Commit
•
fe748eb
1
Parent(s):
c035fdd
Upload 6 files
Browse files- README.md +97 -0
- all_results.json +14 -0
- eval_results.json +10 -0
- train_results.json +7 -0
- trainer_state.json +552 -0
- training_args.bin +3 -0
README.md
ADDED
|
@@ -0,0 +1,97 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
tags:
|
| 4 |
+
- audio-classification
|
| 5 |
+
- generated_from_trainer
|
| 6 |
+
metrics:
|
| 7 |
+
- accuracy
|
| 8 |
+
- precision
|
| 9 |
+
- f1
|
| 10 |
+
model-index:
|
| 11 |
+
- name: wav2vec2-large
|
| 12 |
+
results: []
|
| 13 |
+
---
|
| 14 |
+
|
| 15 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 16 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 17 |
+
|
| 18 |
+
# wav2vec2-large
|
| 19 |
+
|
| 20 |
+
This model is a fine-tuned version of [facebook/wav2vec2-large](https://huggingface.co/facebook/wav2vec2-large) on the galsenai/waxal_dataset dataset.
|
| 21 |
+
It achieves the following results on the evaluation set:
|
| 22 |
+
- Loss: 0.3413
|
| 23 |
+
- Accuracy: 0.9443
|
| 24 |
+
- Precision: 0.9780
|
| 25 |
+
- F1: 0.9604
|
| 26 |
+
|
| 27 |
+
## Model description
|
| 28 |
+
|
| 29 |
+
More information needed
|
| 30 |
+
|
| 31 |
+
## Intended uses & limitations
|
| 32 |
+
|
| 33 |
+
More information needed
|
| 34 |
+
|
| 35 |
+
## Training and evaluation data
|
| 36 |
+
|
| 37 |
+
More information needed
|
| 38 |
+
|
| 39 |
+
## Training procedure
|
| 40 |
+
|
| 41 |
+
### Training hyperparameters
|
| 42 |
+
|
| 43 |
+
The following hyperparameters were used during training:
|
| 44 |
+
- learning_rate: 3e-05
|
| 45 |
+
- train_batch_size: 12
|
| 46 |
+
- eval_batch_size: 12
|
| 47 |
+
- seed: 0
|
| 48 |
+
- gradient_accumulation_steps: 4
|
| 49 |
+
- total_train_batch_size: 48
|
| 50 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 51 |
+
- lr_scheduler_type: linear
|
| 52 |
+
- lr_scheduler_warmup_ratio: 0.1
|
| 53 |
+
- num_epochs: 32.0
|
| 54 |
+
|
| 55 |
+
### Training results
|
| 56 |
+
|
| 57 |
+
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | F1 |
|
| 58 |
+
|:-------------:|:-----:|:-----:|:---------------:|:--------:|:---------:|:------:|
|
| 59 |
+
| 4.6314 | 1.01 | 500 | 4.9165 | 0.0205 | 0.0028 | 0.0049 |
|
| 60 |
+
| 3.7739 | 2.02 | 1000 | 4.4491 | 0.0356 | 0.0750 | 0.0252 |
|
| 61 |
+
| 2.5035 | 3.04 | 1500 | 4.1429 | 0.1129 | 0.2672 | 0.1114 |
|
| 62 |
+
| 1.5633 | 4.05 | 2000 | 3.1973 | 0.3676 | 0.6598 | 0.3830 |
|
| 63 |
+
| 1.0538 | 5.06 | 2500 | 2.5479 | 0.5889 | 0.8417 | 0.6557 |
|
| 64 |
+
| 0.7422 | 6.07 | 3000 | 1.4494 | 0.7825 | 0.8921 | 0.8194 |
|
| 65 |
+
| 0.5762 | 7.08 | 3500 | 1.3168 | 0.7726 | 0.9277 | 0.8267 |
|
| 66 |
+
| 0.46 | 8.1 | 4000 | 0.8783 | 0.8564 | 0.9532 | 0.8982 |
|
| 67 |
+
| 0.4007 | 9.11 | 4500 | 0.7524 | 0.8738 | 0.9637 | 0.9137 |
|
| 68 |
+
| 0.3374 | 10.12 | 5000 | 0.6386 | 0.8852 | 0.9678 | 0.9221 |
|
| 69 |
+
| 0.3108 | 11.13 | 5500 | 0.5049 | 0.9106 | 0.9681 | 0.9373 |
|
| 70 |
+
| 0.2735 | 12.15 | 6000 | 0.6097 | 0.8905 | 0.9624 | 0.9226 |
|
| 71 |
+
| 0.2716 | 13.16 | 6500 | 0.4543 | 0.9000 | 0.9569 | 0.9206 |
|
| 72 |
+
| 0.2484 | 14.17 | 7000 | 0.3965 | 0.9272 | 0.9742 | 0.9489 |
|
| 73 |
+
| 0.228 | 15.18 | 7500 | 0.6807 | 0.8856 | 0.9777 | 0.9257 |
|
| 74 |
+
| 0.2307 | 16.19 | 8000 | 0.5219 | 0.9174 | 0.9802 | 0.9464 |
|
| 75 |
+
| 0.2169 | 17.21 | 8500 | 0.4630 | 0.9121 | 0.9677 | 0.9338 |
|
| 76 |
+
| 0.1997 | 18.22 | 9000 | 0.5152 | 0.9128 | 0.9740 | 0.9398 |
|
| 77 |
+
| 0.1921 | 19.23 | 9500 | 0.5105 | 0.9144 | 0.9867 | 0.9476 |
|
| 78 |
+
| 0.1825 | 20.24 | 10000 | 0.6302 | 0.9053 | 0.9832 | 0.9407 |
|
| 79 |
+
| 0.1786 | 21.25 | 10500 | 0.4602 | 0.9272 | 0.9813 | 0.9524 |
|
| 80 |
+
| 0.1671 | 22.27 | 11000 | 0.5443 | 0.9147 | 0.9794 | 0.9444 |
|
| 81 |
+
| 0.1623 | 23.28 | 11500 | 0.3413 | 0.9443 | 0.9780 | 0.9604 |
|
| 82 |
+
| 0.1595 | 24.29 | 12000 | 0.4478 | 0.9288 | 0.9813 | 0.9531 |
|
| 83 |
+
| 0.151 | 25.3 | 12500 | 0.4178 | 0.9360 | 0.9818 | 0.9571 |
|
| 84 |
+
| 0.1472 | 26.32 | 13000 | 0.4154 | 0.9356 | 0.9833 | 0.9578 |
|
| 85 |
+
| 0.1473 | 27.33 | 13500 | 0.4549 | 0.9318 | 0.9837 | 0.9561 |
|
| 86 |
+
| 0.131 | 28.34 | 14000 | 0.3574 | 0.9424 | 0.9845 | 0.9621 |
|
| 87 |
+
| 0.134 | 29.35 | 14500 | 0.4475 | 0.9333 | 0.9840 | 0.9568 |
|
| 88 |
+
| 0.1282 | 30.36 | 15000 | 0.4012 | 0.9382 | 0.9837 | 0.9591 |
|
| 89 |
+
| 0.1307 | 31.38 | 15500 | 0.3552 | 0.9428 | 0.9847 | 0.9624 |
|
| 90 |
+
|
| 91 |
+
|
| 92 |
+
### Framework versions
|
| 93 |
+
|
| 94 |
+
- Transformers 4.27.0.dev0
|
| 95 |
+
- Pytorch 1.11.0+cu113
|
| 96 |
+
- Datasets 2.9.1.dev0
|
| 97 |
+
- Tokenizers 0.13.2
|
all_results.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 32.0,
|
| 3 |
+
"eval_accuracy": 0.9442970822281167,
|
| 4 |
+
"eval_f1": 0.9604353051727885,
|
| 5 |
+
"eval_loss": 0.3412924110889435,
|
| 6 |
+
"eval_precision": 0.9780145058039889,
|
| 7 |
+
"eval_runtime": 120.0249,
|
| 8 |
+
"eval_samples_per_second": 21.987,
|
| 9 |
+
"eval_steps_per_second": 1.833,
|
| 10 |
+
"train_loss": 0.6361874522950485,
|
| 11 |
+
"train_runtime": 48273.5986,
|
| 12 |
+
"train_samples_per_second": 15.742,
|
| 13 |
+
"train_steps_per_second": 0.327
|
| 14 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 32.0,
|
| 3 |
+
"eval_accuracy": 0.9442970822281167,
|
| 4 |
+
"eval_f1": 0.9604353051727885,
|
| 5 |
+
"eval_loss": 0.3412924110889435,
|
| 6 |
+
"eval_precision": 0.9780145058039889,
|
| 7 |
+
"eval_runtime": 120.0249,
|
| 8 |
+
"eval_samples_per_second": 21.987,
|
| 9 |
+
"eval_steps_per_second": 1.833
|
| 10 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 32.0,
|
| 3 |
+
"train_loss": 0.6361874522950485,
|
| 4 |
+
"train_runtime": 48273.5986,
|
| 5 |
+
"train_samples_per_second": 15.742,
|
| 6 |
+
"train_steps_per_second": 0.327
|
| 7 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,552 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.9442970822281167,
|
| 3 |
+
"best_model_checkpoint": "wav2vec2-large/checkpoint-11500",
|
| 4 |
+
"epoch": 31.998484082870135,
|
| 5 |
+
"global_step": 15808,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 1.01,
|
| 12 |
+
"learning_rate": 9.487666034155598e-06,
|
| 13 |
+
"loss": 4.6314,
|
| 14 |
+
"step": 500
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 1.01,
|
| 18 |
+
"eval_accuracy": 0.02046229632436529,
|
| 19 |
+
"eval_f1": 0.004873418863566646,
|
| 20 |
+
"eval_loss": 4.916522026062012,
|
| 21 |
+
"eval_precision": 0.0027763277100147127,
|
| 22 |
+
"eval_runtime": 123.5513,
|
| 23 |
+
"eval_samples_per_second": 21.36,
|
| 24 |
+
"eval_steps_per_second": 1.781,
|
| 25 |
+
"step": 500
|
| 26 |
+
},
|
| 27 |
+
{
|
| 28 |
+
"epoch": 2.02,
|
| 29 |
+
"learning_rate": 1.8975332068311197e-05,
|
| 30 |
+
"loss": 3.7739,
|
| 31 |
+
"step": 1000
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 2.02,
|
| 35 |
+
"eval_accuracy": 0.03561955286093217,
|
| 36 |
+
"eval_f1": 0.02524670069279936,
|
| 37 |
+
"eval_loss": 4.449059009552002,
|
| 38 |
+
"eval_precision": 0.0750288473053048,
|
| 39 |
+
"eval_runtime": 121.7021,
|
| 40 |
+
"eval_samples_per_second": 21.684,
|
| 41 |
+
"eval_steps_per_second": 1.808,
|
| 42 |
+
"step": 1000
|
| 43 |
+
},
|
| 44 |
+
{
|
| 45 |
+
"epoch": 3.04,
|
| 46 |
+
"learning_rate": 2.846299810246679e-05,
|
| 47 |
+
"loss": 2.5035,
|
| 48 |
+
"step": 1500
|
| 49 |
+
},
|
| 50 |
+
{
|
| 51 |
+
"epoch": 3.04,
|
| 52 |
+
"eval_accuracy": 0.11292156119742326,
|
| 53 |
+
"eval_f1": 0.11144842685774854,
|
| 54 |
+
"eval_loss": 4.142854690551758,
|
| 55 |
+
"eval_precision": 0.2671650922898445,
|
| 56 |
+
"eval_runtime": 120.9962,
|
| 57 |
+
"eval_samples_per_second": 21.811,
|
| 58 |
+
"eval_steps_per_second": 1.818,
|
| 59 |
+
"step": 1500
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 4.05,
|
| 63 |
+
"learning_rate": 2.9116468686300697e-05,
|
| 64 |
+
"loss": 1.5633,
|
| 65 |
+
"step": 2000
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 4.05,
|
| 69 |
+
"eval_accuracy": 0.36756347101174686,
|
| 70 |
+
"eval_f1": 0.3830364851222239,
|
| 71 |
+
"eval_loss": 3.1973092555999756,
|
| 72 |
+
"eval_precision": 0.6598356853987439,
|
| 73 |
+
"eval_runtime": 122.0663,
|
| 74 |
+
"eval_samples_per_second": 21.619,
|
| 75 |
+
"eval_steps_per_second": 1.802,
|
| 76 |
+
"step": 2000
|
| 77 |
+
},
|
| 78 |
+
{
|
| 79 |
+
"epoch": 5.06,
|
| 80 |
+
"learning_rate": 2.8062135376396992e-05,
|
| 81 |
+
"loss": 1.0538,
|
| 82 |
+
"step": 2500
|
| 83 |
+
},
|
| 84 |
+
{
|
| 85 |
+
"epoch": 5.06,
|
| 86 |
+
"eval_accuracy": 0.5888594164456233,
|
| 87 |
+
"eval_f1": 0.6556643570555963,
|
| 88 |
+
"eval_loss": 2.5479371547698975,
|
| 89 |
+
"eval_precision": 0.8416636658887603,
|
| 90 |
+
"eval_runtime": 121.3195,
|
| 91 |
+
"eval_samples_per_second": 21.752,
|
| 92 |
+
"eval_steps_per_second": 1.813,
|
| 93 |
+
"step": 2500
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 6.07,
|
| 97 |
+
"learning_rate": 2.700780206649329e-05,
|
| 98 |
+
"loss": 0.7422,
|
| 99 |
+
"step": 3000
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 6.07,
|
| 103 |
+
"eval_accuracy": 0.7824933687002652,
|
| 104 |
+
"eval_f1": 0.8194352733059617,
|
| 105 |
+
"eval_loss": 1.4494293928146362,
|
| 106 |
+
"eval_precision": 0.8920845853165854,
|
| 107 |
+
"eval_runtime": 120.4455,
|
| 108 |
+
"eval_samples_per_second": 21.91,
|
| 109 |
+
"eval_steps_per_second": 1.827,
|
| 110 |
+
"step": 3000
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 7.08,
|
| 114 |
+
"learning_rate": 2.5953468756589585e-05,
|
| 115 |
+
"loss": 0.5762,
|
| 116 |
+
"step": 3500
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 7.08,
|
| 120 |
+
"eval_accuracy": 0.7726411519514967,
|
| 121 |
+
"eval_f1": 0.8267111468977018,
|
| 122 |
+
"eval_loss": 1.316819190979004,
|
| 123 |
+
"eval_precision": 0.9277444722087854,
|
| 124 |
+
"eval_runtime": 122.3892,
|
| 125 |
+
"eval_samples_per_second": 21.562,
|
| 126 |
+
"eval_steps_per_second": 1.798,
|
| 127 |
+
"step": 3500
|
| 128 |
+
},
|
| 129 |
+
{
|
| 130 |
+
"epoch": 8.1,
|
| 131 |
+
"learning_rate": 2.489913544668588e-05,
|
| 132 |
+
"loss": 0.46,
|
| 133 |
+
"step": 4000
|
| 134 |
+
},
|
| 135 |
+
{
|
| 136 |
+
"epoch": 8.1,
|
| 137 |
+
"eval_accuracy": 0.8563849943160288,
|
| 138 |
+
"eval_f1": 0.898213371629901,
|
| 139 |
+
"eval_loss": 0.8782555460929871,
|
| 140 |
+
"eval_precision": 0.9531998814815654,
|
| 141 |
+
"eval_runtime": 120.9059,
|
| 142 |
+
"eval_samples_per_second": 21.827,
|
| 143 |
+
"eval_steps_per_second": 1.82,
|
| 144 |
+
"step": 4000
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 9.11,
|
| 148 |
+
"learning_rate": 2.3844802136782175e-05,
|
| 149 |
+
"loss": 0.4007,
|
| 150 |
+
"step": 4500
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 9.11,
|
| 154 |
+
"eval_accuracy": 0.8738158393330807,
|
| 155 |
+
"eval_f1": 0.913683470937564,
|
| 156 |
+
"eval_loss": 0.752358615398407,
|
| 157 |
+
"eval_precision": 0.9636991964706034,
|
| 158 |
+
"eval_runtime": 120.9904,
|
| 159 |
+
"eval_samples_per_second": 21.812,
|
| 160 |
+
"eval_steps_per_second": 1.818,
|
| 161 |
+
"step": 4500
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 10.12,
|
| 165 |
+
"learning_rate": 2.279046882687847e-05,
|
| 166 |
+
"loss": 0.3374,
|
| 167 |
+
"step": 5000
|
| 168 |
+
},
|
| 169 |
+
{
|
| 170 |
+
"epoch": 10.12,
|
| 171 |
+
"eval_accuracy": 0.8851837817355058,
|
| 172 |
+
"eval_f1": 0.9220598204792498,
|
| 173 |
+
"eval_loss": 0.6385864615440369,
|
| 174 |
+
"eval_precision": 0.9677694535212527,
|
| 175 |
+
"eval_runtime": 120.467,
|
| 176 |
+
"eval_samples_per_second": 21.906,
|
| 177 |
+
"eval_steps_per_second": 1.826,
|
| 178 |
+
"step": 5000
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 11.13,
|
| 182 |
+
"learning_rate": 2.1736135516974768e-05,
|
| 183 |
+
"loss": 0.3108,
|
| 184 |
+
"step": 5500
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 11.13,
|
| 188 |
+
"eval_accuracy": 0.9105721864342554,
|
| 189 |
+
"eval_f1": 0.9372909559856852,
|
| 190 |
+
"eval_loss": 0.5049420595169067,
|
| 191 |
+
"eval_precision": 0.9680582892579286,
|
| 192 |
+
"eval_runtime": 121.8653,
|
| 193 |
+
"eval_samples_per_second": 21.655,
|
| 194 |
+
"eval_steps_per_second": 1.805,
|
| 195 |
+
"step": 5500
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 12.15,
|
| 199 |
+
"learning_rate": 2.0681802207071063e-05,
|
| 200 |
+
"loss": 0.2735,
|
| 201 |
+
"step": 6000
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 12.15,
|
| 205 |
+
"eval_accuracy": 0.8904888215233043,
|
| 206 |
+
"eval_f1": 0.9225848099506897,
|
| 207 |
+
"eval_loss": 0.6096966862678528,
|
| 208 |
+
"eval_precision": 0.9623634379449385,
|
| 209 |
+
"eval_runtime": 121.4949,
|
| 210 |
+
"eval_samples_per_second": 21.721,
|
| 211 |
+
"eval_steps_per_second": 1.811,
|
| 212 |
+
"step": 6000
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 13.16,
|
| 216 |
+
"learning_rate": 1.9627468897167357e-05,
|
| 217 |
+
"loss": 0.2716,
|
| 218 |
+
"step": 6500
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 13.16,
|
| 222 |
+
"eval_accuracy": 0.8999621068586586,
|
| 223 |
+
"eval_f1": 0.9206365230401724,
|
| 224 |
+
"eval_loss": 0.4542546570301056,
|
| 225 |
+
"eval_precision": 0.9569343688614892,
|
| 226 |
+
"eval_runtime": 120.692,
|
| 227 |
+
"eval_samples_per_second": 21.866,
|
| 228 |
+
"eval_steps_per_second": 1.823,
|
| 229 |
+
"step": 6500
|
| 230 |
+
},
|
| 231 |
+
{
|
| 232 |
+
"epoch": 14.17,
|
| 233 |
+
"learning_rate": 1.8573135587263652e-05,
|
| 234 |
+
"loss": 0.2484,
|
| 235 |
+
"step": 7000
|
| 236 |
+
},
|
| 237 |
+
{
|
| 238 |
+
"epoch": 14.17,
|
| 239 |
+
"eval_accuracy": 0.9272451686244789,
|
| 240 |
+
"eval_f1": 0.9489393378254958,
|
| 241 |
+
"eval_loss": 0.3964819610118866,
|
| 242 |
+
"eval_precision": 0.9742030978909282,
|
| 243 |
+
"eval_runtime": 121.4363,
|
| 244 |
+
"eval_samples_per_second": 21.732,
|
| 245 |
+
"eval_steps_per_second": 1.812,
|
| 246 |
+
"step": 7000
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 15.18,
|
| 250 |
+
"learning_rate": 1.751880227735995e-05,
|
| 251 |
+
"loss": 0.228,
|
| 252 |
+
"step": 7500
|
| 253 |
+
},
|
| 254 |
+
{
|
| 255 |
+
"epoch": 15.18,
|
| 256 |
+
"eval_accuracy": 0.8855627131489201,
|
| 257 |
+
"eval_f1": 0.9256828810141209,
|
| 258 |
+
"eval_loss": 0.6806507110595703,
|
| 259 |
+
"eval_precision": 0.9777330671722411,
|
| 260 |
+
"eval_runtime": 122.0415,
|
| 261 |
+
"eval_samples_per_second": 21.624,
|
| 262 |
+
"eval_steps_per_second": 1.803,
|
| 263 |
+
"step": 7500
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 16.19,
|
| 267 |
+
"learning_rate": 1.6464468967456245e-05,
|
| 268 |
+
"loss": 0.2307,
|
| 269 |
+
"step": 8000
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 16.19,
|
| 273 |
+
"eval_accuracy": 0.9173929518757105,
|
| 274 |
+
"eval_f1": 0.946436431649072,
|
| 275 |
+
"eval_loss": 0.5219454765319824,
|
| 276 |
+
"eval_precision": 0.9801938448683835,
|
| 277 |
+
"eval_runtime": 120.3841,
|
| 278 |
+
"eval_samples_per_second": 21.921,
|
| 279 |
+
"eval_steps_per_second": 1.827,
|
| 280 |
+
"step": 8000
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"epoch": 17.21,
|
| 284 |
+
"learning_rate": 1.541013565755254e-05,
|
| 285 |
+
"loss": 0.2169,
|
| 286 |
+
"step": 8500
|
| 287 |
+
},
|
| 288 |
+
{
|
| 289 |
+
"epoch": 17.21,
|
| 290 |
+
"eval_accuracy": 0.9120879120879121,
|
| 291 |
+
"eval_f1": 0.9337616922587397,
|
| 292 |
+
"eval_loss": 0.4630146920681,
|
| 293 |
+
"eval_precision": 0.9677088927055834,
|
| 294 |
+
"eval_runtime": 121.7745,
|
| 295 |
+
"eval_samples_per_second": 21.671,
|
| 296 |
+
"eval_steps_per_second": 1.807,
|
| 297 |
+
"step": 8500
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 18.22,
|
| 301 |
+
"learning_rate": 1.4355802347648837e-05,
|
| 302 |
+
"loss": 0.1997,
|
| 303 |
+
"step": 9000
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 18.22,
|
| 307 |
+
"eval_accuracy": 0.9128457749147404,
|
| 308 |
+
"eval_f1": 0.9397548887949031,
|
| 309 |
+
"eval_loss": 0.5151729583740234,
|
| 310 |
+
"eval_precision": 0.9740319117972248,
|
| 311 |
+
"eval_runtime": 121.0656,
|
| 312 |
+
"eval_samples_per_second": 21.798,
|
| 313 |
+
"eval_steps_per_second": 1.817,
|
| 314 |
+
"step": 9000
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 19.23,
|
| 318 |
+
"learning_rate": 1.3301469037745133e-05,
|
| 319 |
+
"loss": 0.1921,
|
| 320 |
+
"step": 9500
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 19.23,
|
| 324 |
+
"eval_accuracy": 0.9143615005683972,
|
| 325 |
+
"eval_f1": 0.9475938884984938,
|
| 326 |
+
"eval_loss": 0.510482668876648,
|
| 327 |
+
"eval_precision": 0.9867250624910348,
|
| 328 |
+
"eval_runtime": 122.1644,
|
| 329 |
+
"eval_samples_per_second": 21.602,
|
| 330 |
+
"eval_steps_per_second": 1.801,
|
| 331 |
+
"step": 9500
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 20.24,
|
| 335 |
+
"learning_rate": 1.2247135727841428e-05,
|
| 336 |
+
"loss": 0.1825,
|
| 337 |
+
"step": 10000
|
| 338 |
+
},
|
| 339 |
+
{
|
| 340 |
+
"epoch": 20.24,
|
| 341 |
+
"eval_accuracy": 0.905267146646457,
|
| 342 |
+
"eval_f1": 0.9406773355537659,
|
| 343 |
+
"eval_loss": 0.6302403807640076,
|
| 344 |
+
"eval_precision": 0.9831536512710705,
|
| 345 |
+
"eval_runtime": 121.7536,
|
| 346 |
+
"eval_samples_per_second": 21.675,
|
| 347 |
+
"eval_steps_per_second": 1.807,
|
| 348 |
+
"step": 10000
|
| 349 |
+
},
|
| 350 |
+
{
|
| 351 |
+
"epoch": 21.25,
|
| 352 |
+
"learning_rate": 1.1192802417937724e-05,
|
| 353 |
+
"loss": 0.1786,
|
| 354 |
+
"step": 10500
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 21.25,
|
| 358 |
+
"eval_accuracy": 0.9272451686244789,
|
| 359 |
+
"eval_f1": 0.952370485075943,
|
| 360 |
+
"eval_loss": 0.460215300321579,
|
| 361 |
+
"eval_precision": 0.9813011833529409,
|
| 362 |
+
"eval_runtime": 124.2398,
|
| 363 |
+
"eval_samples_per_second": 21.241,
|
| 364 |
+
"eval_steps_per_second": 1.771,
|
| 365 |
+
"step": 10500
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 22.27,
|
| 369 |
+
"learning_rate": 1.013846910803402e-05,
|
| 370 |
+
"loss": 0.1671,
|
| 371 |
+
"step": 11000
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 22.27,
|
| 375 |
+
"eval_accuracy": 0.9147404319818113,
|
| 376 |
+
"eval_f1": 0.9444178129216141,
|
| 377 |
+
"eval_loss": 0.544323205947876,
|
| 378 |
+
"eval_precision": 0.9794231668433278,
|
| 379 |
+
"eval_runtime": 120.9552,
|
| 380 |
+
"eval_samples_per_second": 21.818,
|
| 381 |
+
"eval_steps_per_second": 1.819,
|
| 382 |
+
"step": 11000
|
| 383 |
+
},
|
| 384 |
+
{
|
| 385 |
+
"epoch": 23.28,
|
| 386 |
+
"learning_rate": 9.084135798130316e-06,
|
| 387 |
+
"loss": 0.1623,
|
| 388 |
+
"step": 11500
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 23.28,
|
| 392 |
+
"eval_accuracy": 0.9442970822281167,
|
| 393 |
+
"eval_f1": 0.9604353051727885,
|
| 394 |
+
"eval_loss": 0.3412924110889435,
|
| 395 |
+
"eval_precision": 0.9780145058039889,
|
| 396 |
+
"eval_runtime": 122.6656,
|
| 397 |
+
"eval_samples_per_second": 21.514,
|
| 398 |
+
"eval_steps_per_second": 1.793,
|
| 399 |
+
"step": 11500
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 24.29,
|
| 403 |
+
"learning_rate": 8.029802488226612e-06,
|
| 404 |
+
"loss": 0.1595,
|
| 405 |
+
"step": 12000
|
| 406 |
+
},
|
| 407 |
+
{
|
| 408 |
+
"epoch": 24.29,
|
| 409 |
+
"eval_accuracy": 0.9287608942781357,
|
| 410 |
+
"eval_f1": 0.9531334781334764,
|
| 411 |
+
"eval_loss": 0.44775113463401794,
|
| 412 |
+
"eval_precision": 0.9812894876888261,
|
| 413 |
+
"eval_runtime": 120.6582,
|
| 414 |
+
"eval_samples_per_second": 21.872,
|
| 415 |
+
"eval_steps_per_second": 1.823,
|
| 416 |
+
"step": 12000
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 25.3,
|
| 420 |
+
"learning_rate": 6.975469178322908e-06,
|
| 421 |
+
"loss": 0.151,
|
| 422 |
+
"step": 12500
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 25.3,
|
| 426 |
+
"eval_accuracy": 0.9359605911330049,
|
| 427 |
+
"eval_f1": 0.9571224982425287,
|
| 428 |
+
"eval_loss": 0.41782599687576294,
|
| 429 |
+
"eval_precision": 0.9818213438674759,
|
| 430 |
+
"eval_runtime": 121.5443,
|
| 431 |
+
"eval_samples_per_second": 21.712,
|
| 432 |
+
"eval_steps_per_second": 1.81,
|
| 433 |
+
"step": 12500
|
| 434 |
+
},
|
| 435 |
+
{
|
| 436 |
+
"epoch": 26.32,
|
| 437 |
+
"learning_rate": 5.9211358684192026e-06,
|
| 438 |
+
"loss": 0.1472,
|
| 439 |
+
"step": 13000
|
| 440 |
+
},
|
| 441 |
+
{
|
| 442 |
+
"epoch": 26.32,
|
| 443 |
+
"eval_accuracy": 0.9355816597195907,
|
| 444 |
+
"eval_f1": 0.9577691906088884,
|
| 445 |
+
"eval_loss": 0.41541919112205505,
|
| 446 |
+
"eval_precision": 0.9832631333451843,
|
| 447 |
+
"eval_runtime": 122.2041,
|
| 448 |
+
"eval_samples_per_second": 21.595,
|
| 449 |
+
"eval_steps_per_second": 1.8,
|
| 450 |
+
"step": 13000
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 27.33,
|
| 454 |
+
"learning_rate": 4.866802558515498e-06,
|
| 455 |
+
"loss": 0.1473,
|
| 456 |
+
"step": 13500
|
| 457 |
+
},
|
| 458 |
+
{
|
| 459 |
+
"epoch": 27.33,
|
| 460 |
+
"eval_accuracy": 0.931792345585449,
|
| 461 |
+
"eval_f1": 0.9560631584126075,
|
| 462 |
+
"eval_loss": 0.4548525810241699,
|
| 463 |
+
"eval_precision": 0.9837464944410116,
|
| 464 |
+
"eval_runtime": 121.1061,
|
| 465 |
+
"eval_samples_per_second": 21.791,
|
| 466 |
+
"eval_steps_per_second": 1.817,
|
| 467 |
+
"step": 13500
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"epoch": 28.34,
|
| 471 |
+
"learning_rate": 3.8124692486117947e-06,
|
| 472 |
+
"loss": 0.131,
|
| 473 |
+
"step": 14000
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 28.34,
|
| 477 |
+
"eval_accuracy": 0.9424024251610459,
|
| 478 |
+
"eval_f1": 0.9620998995027674,
|
| 479 |
+
"eval_loss": 0.3574429154396057,
|
| 480 |
+
"eval_precision": 0.984478231442749,
|
| 481 |
+
"eval_runtime": 120.0382,
|
| 482 |
+
"eval_samples_per_second": 21.985,
|
| 483 |
+
"eval_steps_per_second": 1.833,
|
| 484 |
+
"step": 14000
|
| 485 |
+
},
|
| 486 |
+
{
|
| 487 |
+
"epoch": 29.35,
|
| 488 |
+
"learning_rate": 2.7581359387080904e-06,
|
| 489 |
+
"loss": 0.134,
|
| 490 |
+
"step": 14500
|
| 491 |
+
},
|
| 492 |
+
{
|
| 493 |
+
"epoch": 29.35,
|
| 494 |
+
"eval_accuracy": 0.9333080712391058,
|
| 495 |
+
"eval_f1": 0.9568214460159893,
|
| 496 |
+
"eval_loss": 0.4474771022796631,
|
| 497 |
+
"eval_precision": 0.9840495545740223,
|
| 498 |
+
"eval_runtime": 123.1013,
|
| 499 |
+
"eval_samples_per_second": 21.438,
|
| 500 |
+
"eval_steps_per_second": 1.787,
|
| 501 |
+
"step": 14500
|
| 502 |
+
},
|
| 503 |
+
{
|
| 504 |
+
"epoch": 30.36,
|
| 505 |
+
"learning_rate": 1.7038026288043862e-06,
|
| 506 |
+
"loss": 0.1282,
|
| 507 |
+
"step": 15000
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 30.36,
|
| 511 |
+
"eval_accuracy": 0.93823417961349,
|
| 512 |
+
"eval_f1": 0.959091295015073,
|
| 513 |
+
"eval_loss": 0.401244193315506,
|
| 514 |
+
"eval_precision": 0.9836786405574566,
|
| 515 |
+
"eval_runtime": 121.7667,
|
| 516 |
+
"eval_samples_per_second": 21.673,
|
| 517 |
+
"eval_steps_per_second": 1.807,
|
| 518 |
+
"step": 15000
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 31.38,
|
| 522 |
+
"learning_rate": 6.494693189006819e-07,
|
| 523 |
+
"loss": 0.1307,
|
| 524 |
+
"step": 15500
|
| 525 |
+
},
|
| 526 |
+
{
|
| 527 |
+
"epoch": 31.38,
|
| 528 |
+
"eval_accuracy": 0.94278135657446,
|
| 529 |
+
"eval_f1": 0.9623515495402261,
|
| 530 |
+
"eval_loss": 0.35516688227653503,
|
| 531 |
+
"eval_precision": 0.9847235247308098,
|
| 532 |
+
"eval_runtime": 121.4163,
|
| 533 |
+
"eval_samples_per_second": 21.735,
|
| 534 |
+
"eval_steps_per_second": 1.812,
|
| 535 |
+
"step": 15500
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 32.0,
|
| 539 |
+
"step": 15808,
|
| 540 |
+
"total_flos": 1.2696466131100762e+20,
|
| 541 |
+
"train_loss": 0.6361874522950485,
|
| 542 |
+
"train_runtime": 48273.5986,
|
| 543 |
+
"train_samples_per_second": 15.742,
|
| 544 |
+
"train_steps_per_second": 0.327
|
| 545 |
+
}
|
| 546 |
+
],
|
| 547 |
+
"max_steps": 15808,
|
| 548 |
+
"num_train_epochs": 32,
|
| 549 |
+
"total_flos": 1.2696466131100762e+20,
|
| 550 |
+
"trial_name": null,
|
| 551 |
+
"trial_params": null
|
| 552 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:99acc3ede14d128c64289c5346624e6f7e8ff3c65d983d1aaf7198fbaf146ea6
|
| 3 |
+
size 3503
|