End of training
Browse files- README.md +14 -2
- all_results.json +10 -10
- eval_results.json +6 -6
- train_results.json +5 -5
- trainer_state.json +911 -14
README.md
CHANGED
|
@@ -3,11 +3,23 @@ license: apache-2.0
|
|
| 3 |
base_model: openai/whisper-base
|
| 4 |
tags:
|
| 5 |
- generated_from_trainer
|
|
|
|
|
|
|
| 6 |
metrics:
|
| 7 |
- wer
|
| 8 |
model-index:
|
| 9 |
- name: whisper-base-ckb
|
| 10 |
-
results:
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 11 |
---
|
| 12 |
|
| 13 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
@@ -15,7 +27,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
# whisper-base-ckb
|
| 17 |
|
| 18 |
-
This model is a fine-tuned version of [openai/whisper-base](https://huggingface.co/openai/whisper-base) on
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
- Loss: 0.0641
|
| 21 |
- Wer: 0.1262
|
|
|
|
| 3 |
base_model: openai/whisper-base
|
| 4 |
tags:
|
| 5 |
- generated_from_trainer
|
| 6 |
+
datasets:
|
| 7 |
+
- razhan/common_voice_ckb_16
|
| 8 |
metrics:
|
| 9 |
- wer
|
| 10 |
model-index:
|
| 11 |
- name: whisper-base-ckb
|
| 12 |
+
results:
|
| 13 |
+
- task:
|
| 14 |
+
name: Automatic Speech Recognition
|
| 15 |
+
type: automatic-speech-recognition
|
| 16 |
+
dataset:
|
| 17 |
+
name: razhan/common_voice_ckb_16
|
| 18 |
+
type: razhan/common_voice_ckb_16
|
| 19 |
+
metrics:
|
| 20 |
+
- name: Wer
|
| 21 |
+
type: wer
|
| 22 |
+
value: 0.12623194275685162
|
| 23 |
---
|
| 24 |
|
| 25 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
|
| 27 |
|
| 28 |
# whisper-base-ckb
|
| 29 |
|
| 30 |
+
This model is a fine-tuned version of [openai/whisper-base](https://huggingface.co/openai/whisper-base) on the razhan/common_voice_ckb_16 dataset.
|
| 31 |
It achieves the following results on the evaluation set:
|
| 32 |
- Loss: 0.0641
|
| 33 |
- Wer: 0.1262
|
all_results.json
CHANGED
|
@@ -1,14 +1,14 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_loss": 0.
|
| 4 |
-
"eval_runtime":
|
| 5 |
"eval_samples": 4940,
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second": 0.
|
| 8 |
-
"eval_wer": 0.
|
| 9 |
-
"train_loss": 0.
|
| 10 |
-
"train_runtime":
|
| 11 |
"train_samples": 105929,
|
| 12 |
-
"train_samples_per_second":
|
| 13 |
-
"train_steps_per_second": 0.
|
| 14 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 25.0,
|
| 3 |
+
"eval_loss": 0.0640869140625,
|
| 4 |
+
"eval_runtime": 115.2612,
|
| 5 |
"eval_samples": 4940,
|
| 6 |
+
"eval_samples_per_second": 42.859,
|
| 7 |
+
"eval_steps_per_second": 0.061,
|
| 8 |
+
"eval_wer": 0.12623194275685162,
|
| 9 |
+
"train_loss": 0.0027616678631823995,
|
| 10 |
+
"train_runtime": 6144.0313,
|
| 11 |
"train_samples": 105929,
|
| 12 |
+
"train_samples_per_second": 431.248,
|
| 13 |
+
"train_steps_per_second": 0.374
|
| 14 |
}
|
eval_results.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_loss": 0.
|
| 4 |
-
"eval_runtime":
|
| 5 |
"eval_samples": 4940,
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second": 0.
|
| 8 |
-
"eval_wer": 0.
|
| 9 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 25.0,
|
| 3 |
+
"eval_loss": 0.0640869140625,
|
| 4 |
+
"eval_runtime": 115.2612,
|
| 5 |
"eval_samples": 4940,
|
| 6 |
+
"eval_samples_per_second": 42.859,
|
| 7 |
+
"eval_steps_per_second": 0.061,
|
| 8 |
+
"eval_wer": 0.12623194275685162
|
| 9 |
}
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"train_loss": 0.
|
| 4 |
-
"train_runtime":
|
| 5 |
"train_samples": 105929,
|
| 6 |
-
"train_samples_per_second":
|
| 7 |
-
"train_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 25.0,
|
| 3 |
+
"train_loss": 0.0027616678631823995,
|
| 4 |
+
"train_runtime": 6144.0313,
|
| 5 |
"train_samples": 105929,
|
| 6 |
+
"train_samples_per_second": 431.248,
|
| 7 |
+
"train_steps_per_second": 0.374
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 0.
|
| 3 |
-
"best_model_checkpoint": "./whisper-base-ckb/checkpoint-
|
| 4 |
-
"epoch":
|
| 5 |
"eval_steps": 100,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
@@ -699,21 +699,918 @@
|
|
| 699 |
"step": 1000
|
| 700 |
},
|
| 701 |
{
|
| 702 |
-
"epoch": 10.
|
| 703 |
-
"
|
| 704 |
-
"
|
| 705 |
-
"
|
| 706 |
-
|
| 707 |
-
|
| 708 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 709 |
}
|
| 710 |
],
|
| 711 |
"logging_steps": 10,
|
| 712 |
-
"max_steps":
|
| 713 |
"num_input_tokens_seen": 0,
|
| 714 |
-
"num_train_epochs":
|
| 715 |
"save_steps": 100,
|
| 716 |
-
"total_flos":
|
| 717 |
"train_batch_size": 192,
|
| 718 |
"trial_name": null,
|
| 719 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 0.12623194275685162,
|
| 3 |
+
"best_model_checkpoint": "./whisper-base-ckb/checkpoint-2300",
|
| 4 |
+
"epoch": 25.0,
|
| 5 |
"eval_steps": 100,
|
| 6 |
+
"global_step": 2300,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
|
|
| 699 |
"step": 1000
|
| 700 |
},
|
| 701 |
{
|
| 702 |
+
"epoch": 10.98,
|
| 703 |
+
"learning_rate": 7.889473684210527e-06,
|
| 704 |
+
"loss": 0.0921,
|
| 705 |
+
"step": 1010
|
| 706 |
+
},
|
| 707 |
+
{
|
| 708 |
+
"epoch": 11.09,
|
| 709 |
+
"learning_rate": 7.863157894736842e-06,
|
| 710 |
+
"loss": 0.0904,
|
| 711 |
+
"step": 1020
|
| 712 |
+
},
|
| 713 |
+
{
|
| 714 |
+
"epoch": 11.2,
|
| 715 |
+
"learning_rate": 7.836842105263159e-06,
|
| 716 |
+
"loss": 0.0983,
|
| 717 |
+
"step": 1030
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 11.3,
|
| 721 |
+
"learning_rate": 7.810526315789474e-06,
|
| 722 |
+
"loss": 0.0976,
|
| 723 |
+
"step": 1040
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 11.41,
|
| 727 |
+
"learning_rate": 7.78421052631579e-06,
|
| 728 |
+
"loss": 0.094,
|
| 729 |
+
"step": 1050
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 11.52,
|
| 733 |
+
"learning_rate": 7.757894736842105e-06,
|
| 734 |
+
"loss": 0.0908,
|
| 735 |
+
"step": 1060
|
| 736 |
+
},
|
| 737 |
+
{
|
| 738 |
+
"epoch": 11.63,
|
| 739 |
+
"learning_rate": 7.731578947368422e-06,
|
| 740 |
+
"loss": 0.0882,
|
| 741 |
+
"step": 1070
|
| 742 |
+
},
|
| 743 |
+
{
|
| 744 |
+
"epoch": 11.74,
|
| 745 |
+
"learning_rate": 7.705263157894738e-06,
|
| 746 |
+
"loss": 0.096,
|
| 747 |
+
"step": 1080
|
| 748 |
+
},
|
| 749 |
+
{
|
| 750 |
+
"epoch": 11.85,
|
| 751 |
+
"learning_rate": 7.678947368421053e-06,
|
| 752 |
+
"loss": 0.0949,
|
| 753 |
+
"step": 1090
|
| 754 |
+
},
|
| 755 |
+
{
|
| 756 |
+
"epoch": 11.96,
|
| 757 |
+
"learning_rate": 7.65263157894737e-06,
|
| 758 |
+
"loss": 0.0919,
|
| 759 |
+
"step": 1100
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 11.96,
|
| 763 |
+
"eval_loss": 0.1314697265625,
|
| 764 |
+
"eval_runtime": 111.7237,
|
| 765 |
+
"eval_samples_per_second": 44.216,
|
| 766 |
+
"eval_steps_per_second": 0.063,
|
| 767 |
+
"eval_wer": 0.27420008100445525,
|
| 768 |
+
"step": 1100
|
| 769 |
+
},
|
| 770 |
+
{
|
| 771 |
+
"epoch": 12.07,
|
| 772 |
+
"learning_rate": 7.626315789473685e-06,
|
| 773 |
+
"loss": 0.0866,
|
| 774 |
+
"step": 1110
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 12.17,
|
| 778 |
+
"learning_rate": 7.600000000000001e-06,
|
| 779 |
+
"loss": 0.0797,
|
| 780 |
+
"step": 1120
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 12.28,
|
| 784 |
+
"learning_rate": 7.573684210526317e-06,
|
| 785 |
+
"loss": 0.0859,
|
| 786 |
+
"step": 1130
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 12.39,
|
| 790 |
+
"learning_rate": 7.547368421052632e-06,
|
| 791 |
+
"loss": 0.0869,
|
| 792 |
+
"step": 1140
|
| 793 |
+
},
|
| 794 |
+
{
|
| 795 |
+
"epoch": 12.5,
|
| 796 |
+
"learning_rate": 7.5210526315789475e-06,
|
| 797 |
+
"loss": 0.0843,
|
| 798 |
+
"step": 1150
|
| 799 |
+
},
|
| 800 |
+
{
|
| 801 |
+
"epoch": 12.61,
|
| 802 |
+
"learning_rate": 7.494736842105263e-06,
|
| 803 |
+
"loss": 0.083,
|
| 804 |
+
"step": 1160
|
| 805 |
+
},
|
| 806 |
+
{
|
| 807 |
+
"epoch": 12.72,
|
| 808 |
+
"learning_rate": 7.468421052631579e-06,
|
| 809 |
+
"loss": 0.0784,
|
| 810 |
+
"step": 1170
|
| 811 |
+
},
|
| 812 |
+
{
|
| 813 |
+
"epoch": 12.83,
|
| 814 |
+
"learning_rate": 7.442105263157895e-06,
|
| 815 |
+
"loss": 0.0853,
|
| 816 |
+
"step": 1180
|
| 817 |
+
},
|
| 818 |
+
{
|
| 819 |
+
"epoch": 12.93,
|
| 820 |
+
"learning_rate": 7.415789473684211e-06,
|
| 821 |
+
"loss": 0.0861,
|
| 822 |
+
"step": 1190
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 13.04,
|
| 826 |
+
"learning_rate": 7.3894736842105275e-06,
|
| 827 |
+
"loss": 0.0839,
|
| 828 |
+
"step": 1200
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 13.04,
|
| 832 |
+
"eval_loss": 0.1217041015625,
|
| 833 |
+
"eval_runtime": 110.1755,
|
| 834 |
+
"eval_samples_per_second": 44.838,
|
| 835 |
+
"eval_steps_per_second": 0.064,
|
| 836 |
+
"eval_wer": 0.2596867827730525,
|
| 837 |
+
"step": 1200
|
| 838 |
+
},
|
| 839 |
+
{
|
| 840 |
+
"epoch": 13.15,
|
| 841 |
+
"learning_rate": 7.363157894736843e-06,
|
| 842 |
+
"loss": 0.0759,
|
| 843 |
+
"step": 1210
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 13.26,
|
| 847 |
+
"learning_rate": 7.336842105263159e-06,
|
| 848 |
+
"loss": 0.0733,
|
| 849 |
+
"step": 1220
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 13.37,
|
| 853 |
+
"learning_rate": 7.310526315789475e-06,
|
| 854 |
+
"loss": 0.0768,
|
| 855 |
+
"step": 1230
|
| 856 |
+
},
|
| 857 |
+
{
|
| 858 |
+
"epoch": 13.48,
|
| 859 |
+
"learning_rate": 7.28421052631579e-06,
|
| 860 |
+
"loss": 0.0834,
|
| 861 |
+
"step": 1240
|
| 862 |
+
},
|
| 863 |
+
{
|
| 864 |
+
"epoch": 13.59,
|
| 865 |
+
"learning_rate": 7.257894736842106e-06,
|
| 866 |
+
"loss": 0.077,
|
| 867 |
+
"step": 1250
|
| 868 |
+
},
|
| 869 |
+
{
|
| 870 |
+
"epoch": 13.7,
|
| 871 |
+
"learning_rate": 7.2315789473684215e-06,
|
| 872 |
+
"loss": 0.0738,
|
| 873 |
+
"step": 1260
|
| 874 |
+
},
|
| 875 |
+
{
|
| 876 |
+
"epoch": 13.8,
|
| 877 |
+
"learning_rate": 7.205263157894737e-06,
|
| 878 |
+
"loss": 0.0723,
|
| 879 |
+
"step": 1270
|
| 880 |
+
},
|
| 881 |
+
{
|
| 882 |
+
"epoch": 13.91,
|
| 883 |
+
"learning_rate": 7.178947368421053e-06,
|
| 884 |
+
"loss": 0.0752,
|
| 885 |
+
"step": 1280
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 14.02,
|
| 889 |
+
"learning_rate": 7.152631578947369e-06,
|
| 890 |
+
"loss": 0.0794,
|
| 891 |
+
"step": 1290
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 14.13,
|
| 895 |
+
"learning_rate": 7.126315789473685e-06,
|
| 896 |
+
"loss": 0.0713,
|
| 897 |
+
"step": 1300
|
| 898 |
+
},
|
| 899 |
+
{
|
| 900 |
+
"epoch": 14.13,
|
| 901 |
+
"eval_loss": 0.11322021484375,
|
| 902 |
+
"eval_runtime": 110.9942,
|
| 903 |
+
"eval_samples_per_second": 44.507,
|
| 904 |
+
"eval_steps_per_second": 0.063,
|
| 905 |
+
"eval_wer": 0.23710679087349804,
|
| 906 |
+
"step": 1300
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 14.24,
|
| 910 |
+
"learning_rate": 7.100000000000001e-06,
|
| 911 |
+
"loss": 0.0703,
|
| 912 |
+
"step": 1310
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 14.35,
|
| 916 |
+
"learning_rate": 7.073684210526316e-06,
|
| 917 |
+
"loss": 0.0662,
|
| 918 |
+
"step": 1320
|
| 919 |
+
},
|
| 920 |
+
{
|
| 921 |
+
"epoch": 14.46,
|
| 922 |
+
"learning_rate": 7.047368421052631e-06,
|
| 923 |
+
"loss": 0.0686,
|
| 924 |
+
"step": 1330
|
| 925 |
+
},
|
| 926 |
+
{
|
| 927 |
+
"epoch": 14.57,
|
| 928 |
+
"learning_rate": 7.021052631578948e-06,
|
| 929 |
+
"loss": 0.0771,
|
| 930 |
+
"step": 1340
|
| 931 |
+
},
|
| 932 |
+
{
|
| 933 |
+
"epoch": 14.67,
|
| 934 |
+
"learning_rate": 6.994736842105264e-06,
|
| 935 |
+
"loss": 0.0717,
|
| 936 |
+
"step": 1350
|
| 937 |
+
},
|
| 938 |
+
{
|
| 939 |
+
"epoch": 14.78,
|
| 940 |
+
"learning_rate": 6.96842105263158e-06,
|
| 941 |
+
"loss": 0.071,
|
| 942 |
+
"step": 1360
|
| 943 |
+
},
|
| 944 |
+
{
|
| 945 |
+
"epoch": 14.89,
|
| 946 |
+
"learning_rate": 6.9421052631578955e-06,
|
| 947 |
+
"loss": 0.0674,
|
| 948 |
+
"step": 1370
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 15.0,
|
| 952 |
+
"learning_rate": 6.915789473684211e-06,
|
| 953 |
+
"loss": 0.0694,
|
| 954 |
+
"step": 1380
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 15.11,
|
| 958 |
+
"learning_rate": 6.889473684210527e-06,
|
| 959 |
+
"loss": 0.0732,
|
| 960 |
+
"step": 1390
|
| 961 |
+
},
|
| 962 |
+
{
|
| 963 |
+
"epoch": 15.22,
|
| 964 |
+
"learning_rate": 6.863157894736843e-06,
|
| 965 |
+
"loss": 0.0687,
|
| 966 |
+
"step": 1400
|
| 967 |
+
},
|
| 968 |
+
{
|
| 969 |
+
"epoch": 15.22,
|
| 970 |
+
"eval_loss": 0.109130859375,
|
| 971 |
+
"eval_runtime": 110.2423,
|
| 972 |
+
"eval_samples_per_second": 44.81,
|
| 973 |
+
"eval_steps_per_second": 0.063,
|
| 974 |
+
"eval_wer": 0.23717429458620223,
|
| 975 |
+
"step": 1400
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 15.33,
|
| 979 |
+
"learning_rate": 6.836842105263158e-06,
|
| 980 |
+
"loss": 0.0635,
|
| 981 |
+
"step": 1410
|
| 982 |
+
},
|
| 983 |
+
{
|
| 984 |
+
"epoch": 15.43,
|
| 985 |
+
"learning_rate": 6.810526315789474e-06,
|
| 986 |
+
"loss": 0.0644,
|
| 987 |
+
"step": 1420
|
| 988 |
+
},
|
| 989 |
+
{
|
| 990 |
+
"epoch": 15.54,
|
| 991 |
+
"learning_rate": 6.78421052631579e-06,
|
| 992 |
+
"loss": 0.0583,
|
| 993 |
+
"step": 1430
|
| 994 |
+
},
|
| 995 |
+
{
|
| 996 |
+
"epoch": 15.65,
|
| 997 |
+
"learning_rate": 6.7578947368421054e-06,
|
| 998 |
+
"loss": 0.0743,
|
| 999 |
+
"step": 1440
|
| 1000 |
+
},
|
| 1001 |
+
{
|
| 1002 |
+
"epoch": 15.76,
|
| 1003 |
+
"learning_rate": 6.731578947368421e-06,
|
| 1004 |
+
"loss": 0.069,
|
| 1005 |
+
"step": 1450
|
| 1006 |
+
},
|
| 1007 |
+
{
|
| 1008 |
+
"epoch": 15.87,
|
| 1009 |
+
"learning_rate": 6.705263157894737e-06,
|
| 1010 |
+
"loss": 0.0659,
|
| 1011 |
+
"step": 1460
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 15.98,
|
| 1015 |
+
"learning_rate": 6.678947368421053e-06,
|
| 1016 |
+
"loss": 0.0637,
|
| 1017 |
+
"step": 1470
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 16.09,
|
| 1021 |
+
"learning_rate": 6.6526315789473695e-06,
|
| 1022 |
+
"loss": 0.0586,
|
| 1023 |
+
"step": 1480
|
| 1024 |
+
},
|
| 1025 |
+
{
|
| 1026 |
+
"epoch": 16.2,
|
| 1027 |
+
"learning_rate": 6.626315789473685e-06,
|
| 1028 |
+
"loss": 0.0672,
|
| 1029 |
+
"step": 1490
|
| 1030 |
+
},
|
| 1031 |
+
{
|
| 1032 |
+
"epoch": 16.3,
|
| 1033 |
+
"learning_rate": 6.600000000000001e-06,
|
| 1034 |
+
"loss": 0.0647,
|
| 1035 |
+
"step": 1500
|
| 1036 |
+
},
|
| 1037 |
+
{
|
| 1038 |
+
"epoch": 16.3,
|
| 1039 |
+
"eval_loss": 0.1021728515625,
|
| 1040 |
+
"eval_runtime": 111.1773,
|
| 1041 |
+
"eval_samples_per_second": 44.434,
|
| 1042 |
+
"eval_steps_per_second": 0.063,
|
| 1043 |
+
"eval_wer": 0.21726069933846362,
|
| 1044 |
+
"step": 1500
|
| 1045 |
+
},
|
| 1046 |
+
{
|
| 1047 |
+
"epoch": 16.41,
|
| 1048 |
+
"learning_rate": 6.573684210526316e-06,
|
| 1049 |
+
"loss": 0.0602,
|
| 1050 |
+
"step": 1510
|
| 1051 |
+
},
|
| 1052 |
+
{
|
| 1053 |
+
"epoch": 16.52,
|
| 1054 |
+
"learning_rate": 6.547368421052632e-06,
|
| 1055 |
+
"loss": 0.0586,
|
| 1056 |
+
"step": 1520
|
| 1057 |
+
},
|
| 1058 |
+
{
|
| 1059 |
+
"epoch": 16.63,
|
| 1060 |
+
"learning_rate": 6.521052631578948e-06,
|
| 1061 |
+
"loss": 0.0556,
|
| 1062 |
+
"step": 1530
|
| 1063 |
+
},
|
| 1064 |
+
{
|
| 1065 |
+
"epoch": 16.74,
|
| 1066 |
+
"learning_rate": 6.494736842105264e-06,
|
| 1067 |
+
"loss": 0.0647,
|
| 1068 |
+
"step": 1540
|
| 1069 |
+
},
|
| 1070 |
+
{
|
| 1071 |
+
"epoch": 16.85,
|
| 1072 |
+
"learning_rate": 6.4684210526315794e-06,
|
| 1073 |
+
"loss": 0.064,
|
| 1074 |
+
"step": 1550
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 16.96,
|
| 1078 |
+
"learning_rate": 6.442105263157895e-06,
|
| 1079 |
+
"loss": 0.0623,
|
| 1080 |
+
"step": 1560
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 17.07,
|
| 1084 |
+
"learning_rate": 6.415789473684211e-06,
|
| 1085 |
+
"loss": 0.0551,
|
| 1086 |
+
"step": 1570
|
| 1087 |
+
},
|
| 1088 |
+
{
|
| 1089 |
+
"epoch": 17.17,
|
| 1090 |
+
"learning_rate": 6.389473684210527e-06,
|
| 1091 |
+
"loss": 0.053,
|
| 1092 |
+
"step": 1580
|
| 1093 |
+
},
|
| 1094 |
+
{
|
| 1095 |
+
"epoch": 17.28,
|
| 1096 |
+
"learning_rate": 6.363157894736842e-06,
|
| 1097 |
+
"loss": 0.0598,
|
| 1098 |
+
"step": 1590
|
| 1099 |
+
},
|
| 1100 |
+
{
|
| 1101 |
+
"epoch": 17.39,
|
| 1102 |
+
"learning_rate": 6.336842105263158e-06,
|
| 1103 |
+
"loss": 0.059,
|
| 1104 |
+
"step": 1600
|
| 1105 |
+
},
|
| 1106 |
+
{
|
| 1107 |
+
"epoch": 17.39,
|
| 1108 |
+
"eval_loss": 0.09674072265625,
|
| 1109 |
+
"eval_runtime": 110.518,
|
| 1110 |
+
"eval_samples_per_second": 44.699,
|
| 1111 |
+
"eval_steps_per_second": 0.063,
|
| 1112 |
+
"eval_wer": 0.20433373835560956,
|
| 1113 |
+
"step": 1600
|
| 1114 |
+
},
|
| 1115 |
+
{
|
| 1116 |
+
"epoch": 17.5,
|
| 1117 |
+
"learning_rate": 6.310526315789474e-06,
|
| 1118 |
+
"loss": 0.0564,
|
| 1119 |
+
"step": 1610
|
| 1120 |
+
},
|
| 1121 |
+
{
|
| 1122 |
+
"epoch": 17.61,
|
| 1123 |
+
"learning_rate": 6.28421052631579e-06,
|
| 1124 |
+
"loss": 0.0536,
|
| 1125 |
+
"step": 1620
|
| 1126 |
+
},
|
| 1127 |
+
{
|
| 1128 |
+
"epoch": 17.72,
|
| 1129 |
+
"learning_rate": 6.257894736842106e-06,
|
| 1130 |
+
"loss": 0.0523,
|
| 1131 |
+
"step": 1630
|
| 1132 |
+
},
|
| 1133 |
+
{
|
| 1134 |
+
"epoch": 17.83,
|
| 1135 |
+
"learning_rate": 6.231578947368422e-06,
|
| 1136 |
+
"loss": 0.0583,
|
| 1137 |
+
"step": 1640
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 17.93,
|
| 1141 |
+
"learning_rate": 6.205263157894738e-06,
|
| 1142 |
+
"loss": 0.058,
|
| 1143 |
+
"step": 1650
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 18.04,
|
| 1147 |
+
"learning_rate": 6.1789473684210534e-06,
|
| 1148 |
+
"loss": 0.0549,
|
| 1149 |
+
"step": 1660
|
| 1150 |
+
},
|
| 1151 |
+
{
|
| 1152 |
+
"epoch": 18.15,
|
| 1153 |
+
"learning_rate": 6.152631578947369e-06,
|
| 1154 |
+
"loss": 0.0504,
|
| 1155 |
+
"step": 1670
|
| 1156 |
+
},
|
| 1157 |
+
{
|
| 1158 |
+
"epoch": 18.26,
|
| 1159 |
+
"learning_rate": 6.126315789473685e-06,
|
| 1160 |
+
"loss": 0.0479,
|
| 1161 |
+
"step": 1680
|
| 1162 |
+
},
|
| 1163 |
+
{
|
| 1164 |
+
"epoch": 18.37,
|
| 1165 |
+
"learning_rate": 6.1e-06,
|
| 1166 |
+
"loss": 0.0516,
|
| 1167 |
+
"step": 1690
|
| 1168 |
+
},
|
| 1169 |
+
{
|
| 1170 |
+
"epoch": 18.48,
|
| 1171 |
+
"learning_rate": 6.073684210526316e-06,
|
| 1172 |
+
"loss": 0.0539,
|
| 1173 |
+
"step": 1700
|
| 1174 |
+
},
|
| 1175 |
+
{
|
| 1176 |
+
"epoch": 18.48,
|
| 1177 |
+
"eval_loss": 0.0897216796875,
|
| 1178 |
+
"eval_runtime": 110.8559,
|
| 1179 |
+
"eval_samples_per_second": 44.562,
|
| 1180 |
+
"eval_steps_per_second": 0.063,
|
| 1181 |
+
"eval_wer": 0.19289185905224787,
|
| 1182 |
+
"step": 1700
|
| 1183 |
+
},
|
| 1184 |
+
{
|
| 1185 |
+
"epoch": 18.59,
|
| 1186 |
+
"learning_rate": 6.047368421052632e-06,
|
| 1187 |
+
"loss": 0.0525,
|
| 1188 |
+
"step": 1710
|
| 1189 |
+
},
|
| 1190 |
+
{
|
| 1191 |
+
"epoch": 18.7,
|
| 1192 |
+
"learning_rate": 6.0210526315789475e-06,
|
| 1193 |
+
"loss": 0.0519,
|
| 1194 |
+
"step": 1720
|
| 1195 |
+
},
|
| 1196 |
+
{
|
| 1197 |
+
"epoch": 18.8,
|
| 1198 |
+
"learning_rate": 5.994736842105263e-06,
|
| 1199 |
+
"loss": 0.0463,
|
| 1200 |
+
"step": 1730
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 18.91,
|
| 1204 |
+
"learning_rate": 5.968421052631579e-06,
|
| 1205 |
+
"loss": 0.054,
|
| 1206 |
+
"step": 1740
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 19.02,
|
| 1210 |
+
"learning_rate": 5.942105263157896e-06,
|
| 1211 |
+
"loss": 0.0552,
|
| 1212 |
+
"step": 1750
|
| 1213 |
+
},
|
| 1214 |
+
{
|
| 1215 |
+
"epoch": 19.13,
|
| 1216 |
+
"learning_rate": 5.915789473684212e-06,
|
| 1217 |
+
"loss": 0.0482,
|
| 1218 |
+
"step": 1760
|
| 1219 |
+
},
|
| 1220 |
+
{
|
| 1221 |
+
"epoch": 19.24,
|
| 1222 |
+
"learning_rate": 5.8894736842105274e-06,
|
| 1223 |
+
"loss": 0.0464,
|
| 1224 |
+
"step": 1770
|
| 1225 |
+
},
|
| 1226 |
+
{
|
| 1227 |
+
"epoch": 19.35,
|
| 1228 |
+
"learning_rate": 5.863157894736842e-06,
|
| 1229 |
+
"loss": 0.0433,
|
| 1230 |
+
"step": 1780
|
| 1231 |
+
},
|
| 1232 |
+
{
|
| 1233 |
+
"epoch": 19.46,
|
| 1234 |
+
"learning_rate": 5.836842105263158e-06,
|
| 1235 |
+
"loss": 0.0456,
|
| 1236 |
+
"step": 1790
|
| 1237 |
+
},
|
| 1238 |
+
{
|
| 1239 |
+
"epoch": 19.57,
|
| 1240 |
+
"learning_rate": 5.810526315789474e-06,
|
| 1241 |
+
"loss": 0.0518,
|
| 1242 |
+
"step": 1800
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 19.57,
|
| 1246 |
+
"eval_loss": 0.08270263671875,
|
| 1247 |
+
"eval_runtime": 114.8201,
|
| 1248 |
+
"eval_samples_per_second": 43.024,
|
| 1249 |
+
"eval_steps_per_second": 0.061,
|
| 1250 |
+
"eval_wer": 0.17183070068853787,
|
| 1251 |
+
"step": 1800
|
| 1252 |
+
},
|
| 1253 |
+
{
|
| 1254 |
+
"epoch": 19.67,
|
| 1255 |
+
"learning_rate": 5.78421052631579e-06,
|
| 1256 |
+
"loss": 0.048,
|
| 1257 |
+
"step": 1810
|
| 1258 |
+
},
|
| 1259 |
+
{
|
| 1260 |
+
"epoch": 19.78,
|
| 1261 |
+
"learning_rate": 5.757894736842106e-06,
|
| 1262 |
+
"loss": 0.0453,
|
| 1263 |
+
"step": 1820
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 19.89,
|
| 1267 |
+
"learning_rate": 5.7315789473684215e-06,
|
| 1268 |
+
"loss": 0.0438,
|
| 1269 |
+
"step": 1830
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 20.0,
|
| 1273 |
+
"learning_rate": 5.705263157894737e-06,
|
| 1274 |
+
"loss": 0.0445,
|
| 1275 |
+
"step": 1840
|
| 1276 |
+
},
|
| 1277 |
+
{
|
| 1278 |
+
"epoch": 20.11,
|
| 1279 |
+
"learning_rate": 5.678947368421053e-06,
|
| 1280 |
+
"loss": 0.048,
|
| 1281 |
+
"step": 1850
|
| 1282 |
+
},
|
| 1283 |
+
{
|
| 1284 |
+
"epoch": 20.22,
|
| 1285 |
+
"learning_rate": 5.652631578947368e-06,
|
| 1286 |
+
"loss": 0.0443,
|
| 1287 |
+
"step": 1860
|
| 1288 |
+
},
|
| 1289 |
+
{
|
| 1290 |
+
"epoch": 20.33,
|
| 1291 |
+
"learning_rate": 5.626315789473684e-06,
|
| 1292 |
+
"loss": 0.0435,
|
| 1293 |
+
"step": 1870
|
| 1294 |
+
},
|
| 1295 |
+
{
|
| 1296 |
+
"epoch": 20.43,
|
| 1297 |
+
"learning_rate": 5.600000000000001e-06,
|
| 1298 |
+
"loss": 0.0408,
|
| 1299 |
+
"step": 1880
|
| 1300 |
+
},
|
| 1301 |
+
{
|
| 1302 |
+
"epoch": 20.54,
|
| 1303 |
+
"learning_rate": 5.573684210526316e-06,
|
| 1304 |
+
"loss": 0.0382,
|
| 1305 |
+
"step": 1890
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 20.65,
|
| 1309 |
+
"learning_rate": 5.547368421052632e-06,
|
| 1310 |
+
"loss": 0.0495,
|
| 1311 |
+
"step": 1900
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 20.65,
|
| 1315 |
+
"eval_loss": 0.07867431640625,
|
| 1316 |
+
"eval_runtime": 110.7928,
|
| 1317 |
+
"eval_samples_per_second": 44.588,
|
| 1318 |
+
"eval_steps_per_second": 0.063,
|
| 1319 |
+
"eval_wer": 0.16673417037937086,
|
| 1320 |
+
"step": 1900
|
| 1321 |
+
},
|
| 1322 |
+
{
|
| 1323 |
+
"epoch": 20.76,
|
| 1324 |
+
"learning_rate": 5.521052631578948e-06,
|
| 1325 |
+
"loss": 0.0451,
|
| 1326 |
+
"step": 1910
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 20.87,
|
| 1330 |
+
"learning_rate": 5.494736842105264e-06,
|
| 1331 |
+
"loss": 0.042,
|
| 1332 |
+
"step": 1920
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 20.98,
|
| 1336 |
+
"learning_rate": 5.46842105263158e-06,
|
| 1337 |
+
"loss": 0.0413,
|
| 1338 |
+
"step": 1930
|
| 1339 |
+
},
|
| 1340 |
+
{
|
| 1341 |
+
"epoch": 21.09,
|
| 1342 |
+
"learning_rate": 5.4421052631578955e-06,
|
| 1343 |
+
"loss": 0.0374,
|
| 1344 |
+
"step": 1940
|
| 1345 |
+
},
|
| 1346 |
+
{
|
| 1347 |
+
"epoch": 21.2,
|
| 1348 |
+
"learning_rate": 5.415789473684211e-06,
|
| 1349 |
+
"loss": 0.0445,
|
| 1350 |
+
"step": 1950
|
| 1351 |
+
},
|
| 1352 |
+
{
|
| 1353 |
+
"epoch": 21.3,
|
| 1354 |
+
"learning_rate": 5.389473684210526e-06,
|
| 1355 |
+
"loss": 0.0406,
|
| 1356 |
+
"step": 1960
|
| 1357 |
+
},
|
| 1358 |
+
{
|
| 1359 |
+
"epoch": 21.41,
|
| 1360 |
+
"learning_rate": 5.363157894736842e-06,
|
| 1361 |
+
"loss": 0.0389,
|
| 1362 |
+
"step": 1970
|
| 1363 |
+
},
|
| 1364 |
+
{
|
| 1365 |
+
"epoch": 21.52,
|
| 1366 |
+
"learning_rate": 5.336842105263158e-06,
|
| 1367 |
+
"loss": 0.0373,
|
| 1368 |
+
"step": 1980
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 21.63,
|
| 1372 |
+
"learning_rate": 5.310526315789474e-06,
|
| 1373 |
+
"loss": 0.035,
|
| 1374 |
+
"step": 1990
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 21.74,
|
| 1378 |
+
"learning_rate": 5.2842105263157896e-06,
|
| 1379 |
+
"loss": 0.0444,
|
| 1380 |
+
"step": 2000
|
| 1381 |
+
},
|
| 1382 |
+
{
|
| 1383 |
+
"epoch": 21.74,
|
| 1384 |
+
"eval_loss": 0.07183837890625,
|
| 1385 |
+
"eval_runtime": 112.6262,
|
| 1386 |
+
"eval_samples_per_second": 43.862,
|
| 1387 |
+
"eval_steps_per_second": 0.062,
|
| 1388 |
+
"eval_wer": 0.14692183070068854,
|
| 1389 |
+
"step": 2000
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 21.85,
|
| 1393 |
+
"learning_rate": 5.257894736842105e-06,
|
| 1394 |
+
"loss": 0.04,
|
| 1395 |
+
"step": 2010
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 21.96,
|
| 1399 |
+
"learning_rate": 5.231578947368422e-06,
|
| 1400 |
+
"loss": 0.0391,
|
| 1401 |
+
"step": 2020
|
| 1402 |
+
},
|
| 1403 |
+
{
|
| 1404 |
+
"epoch": 22.07,
|
| 1405 |
+
"learning_rate": 5.205263157894738e-06,
|
| 1406 |
+
"loss": 0.037,
|
| 1407 |
+
"step": 2030
|
| 1408 |
+
},
|
| 1409 |
+
{
|
| 1410 |
+
"epoch": 22.17,
|
| 1411 |
+
"learning_rate": 5.178947368421054e-06,
|
| 1412 |
+
"loss": 0.0332,
|
| 1413 |
+
"step": 2040
|
| 1414 |
+
},
|
| 1415 |
+
{
|
| 1416 |
+
"epoch": 22.28,
|
| 1417 |
+
"learning_rate": 5.1526315789473695e-06,
|
| 1418 |
+
"loss": 0.0385,
|
| 1419 |
+
"step": 2050
|
| 1420 |
+
},
|
| 1421 |
+
{
|
| 1422 |
+
"epoch": 22.39,
|
| 1423 |
+
"learning_rate": 5.1263157894736845e-06,
|
| 1424 |
+
"loss": 0.0377,
|
| 1425 |
+
"step": 2060
|
| 1426 |
+
},
|
| 1427 |
+
{
|
| 1428 |
+
"epoch": 22.5,
|
| 1429 |
+
"learning_rate": 5.1e-06,
|
| 1430 |
+
"loss": 0.0353,
|
| 1431 |
+
"step": 2070
|
| 1432 |
+
},
|
| 1433 |
+
{
|
| 1434 |
+
"epoch": 22.61,
|
| 1435 |
+
"learning_rate": 5.073684210526316e-06,
|
| 1436 |
+
"loss": 0.0338,
|
| 1437 |
+
"step": 2080
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 22.72,
|
| 1441 |
+
"learning_rate": 5.047368421052632e-06,
|
| 1442 |
+
"loss": 0.0327,
|
| 1443 |
+
"step": 2090
|
| 1444 |
+
},
|
| 1445 |
+
{
|
| 1446 |
+
"epoch": 22.83,
|
| 1447 |
+
"learning_rate": 5.021052631578948e-06,
|
| 1448 |
+
"loss": 0.0392,
|
| 1449 |
+
"step": 2100
|
| 1450 |
+
},
|
| 1451 |
+
{
|
| 1452 |
+
"epoch": 22.83,
|
| 1453 |
+
"eval_loss": 0.067138671875,
|
| 1454 |
+
"eval_runtime": 111.3072,
|
| 1455 |
+
"eval_samples_per_second": 44.382,
|
| 1456 |
+
"eval_steps_per_second": 0.063,
|
| 1457 |
+
"eval_wer": 0.13683002565141084,
|
| 1458 |
+
"step": 2100
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 22.93,
|
| 1462 |
+
"learning_rate": 9.42857142857143e-07,
|
| 1463 |
+
"loss": 0.0362,
|
| 1464 |
+
"step": 2110
|
| 1465 |
+
},
|
| 1466 |
+
{
|
| 1467 |
+
"epoch": 23.04,
|
| 1468 |
+
"learning_rate": 8.952380952380953e-07,
|
| 1469 |
+
"loss": 0.032,
|
| 1470 |
+
"step": 2120
|
| 1471 |
+
},
|
| 1472 |
+
{
|
| 1473 |
+
"epoch": 23.15,
|
| 1474 |
+
"learning_rate": 8.476190476190477e-07,
|
| 1475 |
+
"loss": 0.0309,
|
| 1476 |
+
"step": 2130
|
| 1477 |
+
},
|
| 1478 |
+
{
|
| 1479 |
+
"epoch": 23.26,
|
| 1480 |
+
"learning_rate": 8.000000000000001e-07,
|
| 1481 |
+
"loss": 0.0296,
|
| 1482 |
+
"step": 2140
|
| 1483 |
+
},
|
| 1484 |
+
{
|
| 1485 |
+
"epoch": 23.37,
|
| 1486 |
+
"learning_rate": 7.523809523809525e-07,
|
| 1487 |
+
"loss": 0.0338,
|
| 1488 |
+
"step": 2150
|
| 1489 |
+
},
|
| 1490 |
+
{
|
| 1491 |
+
"epoch": 23.48,
|
| 1492 |
+
"learning_rate": 7.047619047619048e-07,
|
| 1493 |
+
"loss": 0.035,
|
| 1494 |
+
"step": 2160
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 23.59,
|
| 1498 |
+
"learning_rate": 6.571428571428571e-07,
|
| 1499 |
+
"loss": 0.0312,
|
| 1500 |
+
"step": 2170
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 23.7,
|
| 1504 |
+
"learning_rate": 6.095238095238095e-07,
|
| 1505 |
+
"loss": 0.0309,
|
| 1506 |
+
"step": 2180
|
| 1507 |
+
},
|
| 1508 |
+
{
|
| 1509 |
+
"epoch": 23.8,
|
| 1510 |
+
"learning_rate": 5.619047619047619e-07,
|
| 1511 |
+
"loss": 0.0292,
|
| 1512 |
+
"step": 2190
|
| 1513 |
+
},
|
| 1514 |
+
{
|
| 1515 |
+
"epoch": 23.91,
|
| 1516 |
+
"learning_rate": 5.142857142857143e-07,
|
| 1517 |
+
"loss": 0.0335,
|
| 1518 |
+
"step": 2200
|
| 1519 |
+
},
|
| 1520 |
+
{
|
| 1521 |
+
"epoch": 23.91,
|
| 1522 |
+
"eval_loss": 0.064453125,
|
| 1523 |
+
"eval_runtime": 113.7883,
|
| 1524 |
+
"eval_samples_per_second": 43.414,
|
| 1525 |
+
"eval_steps_per_second": 0.062,
|
| 1526 |
+
"eval_wer": 0.12626569461320372,
|
| 1527 |
+
"step": 2200
|
| 1528 |
+
},
|
| 1529 |
+
{
|
| 1530 |
+
"epoch": 24.02,
|
| 1531 |
+
"learning_rate": 4.666666666666667e-07,
|
| 1532 |
+
"loss": 0.0349,
|
| 1533 |
+
"step": 2210
|
| 1534 |
+
},
|
| 1535 |
+
{
|
| 1536 |
+
"epoch": 24.13,
|
| 1537 |
+
"learning_rate": 4.1904761904761906e-07,
|
| 1538 |
+
"loss": 0.0327,
|
| 1539 |
+
"step": 2220
|
| 1540 |
+
},
|
| 1541 |
+
{
|
| 1542 |
+
"epoch": 24.24,
|
| 1543 |
+
"learning_rate": 3.7142857142857145e-07,
|
| 1544 |
+
"loss": 0.0317,
|
| 1545 |
+
"step": 2230
|
| 1546 |
+
},
|
| 1547 |
+
{
|
| 1548 |
+
"epoch": 24.35,
|
| 1549 |
+
"learning_rate": 3.238095238095238e-07,
|
| 1550 |
+
"loss": 0.0298,
|
| 1551 |
+
"step": 2240
|
| 1552 |
+
},
|
| 1553 |
+
{
|
| 1554 |
+
"epoch": 24.46,
|
| 1555 |
+
"learning_rate": 2.7619047619047624e-07,
|
| 1556 |
+
"loss": 0.0302,
|
| 1557 |
+
"step": 2250
|
| 1558 |
+
},
|
| 1559 |
+
{
|
| 1560 |
+
"epoch": 24.57,
|
| 1561 |
+
"learning_rate": 2.285714285714286e-07,
|
| 1562 |
+
"loss": 0.0355,
|
| 1563 |
+
"step": 2260
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 24.67,
|
| 1567 |
+
"learning_rate": 1.8095238095238097e-07,
|
| 1568 |
+
"loss": 0.0309,
|
| 1569 |
+
"step": 2270
|
| 1570 |
+
},
|
| 1571 |
+
{
|
| 1572 |
+
"epoch": 24.78,
|
| 1573 |
+
"learning_rate": 1.3333333333333336e-07,
|
| 1574 |
+
"loss": 0.0299,
|
| 1575 |
+
"step": 2280
|
| 1576 |
+
},
|
| 1577 |
+
{
|
| 1578 |
+
"epoch": 24.89,
|
| 1579 |
+
"learning_rate": 8.571428571428573e-08,
|
| 1580 |
+
"loss": 0.028,
|
| 1581 |
+
"step": 2290
|
| 1582 |
+
},
|
| 1583 |
+
{
|
| 1584 |
+
"epoch": 25.0,
|
| 1585 |
+
"learning_rate": 3.8095238095238096e-08,
|
| 1586 |
+
"loss": 0.0292,
|
| 1587 |
+
"step": 2300
|
| 1588 |
+
},
|
| 1589 |
+
{
|
| 1590 |
+
"epoch": 25.0,
|
| 1591 |
+
"eval_loss": 0.0640869140625,
|
| 1592 |
+
"eval_runtime": 116.9844,
|
| 1593 |
+
"eval_samples_per_second": 42.228,
|
| 1594 |
+
"eval_steps_per_second": 0.06,
|
| 1595 |
+
"eval_wer": 0.12623194275685162,
|
| 1596 |
+
"step": 2300
|
| 1597 |
+
},
|
| 1598 |
+
{
|
| 1599 |
+
"epoch": 25.0,
|
| 1600 |
+
"step": 2300,
|
| 1601 |
+
"total_flos": 1.7185304344854004e+20,
|
| 1602 |
+
"train_loss": 0.0027616678631823995,
|
| 1603 |
+
"train_runtime": 6144.0313,
|
| 1604 |
+
"train_samples_per_second": 431.248,
|
| 1605 |
+
"train_steps_per_second": 0.374
|
| 1606 |
}
|
| 1607 |
],
|
| 1608 |
"logging_steps": 10,
|
| 1609 |
+
"max_steps": 2300,
|
| 1610 |
"num_input_tokens_seen": 0,
|
| 1611 |
+
"num_train_epochs": 25,
|
| 1612 |
"save_steps": 100,
|
| 1613 |
+
"total_flos": 1.7185304344854004e+20,
|
| 1614 |
"train_batch_size": 192,
|
| 1615 |
"trial_name": null,
|
| 1616 |
"trial_params": null
|