_rate": 0.0006666666666666666,
|
| 640 |
+
"loss": 1.2617,
|
| 641 |
+
"step": 4342
|
| 642 |
+
},
|
| 643 |
+
{
|
| 644 |
+
"epoch": 26.0,
|
| 645 |
+
"eval_accuracy": 0.5035087719298246,
|
| 646 |
+
"eval_f1_macro": 0.4212203299346101,
|
| 647 |
+
"eval_f1_micro": 0.5035087719298246,
|
| 648 |
+
"eval_f1_weighted": 0.4941328967539604,
|
| 649 |
+
"eval_loss": 1.632319450378418,
|
| 650 |
+
"eval_precision_macro": 0.44361812797226563,
|
| 651 |
+
"eval_precision_micro": 0.5035087719298246,
|
| 652 |
+
"eval_precision_weighted": 0.5004178461640878,
|
| 653 |
+
"eval_recall_macro": 0.4138272297786372,
|
| 654 |
+
"eval_recall_micro": 0.5035087719298246,
|
| 655 |
+
"eval_recall_weighted": 0.5035087719298246,
|
| 656 |
+
"eval_runtime": 387.432,
|
| 657 |
+
"eval_samples_per_second": 2.942,
|
| 658 |
+
"eval_steps_per_second": 0.186,
|
| 659 |
+
"step": 4342
|
| 660 |
+
},
|
| 661 |
+
{
|
| 662 |
+
"epoch": 27.0,
|
| 663 |
+
"grad_norm": 6.3707780838012695,
|
| 664 |
+
"learning_rate": 0.0005,
|
| 665 |
+
"loss": 1.232,
|
| 666 |
+
"step": 4509
|
| 667 |
+
},
|
| 668 |
+
{
|
| 669 |
+
"epoch": 27.0,
|
| 670 |
+
"eval_accuracy": 0.5,
|
| 671 |
+
"eval_f1_macro": 0.4177911105472344,
|
| 672 |
+
"eval_f1_micro": 0.5,
|
| 673 |
+
"eval_f1_weighted": 0.489643641514919,
|
| 674 |
+
"eval_loss": 1.658205270767212,
|
| 675 |
+
"eval_precision_macro": 0.4496790851883594,
|
| 676 |
+
"eval_precision_micro": 0.5,
|
| 677 |
+
"eval_precision_weighted": 0.5029776476941141,
|
| 678 |
+
"eval_recall_macro": 0.4138497914465852,
|
| 679 |
+
"eval_recall_micro": 0.5,
|
| 680 |
+
"eval_recall_weighted": 0.5,
|
| 681 |
+
"eval_runtime": 387.4035,
|
| 682 |
+
"eval_samples_per_second": 2.943,
|
| 683 |
+
"eval_steps_per_second": 0.186,
|
| 684 |
+
"step": 4509
|
| 685 |
+
},
|
| 686 |
+
{
|
| 687 |
+
"epoch": 28.0,
|
| 688 |
+
"grad_norm": 10.817475318908691,
|
| 689 |
+
"learning_rate": 0.0003333333333333333,
|
| 690 |
+
"loss": 1.2779,
|
| 691 |
+
"step": 4676
|
| 692 |
+
},
|
| 693 |
+
{
|
| 694 |
+
"epoch": 28.0,
|
| 695 |
+
"eval_accuracy": 0.506140350877193,
|
| 696 |
+
"eval_f1_macro": 0.43333273047004633,
|
| 697 |
+
"eval_f1_micro": 0.506140350877193,
|
| 698 |
+
"eval_f1_weighted": 0.5013645953971182,
|
| 699 |
+
"eval_loss": 1.5920156240463257,
|
| 700 |
+
"eval_precision_macro": 0.45066820886948744,
|
| 701 |
+
"eval_precision_micro": 0.506140350877193,
|
| 702 |
+
"eval_precision_weighted": 0.5098806030910483,
|
| 703 |
+
"eval_recall_macro": 0.4307563083655982,
|
| 704 |
+
"eval_recall_micro": 0.506140350877193,
|
| 705 |
+
"eval_recall_weighted": 0.506140350877193,
|
| 706 |
+
"eval_runtime": 387.7867,
|
| 707 |
+
"eval_samples_per_second": 2.94,
|
| 708 |
+
"eval_steps_per_second": 0.186,
|
| 709 |
+
"step": 4676
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 29.0,
|
| 713 |
+
"grad_norm": 8.93726634979248,
|
| 714 |
+
"learning_rate": 0.00016666666666666666,
|
| 715 |
+
"loss": 1.2478,
|
| 716 |
+
"step": 4843
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 29.0,
|
| 720 |
+
"eval_accuracy": 0.5140350877192983,
|
| 721 |
+
"eval_f1_macro": 0.4370072407808227,
|
| 722 |
+
"eval_f1_micro": 0.5140350877192983,
|
| 723 |
+
"eval_f1_weighted": 0.5079223850593363,
|
| 724 |
+
"eval_loss": 1.6204602718353271,
|
| 725 |
+
"eval_precision_macro": 0.45427166811835185,
|
| 726 |
+
"eval_precision_micro": 0.5140350877192983,
|
| 727 |
+
"eval_precision_weighted": 0.5134498813977569,
|
| 728 |
+
"eval_recall_macro": 0.4348661701787724,
|
| 729 |
+
"eval_recall_micro": 0.5140350877192983,
|
| 730 |
+
"eval_recall_weighted": 0.5140350877192983,
|
| 731 |
+
"eval_runtime": 387.6318,
|
| 732 |
+
"eval_samples_per_second": 2.941,
|
| 733 |
+
"eval_steps_per_second": 0.186,
|
| 734 |
+
"step": 4843
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 30.0,
|
| 738 |
+
"grad_norm": 11.193902015686035,
|
| 739 |
+
"learning_rate": 0.0,
|
| 740 |
+
"loss": 1.2482,
|
| 741 |
+
"step": 5010
|
| 742 |
+
},
|
| 743 |
+
{
|
| 744 |
+
"epoch": 30.0,
|
| 745 |
+
"eval_accuracy": 0.5166666666666667,
|
| 746 |
+
"eval_f1_macro": 0.4422232782296543,
|
| 747 |
+
"eval_f1_micro": 0.5166666666666667,
|
| 748 |
+
"eval_f1_weighted": 0.5099377352687338,
|
| 749 |
+
"eval_loss": 1.6233675479888916,
|
| 750 |
+
"eval_precision_macro": 0.46481125646765853,
|
| 751 |
+
"eval_precision_micro": 0.5166666666666667,
|
| 752 |
+
"eval_precision_weighted": 0.5182626681333475,
|
| 753 |
+
"eval_recall_macro": 0.4380791377289606,
|
| 754 |
+
"eval_recall_micro": 0.5166666666666667,
|
| 755 |
+
"eval_recall_weighted": 0.5166666666666667,
|
| 756 |
+
"eval_runtime": 388.1928,
|
| 757 |
+
"eval_samples_per_second": 2.937,
|
| 758 |
+
"eval_steps_per_second": 0.185,
|
| 759 |
+
"step": 5010
|
| 760 |
+
}
|
| 761 |
+
],
|
| 762 |
+
"logging_steps": 500,
|
| 763 |
+
"max_steps": 5010,
|
| 764 |
+
"num_input_tokens_seen": 0,
|
| 765 |
+
"num_train_epochs": 30,
|
| 766 |
+
"save_steps": 500,
|
| 767 |
+
"stateful_callbacks": {
|
| 768 |
+
"TrainerControl": {
|
| 769 |
+
"args": {
|
| 770 |
+
"should_epoch_stop": false,
|
| 771 |
+
"should_evaluate": false,
|
| 772 |
+
"should_log": false,
|
| 773 |
+
"should_save": true,
|
| 774 |
+
"should_training_stop": true
|
| 775 |
+
},
|
| 776 |
+
"attributes": {}
|
| 777 |
+
}
|
| 778 |
+
},
|
| 779 |
+
"total_flos": 0.0,
|
| 780 |
+
"train_batch_size": 16,
|
| 781 |
+
"trial_name": null,
|
| 782 |
+
"trial_params": null
|
| 783 |
+
}
|
release/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5fdbd74cff7a3058b8263eb2e3b9ffdce201f26358dd3f7b965737a2cf02778b
|
| 3 |
+
size 392738908
|
release/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4b60bf49cfa653ac8ee67cc8f8dd524511edf8879cc03616e46ab9d1360862d6
|
| 3 |
+
size 30462205
|
release/preprocessor_config.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"do_normalize": true,
|
| 3 |
+
"feature_extractor_type": "Wav2Vec2FeatureExtractor",
|
| 4 |
+
"feature_size": 1,
|
| 5 |
+
"padding_side": "right",
|
| 6 |
+
"padding_value": 0.0,
|
| 7 |
+
"return_attention_mask": true,
|
| 8 |
+
"sampling_rate": 16000
|
| 9 |
+
}
|
release/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:eee22b406fc2f6df1c483299aef7cc518d1b36057521369615d78d6c238e26c5
|
| 3 |
+
size 14244
|
release/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:09b99f5e551331099d97260b90e5773355f17e37027489bb3835ef87b5f7d83e
|
| 3 |
+
size 1064
|
release/trainer_state.json
ADDED
|
@@ -0,0 +1,783 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.5166666666666667,
|
| 3 |
+
"best_model_checkpoint": "model/mert_cmp_single/checkpoint-5010",
|
| 4 |
+
"epoch": 30.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 5010,
|
| 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": 1.972764253616333,
|
| 14 |
+
"learning_rate": 0.004833333333333334,
|
| 15 |
+
"loss": 2.0548,
|
| 16 |
+
"step": 167
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 1.0,
|
| 20 |
+
"eval_accuracy": 0.3719298245614035,
|
| 21 |
+
"eval_f1_macro": 0.1752339726946509,
|
| 22 |
+
"eval_f1_micro": 0.3719298245614035,
|
| 23 |
+
"eval_f1_weighted": 0.3032743408676906,
|
| 24 |
+
"eval_loss": 1.8728982210159302,
|
| 25 |
+
"eval_precision_macro": 0.15349976094634282,
|
| 26 |
+
"eval_precision_micro": 0.3719298245614035,
|
| 27 |
+
"eval_precision_weighted": 0.2655915542913314,
|
| 28 |
+
"eval_recall_macro": 0.2199586762280261,
|
| 29 |
+
"eval_recall_micro": 0.3719298245614035,
|
| 30 |
+
"eval_recall_weighted": 0.3719298245614035,
|
| 31 |
+
"eval_runtime": 390.667,
|
| 32 |
+
"eval_samples_per_second": 2.918,
|
| 33 |
+
"eval_steps_per_second": 0.184,
|
| 34 |
+
"step": 167
|
| 35 |
+
},
|
| 36 |
+
{
|
| 37 |
+
"epoch": 2.0,
|
| 38 |
+
"grad_norm": 3.7620842456817627,
|
| 39 |
+
"learning_rate": 0.004666666666666667,
|
| 40 |
+
"loss": 1.8547,
|
| 41 |
+
"step": 334
|
| 42 |
+
},
|
| 43 |
+
{
|
| 44 |
+
"epoch": 2.0,
|
| 45 |
+
"eval_accuracy": 0.38596491228070173,
|
| 46 |
+
"eval_f1_macro": 0.20308576109052026,
|
| 47 |
+
"eval_f1_micro": 0.38596491228070173,
|
| 48 |
+
"eval_f1_weighted": 0.34326825024510915,
|
| 49 |
+
"eval_loss": 1.7156952619552612,
|
| 50 |
+
"eval_precision_macro": 0.21519653478856515,
|
| 51 |
+
"eval_precision_micro": 0.38596491228070173,
|
| 52 |
+
"eval_precision_weighted": 0.34210736396220265,
|
| 53 |
+
"eval_recall_macro": 0.22995235116630916,
|
| 54 |
+
"eval_recall_micro": 0.38596491228070173,
|
| 55 |
+
"eval_recall_weighted": 0.38596491228070173,
|
| 56 |
+
"eval_runtime": 388.1478,
|
| 57 |
+
"eval_samples_per_second": 2.937,
|
| 58 |
+
"eval_steps_per_second": 0.185,
|
| 59 |
+
"step": 334
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 3.0,
|
| 63 |
+
"grad_norm": 2.893453359603882,
|
| 64 |
+
"learning_rate": 0.0045000000000000005,
|
| 65 |
+
"loss": 1.7595,
|
| 66 |
+
"step": 501
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 3.0,
|
| 70 |
+
"eval_accuracy": 0.39035087719298245,
|
| 71 |
+
"eval_f1_macro": 0.27399951358210684,
|
| 72 |
+
"eval_f1_micro": 0.39035087719298245,
|
| 73 |
+
"eval_f1_weighted": 0.3326450313532406,
|
| 74 |
+
"eval_loss": 1.649847388267517,
|
| 75 |
+
"eval_precision_macro": 0.32755489540557126,
|
| 76 |
+
"eval_precision_micro": 0.39035087719298245,
|
| 77 |
+
"eval_precision_weighted": 0.3633738601389556,
|
| 78 |
+
"eval_recall_macro": 0.293769851069751,
|
| 79 |
+
"eval_recall_micro": 0.39035087719298245,
|
| 80 |
+
"eval_recall_weighted": 0.39035087719298245,
|
| 81 |
+
"eval_runtime": 387.7811,
|
| 82 |
+
"eval_samples_per_second": 2.94,
|
| 83 |
+
"eval_steps_per_second": 0.186,
|
| 84 |
+
"step": 501
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 4.0,
|
| 88 |
+
"grad_norm": 4.628668785095215,
|
| 89 |
+
"learning_rate": 0.004333333333333334,
|
| 90 |
+
"loss": 1.6863,
|
| 91 |
+
"step": 668
|
| 92 |
+
},
|
| 93 |
+
{
|
| 94 |
+
"epoch": 4.0,
|
| 95 |
+
"eval_accuracy": 0.4456140350877193,
|
| 96 |
+
"eval_f1_macro": 0.2949654893702652,
|
| 97 |
+
"eval_f1_micro": 0.4456140350877193,
|
| 98 |
+
"eval_f1_weighted": 0.4051374424710547,
|
| 99 |
+
"eval_loss": 1.4840811491012573,
|
| 100 |
+
"eval_precision_macro": 0.2723679270138862,
|
| 101 |
+
"eval_precision_micro": 0.4456140350877193,
|
| 102 |
+
"eval_precision_weighted": 0.3877325579722141,
|
| 103 |
+
"eval_recall_macro": 0.3506862934107516,
|
| 104 |
+
"eval_recall_micro": 0.4456140350877193,
|
| 105 |
+
"eval_recall_weighted": 0.4456140350877193,
|
| 106 |
+
"eval_runtime": 387.302,
|
| 107 |
+
"eval_samples_per_second": 2.943,
|
| 108 |
+
"eval_steps_per_second": 0.186,
|
| 109 |
+
"step": 668
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 5.0,
|
| 113 |
+
"grad_norm": 2.4614219665527344,
|
| 114 |
+
"learning_rate": 0.004166666666666667,
|
| 115 |
+
"loss": 1.6545,
|
| 116 |
+
"step": 835
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 5.0,
|
| 120 |
+
"eval_accuracy": 0.4710526315789474,
|
| 121 |
+
"eval_f1_macro": 0.32061989017679554,
|
| 122 |
+
"eval_f1_micro": 0.4710526315789474,
|
| 123 |
+
"eval_f1_weighted": 0.43804477640411854,
|
| 124 |
+
"eval_loss": 1.5168241262435913,
|
| 125 |
+
"eval_precision_macro": 0.3093839470299629,
|
| 126 |
+
"eval_precision_micro": 0.4710526315789474,
|
| 127 |
+
"eval_precision_weighted": 0.4152776413552219,
|
| 128 |
+
"eval_recall_macro": 0.3463593477939053,
|
| 129 |
+
"eval_recall_micro": 0.4710526315789474,
|
| 130 |
+
"eval_recall_weighted": 0.4710526315789474,
|
| 131 |
+
"eval_runtime": 388.1938,
|
| 132 |
+
"eval_samples_per_second": 2.937,
|
| 133 |
+
"eval_steps_per_second": 0.185,
|
| 134 |
+
"step": 835
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 6.0,
|
| 138 |
+
"grad_norm": 3.0677852630615234,
|
| 139 |
+
"learning_rate": 0.004,
|
| 140 |
+
"loss": 1.6326,
|
| 141 |
+
"step": 1002
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 6.0,
|
| 145 |
+
"eval_accuracy": 0.4324561403508772,
|
| 146 |
+
"eval_f1_macro": 0.3009858659534055,
|
| 147 |
+
"eval_f1_micro": 0.4324561403508772,
|
| 148 |
+
"eval_f1_weighted": 0.40163961306373386,
|
| 149 |
+
"eval_loss": 1.5056780576705933,
|
| 150 |
+
"eval_precision_macro": 0.3533036543942354,
|
| 151 |
+
"eval_precision_micro": 0.4324561403508772,
|
| 152 |
+
"eval_precision_weighted": 0.41979003159359185,
|
| 153 |
+
"eval_recall_macro": 0.33496368714252633,
|
| 154 |
+
"eval_recall_micro": 0.4324561403508772,
|
| 155 |
+
"eval_recall_weighted": 0.4324561403508772,
|
| 156 |
+
"eval_runtime": 389.6423,
|
| 157 |
+
"eval_samples_per_second": 2.926,
|
| 158 |
+
"eval_steps_per_second": 0.185,
|
| 159 |
+
"step": 1002
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 7.0,
|
| 163 |
+
"grad_norm": 4.243473529815674,
|
| 164 |
+
"learning_rate": 0.0038333333333333336,
|
| 165 |
+
"loss": 1.599,
|
| 166 |
+
"step": 1169
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"epoch": 7.0,
|
| 170 |
+
"eval_accuracy": 0.4789473684210526,
|
| 171 |
+
"eval_f1_macro": 0.3417367991830593,
|
| 172 |
+
"eval_f1_micro": 0.4789473684210526,
|
| 173 |
+
"eval_f1_weighted": 0.44204068485349196,
|
| 174 |
+
"eval_loss": 1.485596776008606,
|
| 175 |
+
"eval_precision_macro": 0.35341928050514015,
|
| 176 |
+
"eval_precision_micro": 0.4789473684210526,
|
| 177 |
+
"eval_precision_weighted": 0.43268574425886164,
|
| 178 |
+
"eval_recall_macro": 0.3618233001813015,
|
| 179 |
+
"eval_recall_micro": 0.4789473684210526,
|
| 180 |
+
"eval_recall_weighted": 0.4789473684210526,
|
| 181 |
+
"eval_runtime": 387.4622,
|
| 182 |
+
"eval_samples_per_second": 2.942,
|
| 183 |
+
"eval_steps_per_second": 0.186,
|
| 184 |
+
"step": 1169
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 8.0,
|
| 188 |
+
"grad_norm": 3.4915621280670166,
|
| 189 |
+
"learning_rate": 0.0036666666666666666,
|
| 190 |
+
"loss": 1.5571,
|
| 191 |
+
"step": 1336
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 8.0,
|
| 195 |
+
"eval_accuracy": 0.4649122807017544,
|
| 196 |
+
"eval_f1_macro": 0.33185435154616927,
|
| 197 |
+
"eval_f1_micro": 0.4649122807017544,
|
| 198 |
+
"eval_f1_weighted": 0.4375194270323329,
|
| 199 |
+
"eval_loss": 1.5075509548187256,
|
| 200 |
+
"eval_precision_macro": 0.402563382174571,
|
| 201 |
+
"eval_precision_micro": 0.4649122807017544,
|
| 202 |
+
"eval_precision_weighted": 0.4595468667733386,
|
| 203 |
+
"eval_recall_macro": 0.3376072239589585,
|
| 204 |
+
"eval_recall_micro": 0.4649122807017544,
|
| 205 |
+
"eval_recall_weighted": 0.4649122807017544,
|
| 206 |
+
"eval_runtime": 387.2913,
|
| 207 |
+
"eval_samples_per_second": 2.944,
|
| 208 |
+
"eval_steps_per_second": 0.186,
|
| 209 |
+
"step": 1336
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 9.0,
|
| 213 |
+
"grad_norm": 5.648871421813965,
|
| 214 |
+
"learning_rate": 0.0034999999999999996,
|
| 215 |
+
"loss": 1.5423,
|
| 216 |
+
"step": 1503
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 9.0,
|
| 220 |
+
"eval_accuracy": 0.4473684210526316,
|
| 221 |
+
"eval_f1_macro": 0.3633193692983963,
|
| 222 |
+
"eval_f1_micro": 0.4473684210526316,
|
| 223 |
+
"eval_f1_weighted": 0.43457480187919556,
|
| 224 |
+
"eval_loss": 1.5081905126571655,
|
| 225 |
+
"eval_precision_macro": 0.3975086633004074,
|
| 226 |
+
"eval_precision_micro": 0.4473684210526316,
|
| 227 |
+
"eval_precision_weighted": 0.453531478964179,
|
| 228 |
+
"eval_recall_macro": 0.35673163734767305,
|
| 229 |
+
"eval_recall_micro": 0.4473684210526316,
|
| 230 |
+
"eval_recall_weighted": 0.4473684210526316,
|
| 231 |
+
"eval_runtime": 387.3663,
|
| 232 |
+
"eval_samples_per_second": 2.943,
|
| 233 |
+
"eval_steps_per_second": 0.186,
|
| 234 |
+
"step": 1503
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 10.0,
|
| 238 |
+
"grad_norm": 3.1236555576324463,
|
| 239 |
+
"learning_rate": 0.003333333333333333,
|
| 240 |
+
"loss": 1.5356,
|
| 241 |
+
"step": 1670
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 10.0,
|
| 245 |
+
"eval_accuracy": 0.4649122807017544,
|
| 246 |
+
"eval_f1_macro": 0.32504503507830124,
|
| 247 |
+
"eval_f1_micro": 0.4649122807017544,
|
| 248 |
+
"eval_f1_weighted": 0.4232674688894728,
|
| 249 |
+
"eval_loss": 1.5755454301834106,
|
| 250 |
+
"eval_precision_macro": 0.361163616907195,
|
| 251 |
+
"eval_precision_micro": 0.4649122807017544,
|
| 252 |
+
"eval_precision_weighted": 0.4247075744687934,
|
| 253 |
+
"eval_recall_macro": 0.3384149762412627,
|
| 254 |
+
"eval_recall_micro": 0.4649122807017544,
|
| 255 |
+
"eval_recall_weighted": 0.4649122807017544,
|
| 256 |
+
"eval_runtime": 389.1667,
|
| 257 |
+
"eval_samples_per_second": 2.929,
|
| 258 |
+
"eval_steps_per_second": 0.185,
|
| 259 |
+
"step": 1670
|
| 260 |
+
},
|
| 261 |
+
{
|
| 262 |
+
"epoch": 11.0,
|
| 263 |
+
"grad_norm": 3.6445953845977783,
|
| 264 |
+
"learning_rate": 0.0031666666666666666,
|
| 265 |
+
"loss": 1.5174,
|
| 266 |
+
"step": 1837
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 11.0,
|
| 270 |
+
"eval_accuracy": 0.4666666666666667,
|
| 271 |
+
"eval_f1_macro": 0.35976985093947306,
|
| 272 |
+
"eval_f1_micro": 0.4666666666666667,
|
| 273 |
+
"eval_f1_weighted": 0.4458889953148088,
|
| 274 |
+
"eval_loss": 1.4875001907348633,
|
| 275 |
+
"eval_precision_macro": 0.4802628528097644,
|
| 276 |
+
"eval_precision_micro": 0.4666666666666667,
|
| 277 |
+
"eval_precision_weighted": 0.5174355391927419,
|
| 278 |
+
"eval_recall_macro": 0.3642094502216486,
|
| 279 |
+
"eval_recall_micro": 0.4666666666666667,
|
| 280 |
+
"eval_recall_weighted": 0.4666666666666667,
|
| 281 |
+
"eval_runtime": 388.4896,
|
| 282 |
+
"eval_samples_per_second": 2.934,
|
| 283 |
+
"eval_steps_per_second": 0.185,
|
| 284 |
+
"step": 1837
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 12.0,
|
| 288 |
+
"grad_norm": 3.5447826385498047,
|
| 289 |
+
"learning_rate": 0.003,
|
| 290 |
+
"loss": 1.4717,
|
| 291 |
+
"step": 2004
|
| 292 |
+
},
|
| 293 |
+
{
|
| 294 |
+
"epoch": 12.0,
|
| 295 |
+
"eval_accuracy": 0.4666666666666667,
|
| 296 |
+
"eval_f1_macro": 0.36601636096965706,
|
| 297 |
+
"eval_f1_micro": 0.4666666666666667,
|
| 298 |
+
"eval_f1_weighted": 0.44027602426120266,
|
| 299 |
+
"eval_loss": 1.4927704334259033,
|
| 300 |
+
"eval_precision_macro": 0.43193780369068935,
|
| 301 |
+
"eval_precision_micro": 0.4666666666666667,
|
| 302 |
+
"eval_precision_weighted": 0.47269213977673047,
|
| 303 |
+
"eval_recall_macro": 0.3798912460539077,
|
| 304 |
+
"eval_recall_micro": 0.4666666666666667,
|
| 305 |
+
"eval_recall_weighted": 0.4666666666666667,
|
| 306 |
+
"eval_runtime": 387.823,
|
| 307 |
+
"eval_samples_per_second": 2.939,
|
| 308 |
+
"eval_steps_per_second": 0.186,
|
| 309 |
+
"step": 2004
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 13.0,
|
| 313 |
+
"grad_norm": 4.328464508056641,
|
| 314 |
+
"learning_rate": 0.002833333333333333,
|
| 315 |
+
"loss": 1.4669,
|
| 316 |
+
"step": 2171
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 13.0,
|
| 320 |
+
"eval_accuracy": 0.4745614035087719,
|
| 321 |
+
"eval_f1_macro": 0.38859755562541454,
|
| 322 |
+
"eval_f1_micro": 0.4745614035087719,
|
| 323 |
+
"eval_f1_weighted": 0.46671787231935036,
|
| 324 |
+
"eval_loss": 1.4511743783950806,
|
| 325 |
+
"eval_precision_macro": 0.4015529868639453,
|
| 326 |
+
"eval_precision_micro": 0.4745614035087719,
|
| 327 |
+
"eval_precision_weighted": 0.4732841181218933,
|
| 328 |
+
"eval_recall_macro": 0.3975362907769957,
|
| 329 |
+
"eval_recall_micro": 0.4745614035087719,
|
| 330 |
+
"eval_recall_weighted": 0.4745614035087719,
|
| 331 |
+
"eval_runtime": 387.3991,
|
| 332 |
+
"eval_samples_per_second": 2.943,
|
| 333 |
+
"eval_steps_per_second": 0.186,
|
| 334 |
+
"step": 2171
|
| 335 |
+
},
|
| 336 |
+
{
|
| 337 |
+
"epoch": 14.0,
|
| 338 |
+
"grad_norm": 8.561325073242188,
|
| 339 |
+
"learning_rate": 0.0026666666666666666,
|
| 340 |
+
"loss": 1.4516,
|
| 341 |
+
"step": 2338
|
| 342 |
+
},
|
| 343 |
+
{
|
| 344 |
+
"epoch": 14.0,
|
| 345 |
+
"eval_accuracy": 0.47719298245614034,
|
| 346 |
+
"eval_f1_macro": 0.38778824262319195,
|
| 347 |
+
"eval_f1_micro": 0.47719298245614034,
|
| 348 |
+
"eval_f1_weighted": 0.4521071312833841,
|
| 349 |
+
"eval_loss": 1.5683324337005615,
|
| 350 |
+
"eval_precision_macro": 0.45015826319899144,
|
| 351 |
+
"eval_precision_micro": 0.47719298245614034,
|
| 352 |
+
"eval_precision_weighted": 0.48469205065002274,
|
| 353 |
+
"eval_recall_macro": 0.3807191719284666,
|
| 354 |
+
"eval_recall_micro": 0.47719298245614034,
|
| 355 |
+
"eval_recall_weighted": 0.47719298245614034,
|
| 356 |
+
"eval_runtime": 387.3112,
|
| 357 |
+
"eval_samples_per_second": 2.943,
|
| 358 |
+
"eval_steps_per_second": 0.186,
|
| 359 |
+
"step": 2338
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 15.0,
|
| 363 |
+
"grad_norm": 5.659043788909912,
|
| 364 |
+
"learning_rate": 0.0025,
|
| 365 |
+
"loss": 1.4468,
|
| 366 |
+
"step": 2505
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 15.0,
|
| 370 |
+
"eval_accuracy": 0.4649122807017544,
|
| 371 |
+
"eval_f1_macro": 0.3746786393581578,
|
| 372 |
+
"eval_f1_micro": 0.4649122807017544,
|
| 373 |
+
"eval_f1_weighted": 0.4490106368369649,
|
| 374 |
+
"eval_loss": 1.522507905960083,
|
| 375 |
+
"eval_precision_macro": 0.42440688015824424,
|
| 376 |
+
"eval_precision_micro": 0.4649122807017544,
|
| 377 |
+
"eval_precision_weighted": 0.48542408043393825,
|
| 378 |
+
"eval_recall_macro": 0.3792014977295045,
|
| 379 |
+
"eval_recall_micro": 0.4649122807017544,
|
| 380 |
+
"eval_recall_weighted": 0.4649122807017544,
|
| 381 |
+
"eval_runtime": 387.3309,
|
| 382 |
+
"eval_samples_per_second": 2.943,
|
| 383 |
+
"eval_steps_per_second": 0.186,
|
| 384 |
+
"step": 2505
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 16.0,
|
| 388 |
+
"grad_norm": 3.6838574409484863,
|
| 389 |
+
"learning_rate": 0.0023333333333333335,
|
| 390 |
+
"loss": 1.3935,
|
| 391 |
+
"step": 2672
|
| 392 |
+
},
|
| 393 |
+
{
|
| 394 |
+
"epoch": 16.0,
|
| 395 |
+
"eval_accuracy": 0.4850877192982456,
|
| 396 |
+
"eval_f1_macro": 0.41079129743155646,
|
| 397 |
+
"eval_f1_micro": 0.4850877192982456,
|
| 398 |
+
"eval_f1_weighted": 0.4818917262202846,
|
| 399 |
+
"eval_loss": 1.4960697889328003,
|
| 400 |
+
"eval_precision_macro": 0.42311229328764643,
|
| 401 |
+
"eval_precision_micro": 0.4850877192982456,
|
| 402 |
+
"eval_precision_weighted": 0.49189296461866266,
|
| 403 |
+
"eval_recall_macro": 0.4115979824480055,
|
| 404 |
+
"eval_recall_micro": 0.4850877192982456,
|
| 405 |
+
"eval_recall_weighted": 0.4850877192982456,
|
| 406 |
+
"eval_runtime": 386.9296,
|
| 407 |
+
"eval_samples_per_second": 2.946,
|
| 408 |
+
"eval_steps_per_second": 0.186,
|
| 409 |
+
"step": 2672
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 17.0,
|
| 413 |
+
"grad_norm": 6.394424915313721,
|
| 414 |
+
"learning_rate": 0.002166666666666667,
|
| 415 |
+
"loss": 1.4217,
|
| 416 |
+
"step": 2839
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 17.0,
|
| 420 |
+
"eval_accuracy": 0.48771929824561405,
|
| 421 |
+
"eval_f1_macro": 0.38735566453611514,
|
| 422 |
+
"eval_f1_micro": 0.48771929824561405,
|
| 423 |
+
"eval_f1_weighted": 0.4661251566654804,
|
| 424 |
+
"eval_loss": 1.524192214012146,
|
| 425 |
+
"eval_precision_macro": 0.4543356420217285,
|
| 426 |
+
"eval_precision_micro": 0.48771929824561405,
|
| 427 |
+
"eval_precision_weighted": 0.5054670595798253,
|
| 428 |
+
"eval_recall_macro": 0.4000549048543122,
|
| 429 |
+
"eval_recall_micro": 0.48771929824561405,
|
| 430 |
+
"eval_recall_weighted": 0.48771929824561405,
|
| 431 |
+
"eval_runtime": 387.3753,
|
| 432 |
+
"eval_samples_per_second": 2.943,
|
| 433 |
+
"eval_steps_per_second": 0.186,
|
| 434 |
+
"step": 2839
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 18.0,
|
| 438 |
+
"grad_norm": 5.048763751983643,
|
| 439 |
+
"learning_rate": 0.002,
|
| 440 |
+
"loss": 1.3831,
|
| 441 |
+
"step": 3006
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 18.0,
|
| 445 |
+
"eval_accuracy": 0.49298245614035086,
|
| 446 |
+
"eval_f1_macro": 0.3945057051668317,
|
| 447 |
+
"eval_f1_micro": 0.49298245614035086,
|
| 448 |
+
"eval_f1_weighted": 0.4768674757940579,
|
| 449 |
+
"eval_loss": 1.4723140001296997,
|
| 450 |
+
"eval_precision_macro": 0.45564765495926707,
|
| 451 |
+
"eval_precision_micro": 0.49298245614035086,
|
| 452 |
+
"eval_precision_weighted": 0.5182791073380009,
|
| 453 |
+
"eval_recall_macro": 0.4229398794166392,
|
| 454 |
+
"eval_recall_micro": 0.49298245614035086,
|
| 455 |
+
"eval_recall_weighted": 0.49298245614035086,
|
| 456 |
+
"eval_runtime": 387.6312,
|
| 457 |
+
"eval_samples_per_second": 2.941,
|
| 458 |
+
"eval_steps_per_second": 0.186,
|
| 459 |
+
"step": 3006
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 19.0,
|
| 463 |
+
"grad_norm": 8.706170082092285,
|
| 464 |
+
"learning_rate": 0.0018333333333333333,
|
| 465 |
+
"loss": 1.4249,
|
| 466 |
+
"step": 3173
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 19.0,
|
| 470 |
+
"eval_accuracy": 0.5026315789473684,
|
| 471 |
+
"eval_f1_macro": 0.3967450923381979,
|
| 472 |
+
"eval_f1_micro": 0.5026315789473684,
|
| 473 |
+
"eval_f1_weighted": 0.48467826189256524,
|
| 474 |
+
"eval_loss": 1.4886491298675537,
|
| 475 |
+
"eval_precision_macro": 0.4442558168809208,
|
| 476 |
+
"eval_precision_micro": 0.5026315789473684,
|
| 477 |
+
"eval_precision_weighted": 0.49883534500451954,
|
| 478 |
+
"eval_recall_macro": 0.39853885191451394,
|
| 479 |
+
"eval_recall_micro": 0.5026315789473684,
|
| 480 |
+
"eval_recall_weighted": 0.5026315789473684,
|
| 481 |
+
"eval_runtime": 387.008,
|
| 482 |
+
"eval_samples_per_second": 2.946,
|
| 483 |
+
"eval_steps_per_second": 0.186,
|
| 484 |
+
"step": 3173
|
| 485 |
+
},
|
| 486 |
+
{
|
| 487 |
+
"epoch": 20.0,
|
| 488 |
+
"grad_norm": 9.670186042785645,
|
| 489 |
+
"learning_rate": 0.0016666666666666666,
|
| 490 |
+
"loss": 1.3931,
|
| 491 |
+
"step": 3340
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 20.0,
|
| 495 |
+
"eval_accuracy": 0.49122807017543857,
|
| 496 |
+
"eval_f1_macro": 0.4000545127776559,
|
| 497 |
+
"eval_f1_micro": 0.49122807017543857,
|
| 498 |
+
"eval_f1_weighted": 0.47950452147445755,
|
| 499 |
+
"eval_loss": 1.5036280155181885,
|
| 500 |
+
"eval_precision_macro": 0.443839992334728,
|
| 501 |
+
"eval_precision_micro": 0.49122807017543857,
|
| 502 |
+
"eval_precision_weighted": 0.499565206319528,
|
| 503 |
+
"eval_recall_macro": 0.398944898086383,
|
| 504 |
+
"eval_recall_micro": 0.49122807017543857,
|
| 505 |
+
"eval_recall_weighted": 0.49122807017543857,
|
| 506 |
+
"eval_runtime": 387.8055,
|
| 507 |
+
"eval_samples_per_second": 2.94,
|
| 508 |
+
"eval_steps_per_second": 0.186,
|
| 509 |
+
"step": 3340
|
| 510 |
+
},
|
| 511 |
+
{
|
| 512 |
+
"epoch": 21.0,
|
| 513 |
+
"grad_norm": 3.482584238052368,
|
| 514 |
+
"learning_rate": 0.0015,
|
| 515 |
+
"loss": 1.3695,
|
| 516 |
+
"step": 3507
|
| 517 |
+
},
|
| 518 |
+
{
|
| 519 |
+
"epoch": 21.0,
|
| 520 |
+
"eval_accuracy": 0.4675438596491228,
|
| 521 |
+
"eval_f1_macro": 0.37912447794813103,
|
| 522 |
+
"eval_f1_micro": 0.4675438596491228,
|
| 523 |
+
"eval_f1_weighted": 0.4472561141761033,
|
| 524 |
+
"eval_loss": 1.5602822303771973,
|
| 525 |
+
"eval_precision_macro": 0.409817315097036,
|
| 526 |
+
"eval_precision_micro": 0.4675438596491228,
|
| 527 |
+
"eval_precision_weighted": 0.46947315014830876,
|
| 528 |
+
"eval_recall_macro": 0.39767343278642575,
|
| 529 |
+
"eval_recall_micro": 0.4675438596491228,
|
| 530 |
+
"eval_recall_weighted": 0.4675438596491228,
|
| 531 |
+
"eval_runtime": 387.5678,
|
| 532 |
+
"eval_samples_per_second": 2.941,
|
| 533 |
+
"eval_steps_per_second": 0.186,
|
| 534 |
+
"step": 3507
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 22.0,
|
| 538 |
+
"grad_norm": 5.20711088180542,
|
| 539 |
+
"learning_rate": 0.0013333333333333333,
|
| 540 |
+
"loss": 1.3424,
|
| 541 |
+
"step": 3674
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 22.0,
|
| 545 |
+
"eval_accuracy": 0.5026315789473684,
|
| 546 |
+
"eval_f1_macro": 0.41946040337720514,
|
| 547 |
+
"eval_f1_micro": 0.5026315789473684,
|
| 548 |
+
"eval_f1_weighted": 0.4942558483892827,
|
| 549 |
+
"eval_loss": 1.5268137454986572,
|
| 550 |
+
"eval_precision_macro": 0.45115966752712633,
|
| 551 |
+
"eval_precision_micro": 0.5026315789473684,
|
| 552 |
+
"eval_precision_weighted": 0.5051162872222296,
|
| 553 |
+
"eval_recall_macro": 0.41687400337731273,
|
| 554 |
+
"eval_recall_micro": 0.5026315789473684,
|
| 555 |
+
"eval_recall_weighted": 0.5026315789473684,
|
| 556 |
+
"eval_runtime": 389.0165,
|
| 557 |
+
"eval_samples_per_second": 2.93,
|
| 558 |
+
"eval_steps_per_second": 0.185,
|
| 559 |
+
"step": 3674
|
| 560 |
+
},
|
| 561 |
+
{
|
| 562 |
+
"epoch": 23.0,
|
| 563 |
+
"grad_norm": 3.0958826541900635,
|
| 564 |
+
"learning_rate": 0.0011666666666666668,
|
| 565 |
+
"loss": 1.3182,
|
| 566 |
+
"step": 3841
|
| 567 |
+
},
|
| 568 |
+
{
|
| 569 |
+
"epoch": 23.0,
|
| 570 |
+
"eval_accuracy": 0.5043859649122807,
|
| 571 |
+
"eval_f1_macro": 0.4250257135657705,
|
| 572 |
+
"eval_f1_micro": 0.5043859649122807,
|
| 573 |
+
"eval_f1_weighted": 0.49650300655947666,
|
| 574 |
+
"eval_loss": 1.5448613166809082,
|
| 575 |
+
"eval_precision_macro": 0.4553361061245826,
|
| 576 |
+
"eval_precision_micro": 0.5043859649122807,
|
| 577 |
+
"eval_precision_weighted": 0.5092558110886319,
|
| 578 |
+
"eval_recall_macro": 0.42373341360749667,
|
| 579 |
+
"eval_recall_micro": 0.5043859649122807,
|
| 580 |
+
"eval_recall_weighted": 0.5043859649122807,
|
| 581 |
+
"eval_runtime": 387.3958,
|
| 582 |
+
"eval_samples_per_second": 2.943,
|
| 583 |
+
"eval_steps_per_second": 0.186,
|
| 584 |
+
"step": 3841
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 24.0,
|
| 588 |
+
"grad_norm": 10.11511516571045,
|
| 589 |
+
"learning_rate": 0.001,
|
| 590 |
+
"loss": 1.3203,
|
| 591 |
+
"step": 4008
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 24.0,
|
| 595 |
+
"eval_accuracy": 0.5096491228070176,
|
| 596 |
+
"eval_f1_macro": 0.4226791843362771,
|
| 597 |
+
"eval_f1_micro": 0.5096491228070176,
|
| 598 |
+
"eval_f1_weighted": 0.49650965589318546,
|
| 599 |
+
"eval_loss": 1.5966026782989502,
|
| 600 |
+
"eval_precision_macro": 0.44893183685690363,
|
| 601 |
+
"eval_precision_micro": 0.5096491228070176,
|
| 602 |
+
"eval_precision_weighted": 0.5017696631956957,
|
| 603 |
+
"eval_recall_macro": 0.4174735986452103,
|
| 604 |
+
"eval_recall_micro": 0.5096491228070176,
|
| 605 |
+
"eval_recall_weighted": 0.5096491228070176,
|
| 606 |
+
"eval_runtime": 387.7731,
|
| 607 |
+
"eval_samples_per_second": 2.94,
|
| 608 |
+
"eval_steps_per_second": 0.186,
|
| 609 |
+
"step": 4008
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 25.0,
|
| 613 |
+
"grad_norm": 9.380126953125,
|
| 614 |
+
"learning_rate": 0.0008333333333333333,
|
| 615 |
+
"loss": 1.2721,
|
| 616 |
+
"step": 4175
|
| 617 |
+
},
|
| 618 |
+
{
|
| 619 |
+
"epoch": 25.0,
|
| 620 |
+
"eval_accuracy": 0.5070175438596491,
|
| 621 |
+
"eval_f1_macro": 0.42100594112908074,
|
| 622 |
+
"eval_f1_micro": 0.5070175438596491,
|
| 623 |
+
"eval_f1_weighted": 0.49391608845007934,
|
| 624 |
+
"eval_loss": 1.7380679845809937,
|
| 625 |
+
"eval_precision_macro": 0.45476177682378327,
|
| 626 |
+
"eval_precision_micro": 0.5070175438596491,
|
| 627 |
+
"eval_precision_weighted": 0.5038725473468296,
|
| 628 |
+
"eval_recall_macro": 0.41455121185805693,
|
| 629 |
+
"eval_recall_micro": 0.5070175438596491,
|
| 630 |
+
"eval_recall_weighted": 0.5070175438596491,
|
| 631 |
+
"eval_runtime": 387.0289,
|
| 632 |
+
"eval_samples_per_second": 2.946,
|
| 633 |
+
"eval_steps_per_second": 0.186,
|
| 634 |
+
"step": 4175
|
| 635 |
+
},
|
| 636 |
+
{
|
| 637 |
+
"epoch": 26.0,
|
| 638 |
+
"grad_norm": 9.300443649291992,
|
| 639 |
+
"learning_rate": 0.0006666666666666666,
|
| 640 |
+
"loss": 1.2617,
|
| 641 |
+
"step": 4342
|
| 642 |
+
},
|
| 643 |
+
{
|
| 644 |
+
"epoch": 26.0,
|
| 645 |
+
"eval_accuracy": 0.5035087719298246,
|
| 646 |
+
"eval_f1_macro": 0.4212203299346101,
|
| 647 |
+
"eval_f1_micro": 0.5035087719298246,
|
| 648 |
+
"eval_f1_weighted": 0.4941328967539604,
|
| 649 |
+
"eval_loss": 1.632319450378418,
|
| 650 |
+
"eval_precision_macro": 0.44361812797226563,
|
| 651 |
+
"eval_precision_micro": 0.5035087719298246,
|
| 652 |
+
"eval_precision_weighted": 0.5004178461640878,
|
| 653 |
+
"eval_recall_macro": 0.4138272297786372,
|
| 654 |
+
"eval_recall_micro": 0.5035087719298246,
|
| 655 |
+
"eval_recall_weighted": 0.5035087719298246,
|
| 656 |
+
"eval_runtime": 387.432,
|
| 657 |
+
"eval_samples_per_second": 2.942,
|
| 658 |
+
"eval_steps_per_second": 0.186,
|
| 659 |
+
"step": 4342
|
| 660 |
+
},
|
| 661 |
+
{
|
| 662 |
+
"epoch": 27.0,
|
| 663 |
+
"grad_norm": 6.3707780838012695,
|
| 664 |
+
"learning_rate": 0.0005,
|
| 665 |
+
"loss": 1.232,
|
| 666 |
+
"step": 4509
|
| 667 |
+
},
|
| 668 |
+
{
|
| 669 |
+
"epoch": 27.0,
|
| 670 |
+
"eval_accuracy": 0.5,
|
| 671 |
+
"eval_f1_macro": 0.4177911105472344,
|
| 672 |
+
"eval_f1_micro": 0.5,
|
| 673 |
+
"eval_f1_weighted": 0.489643641514919,
|
| 674 |
+
"eval_loss": 1.658205270767212,
|
| 675 |
+
"eval_precision_macro": 0.4496790851883594,
|
| 676 |
+
"eval_precision_micro": 0.5,
|
| 677 |
+
"eval_precision_weighted": 0.5029776476941141,
|
| 678 |
+
"eval_recall_macro": 0.4138497914465852,
|
| 679 |
+
"eval_recall_micro": 0.5,
|
| 680 |
+
"eval_recall_weighted": 0.5,
|
| 681 |
+
"eval_runtime": 387.4035,
|
| 682 |
+
"eval_samples_per_second": 2.943,
|
| 683 |
+
"eval_steps_per_second": 0.186,
|
| 684 |
+
"step": 4509
|
| 685 |
+
},
|
| 686 |
+
{
|
| 687 |
+
"epoch": 28.0,
|
| 688 |
+
"grad_norm": 10.817475318908691,
|
| 689 |
+
"learning_rate": 0.0003333333333333333,
|
| 690 |
+
"loss": 1.2779,
|
| 691 |
+
"step": 4676
|
| 692 |
+
},
|
| 693 |
+
{
|
| 694 |
+
"epoch": 28.0,
|
| 695 |
+
"eval_accuracy": 0.506140350877193,
|
| 696 |
+
"eval_f1_macro": 0.43333273047004633,
|
| 697 |
+
"eval_f1_micro": 0.506140350877193,
|
| 698 |
+
"eval_f1_weighted": 0.5013645953971182,
|
| 699 |
+
"eval_loss": 1.5920156240463257,
|
| 700 |
+
"eval_precision_macro": 0.45066820886948744,
|
| 701 |
+
"eval_precision_micro": 0.506140350877193,
|
| 702 |
+
"eval_precision_weighted": 0.5098806030910483,
|
| 703 |
+
"eval_recall_macro": 0.4307563083655982,
|
| 704 |
+
"eval_recall_micro": 0.506140350877193,
|
| 705 |
+
"eval_recall_weighted": 0.506140350877193,
|
| 706 |
+
"eval_runtime": 387.7867,
|
| 707 |
+
"eval_samples_per_second": 2.94,
|
| 708 |
+
"eval_steps_per_second": 0.186,
|
| 709 |
+
"step": 4676
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 29.0,
|
| 713 |
+
"grad_norm": 8.93726634979248,
|
| 714 |
+
"learning_rate": 0.00016666666666666666,
|
| 715 |
+
"loss": 1.2478,
|
| 716 |
+
"step": 4843
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 29.0,
|
| 720 |
+
"eval_accuracy": 0.5140350877192983,
|
| 721 |
+
"eval_f1_macro": 0.4370072407808227,
|
| 722 |
+
"eval_f1_micro": 0.5140350877192983,
|
| 723 |
+
"eval_f1_weighted": 0.5079223850593363,
|
| 724 |
+
"eval_loss": 1.6204602718353271,
|
| 725 |
+
"eval_precision_macro": 0.45427166811835185,
|
| 726 |
+
"eval_precision_micro": 0.5140350877192983,
|
| 727 |
+
"eval_precision_weighted": 0.5134498813977569,
|
| 728 |
+
"eval_recall_macro": 0.4348661701787724,
|
| 729 |
+
"eval_recall_micro": 0.5140350877192983,
|
| 730 |
+
"eval_recall_weighted": 0.5140350877192983,
|
| 731 |
+
"eval_runtime": 387.6318,
|
| 732 |
+
"eval_samples_per_second": 2.941,
|
| 733 |
+
"eval_steps_per_second": 0.186,
|
| 734 |
+
"step": 4843
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 30.0,
|
| 738 |
+
"grad_norm": 11.193902015686035,
|
| 739 |
+
"learning_rate": 0.0,
|
| 740 |
+
"loss": 1.2482,
|
| 741 |
+
"step": 5010
|
| 742 |
+
},
|
| 743 |
+
{
|
| 744 |
+
"epoch": 30.0,
|
| 745 |
+
"eval_accuracy": 0.5166666666666667,
|
| 746 |
+
"eval_f1_macro": 0.4422232782296543,
|
| 747 |
+
"eval_f1_micro": 0.5166666666666667,
|
| 748 |
+
"eval_f1_weighted": 0.5099377352687338,
|
| 749 |
+
"eval_loss": 1.6233675479888916,
|
| 750 |
+
"eval_precision_macro": 0.46481125646765853,
|
| 751 |
+
"eval_precision_micro": 0.5166666666666667,
|
| 752 |
+
"eval_precision_weighted": 0.5182626681333475,
|
| 753 |
+
"eval_recall_macro": 0.4380791377289606,
|
| 754 |
+
"eval_recall_micro": 0.5166666666666667,
|
| 755 |
+
"eval_recall_weighted": 0.5166666666666667,
|
| 756 |
+
"eval_runtime": 388.1928,
|
| 757 |
+
"eval_samples_per_second": 2.937,
|
| 758 |
+
"eval_steps_per_second": 0.185,
|
| 759 |
+
"step": 5010
|
| 760 |
+
}
|
| 761 |
+
],
|
| 762 |
+
"logging_steps": 500,
|
| 763 |
+
"max_steps": 5010,
|
| 764 |
+
"num_input_tokens_seen": 0,
|
| 765 |
+
"num_train_epochs": 30,
|
| 766 |
+
"save_steps": 500,
|
| 767 |
+
"stateful_callbacks": {
|
| 768 |
+
"TrainerControl": {
|
| 769 |
+
"args": {
|
| 770 |
+
"should_epoch_stop": false,
|
| 771 |
+
"should_evaluate": false,
|
| 772 |
+
"should_log": false,
|
| 773 |
+
"should_save": true,
|
| 774 |
+
"should_training_stop": true
|
| 775 |
+
},
|
| 776 |
+
"attributes": {}
|
| 777 |
+
}
|
| 778 |
+
},
|
| 779 |
+
"total_flos": 0.0,
|
| 780 |
+
"train_batch_size": 16,
|
| 781 |
+
"trial_name": null,
|
| 782 |
+
"trial_params": null
|
| 783 |
+
}
|
release/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:12e3bed5008004956a38638a255eb3800fff5090c52841ce77f48feecb30343b
|
| 3 |
+
size 5176
|
trainer_state.json
ADDED
|
@@ -0,0 +1,67 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.08695652173913043,
|
| 3 |
+
"best_model_checkpoint": "model/mert_cmp_single/checkpoint-4",
|
| 4 |
+
"epoch": 1.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 4,
|
| 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": 3.237967014312744,
|
| 14 |
+
"learning_rate": 0.0,
|
| 15 |
+
"loss": 2.2433,
|
| 16 |
+
"step": 4
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 1.0,
|
| 20 |
+
"eval_accuracy": 0.08695652173913043,
|
| 21 |
+
"eval_f1_macro": 0.02,
|
| 22 |
+
"eval_f1_micro": 0.08695652173913043,
|
| 23 |
+
"eval_f1_weighted": 0.01391304347826087,
|
| 24 |
+
"eval_loss": 2.322828531265259,
|
| 25 |
+
"eval_precision_macro": 0.010869565217391304,
|
| 26 |
+
"eval_precision_micro": 0.08695652173913043,
|
| 27 |
+
"eval_precision_weighted": 0.007561436672967864,
|
| 28 |
+
"eval_recall_macro": 0.125,
|
| 29 |
+
"eval_recall_micro": 0.08695652173913043,
|
| 30 |
+
"eval_recall_weighted": 0.08695652173913043,
|
| 31 |
+
"eval_runtime": 8.2286,
|
| 32 |
+
"eval_samples_per_second": 2.795,
|
| 33 |
+
"eval_steps_per_second": 0.243,
|
| 34 |
+
"step": 4
|
| 35 |
+
},
|
| 36 |
+
{
|
| 37 |
+
"epoch": 1.0,
|
| 38 |
+
"step": 4,
|
| 39 |
+
"total_flos": 0.0,
|
| 40 |
+
"train_loss": 2.243255853652954,
|
| 41 |
+
"train_runtime": 35.1877,
|
| 42 |
+
"train_samples_per_second": 1.449,
|
| 43 |
+
"train_steps_per_second": 0.114
|
| 44 |
+
}
|
| 45 |
+
],
|
| 46 |
+
"logging_steps": 500,
|
| 47 |
+
"max_steps": 4,
|
| 48 |
+
"num_input_tokens_seen": 0,
|
| 49 |
+
"num_train_epochs": 1,
|
| 50 |
+
"save_steps": 500,
|
| 51 |
+
"stateful_callbacks": {
|
| 52 |
+
"TrainerControl": {
|
| 53 |
+
"args": {
|
| 54 |
+
"should_epoch_stop": false,
|
| 55 |
+
"should_evaluate": false,
|
| 56 |
+
"should_log": false,
|
| 57 |
+
"should_save": true,
|
| 58 |
+
"should_training_stop": true
|
| 59 |
+
},
|
| 60 |
+
"attributes": {}
|
| 61 |
+
}
|
| 62 |
+
},
|
| 63 |
+
"total_flos": 0.0,
|
| 64 |
+
"train_batch_size": 16,
|
| 65 |
+
"trial_name": null,
|
| 66 |
+
"trial_params": null
|
| 67 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:942f9785a129cbea37b4679f4a7bcef4fefb592f356047e5c0984175d49f35f3
|
| 3 |
+
size 5176
|