Masaki Hattori
commited on
Commit
·
bcf39a7
1
Parent(s):
93746ff
Model save
Browse files- README.md +83 -0
- trainer_state.json +1194 -0
README.md
ADDED
|
@@ -0,0 +1,83 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
base_model: rinna/japanese-hubert-base
|
| 4 |
+
tags:
|
| 5 |
+
- generated_from_trainer
|
| 6 |
+
metrics:
|
| 7 |
+
- wer
|
| 8 |
+
model-index:
|
| 9 |
+
- name: hubert-rinnna-jp-jdrtsp-fw07sp-13
|
| 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 |
+
# hubert-rinnna-jp-jdrtsp-fw07sp-13
|
| 17 |
+
|
| 18 |
+
This model is a fine-tuned version of [rinna/japanese-hubert-base](https://huggingface.co/rinna/japanese-hubert-base) on the None dataset.
|
| 19 |
+
It achieves the following results on the evaluation set:
|
| 20 |
+
- Loss: 0.1606
|
| 21 |
+
- Wer: 0.3004
|
| 22 |
+
- Cer: 0.1786
|
| 23 |
+
|
| 24 |
+
## Model description
|
| 25 |
+
|
| 26 |
+
More information needed
|
| 27 |
+
|
| 28 |
+
## Intended uses & limitations
|
| 29 |
+
|
| 30 |
+
More information needed
|
| 31 |
+
|
| 32 |
+
## Training and evaluation data
|
| 33 |
+
|
| 34 |
+
More information needed
|
| 35 |
+
|
| 36 |
+
## Training procedure
|
| 37 |
+
|
| 38 |
+
### Training hyperparameters
|
| 39 |
+
|
| 40 |
+
The following hyperparameters were used during training:
|
| 41 |
+
- learning_rate: 0.0005
|
| 42 |
+
- train_batch_size: 32
|
| 43 |
+
- eval_batch_size: 16
|
| 44 |
+
- seed: 42
|
| 45 |
+
- gradient_accumulation_steps: 2
|
| 46 |
+
- total_train_batch_size: 64
|
| 47 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 48 |
+
- lr_scheduler_type: linear
|
| 49 |
+
- lr_scheduler_warmup_steps: 1000
|
| 50 |
+
- num_epochs: 20
|
| 51 |
+
|
| 52 |
+
### Training results
|
| 53 |
+
|
| 54 |
+
| Training Loss | Epoch | Step | Validation Loss | Wer | Cer |
|
| 55 |
+
|:-------------:|:-----:|:----:|:---------------:|:------:|:------:|
|
| 56 |
+
| 1.352 | 1.0 | 404 | 0.9913 | 0.6021 | 0.4479 |
|
| 57 |
+
| 0.9044 | 2.0 | 808 | 0.5053 | 0.4261 | 0.2774 |
|
| 58 |
+
| 0.9001 | 3.0 | 1212 | 0.8458 | 0.4848 | 0.3267 |
|
| 59 |
+
| 0.8425 | 4.0 | 1616 | 0.5311 | 0.4577 | 0.3053 |
|
| 60 |
+
| 0.8408 | 5.0 | 2020 | 0.4328 | 0.4075 | 0.2776 |
|
| 61 |
+
| 0.7759 | 6.0 | 2424 | 0.4736 | 0.4394 | 0.3363 |
|
| 62 |
+
| 0.7228 | 7.0 | 2828 | 0.4667 | 0.4173 | 0.2862 |
|
| 63 |
+
| 0.6755 | 8.0 | 3232 | 0.4190 | 0.4114 | 0.2611 |
|
| 64 |
+
| 0.634 | 9.0 | 3636 | 0.4252 | 0.3993 | 0.2612 |
|
| 65 |
+
| 0.6267 | 10.0 | 4040 | 0.3275 | 0.3734 | 0.2362 |
|
| 66 |
+
| 0.6199 | 11.0 | 4444 | 0.2786 | 0.3543 | 0.2222 |
|
| 67 |
+
| 0.5396 | 12.0 | 4848 | 0.2851 | 0.3501 | 0.2146 |
|
| 68 |
+
| 0.5343 | 13.0 | 5252 | 0.2527 | 0.3448 | 0.2106 |
|
| 69 |
+
| 0.5488 | 14.0 | 5656 | 0.2725 | 0.3431 | 0.2100 |
|
| 70 |
+
| 0.4606 | 15.0 | 6060 | 0.2293 | 0.3259 | 0.1962 |
|
| 71 |
+
| 0.4229 | 16.0 | 6464 | 0.2043 | 0.3172 | 0.1914 |
|
| 72 |
+
| 0.4078 | 17.0 | 6868 | 0.1891 | 0.3128 | 0.1862 |
|
| 73 |
+
| 0.4017 | 18.0 | 7272 | 0.1785 | 0.3075 | 0.1833 |
|
| 74 |
+
| 0.3618 | 19.0 | 7676 | 0.1673 | 0.3035 | 0.1803 |
|
| 75 |
+
| 0.3739 | 20.0 | 8080 | 0.1606 | 0.3004 | 0.1786 |
|
| 76 |
+
|
| 77 |
+
|
| 78 |
+
### Framework versions
|
| 79 |
+
|
| 80 |
+
- Transformers 4.34.0.dev0
|
| 81 |
+
- Pytorch 2.0.1+cu118
|
| 82 |
+
- Datasets 2.14.5
|
| 83 |
+
- Tokenizers 0.13.3
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1194 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 20.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 8080,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.12,
|
| 13 |
+
"learning_rate": 2.5e-05,
|
| 14 |
+
"loss": 9.1111,
|
| 15 |
+
"step": 50
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 0.25,
|
| 19 |
+
"learning_rate": 5e-05,
|
| 20 |
+
"loss": 4.3213,
|
| 21 |
+
"step": 100
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
"epoch": 0.37,
|
| 25 |
+
"learning_rate": 7.5e-05,
|
| 26 |
+
"loss": 3.2869,
|
| 27 |
+
"step": 150
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 0.5,
|
| 31 |
+
"learning_rate": 0.0001,
|
| 32 |
+
"loss": 2.7342,
|
| 33 |
+
"step": 200
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 0.62,
|
| 37 |
+
"learning_rate": 0.000125,
|
| 38 |
+
"loss": 2.5924,
|
| 39 |
+
"step": 250
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.74,
|
| 43 |
+
"learning_rate": 0.00015,
|
| 44 |
+
"loss": 2.3303,
|
| 45 |
+
"step": 300
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.87,
|
| 49 |
+
"learning_rate": 0.000175,
|
| 50 |
+
"loss": 1.75,
|
| 51 |
+
"step": 350
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.99,
|
| 55 |
+
"learning_rate": 0.0002,
|
| 56 |
+
"loss": 1.352,
|
| 57 |
+
"step": 400
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 1.0,
|
| 61 |
+
"eval_cer": 0.4478702548565318,
|
| 62 |
+
"eval_loss": 0.9912864565849304,
|
| 63 |
+
"eval_runtime": 37.3851,
|
| 64 |
+
"eval_samples_per_second": 172.903,
|
| 65 |
+
"eval_steps_per_second": 10.806,
|
| 66 |
+
"eval_wer": 0.6020902919943971,
|
| 67 |
+
"step": 404
|
| 68 |
+
},
|
| 69 |
+
{
|
| 70 |
+
"epoch": 1.11,
|
| 71 |
+
"learning_rate": 0.00022500000000000002,
|
| 72 |
+
"loss": 1.2041,
|
| 73 |
+
"step": 450
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 1.24,
|
| 77 |
+
"learning_rate": 0.00025,
|
| 78 |
+
"loss": 1.0738,
|
| 79 |
+
"step": 500
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 1.36,
|
| 83 |
+
"learning_rate": 0.000275,
|
| 84 |
+
"loss": 1.0552,
|
| 85 |
+
"step": 550
|
| 86 |
+
},
|
| 87 |
+
{
|
| 88 |
+
"epoch": 1.49,
|
| 89 |
+
"learning_rate": 0.0003,
|
| 90 |
+
"loss": 0.9818,
|
| 91 |
+
"step": 600
|
| 92 |
+
},
|
| 93 |
+
{
|
| 94 |
+
"epoch": 1.61,
|
| 95 |
+
"learning_rate": 0.00032500000000000004,
|
| 96 |
+
"loss": 1.0827,
|
| 97 |
+
"step": 650
|
| 98 |
+
},
|
| 99 |
+
{
|
| 100 |
+
"epoch": 1.73,
|
| 101 |
+
"learning_rate": 0.00035,
|
| 102 |
+
"loss": 0.944,
|
| 103 |
+
"step": 700
|
| 104 |
+
},
|
| 105 |
+
{
|
| 106 |
+
"epoch": 1.86,
|
| 107 |
+
"learning_rate": 0.000375,
|
| 108 |
+
"loss": 0.9694,
|
| 109 |
+
"step": 750
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 1.98,
|
| 113 |
+
"learning_rate": 0.0004,
|
| 114 |
+
"loss": 0.9044,
|
| 115 |
+
"step": 800
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 2.0,
|
| 119 |
+
"eval_cer": 0.2773916315174558,
|
| 120 |
+
"eval_loss": 0.5053090453147888,
|
| 121 |
+
"eval_runtime": 36.8215,
|
| 122 |
+
"eval_samples_per_second": 175.55,
|
| 123 |
+
"eval_steps_per_second": 10.972,
|
| 124 |
+
"eval_wer": 0.42606759329095284,
|
| 125 |
+
"step": 808
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 2.1,
|
| 129 |
+
"learning_rate": 0.000425,
|
| 130 |
+
"loss": 0.9818,
|
| 131 |
+
"step": 850
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 2.23,
|
| 135 |
+
"learning_rate": 0.00045000000000000004,
|
| 136 |
+
"loss": 0.9531,
|
| 137 |
+
"step": 900
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 2.35,
|
| 141 |
+
"learning_rate": 0.000475,
|
| 142 |
+
"loss": 0.919,
|
| 143 |
+
"step": 950
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 2.48,
|
| 147 |
+
"learning_rate": 0.0005,
|
| 148 |
+
"loss": 0.9661,
|
| 149 |
+
"step": 1000
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 2.6,
|
| 153 |
+
"learning_rate": 0.0004964689265536723,
|
| 154 |
+
"loss": 0.993,
|
| 155 |
+
"step": 1050
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 2.72,
|
| 159 |
+
"learning_rate": 0.0004929378531073446,
|
| 160 |
+
"loss": 0.9892,
|
| 161 |
+
"step": 1100
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 2.85,
|
| 165 |
+
"learning_rate": 0.0004894067796610169,
|
| 166 |
+
"loss": 0.9471,
|
| 167 |
+
"step": 1150
|
| 168 |
+
},
|
| 169 |
+
{
|
| 170 |
+
"epoch": 2.97,
|
| 171 |
+
"learning_rate": 0.0004858757062146893,
|
| 172 |
+
"loss": 0.9001,
|
| 173 |
+
"step": 1200
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 3.0,
|
| 177 |
+
"eval_cer": 0.32669953860472484,
|
| 178 |
+
"eval_loss": 0.845825731754303,
|
| 179 |
+
"eval_runtime": 36.8369,
|
| 180 |
+
"eval_samples_per_second": 175.476,
|
| 181 |
+
"eval_steps_per_second": 10.967,
|
| 182 |
+
"eval_wer": 0.4847537980821032,
|
| 183 |
+
"step": 1212
|
| 184 |
+
},
|
| 185 |
+
{
|
| 186 |
+
"epoch": 3.09,
|
| 187 |
+
"learning_rate": 0.0004823446327683616,
|
| 188 |
+
"loss": 0.8903,
|
| 189 |
+
"step": 1250
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 3.22,
|
| 193 |
+
"learning_rate": 0.0004788135593220339,
|
| 194 |
+
"loss": 0.9181,
|
| 195 |
+
"step": 1300
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 3.34,
|
| 199 |
+
"learning_rate": 0.0004752824858757062,
|
| 200 |
+
"loss": 0.858,
|
| 201 |
+
"step": 1350
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 3.47,
|
| 205 |
+
"learning_rate": 0.00047175141242937854,
|
| 206 |
+
"loss": 0.917,
|
| 207 |
+
"step": 1400
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 3.59,
|
| 211 |
+
"learning_rate": 0.0004682203389830508,
|
| 212 |
+
"loss": 0.915,
|
| 213 |
+
"step": 1450
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 3.71,
|
| 217 |
+
"learning_rate": 0.0004646892655367232,
|
| 218 |
+
"loss": 0.8631,
|
| 219 |
+
"step": 1500
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 3.84,
|
| 223 |
+
"learning_rate": 0.0004611581920903955,
|
| 224 |
+
"loss": 0.8267,
|
| 225 |
+
"step": 1550
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 3.96,
|
| 229 |
+
"learning_rate": 0.0004576271186440678,
|
| 230 |
+
"loss": 0.8425,
|
| 231 |
+
"step": 1600
|
| 232 |
+
},
|
| 233 |
+
{
|
| 234 |
+
"epoch": 4.0,
|
| 235 |
+
"eval_cer": 0.30525356937761144,
|
| 236 |
+
"eval_loss": 0.5311021208763123,
|
| 237 |
+
"eval_runtime": 36.7576,
|
| 238 |
+
"eval_samples_per_second": 175.855,
|
| 239 |
+
"eval_steps_per_second": 10.991,
|
| 240 |
+
"eval_wer": 0.45774521423697156,
|
| 241 |
+
"step": 1616
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 4.08,
|
| 245 |
+
"learning_rate": 0.0004540960451977401,
|
| 246 |
+
"loss": 0.8399,
|
| 247 |
+
"step": 1650
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 4.21,
|
| 251 |
+
"learning_rate": 0.00045056497175141244,
|
| 252 |
+
"loss": 0.8166,
|
| 253 |
+
"step": 1700
|
| 254 |
+
},
|
| 255 |
+
{
|
| 256 |
+
"epoch": 4.33,
|
| 257 |
+
"learning_rate": 0.0004470338983050847,
|
| 258 |
+
"loss": 0.8089,
|
| 259 |
+
"step": 1750
|
| 260 |
+
},
|
| 261 |
+
{
|
| 262 |
+
"epoch": 4.46,
|
| 263 |
+
"learning_rate": 0.0004435028248587571,
|
| 264 |
+
"loss": 0.8094,
|
| 265 |
+
"step": 1800
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 4.58,
|
| 269 |
+
"learning_rate": 0.0004399717514124294,
|
| 270 |
+
"loss": 0.8622,
|
| 271 |
+
"step": 1850
|
| 272 |
+
},
|
| 273 |
+
{
|
| 274 |
+
"epoch": 4.7,
|
| 275 |
+
"learning_rate": 0.00043644067796610173,
|
| 276 |
+
"loss": 0.8161,
|
| 277 |
+
"step": 1900
|
| 278 |
+
},
|
| 279 |
+
{
|
| 280 |
+
"epoch": 4.83,
|
| 281 |
+
"learning_rate": 0.000432909604519774,
|
| 282 |
+
"loss": 0.8011,
|
| 283 |
+
"step": 1950
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 4.95,
|
| 287 |
+
"learning_rate": 0.0004293785310734463,
|
| 288 |
+
"loss": 0.8408,
|
| 289 |
+
"step": 2000
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 5.0,
|
| 293 |
+
"eval_cer": 0.27762925998534627,
|
| 294 |
+
"eval_loss": 0.43282997608184814,
|
| 295 |
+
"eval_runtime": 36.8381,
|
| 296 |
+
"eval_samples_per_second": 175.47,
|
| 297 |
+
"eval_steps_per_second": 10.967,
|
| 298 |
+
"eval_wer": 0.40753510756743166,
|
| 299 |
+
"step": 2020
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 5.07,
|
| 303 |
+
"learning_rate": 0.00042584745762711863,
|
| 304 |
+
"loss": 0.8286,
|
| 305 |
+
"step": 2050
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 5.2,
|
| 309 |
+
"learning_rate": 0.00042231638418079097,
|
| 310 |
+
"loss": 0.8636,
|
| 311 |
+
"step": 2100
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 5.32,
|
| 315 |
+
"learning_rate": 0.0004187853107344633,
|
| 316 |
+
"loss": 0.8357,
|
| 317 |
+
"step": 2150
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 5.45,
|
| 321 |
+
"learning_rate": 0.0004152542372881356,
|
| 322 |
+
"loss": 0.7934,
|
| 323 |
+
"step": 2200
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 5.57,
|
| 327 |
+
"learning_rate": 0.0004117231638418079,
|
| 328 |
+
"loss": 0.8625,
|
| 329 |
+
"step": 2250
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 5.69,
|
| 333 |
+
"learning_rate": 0.0004081920903954802,
|
| 334 |
+
"loss": 0.7641,
|
| 335 |
+
"step": 2300
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"epoch": 5.82,
|
| 339 |
+
"learning_rate": 0.00040466101694915254,
|
| 340 |
+
"loss": 0.7718,
|
| 341 |
+
"step": 2350
|
| 342 |
+
},
|
| 343 |
+
{
|
| 344 |
+
"epoch": 5.94,
|
| 345 |
+
"learning_rate": 0.0004011299435028249,
|
| 346 |
+
"loss": 0.7759,
|
| 347 |
+
"step": 2400
|
| 348 |
+
},
|
| 349 |
+
{
|
| 350 |
+
"epoch": 6.0,
|
| 351 |
+
"eval_cer": 0.3363432939266124,
|
| 352 |
+
"eval_loss": 0.4735947549343109,
|
| 353 |
+
"eval_runtime": 37.0464,
|
| 354 |
+
"eval_samples_per_second": 174.484,
|
| 355 |
+
"eval_steps_per_second": 10.905,
|
| 356 |
+
"eval_wer": 0.4393563911934777,
|
| 357 |
+
"step": 2424
|
| 358 |
+
},
|
| 359 |
+
{
|
| 360 |
+
"epoch": 6.06,
|
| 361 |
+
"learning_rate": 0.0003975988700564972,
|
| 362 |
+
"loss": 0.7466,
|
| 363 |
+
"step": 2450
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 6.19,
|
| 367 |
+
"learning_rate": 0.0003940677966101695,
|
| 368 |
+
"loss": 0.8045,
|
| 369 |
+
"step": 2500
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 6.31,
|
| 373 |
+
"learning_rate": 0.00039053672316384183,
|
| 374 |
+
"loss": 0.7949,
|
| 375 |
+
"step": 2550
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 6.44,
|
| 379 |
+
"learning_rate": 0.0003870056497175141,
|
| 380 |
+
"loss": 0.7749,
|
| 381 |
+
"step": 2600
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 6.56,
|
| 385 |
+
"learning_rate": 0.00038347457627118644,
|
| 386 |
+
"loss": 0.7226,
|
| 387 |
+
"step": 2650
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 6.68,
|
| 391 |
+
"learning_rate": 0.0003799435028248588,
|
| 392 |
+
"loss": 0.7424,
|
| 393 |
+
"step": 2700
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 6.81,
|
| 397 |
+
"learning_rate": 0.0003764124293785311,
|
| 398 |
+
"loss": 0.7302,
|
| 399 |
+
"step": 2750
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 6.93,
|
| 403 |
+
"learning_rate": 0.0003728813559322034,
|
| 404 |
+
"loss": 0.7228,
|
| 405 |
+
"step": 2800
|
| 406 |
+
},
|
| 407 |
+
{
|
| 408 |
+
"epoch": 7.0,
|
| 409 |
+
"eval_cer": 0.28624329194637516,
|
| 410 |
+
"eval_loss": 0.466670960187912,
|
| 411 |
+
"eval_runtime": 37.148,
|
| 412 |
+
"eval_samples_per_second": 174.007,
|
| 413 |
+
"eval_steps_per_second": 10.875,
|
| 414 |
+
"eval_wer": 0.417268254139281,
|
| 415 |
+
"step": 2828
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 7.05,
|
| 419 |
+
"learning_rate": 0.00036935028248587573,
|
| 420 |
+
"loss": 0.7437,
|
| 421 |
+
"step": 2850
|
| 422 |
+
},
|
| 423 |
+
{
|
| 424 |
+
"epoch": 7.18,
|
| 425 |
+
"learning_rate": 0.000365819209039548,
|
| 426 |
+
"loss": 0.7815,
|
| 427 |
+
"step": 2900
|
| 428 |
+
},
|
| 429 |
+
{
|
| 430 |
+
"epoch": 7.3,
|
| 431 |
+
"learning_rate": 0.0003622881355932203,
|
| 432 |
+
"loss": 0.7218,
|
| 433 |
+
"step": 2950
|
| 434 |
+
},
|
| 435 |
+
{
|
| 436 |
+
"epoch": 7.43,
|
| 437 |
+
"learning_rate": 0.0003587570621468927,
|
| 438 |
+
"loss": 0.7347,
|
| 439 |
+
"step": 3000
|
| 440 |
+
},
|
| 441 |
+
{
|
| 442 |
+
"epoch": 7.55,
|
| 443 |
+
"learning_rate": 0.00035522598870056497,
|
| 444 |
+
"loss": 0.7033,
|
| 445 |
+
"step": 3050
|
| 446 |
+
},
|
| 447 |
+
{
|
| 448 |
+
"epoch": 7.67,
|
| 449 |
+
"learning_rate": 0.0003516949152542373,
|
| 450 |
+
"loss": 0.6966,
|
| 451 |
+
"step": 3100
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 7.8,
|
| 455 |
+
"learning_rate": 0.0003481638418079096,
|
| 456 |
+
"loss": 0.6914,
|
| 457 |
+
"step": 3150
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 7.92,
|
| 461 |
+
"learning_rate": 0.0003446327683615819,
|
| 462 |
+
"loss": 0.6755,
|
| 463 |
+
"step": 3200
|
| 464 |
+
},
|
| 465 |
+
{
|
| 466 |
+
"epoch": 8.0,
|
| 467 |
+
"eval_cer": 0.26113388383928393,
|
| 468 |
+
"eval_loss": 0.4190239906311035,
|
| 469 |
+
"eval_runtime": 37.0113,
|
| 470 |
+
"eval_samples_per_second": 174.649,
|
| 471 |
+
"eval_steps_per_second": 10.916,
|
| 472 |
+
"eval_wer": 0.4114499155981755,
|
| 473 |
+
"step": 3232
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 8.04,
|
| 477 |
+
"learning_rate": 0.0003411016949152542,
|
| 478 |
+
"loss": 0.7279,
|
| 479 |
+
"step": 3250
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 8.17,
|
| 483 |
+
"learning_rate": 0.0003375706214689266,
|
| 484 |
+
"loss": 0.6756,
|
| 485 |
+
"step": 3300
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 8.29,
|
| 489 |
+
"learning_rate": 0.0003340395480225989,
|
| 490 |
+
"loss": 0.6758,
|
| 491 |
+
"step": 3350
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 8.42,
|
| 495 |
+
"learning_rate": 0.0003305084745762712,
|
| 496 |
+
"loss": 0.7251,
|
| 497 |
+
"step": 3400
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 8.54,
|
| 501 |
+
"learning_rate": 0.0003269774011299435,
|
| 502 |
+
"loss": 0.6824,
|
| 503 |
+
"step": 3450
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 8.66,
|
| 507 |
+
"learning_rate": 0.00032344632768361583,
|
| 508 |
+
"loss": 0.7119,
|
| 509 |
+
"step": 3500
|
| 510 |
+
},
|
| 511 |
+
{
|
| 512 |
+
"epoch": 8.79,
|
| 513 |
+
"learning_rate": 0.0003199152542372881,
|
| 514 |
+
"loss": 0.7501,
|
| 515 |
+
"step": 3550
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 8.91,
|
| 519 |
+
"learning_rate": 0.0003163841807909605,
|
| 520 |
+
"loss": 0.634,
|
| 521 |
+
"step": 3600
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 9.0,
|
| 525 |
+
"eval_cer": 0.26115368621160817,
|
| 526 |
+
"eval_loss": 0.4252440333366394,
|
| 527 |
+
"eval_runtime": 37.3698,
|
| 528 |
+
"eval_samples_per_second": 172.974,
|
| 529 |
+
"eval_steps_per_second": 10.811,
|
| 530 |
+
"eval_wer": 0.39927450346586213,
|
| 531 |
+
"step": 3636
|
| 532 |
+
},
|
| 533 |
+
{
|
| 534 |
+
"epoch": 9.03,
|
| 535 |
+
"learning_rate": 0.0003128531073446328,
|
| 536 |
+
"loss": 0.6416,
|
| 537 |
+
"step": 3650
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 9.16,
|
| 541 |
+
"learning_rate": 0.0003093220338983051,
|
| 542 |
+
"loss": 0.6355,
|
| 543 |
+
"step": 3700
|
| 544 |
+
},
|
| 545 |
+
{
|
| 546 |
+
"epoch": 9.28,
|
| 547 |
+
"learning_rate": 0.0003057909604519774,
|
| 548 |
+
"loss": 0.6204,
|
| 549 |
+
"step": 3750
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 9.41,
|
| 553 |
+
"learning_rate": 0.00030225988700564974,
|
| 554 |
+
"loss": 0.6604,
|
| 555 |
+
"step": 3800
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 9.53,
|
| 559 |
+
"learning_rate": 0.000298728813559322,
|
| 560 |
+
"loss": 0.6411,
|
| 561 |
+
"step": 3850
|
| 562 |
+
},
|
| 563 |
+
{
|
| 564 |
+
"epoch": 9.65,
|
| 565 |
+
"learning_rate": 0.0002951977401129944,
|
| 566 |
+
"loss": 0.6156,
|
| 567 |
+
"step": 3900
|
| 568 |
+
},
|
| 569 |
+
{
|
| 570 |
+
"epoch": 9.78,
|
| 571 |
+
"learning_rate": 0.0002916666666666667,
|
| 572 |
+
"loss": 0.6293,
|
| 573 |
+
"step": 3950
|
| 574 |
+
},
|
| 575 |
+
{
|
| 576 |
+
"epoch": 9.9,
|
| 577 |
+
"learning_rate": 0.00028813559322033897,
|
| 578 |
+
"loss": 0.6267,
|
| 579 |
+
"step": 4000
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 10.0,
|
| 583 |
+
"eval_cer": 0.23618289471078635,
|
| 584 |
+
"eval_loss": 0.3275207579135895,
|
| 585 |
+
"eval_runtime": 37.2788,
|
| 586 |
+
"eval_samples_per_second": 173.396,
|
| 587 |
+
"eval_steps_per_second": 10.837,
|
| 588 |
+
"eval_wer": 0.37337930539094205,
|
| 589 |
+
"step": 4040
|
| 590 |
+
},
|
| 591 |
+
{
|
| 592 |
+
"epoch": 10.02,
|
| 593 |
+
"learning_rate": 0.0002846045197740113,
|
| 594 |
+
"loss": 0.6276,
|
| 595 |
+
"step": 4050
|
| 596 |
+
},
|
| 597 |
+
{
|
| 598 |
+
"epoch": 10.15,
|
| 599 |
+
"learning_rate": 0.0002810734463276836,
|
| 600 |
+
"loss": 0.6043,
|
| 601 |
+
"step": 4100
|
| 602 |
+
},
|
| 603 |
+
{
|
| 604 |
+
"epoch": 10.27,
|
| 605 |
+
"learning_rate": 0.0002775423728813559,
|
| 606 |
+
"loss": 0.6354,
|
| 607 |
+
"step": 4150
|
| 608 |
+
},
|
| 609 |
+
{
|
| 610 |
+
"epoch": 10.4,
|
| 611 |
+
"learning_rate": 0.0002740112994350282,
|
| 612 |
+
"loss": 0.6651,
|
| 613 |
+
"step": 4200
|
| 614 |
+
},
|
| 615 |
+
{
|
| 616 |
+
"epoch": 10.52,
|
| 617 |
+
"learning_rate": 0.0002704802259887006,
|
| 618 |
+
"loss": 0.6667,
|
| 619 |
+
"step": 4250
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 10.64,
|
| 623 |
+
"learning_rate": 0.0002669491525423729,
|
| 624 |
+
"loss": 0.6108,
|
| 625 |
+
"step": 4300
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 10.77,
|
| 629 |
+
"learning_rate": 0.0002634180790960452,
|
| 630 |
+
"loss": 0.6034,
|
| 631 |
+
"step": 4350
|
| 632 |
+
},
|
| 633 |
+
{
|
| 634 |
+
"epoch": 10.89,
|
| 635 |
+
"learning_rate": 0.0002598870056497175,
|
| 636 |
+
"loss": 0.6199,
|
| 637 |
+
"step": 4400
|
| 638 |
+
},
|
| 639 |
+
{
|
| 640 |
+
"epoch": 11.0,
|
| 641 |
+
"eval_cer": 0.22224202459454642,
|
| 642 |
+
"eval_loss": 0.2785743474960327,
|
| 643 |
+
"eval_runtime": 37.3388,
|
| 644 |
+
"eval_samples_per_second": 173.117,
|
| 645 |
+
"eval_steps_per_second": 10.82,
|
| 646 |
+
"eval_wer": 0.3543440002873254,
|
| 647 |
+
"step": 4444
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 11.01,
|
| 651 |
+
"learning_rate": 0.00025635593220338983,
|
| 652 |
+
"loss": 0.5854,
|
| 653 |
+
"step": 4450
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 11.14,
|
| 657 |
+
"learning_rate": 0.0002528248587570621,
|
| 658 |
+
"loss": 0.5605,
|
| 659 |
+
"step": 4500
|
| 660 |
+
},
|
| 661 |
+
{
|
| 662 |
+
"epoch": 11.26,
|
| 663 |
+
"learning_rate": 0.00024929378531073445,
|
| 664 |
+
"loss": 0.5665,
|
| 665 |
+
"step": 4550
|
| 666 |
+
},
|
| 667 |
+
{
|
| 668 |
+
"epoch": 11.39,
|
| 669 |
+
"learning_rate": 0.0002457627118644068,
|
| 670 |
+
"loss": 0.6504,
|
| 671 |
+
"step": 4600
|
| 672 |
+
},
|
| 673 |
+
{
|
| 674 |
+
"epoch": 11.51,
|
| 675 |
+
"learning_rate": 0.0002422316384180791,
|
| 676 |
+
"loss": 0.5566,
|
| 677 |
+
"step": 4650
|
| 678 |
+
},
|
| 679 |
+
{
|
| 680 |
+
"epoch": 11.63,
|
| 681 |
+
"learning_rate": 0.0002387005649717514,
|
| 682 |
+
"loss": 0.5574,
|
| 683 |
+
"step": 4700
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 11.76,
|
| 687 |
+
"learning_rate": 0.00023516949152542374,
|
| 688 |
+
"loss": 0.5524,
|
| 689 |
+
"step": 4750
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 11.88,
|
| 693 |
+
"learning_rate": 0.00023163841807909605,
|
| 694 |
+
"loss": 0.5396,
|
| 695 |
+
"step": 4800
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 12.0,
|
| 699 |
+
"eval_cer": 0.2145587041327551,
|
| 700 |
+
"eval_loss": 0.2851468026638031,
|
| 701 |
+
"eval_runtime": 37.0849,
|
| 702 |
+
"eval_samples_per_second": 174.303,
|
| 703 |
+
"eval_steps_per_second": 10.894,
|
| 704 |
+
"eval_wer": 0.3500700355565133,
|
| 705 |
+
"step": 4848
|
| 706 |
+
},
|
| 707 |
+
{
|
| 708 |
+
"epoch": 12.0,
|
| 709 |
+
"learning_rate": 0.00022810734463276836,
|
| 710 |
+
"loss": 0.5405,
|
| 711 |
+
"step": 4850
|
| 712 |
+
},
|
| 713 |
+
{
|
| 714 |
+
"epoch": 12.13,
|
| 715 |
+
"learning_rate": 0.0002245762711864407,
|
| 716 |
+
"loss": 0.5451,
|
| 717 |
+
"step": 4900
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 12.25,
|
| 721 |
+
"learning_rate": 0.000221045197740113,
|
| 722 |
+
"loss": 0.5538,
|
| 723 |
+
"step": 4950
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 12.38,
|
| 727 |
+
"learning_rate": 0.0002175141242937853,
|
| 728 |
+
"loss": 0.5403,
|
| 729 |
+
"step": 5000
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 12.5,
|
| 733 |
+
"learning_rate": 0.00021398305084745765,
|
| 734 |
+
"loss": 0.5506,
|
| 735 |
+
"step": 5050
|
| 736 |
+
},
|
| 737 |
+
{
|
| 738 |
+
"epoch": 12.62,
|
| 739 |
+
"learning_rate": 0.00021045197740112996,
|
| 740 |
+
"loss": 0.54,
|
| 741 |
+
"step": 5100
|
| 742 |
+
},
|
| 743 |
+
{
|
| 744 |
+
"epoch": 12.75,
|
| 745 |
+
"learning_rate": 0.00020692090395480226,
|
| 746 |
+
"loss": 0.4959,
|
| 747 |
+
"step": 5150
|
| 748 |
+
},
|
| 749 |
+
{
|
| 750 |
+
"epoch": 12.87,
|
| 751 |
+
"learning_rate": 0.0002033898305084746,
|
| 752 |
+
"loss": 0.5268,
|
| 753 |
+
"step": 5200
|
| 754 |
+
},
|
| 755 |
+
{
|
| 756 |
+
"epoch": 13.0,
|
| 757 |
+
"learning_rate": 0.0001998587570621469,
|
| 758 |
+
"loss": 0.5343,
|
| 759 |
+
"step": 5250
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 13.0,
|
| 763 |
+
"eval_cer": 0.2105586249232658,
|
| 764 |
+
"eval_loss": 0.25269845128059387,
|
| 765 |
+
"eval_runtime": 37.2453,
|
| 766 |
+
"eval_samples_per_second": 173.552,
|
| 767 |
+
"eval_steps_per_second": 10.847,
|
| 768 |
+
"eval_wer": 0.344826347735517,
|
| 769 |
+
"step": 5252
|
| 770 |
+
},
|
| 771 |
+
{
|
| 772 |
+
"epoch": 13.12,
|
| 773 |
+
"learning_rate": 0.00019632768361581922,
|
| 774 |
+
"loss": 0.4924,
|
| 775 |
+
"step": 5300
|
| 776 |
+
},
|
| 777 |
+
{
|
| 778 |
+
"epoch": 13.24,
|
| 779 |
+
"learning_rate": 0.00019279661016949155,
|
| 780 |
+
"loss": 0.5308,
|
| 781 |
+
"step": 5350
|
| 782 |
+
},
|
| 783 |
+
{
|
| 784 |
+
"epoch": 13.37,
|
| 785 |
+
"learning_rate": 0.00018926553672316386,
|
| 786 |
+
"loss": 0.5263,
|
| 787 |
+
"step": 5400
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 13.49,
|
| 791 |
+
"learning_rate": 0.00018573446327683614,
|
| 792 |
+
"loss": 0.498,
|
| 793 |
+
"step": 5450
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 13.61,
|
| 797 |
+
"learning_rate": 0.00018220338983050845,
|
| 798 |
+
"loss": 0.5096,
|
| 799 |
+
"step": 5500
|
| 800 |
+
},
|
| 801 |
+
{
|
| 802 |
+
"epoch": 13.74,
|
| 803 |
+
"learning_rate": 0.0001786723163841808,
|
| 804 |
+
"loss": 0.5077,
|
| 805 |
+
"step": 5550
|
| 806 |
+
},
|
| 807 |
+
{
|
| 808 |
+
"epoch": 13.86,
|
| 809 |
+
"learning_rate": 0.0001751412429378531,
|
| 810 |
+
"loss": 0.5295,
|
| 811 |
+
"step": 5600
|
| 812 |
+
},
|
| 813 |
+
{
|
| 814 |
+
"epoch": 13.99,
|
| 815 |
+
"learning_rate": 0.0001716101694915254,
|
| 816 |
+
"loss": 0.5488,
|
| 817 |
+
"step": 5650
|
| 818 |
+
},
|
| 819 |
+
{
|
| 820 |
+
"epoch": 14.0,
|
| 821 |
+
"eval_cer": 0.20998435612586389,
|
| 822 |
+
"eval_loss": 0.2725040316581726,
|
| 823 |
+
"eval_runtime": 36.9929,
|
| 824 |
+
"eval_samples_per_second": 174.736,
|
| 825 |
+
"eval_steps_per_second": 10.921,
|
| 826 |
+
"eval_wer": 0.3431383112451963,
|
| 827 |
+
"step": 5656
|
| 828 |
+
},
|
| 829 |
+
{
|
| 830 |
+
"epoch": 14.11,
|
| 831 |
+
"learning_rate": 0.00016807909604519774,
|
| 832 |
+
"loss": 0.4966,
|
| 833 |
+
"step": 5700
|
| 834 |
+
},
|
| 835 |
+
{
|
| 836 |
+
"epoch": 14.23,
|
| 837 |
+
"learning_rate": 0.00016454802259887005,
|
| 838 |
+
"loss": 0.5045,
|
| 839 |
+
"step": 5750
|
| 840 |
+
},
|
| 841 |
+
{
|
| 842 |
+
"epoch": 14.36,
|
| 843 |
+
"learning_rate": 0.00016101694915254236,
|
| 844 |
+
"loss": 0.4877,
|
| 845 |
+
"step": 5800
|
| 846 |
+
},
|
| 847 |
+
{
|
| 848 |
+
"epoch": 14.48,
|
| 849 |
+
"learning_rate": 0.0001574858757062147,
|
| 850 |
+
"loss": 0.4886,
|
| 851 |
+
"step": 5850
|
| 852 |
+
},
|
| 853 |
+
{
|
| 854 |
+
"epoch": 14.6,
|
| 855 |
+
"learning_rate": 0.000153954802259887,
|
| 856 |
+
"loss": 0.5026,
|
| 857 |
+
"step": 5900
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 14.73,
|
| 861 |
+
"learning_rate": 0.00015042372881355931,
|
| 862 |
+
"loss": 0.4759,
|
| 863 |
+
"step": 5950
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 14.85,
|
| 867 |
+
"learning_rate": 0.00014689265536723165,
|
| 868 |
+
"loss": 0.475,
|
| 869 |
+
"step": 6000
|
| 870 |
+
},
|
| 871 |
+
{
|
| 872 |
+
"epoch": 14.98,
|
| 873 |
+
"learning_rate": 0.00014336158192090396,
|
| 874 |
+
"loss": 0.4606,
|
| 875 |
+
"step": 6050
|
| 876 |
+
},
|
| 877 |
+
{
|
| 878 |
+
"epoch": 15.0,
|
| 879 |
+
"eval_cer": 0.1962019049882176,
|
| 880 |
+
"eval_loss": 0.22929197549819946,
|
| 881 |
+
"eval_runtime": 37.0601,
|
| 882 |
+
"eval_samples_per_second": 174.419,
|
| 883 |
+
"eval_steps_per_second": 10.901,
|
| 884 |
+
"eval_wer": 0.32589878964192076,
|
| 885 |
+
"step": 6060
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 15.1,
|
| 889 |
+
"learning_rate": 0.00013983050847457627,
|
| 890 |
+
"loss": 0.4564,
|
| 891 |
+
"step": 6100
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 15.22,
|
| 895 |
+
"learning_rate": 0.0001362994350282486,
|
| 896 |
+
"loss": 0.4658,
|
| 897 |
+
"step": 6150
|
| 898 |
+
},
|
| 899 |
+
{
|
| 900 |
+
"epoch": 15.35,
|
| 901 |
+
"learning_rate": 0.0001327683615819209,
|
| 902 |
+
"loss": 0.4465,
|
| 903 |
+
"step": 6200
|
| 904 |
+
},
|
| 905 |
+
{
|
| 906 |
+
"epoch": 15.47,
|
| 907 |
+
"learning_rate": 0.00012923728813559322,
|
| 908 |
+
"loss": 0.4867,
|
| 909 |
+
"step": 6250
|
| 910 |
+
},
|
| 911 |
+
{
|
| 912 |
+
"epoch": 15.59,
|
| 913 |
+
"learning_rate": 0.00012570621468926556,
|
| 914 |
+
"loss": 0.4349,
|
| 915 |
+
"step": 6300
|
| 916 |
+
},
|
| 917 |
+
{
|
| 918 |
+
"epoch": 15.72,
|
| 919 |
+
"learning_rate": 0.00012217514124293787,
|
| 920 |
+
"loss": 0.475,
|
| 921 |
+
"step": 6350
|
| 922 |
+
},
|
| 923 |
+
{
|
| 924 |
+
"epoch": 15.84,
|
| 925 |
+
"learning_rate": 0.00011864406779661017,
|
| 926 |
+
"loss": 0.4241,
|
| 927 |
+
"step": 6400
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 15.97,
|
| 931 |
+
"learning_rate": 0.00011511299435028248,
|
| 932 |
+
"loss": 0.4229,
|
| 933 |
+
"step": 6450
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 16.0,
|
| 937 |
+
"eval_cer": 0.19142953325808432,
|
| 938 |
+
"eval_loss": 0.20434732735157013,
|
| 939 |
+
"eval_runtime": 36.9841,
|
| 940 |
+
"eval_samples_per_second": 174.778,
|
| 941 |
+
"eval_steps_per_second": 10.924,
|
| 942 |
+
"eval_wer": 0.3172071975002694,
|
| 943 |
+
"step": 6464
|
| 944 |
+
},
|
| 945 |
+
{
|
| 946 |
+
"epoch": 16.09,
|
| 947 |
+
"learning_rate": 0.0001115819209039548,
|
| 948 |
+
"loss": 0.429,
|
| 949 |
+
"step": 6500
|
| 950 |
+
},
|
| 951 |
+
{
|
| 952 |
+
"epoch": 16.21,
|
| 953 |
+
"learning_rate": 0.00010805084745762713,
|
| 954 |
+
"loss": 0.409,
|
| 955 |
+
"step": 6550
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 16.34,
|
| 959 |
+
"learning_rate": 0.00010451977401129944,
|
| 960 |
+
"loss": 0.3961,
|
| 961 |
+
"step": 6600
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 16.46,
|
| 965 |
+
"learning_rate": 0.00010098870056497176,
|
| 966 |
+
"loss": 0.4047,
|
| 967 |
+
"step": 6650
|
| 968 |
+
},
|
| 969 |
+
{
|
| 970 |
+
"epoch": 16.58,
|
| 971 |
+
"learning_rate": 9.745762711864407e-05,
|
| 972 |
+
"loss": 0.4344,
|
| 973 |
+
"step": 6700
|
| 974 |
+
},
|
| 975 |
+
{
|
| 976 |
+
"epoch": 16.71,
|
| 977 |
+
"learning_rate": 9.392655367231638e-05,
|
| 978 |
+
"loss": 0.4268,
|
| 979 |
+
"step": 6750
|
| 980 |
+
},
|
| 981 |
+
{
|
| 982 |
+
"epoch": 16.83,
|
| 983 |
+
"learning_rate": 9.03954802259887e-05,
|
| 984 |
+
"loss": 0.3991,
|
| 985 |
+
"step": 6800
|
| 986 |
+
},
|
| 987 |
+
{
|
| 988 |
+
"epoch": 16.96,
|
| 989 |
+
"learning_rate": 8.686440677966102e-05,
|
| 990 |
+
"loss": 0.4078,
|
| 991 |
+
"step": 6850
|
| 992 |
+
},
|
| 993 |
+
{
|
| 994 |
+
"epoch": 17.0,
|
| 995 |
+
"eval_cer": 0.18618190459217016,
|
| 996 |
+
"eval_loss": 0.1891370713710785,
|
| 997 |
+
"eval_runtime": 37.0446,
|
| 998 |
+
"eval_samples_per_second": 174.492,
|
| 999 |
+
"eval_steps_per_second": 10.906,
|
| 1000 |
+
"eval_wer": 0.3127536544194232,
|
| 1001 |
+
"step": 6868
|
| 1002 |
+
},
|
| 1003 |
+
{
|
| 1004 |
+
"epoch": 17.08,
|
| 1005 |
+
"learning_rate": 8.333333333333333e-05,
|
| 1006 |
+
"loss": 0.3941,
|
| 1007 |
+
"step": 6900
|
| 1008 |
+
},
|
| 1009 |
+
{
|
| 1010 |
+
"epoch": 17.2,
|
| 1011 |
+
"learning_rate": 7.980225988700565e-05,
|
| 1012 |
+
"loss": 0.382,
|
| 1013 |
+
"step": 6950
|
| 1014 |
+
},
|
| 1015 |
+
{
|
| 1016 |
+
"epoch": 17.33,
|
| 1017 |
+
"learning_rate": 7.627118644067798e-05,
|
| 1018 |
+
"loss": 0.3898,
|
| 1019 |
+
"step": 7000
|
| 1020 |
+
},
|
| 1021 |
+
{
|
| 1022 |
+
"epoch": 17.45,
|
| 1023 |
+
"learning_rate": 7.274011299435028e-05,
|
| 1024 |
+
"loss": 0.3941,
|
| 1025 |
+
"step": 7050
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 17.57,
|
| 1029 |
+
"learning_rate": 6.92090395480226e-05,
|
| 1030 |
+
"loss": 0.3804,
|
| 1031 |
+
"step": 7100
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 17.7,
|
| 1035 |
+
"learning_rate": 6.567796610169493e-05,
|
| 1036 |
+
"loss": 0.3924,
|
| 1037 |
+
"step": 7150
|
| 1038 |
+
},
|
| 1039 |
+
{
|
| 1040 |
+
"epoch": 17.82,
|
| 1041 |
+
"learning_rate": 6.214689265536724e-05,
|
| 1042 |
+
"loss": 0.3849,
|
| 1043 |
+
"step": 7200
|
| 1044 |
+
},
|
| 1045 |
+
{
|
| 1046 |
+
"epoch": 17.95,
|
| 1047 |
+
"learning_rate": 5.861581920903955e-05,
|
| 1048 |
+
"loss": 0.4017,
|
| 1049 |
+
"step": 7250
|
| 1050 |
+
},
|
| 1051 |
+
{
|
| 1052 |
+
"epoch": 18.0,
|
| 1053 |
+
"eval_cer": 0.1833105606051605,
|
| 1054 |
+
"eval_loss": 0.17852339148521423,
|
| 1055 |
+
"eval_runtime": 37.3017,
|
| 1056 |
+
"eval_samples_per_second": 173.29,
|
| 1057 |
+
"eval_steps_per_second": 10.831,
|
| 1058 |
+
"eval_wer": 0.3075458822684337,
|
| 1059 |
+
"step": 7272
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 18.07,
|
| 1063 |
+
"learning_rate": 5.508474576271186e-05,
|
| 1064 |
+
"loss": 0.4008,
|
| 1065 |
+
"step": 7300
|
| 1066 |
+
},
|
| 1067 |
+
{
|
| 1068 |
+
"epoch": 18.19,
|
| 1069 |
+
"learning_rate": 5.155367231638418e-05,
|
| 1070 |
+
"loss": 0.3744,
|
| 1071 |
+
"step": 7350
|
| 1072 |
+
},
|
| 1073 |
+
{
|
| 1074 |
+
"epoch": 18.32,
|
| 1075 |
+
"learning_rate": 4.80225988700565e-05,
|
| 1076 |
+
"loss": 0.3699,
|
| 1077 |
+
"step": 7400
|
| 1078 |
+
},
|
| 1079 |
+
{
|
| 1080 |
+
"epoch": 18.44,
|
| 1081 |
+
"learning_rate": 4.4491525423728816e-05,
|
| 1082 |
+
"loss": 0.3922,
|
| 1083 |
+
"step": 7450
|
| 1084 |
+
},
|
| 1085 |
+
{
|
| 1086 |
+
"epoch": 18.56,
|
| 1087 |
+
"learning_rate": 4.096045197740113e-05,
|
| 1088 |
+
"loss": 0.3734,
|
| 1089 |
+
"step": 7500
|
| 1090 |
+
},
|
| 1091 |
+
{
|
| 1092 |
+
"epoch": 18.69,
|
| 1093 |
+
"learning_rate": 3.742937853107345e-05,
|
| 1094 |
+
"loss": 0.3822,
|
| 1095 |
+
"step": 7550
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 18.81,
|
| 1099 |
+
"learning_rate": 3.389830508474576e-05,
|
| 1100 |
+
"loss": 0.3707,
|
| 1101 |
+
"step": 7600
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 18.94,
|
| 1105 |
+
"learning_rate": 3.036723163841808e-05,
|
| 1106 |
+
"loss": 0.3618,
|
| 1107 |
+
"step": 7650
|
| 1108 |
+
},
|
| 1109 |
+
{
|
| 1110 |
+
"epoch": 19.0,
|
| 1111 |
+
"eval_cer": 0.18034020475652984,
|
| 1112 |
+
"eval_loss": 0.16725142300128937,
|
| 1113 |
+
"eval_runtime": 36.9107,
|
| 1114 |
+
"eval_samples_per_second": 175.125,
|
| 1115 |
+
"eval_steps_per_second": 10.945,
|
| 1116 |
+
"eval_wer": 0.3034514958876558,
|
| 1117 |
+
"step": 7676
|
| 1118 |
+
},
|
| 1119 |
+
{
|
| 1120 |
+
"epoch": 19.06,
|
| 1121 |
+
"learning_rate": 2.6836158192090393e-05,
|
| 1122 |
+
"loss": 0.3507,
|
| 1123 |
+
"step": 7700
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 19.18,
|
| 1127 |
+
"learning_rate": 2.3305084745762712e-05,
|
| 1128 |
+
"loss": 0.3638,
|
| 1129 |
+
"step": 7750
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 19.31,
|
| 1133 |
+
"learning_rate": 1.977401129943503e-05,
|
| 1134 |
+
"loss": 0.3819,
|
| 1135 |
+
"step": 7800
|
| 1136 |
+
},
|
| 1137 |
+
{
|
| 1138 |
+
"epoch": 19.43,
|
| 1139 |
+
"learning_rate": 1.6242937853107344e-05,
|
| 1140 |
+
"loss": 0.3713,
|
| 1141 |
+
"step": 7850
|
| 1142 |
+
},
|
| 1143 |
+
{
|
| 1144 |
+
"epoch": 19.55,
|
| 1145 |
+
"learning_rate": 1.2711864406779663e-05,
|
| 1146 |
+
"loss": 0.3702,
|
| 1147 |
+
"step": 7900
|
| 1148 |
+
},
|
| 1149 |
+
{
|
| 1150 |
+
"epoch": 19.68,
|
| 1151 |
+
"learning_rate": 9.180790960451978e-06,
|
| 1152 |
+
"loss": 0.3559,
|
| 1153 |
+
"step": 7950
|
| 1154 |
+
},
|
| 1155 |
+
{
|
| 1156 |
+
"epoch": 19.8,
|
| 1157 |
+
"learning_rate": 5.649717514124294e-06,
|
| 1158 |
+
"loss": 0.3567,
|
| 1159 |
+
"step": 8000
|
| 1160 |
+
},
|
| 1161 |
+
{
|
| 1162 |
+
"epoch": 19.93,
|
| 1163 |
+
"learning_rate": 2.11864406779661e-06,
|
| 1164 |
+
"loss": 0.3739,
|
| 1165 |
+
"step": 8050
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 20.0,
|
| 1169 |
+
"eval_cer": 0.17857779361967563,
|
| 1170 |
+
"eval_loss": 0.16060513257980347,
|
| 1171 |
+
"eval_runtime": 37.2075,
|
| 1172 |
+
"eval_samples_per_second": 173.728,
|
| 1173 |
+
"eval_steps_per_second": 10.858,
|
| 1174 |
+
"eval_wer": 0.30036274826706894,
|
| 1175 |
+
"step": 8080
|
| 1176 |
+
},
|
| 1177 |
+
{
|
| 1178 |
+
"epoch": 20.0,
|
| 1179 |
+
"step": 8080,
|
| 1180 |
+
"total_flos": 4.3127988037094564e+18,
|
| 1181 |
+
"train_loss": 0.7708885439551703,
|
| 1182 |
+
"train_runtime": 17348.5062,
|
| 1183 |
+
"train_samples_per_second": 29.808,
|
| 1184 |
+
"train_steps_per_second": 0.466
|
| 1185 |
+
}
|
| 1186 |
+
],
|
| 1187 |
+
"logging_steps": 50,
|
| 1188 |
+
"max_steps": 8080,
|
| 1189 |
+
"num_train_epochs": 20,
|
| 1190 |
+
"save_steps": 500,
|
| 1191 |
+
"total_flos": 4.3127988037094564e+18,
|
| 1192 |
+
"trial_name": null,
|
| 1193 |
+
"trial_params": null
|
| 1194 |
+
}
|