updated the model with more data
Browse files- README.md +22 -16
- all_results.json +10 -10
- config.json +2 -2
- eval_results.json +6 -6
- pytorch_model.bin +1 -1
- runs/Dec14_07-09-13_4848315c9cd8/1671008819.384087/events.out.tfevents.1671008819.4848315c9cd8.75049.1 +3 -0
- runs/Dec14_07-09-13_4848315c9cd8/events.out.tfevents.1671008819.4848315c9cd8.75049.0 +3 -0
- runs/Dec14_10-33-04_4848315c9cd8/1671010890.9726396/events.out.tfevents.1671010890.4848315c9cd8.31516.1 +3 -0
- runs/Dec14_10-33-04_4848315c9cd8/events.out.tfevents.1671010890.4848315c9cd8.31516.0 +3 -0
- runs/Dec14_10-33-04_4848315c9cd8/events.out.tfevents.1671031901.4848315c9cd8.31516.2 +3 -0
- runs/Dec14_17-19-00_4848315c9cd8/1671045128.7881477/events.out.tfevents.1671045128.4848315c9cd8.77445.1 +3 -0
- runs/Dec14_17-19-00_4848315c9cd8/events.out.tfevents.1671045128.4848315c9cd8.77445.0 +3 -0
- runs/Dec14_17-19-00_4848315c9cd8/events.out.tfevents.1671108519.4848315c9cd8.77445.2 +3 -0
- tokenizer_config.json +1 -1
- train_results.json +5 -5
- trainer_state.json +1888 -643
- training_args.bin +1 -1
README.md
CHANGED
|
@@ -7,6 +7,8 @@ tags:
|
|
| 7 |
- generated_from_trainer
|
| 8 |
datasets:
|
| 9 |
- mozilla-foundation/common_voice_11_0
|
|
|
|
|
|
|
| 10 |
metrics:
|
| 11 |
- wer
|
| 12 |
model-index:
|
|
@@ -20,11 +22,10 @@ model-index:
|
|
| 20 |
type: mozilla-foundation/common_voice_11_0
|
| 21 |
config: id
|
| 22 |
split: test
|
| 23 |
-
args: id
|
| 24 |
metrics:
|
| 25 |
- name: Wer
|
| 26 |
type: wer
|
| 27 |
-
value:
|
| 28 |
---
|
| 29 |
|
| 30 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
@@ -32,10 +33,10 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 32 |
|
| 33 |
# Whisper Medium Indonesian
|
| 34 |
|
| 35 |
-
This model is a fine-tuned version of [
|
| 36 |
It achieves the following results on the evaluation set:
|
| 37 |
-
- Loss: 0.
|
| 38 |
-
- Wer:
|
| 39 |
|
| 40 |
## Model description
|
| 41 |
|
|
@@ -54,25 +55,30 @@ More information needed
|
|
| 54 |
### Training hyperparameters
|
| 55 |
|
| 56 |
The following hyperparameters were used during training:
|
| 57 |
-
- learning_rate: 1e-
|
| 58 |
-
- train_batch_size:
|
| 59 |
-
- eval_batch_size:
|
| 60 |
- seed: 42
|
| 61 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 62 |
- lr_scheduler_type: linear
|
| 63 |
- lr_scheduler_warmup_steps: 500
|
| 64 |
-
- training_steps:
|
| 65 |
- mixed_precision_training: Native AMP
|
| 66 |
|
| 67 |
### Training results
|
| 68 |
|
| 69 |
-
| Training Loss | Epoch | Step
|
| 70 |
-
|
| 71 |
-
| 0.
|
| 72 |
-
| 0.
|
| 73 |
-
| 0.
|
| 74 |
-
| 0.
|
| 75 |
-
| 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 76 |
|
| 77 |
|
| 78 |
### Framework versions
|
|
|
|
| 7 |
- generated_from_trainer
|
| 8 |
datasets:
|
| 9 |
- mozilla-foundation/common_voice_11_0
|
| 10 |
+
- magic_data
|
| 11 |
+
- TITML
|
| 12 |
metrics:
|
| 13 |
- wer
|
| 14 |
model-index:
|
|
|
|
| 22 |
type: mozilla-foundation/common_voice_11_0
|
| 23 |
config: id
|
| 24 |
split: test
|
|
|
|
| 25 |
metrics:
|
| 26 |
- name: Wer
|
| 27 |
type: wer
|
| 28 |
+
value: 3.993359771281011
|
| 29 |
---
|
| 30 |
|
| 31 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
|
| 33 |
|
| 34 |
# Whisper Medium Indonesian
|
| 35 |
|
| 36 |
+
This model is a fine-tuned version of [./whisper-medium-id](https://huggingface.co/./whisper-medium-id) on the mozilla-foundation/common_voice_11_0, local_datasets/magic_data, local_datasets/magic_data, local_datasets/titml id, id, id, id dataset.
|
| 37 |
It achieves the following results on the evaluation set:
|
| 38 |
+
- Loss: 0.0698
|
| 39 |
+
- Wer: 3.9934
|
| 40 |
|
| 41 |
## Model description
|
| 42 |
|
|
|
|
| 55 |
### Training hyperparameters
|
| 56 |
|
| 57 |
The following hyperparameters were used during training:
|
| 58 |
+
- learning_rate: 1e-06
|
| 59 |
+
- train_batch_size: 16
|
| 60 |
+
- eval_batch_size: 16
|
| 61 |
- seed: 42
|
| 62 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 63 |
- lr_scheduler_type: linear
|
| 64 |
- lr_scheduler_warmup_steps: 500
|
| 65 |
+
- training_steps: 10000
|
| 66 |
- mixed_precision_training: Native AMP
|
| 67 |
|
| 68 |
### Training results
|
| 69 |
|
| 70 |
+
| Training Loss | Epoch | Step | Validation Loss | Wer |
|
| 71 |
+
|:-------------:|:-----:|:-----:|:---------------:|:------:|
|
| 72 |
+
| 0.0252 | 0.35 | 1000 | 0.0733 | 4.4822 |
|
| 73 |
+
| 0.0222 | 0.7 | 2000 | 0.0698 | 4.3392 |
|
| 74 |
+
| 0.0166 | 1.06 | 3000 | 0.0696 | 4.2378 |
|
| 75 |
+
| 0.0117 | 1.41 | 4000 | 0.0679 | 4.0810 |
|
| 76 |
+
| 0.0295 | 1.76 | 5000 | 0.0671 | 4.0856 |
|
| 77 |
+
| 0.0147 | 2.11 | 6000 | 0.0690 | 4.0302 |
|
| 78 |
+
| 0.0157 | 2.47 | 7000 | 0.0698 | 3.9934 |
|
| 79 |
+
| 0.0113 | 2.82 | 8000 | 0.0698 | 4.0302 |
|
| 80 |
+
| 0.0101 | 3.17 | 9000 | 0.0707 | 4.0579 |
|
| 81 |
+
| 0.0076 | 3.52 | 10000 | 0.0708 | 4.0625 |
|
| 82 |
|
| 83 |
|
| 84 |
### Framework versions
|
all_results.json
CHANGED
|
@@ -1,12 +1,12 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_loss": 0.
|
| 4 |
-
"eval_runtime":
|
| 5 |
-
"eval_samples_per_second":
|
| 6 |
-
"eval_steps_per_second": 0.
|
| 7 |
-
"eval_wer":
|
| 8 |
-
"train_loss": 0.
|
| 9 |
-
"train_runtime":
|
| 10 |
-
"train_samples_per_second":
|
| 11 |
-
"train_steps_per_second": 0.
|
| 12 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 3.52,
|
| 3 |
+
"eval_loss": 0.06976281106472015,
|
| 4 |
+
"eval_runtime": 2273.5419,
|
| 5 |
+
"eval_samples_per_second": 1.591,
|
| 6 |
+
"eval_steps_per_second": 0.1,
|
| 7 |
+
"eval_wer": 3.993359771281011,
|
| 8 |
+
"train_loss": 0.019241938863694668,
|
| 9 |
+
"train_runtime": 61111.4423,
|
| 10 |
+
"train_samples_per_second": 2.618,
|
| 11 |
+
"train_steps_per_second": 0.164
|
| 12 |
}
|
config.json
CHANGED
|
@@ -1,5 +1,5 @@
|
|
| 1 |
{
|
| 2 |
-
"_name_or_path": "
|
| 3 |
"activation_dropout": 0.0,
|
| 4 |
"activation_function": "gelu",
|
| 5 |
"architectures": [
|
|
@@ -36,6 +36,6 @@
|
|
| 36 |
"scale_embedding": false,
|
| 37 |
"torch_dtype": "float32",
|
| 38 |
"transformers_version": "4.26.0.dev0",
|
| 39 |
-
"use_cache":
|
| 40 |
"vocab_size": 51865
|
| 41 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"_name_or_path": "./whisper-medium-id",
|
| 3 |
"activation_dropout": 0.0,
|
| 4 |
"activation_function": "gelu",
|
| 5 |
"architectures": [
|
|
|
|
| 36 |
"scale_embedding": false,
|
| 37 |
"torch_dtype": "float32",
|
| 38 |
"transformers_version": "4.26.0.dev0",
|
| 39 |
+
"use_cache": false,
|
| 40 |
"vocab_size": 51865
|
| 41 |
}
|
eval_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_loss": 0.
|
| 4 |
-
"eval_runtime":
|
| 5 |
-
"eval_samples_per_second":
|
| 6 |
-
"eval_steps_per_second": 0.
|
| 7 |
-
"eval_wer":
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 3.52,
|
| 3 |
+
"eval_loss": 0.06976281106472015,
|
| 4 |
+
"eval_runtime": 2273.5419,
|
| 5 |
+
"eval_samples_per_second": 1.591,
|
| 6 |
+
"eval_steps_per_second": 0.1,
|
| 7 |
+
"eval_wer": 3.993359771281011
|
| 8 |
}
|
pytorch_model.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 3055754841
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5517ba8446b067798fcdc8a83e3f761e943a88fdc08ecb3dd872312606c24ee4
|
| 3 |
size 3055754841
|
runs/Dec14_07-09-13_4848315c9cd8/1671008819.384087/events.out.tfevents.1671008819.4848315c9cd8.75049.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3f4f0e96b7e80a6dfe2b114caf59cad9273394e8687f0b6cdcb29b16fbd9c49c
|
| 3 |
+
size 5759
|
runs/Dec14_07-09-13_4848315c9cd8/events.out.tfevents.1671008819.4848315c9cd8.75049.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b38405ab158ad86d24c82adf3cca49811fe26e758967c5ec929066d0ab32dea1
|
| 3 |
+
size 4232
|
runs/Dec14_10-33-04_4848315c9cd8/1671010890.9726396/events.out.tfevents.1671010890.4848315c9cd8.31516.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1f3f7690ae45106ae76a4348c90d787660b45bd29909b4e4238262e94178725d
|
| 3 |
+
size 5760
|
runs/Dec14_10-33-04_4848315c9cd8/events.out.tfevents.1671010890.4848315c9cd8.31516.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:649c39bad62da63d7809745aaa53b3f16ad4c1bff7a1bb0e93721e177e1e9605
|
| 3 |
+
size 24367
|
runs/Dec14_10-33-04_4848315c9cd8/events.out.tfevents.1671031901.4848315c9cd8.31516.2
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:93dd13979907d98f2b7f4daa9e30107cf705bad4e43f0f41217ffa5eabe104d2
|
| 3 |
+
size 358
|
runs/Dec14_17-19-00_4848315c9cd8/1671045128.7881477/events.out.tfevents.1671045128.4848315c9cd8.77445.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d322fd87673352dec294c8c5ba2f0edf15f6951c17e87f6963d0c4cace1a12fd
|
| 3 |
+
size 5760
|
runs/Dec14_17-19-00_4848315c9cd8/events.out.tfevents.1671045128.4848315c9cd8.77445.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:dd6d62ce7a8cdd7faf8d98248c5d6f9e7820193019f86239817fd667b9a8d761
|
| 3 |
+
size 70550
|
runs/Dec14_17-19-00_4848315c9cd8/events.out.tfevents.1671108519.4848315c9cd8.77445.2
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ee0eb35c70b0fbb0fde1dd0669ba06b64b3568ab282db556f2d472629178598b
|
| 3 |
+
size 358
|
tokenizer_config.json
CHANGED
|
@@ -19,7 +19,7 @@
|
|
| 19 |
},
|
| 20 |
"errors": "replace",
|
| 21 |
"model_max_length": 1024,
|
| 22 |
-
"name_or_path": "
|
| 23 |
"pad_token": null,
|
| 24 |
"processor_class": "WhisperProcessor",
|
| 25 |
"return_attention_mask": false,
|
|
|
|
| 19 |
},
|
| 20 |
"errors": "replace",
|
| 21 |
"model_max_length": 1024,
|
| 22 |
+
"name_or_path": "./whisper-medium-id",
|
| 23 |
"pad_token": null,
|
| 24 |
"processor_class": "WhisperProcessor",
|
| 25 |
"return_attention_mask": false,
|
train_results.json
CHANGED
|
@@ -1,7 +1,7 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"train_loss": 0.
|
| 4 |
-
"train_runtime":
|
| 5 |
-
"train_samples_per_second":
|
| 6 |
-
"train_steps_per_second": 0.
|
| 7 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 3.52,
|
| 3 |
+
"train_loss": 0.019241938863694668,
|
| 4 |
+
"train_runtime": 61111.4423,
|
| 5 |
+
"train_samples_per_second": 2.618,
|
| 6 |
+
"train_steps_per_second": 0.164
|
| 7 |
}
|
trainer_state.json
CHANGED
|
@@ -1,1270 +1,2515 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric":
|
| 3 |
-
"best_model_checkpoint": "./checkpoint-
|
| 4 |
-
"epoch":
|
| 5 |
-
"global_step":
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
| 9 |
"log_history": [
|
| 10 |
{
|
| 11 |
"epoch": 0.01,
|
| 12 |
-
"learning_rate": 4.
|
| 13 |
-
"loss":
|
| 14 |
"step": 25
|
| 15 |
},
|
| 16 |
{
|
| 17 |
-
"epoch": 0.
|
| 18 |
-
"learning_rate": 9.
|
| 19 |
-
"loss": 0.
|
| 20 |
"step": 50
|
| 21 |
},
|
| 22 |
{
|
| 23 |
-
"epoch": 0.
|
| 24 |
-
"learning_rate": 1.
|
| 25 |
-
"loss": 0.
|
| 26 |
"step": 75
|
| 27 |
},
|
| 28 |
{
|
| 29 |
-
"epoch": 0.
|
| 30 |
-
"learning_rate": 1.
|
| 31 |
-
"loss": 0.
|
| 32 |
"step": 100
|
| 33 |
},
|
| 34 |
{
|
| 35 |
-
"epoch": 0.
|
| 36 |
-
"learning_rate": 2.46e-
|
| 37 |
-
"loss": 0.
|
| 38 |
"step": 125
|
| 39 |
},
|
| 40 |
{
|
| 41 |
-
"epoch": 0.
|
| 42 |
-
"learning_rate": 2.
|
| 43 |
-
"loss": 0.
|
| 44 |
"step": 150
|
| 45 |
},
|
| 46 |
{
|
| 47 |
-
"epoch": 0.
|
| 48 |
-
"learning_rate": 3.
|
| 49 |
-
"loss": 0.
|
| 50 |
"step": 175
|
| 51 |
},
|
| 52 |
{
|
| 53 |
-
"epoch": 0.
|
| 54 |
-
"learning_rate": 3.96e-
|
| 55 |
-
"loss": 0.
|
| 56 |
"step": 200
|
| 57 |
},
|
| 58 |
{
|
| 59 |
-
"epoch":
|
| 60 |
-
"learning_rate": 4.
|
| 61 |
-
"loss": 0.
|
| 62 |
"step": 225
|
| 63 |
},
|
| 64 |
{
|
| 65 |
-
"epoch":
|
| 66 |
-
"learning_rate": 4.
|
| 67 |
-
"loss": 0.
|
| 68 |
"step": 250
|
| 69 |
},
|
| 70 |
{
|
| 71 |
-
"epoch": 1
|
| 72 |
-
"learning_rate": 5.
|
| 73 |
-
"loss": 0.
|
| 74 |
"step": 275
|
| 75 |
},
|
| 76 |
{
|
| 77 |
-
"epoch":
|
| 78 |
-
"learning_rate": 5.
|
| 79 |
-
"loss": 0.
|
| 80 |
"step": 300
|
| 81 |
},
|
| 82 |
{
|
| 83 |
-
"epoch":
|
| 84 |
-
"learning_rate": 6.
|
| 85 |
-
"loss": 0.
|
| 86 |
"step": 325
|
| 87 |
},
|
| 88 |
{
|
| 89 |
-
"epoch":
|
| 90 |
-
"learning_rate": 6.
|
| 91 |
-
"loss": 0.
|
| 92 |
"step": 350
|
| 93 |
},
|
| 94 |
{
|
| 95 |
-
"epoch":
|
| 96 |
-
"learning_rate": 7.
|
| 97 |
-
"loss": 0.
|
| 98 |
"step": 375
|
| 99 |
},
|
| 100 |
{
|
| 101 |
-
"epoch":
|
| 102 |
-
"learning_rate": 7.
|
| 103 |
-
"loss": 0.
|
| 104 |
"step": 400
|
| 105 |
},
|
| 106 |
{
|
| 107 |
-
"epoch":
|
| 108 |
-
"learning_rate": 8.
|
| 109 |
-
"loss": 0.
|
| 110 |
"step": 425
|
| 111 |
},
|
| 112 |
{
|
| 113 |
-
"epoch":
|
| 114 |
-
"learning_rate": 8.96e-
|
| 115 |
-
"loss": 0.
|
| 116 |
"step": 450
|
| 117 |
},
|
| 118 |
{
|
| 119 |
-
"epoch":
|
| 120 |
-
"learning_rate": 9.
|
| 121 |
-
"loss": 0.
|
| 122 |
"step": 475
|
| 123 |
},
|
| 124 |
{
|
| 125 |
-
"epoch":
|
| 126 |
-
"learning_rate": 9.
|
| 127 |
-
"loss": 0.
|
| 128 |
"step": 500
|
| 129 |
},
|
| 130 |
{
|
| 131 |
-
"epoch":
|
| 132 |
-
"learning_rate": 9.
|
| 133 |
-
"loss": 0.
|
| 134 |
"step": 525
|
| 135 |
},
|
| 136 |
{
|
| 137 |
-
"epoch":
|
| 138 |
-
"learning_rate": 9.
|
| 139 |
-
"loss": 0.
|
| 140 |
"step": 550
|
| 141 |
},
|
| 142 |
{
|
| 143 |
-
"epoch": 2
|
| 144 |
-
"learning_rate": 9.
|
| 145 |
-
"loss": 0.
|
| 146 |
"step": 575
|
| 147 |
},
|
| 148 |
{
|
| 149 |
-
"epoch":
|
| 150 |
-
"learning_rate": 9.
|
| 151 |
-
"loss": 0.
|
| 152 |
"step": 600
|
| 153 |
},
|
| 154 |
{
|
| 155 |
-
"epoch":
|
| 156 |
-
"learning_rate": 9.
|
| 157 |
-
"loss": 0.
|
| 158 |
"step": 625
|
| 159 |
},
|
| 160 |
{
|
| 161 |
-
"epoch":
|
| 162 |
-
"learning_rate": 9.
|
| 163 |
-
"loss": 0.
|
| 164 |
"step": 650
|
| 165 |
},
|
| 166 |
{
|
| 167 |
-
"epoch":
|
| 168 |
-
"learning_rate": 9.
|
| 169 |
-
"loss": 0.
|
| 170 |
"step": 675
|
| 171 |
},
|
| 172 |
{
|
| 173 |
-
"epoch":
|
| 174 |
-
"learning_rate": 9.
|
| 175 |
-
"loss": 0.
|
| 176 |
"step": 700
|
| 177 |
},
|
| 178 |
{
|
| 179 |
-
"epoch":
|
| 180 |
-
"learning_rate": 9.
|
| 181 |
-
"loss": 0.
|
| 182 |
"step": 725
|
| 183 |
},
|
| 184 |
{
|
| 185 |
-
"epoch":
|
| 186 |
-
"learning_rate": 9.
|
| 187 |
-
"loss": 0.
|
| 188 |
"step": 750
|
| 189 |
},
|
| 190 |
{
|
| 191 |
-
"epoch":
|
| 192 |
-
"learning_rate": 9.
|
| 193 |
-
"loss": 0.
|
| 194 |
"step": 775
|
| 195 |
},
|
| 196 |
{
|
| 197 |
-
"epoch":
|
| 198 |
-
"learning_rate": 9.
|
| 199 |
-
"loss": 0.
|
| 200 |
"step": 800
|
| 201 |
},
|
| 202 |
{
|
| 203 |
-
"epoch":
|
| 204 |
-
"learning_rate": 9.
|
| 205 |
-
"loss": 0.
|
| 206 |
"step": 825
|
| 207 |
},
|
| 208 |
{
|
| 209 |
-
"epoch":
|
| 210 |
-
"learning_rate": 9.
|
| 211 |
-
"loss": 0.
|
| 212 |
"step": 850
|
| 213 |
},
|
| 214 |
{
|
| 215 |
-
"epoch":
|
| 216 |
-
"learning_rate": 9.
|
| 217 |
-
"loss": 0.
|
| 218 |
"step": 875
|
| 219 |
},
|
| 220 |
{
|
| 221 |
-
"epoch":
|
| 222 |
-
"learning_rate": 9.
|
| 223 |
-
"loss": 0.
|
| 224 |
"step": 900
|
| 225 |
},
|
| 226 |
{
|
| 227 |
-
"epoch":
|
| 228 |
-
"learning_rate": 9.
|
| 229 |
-
"loss": 0.
|
| 230 |
"step": 925
|
| 231 |
},
|
| 232 |
{
|
| 233 |
-
"epoch":
|
| 234 |
-
"learning_rate": 9.
|
| 235 |
-
"loss": 0.
|
| 236 |
"step": 950
|
| 237 |
},
|
| 238 |
{
|
| 239 |
-
"epoch":
|
| 240 |
-
"learning_rate":
|
| 241 |
-
"loss": 0.
|
| 242 |
"step": 975
|
| 243 |
},
|
| 244 |
{
|
| 245 |
-
"epoch":
|
| 246 |
-
"learning_rate":
|
| 247 |
-
"loss": 0.
|
| 248 |
"step": 1000
|
| 249 |
},
|
| 250 |
{
|
| 251 |
-
"epoch":
|
| 252 |
-
"eval_loss": 0.
|
| 253 |
-
"eval_runtime":
|
| 254 |
-
"eval_samples_per_second":
|
| 255 |
-
"eval_steps_per_second": 0.
|
| 256 |
-
"eval_wer":
|
| 257 |
"step": 1000
|
| 258 |
},
|
| 259 |
{
|
| 260 |
-
"epoch":
|
| 261 |
-
"learning_rate":
|
| 262 |
-
"loss": 0.
|
| 263 |
"step": 1025
|
| 264 |
},
|
| 265 |
{
|
| 266 |
-
"epoch":
|
| 267 |
-
"learning_rate":
|
| 268 |
-
"loss": 0.
|
| 269 |
"step": 1050
|
| 270 |
},
|
| 271 |
{
|
| 272 |
-
"epoch":
|
| 273 |
-
"learning_rate":
|
| 274 |
-
"loss": 0.
|
| 275 |
"step": 1075
|
| 276 |
},
|
| 277 |
{
|
| 278 |
-
"epoch":
|
| 279 |
-
"learning_rate":
|
| 280 |
-
"loss": 0.
|
| 281 |
"step": 1100
|
| 282 |
},
|
| 283 |
{
|
| 284 |
-
"epoch":
|
| 285 |
-
"learning_rate":
|
| 286 |
-
"loss": 0.
|
| 287 |
"step": 1125
|
| 288 |
},
|
| 289 |
{
|
| 290 |
-
"epoch":
|
| 291 |
-
"learning_rate":
|
| 292 |
-
"loss": 0.
|
| 293 |
"step": 1150
|
| 294 |
},
|
| 295 |
{
|
| 296 |
-
"epoch":
|
| 297 |
-
"learning_rate":
|
| 298 |
-
"loss": 0.
|
| 299 |
"step": 1175
|
| 300 |
},
|
| 301 |
{
|
| 302 |
-
"epoch":
|
| 303 |
-
"learning_rate":
|
| 304 |
-
"loss": 0.
|
| 305 |
"step": 1200
|
| 306 |
},
|
| 307 |
{
|
| 308 |
-
"epoch":
|
| 309 |
-
"learning_rate":
|
| 310 |
-
"loss": 0.
|
| 311 |
"step": 1225
|
| 312 |
},
|
| 313 |
{
|
| 314 |
-
"epoch":
|
| 315 |
-
"learning_rate":
|
| 316 |
-
"loss": 0.
|
| 317 |
"step": 1250
|
| 318 |
},
|
| 319 |
{
|
| 320 |
-
"epoch":
|
| 321 |
-
"learning_rate":
|
| 322 |
-
"loss": 0.
|
| 323 |
"step": 1275
|
| 324 |
},
|
| 325 |
{
|
| 326 |
-
"epoch":
|
| 327 |
-
"learning_rate":
|
| 328 |
-
"loss": 0.
|
| 329 |
"step": 1300
|
| 330 |
},
|
| 331 |
{
|
| 332 |
-
"epoch":
|
| 333 |
-
"learning_rate":
|
| 334 |
-
"loss": 0.
|
| 335 |
"step": 1325
|
| 336 |
},
|
| 337 |
{
|
| 338 |
-
"epoch":
|
| 339 |
-
"learning_rate":
|
| 340 |
-
"loss": 0.
|
| 341 |
"step": 1350
|
| 342 |
},
|
| 343 |
{
|
| 344 |
-
"epoch":
|
| 345 |
-
"learning_rate":
|
| 346 |
-
"loss": 0.
|
| 347 |
"step": 1375
|
| 348 |
},
|
| 349 |
{
|
| 350 |
-
"epoch":
|
| 351 |
-
"learning_rate":
|
| 352 |
-
"loss": 0.
|
| 353 |
"step": 1400
|
| 354 |
},
|
| 355 |
{
|
| 356 |
-
"epoch":
|
| 357 |
-
"learning_rate":
|
| 358 |
-
"loss": 0.
|
| 359 |
"step": 1425
|
| 360 |
},
|
| 361 |
{
|
| 362 |
-
"epoch":
|
| 363 |
-
"learning_rate":
|
| 364 |
-
"loss": 0.
|
| 365 |
"step": 1450
|
| 366 |
},
|
| 367 |
{
|
| 368 |
-
"epoch":
|
| 369 |
-
"learning_rate":
|
| 370 |
-
"loss": 0.
|
| 371 |
"step": 1475
|
| 372 |
},
|
| 373 |
{
|
| 374 |
-
"epoch":
|
| 375 |
-
"learning_rate":
|
| 376 |
-
"loss": 0.
|
| 377 |
"step": 1500
|
| 378 |
},
|
| 379 |
{
|
| 380 |
-
"epoch":
|
| 381 |
-
"learning_rate":
|
| 382 |
-
"loss": 0.
|
| 383 |
"step": 1525
|
| 384 |
},
|
| 385 |
{
|
| 386 |
-
"epoch":
|
| 387 |
-
"learning_rate":
|
| 388 |
-
"loss": 0.
|
| 389 |
"step": 1550
|
| 390 |
},
|
| 391 |
{
|
| 392 |
-
"epoch":
|
| 393 |
-
"learning_rate":
|
| 394 |
-
"loss": 0.
|
| 395 |
"step": 1575
|
| 396 |
},
|
| 397 |
{
|
| 398 |
-
"epoch":
|
| 399 |
-
"learning_rate":
|
| 400 |
-
"loss": 0.
|
| 401 |
"step": 1600
|
| 402 |
},
|
| 403 |
{
|
| 404 |
-
"epoch":
|
| 405 |
-
"learning_rate":
|
| 406 |
-
"loss": 0.
|
| 407 |
"step": 1625
|
| 408 |
},
|
| 409 |
{
|
| 410 |
-
"epoch":
|
| 411 |
-
"learning_rate":
|
| 412 |
-
"loss": 0.
|
| 413 |
"step": 1650
|
| 414 |
},
|
| 415 |
{
|
| 416 |
-
"epoch":
|
| 417 |
-
"learning_rate":
|
| 418 |
-
"loss": 0.
|
| 419 |
"step": 1675
|
| 420 |
},
|
| 421 |
{
|
| 422 |
-
"epoch":
|
| 423 |
-
"learning_rate":
|
| 424 |
-
"loss": 0.
|
| 425 |
"step": 1700
|
| 426 |
},
|
| 427 |
{
|
| 428 |
-
"epoch":
|
| 429 |
-
"learning_rate":
|
| 430 |
-
"loss": 0.
|
| 431 |
"step": 1725
|
| 432 |
},
|
| 433 |
{
|
| 434 |
-
"epoch":
|
| 435 |
-
"learning_rate":
|
| 436 |
-
"loss": 0.
|
| 437 |
"step": 1750
|
| 438 |
},
|
| 439 |
{
|
| 440 |
-
"epoch":
|
| 441 |
-
"learning_rate":
|
| 442 |
-
"loss": 0.
|
| 443 |
"step": 1775
|
| 444 |
},
|
| 445 |
{
|
| 446 |
-
"epoch":
|
| 447 |
-
"learning_rate":
|
| 448 |
-
"loss": 0.
|
| 449 |
"step": 1800
|
| 450 |
},
|
| 451 |
{
|
| 452 |
-
"epoch":
|
| 453 |
-
"learning_rate":
|
| 454 |
-
"loss": 0.
|
| 455 |
"step": 1825
|
| 456 |
},
|
| 457 |
{
|
| 458 |
-
"epoch":
|
| 459 |
-
"learning_rate":
|
| 460 |
-
"loss": 0.
|
| 461 |
"step": 1850
|
| 462 |
},
|
| 463 |
{
|
| 464 |
-
"epoch":
|
| 465 |
-
"learning_rate":
|
| 466 |
-
"loss": 0.
|
| 467 |
"step": 1875
|
| 468 |
},
|
| 469 |
{
|
| 470 |
-
"epoch":
|
| 471 |
-
"learning_rate":
|
| 472 |
-
"loss": 0.
|
| 473 |
"step": 1900
|
| 474 |
},
|
| 475 |
{
|
| 476 |
-
"epoch":
|
| 477 |
-
"learning_rate":
|
| 478 |
-
"loss": 0.
|
| 479 |
"step": 1925
|
| 480 |
},
|
| 481 |
{
|
| 482 |
-
"epoch":
|
| 483 |
-
"learning_rate":
|
| 484 |
-
"loss": 0.
|
| 485 |
"step": 1950
|
| 486 |
},
|
| 487 |
{
|
| 488 |
-
"epoch":
|
| 489 |
-
"learning_rate":
|
| 490 |
-
"loss": 0.
|
| 491 |
"step": 1975
|
| 492 |
},
|
| 493 |
{
|
| 494 |
-
"epoch":
|
| 495 |
-
"learning_rate":
|
| 496 |
-
"loss": 0.
|
| 497 |
"step": 2000
|
| 498 |
},
|
| 499 |
{
|
| 500 |
-
"epoch":
|
| 501 |
-
"eval_loss": 0.
|
| 502 |
-
"eval_runtime":
|
| 503 |
-
"eval_samples_per_second":
|
| 504 |
-
"eval_steps_per_second": 0.
|
| 505 |
-
"eval_wer":
|
| 506 |
"step": 2000
|
| 507 |
},
|
| 508 |
{
|
| 509 |
-
"epoch":
|
| 510 |
-
"learning_rate":
|
| 511 |
-
"loss": 0.
|
| 512 |
"step": 2025
|
| 513 |
},
|
| 514 |
{
|
| 515 |
-
"epoch":
|
| 516 |
-
"learning_rate":
|
| 517 |
-
"loss": 0.
|
| 518 |
"step": 2050
|
| 519 |
},
|
| 520 |
{
|
| 521 |
-
"epoch":
|
| 522 |
-
"learning_rate":
|
| 523 |
-
"loss": 0.
|
| 524 |
"step": 2075
|
| 525 |
},
|
| 526 |
{
|
| 527 |
-
"epoch":
|
| 528 |
-
"learning_rate":
|
| 529 |
-
"loss": 0.
|
| 530 |
"step": 2100
|
| 531 |
},
|
| 532 |
{
|
| 533 |
-
"epoch":
|
| 534 |
-
"learning_rate":
|
| 535 |
-
"loss": 0.
|
| 536 |
"step": 2125
|
| 537 |
},
|
| 538 |
{
|
| 539 |
-
"epoch":
|
| 540 |
-
"learning_rate":
|
| 541 |
-
"loss": 0.
|
| 542 |
"step": 2150
|
| 543 |
},
|
| 544 |
{
|
| 545 |
-
"epoch":
|
| 546 |
-
"learning_rate":
|
| 547 |
-
"loss": 0.
|
| 548 |
"step": 2175
|
| 549 |
},
|
| 550 |
{
|
| 551 |
-
"epoch":
|
| 552 |
-
"learning_rate":
|
| 553 |
-
"loss": 0.
|
| 554 |
"step": 2200
|
| 555 |
},
|
| 556 |
{
|
| 557 |
-
"epoch":
|
| 558 |
-
"learning_rate":
|
| 559 |
-
"loss": 0.
|
| 560 |
"step": 2225
|
| 561 |
},
|
| 562 |
{
|
| 563 |
-
"epoch":
|
| 564 |
-
"learning_rate":
|
| 565 |
-
"loss": 0.
|
| 566 |
"step": 2250
|
| 567 |
},
|
| 568 |
{
|
| 569 |
-
"epoch":
|
| 570 |
-
"learning_rate":
|
| 571 |
-
"loss": 0.
|
| 572 |
"step": 2275
|
| 573 |
},
|
| 574 |
{
|
| 575 |
-
"epoch":
|
| 576 |
-
"learning_rate":
|
| 577 |
-
"loss": 0.
|
| 578 |
"step": 2300
|
| 579 |
},
|
| 580 |
{
|
| 581 |
-
"epoch":
|
| 582 |
-
"learning_rate":
|
| 583 |
-
"loss": 0.
|
| 584 |
"step": 2325
|
| 585 |
},
|
| 586 |
{
|
| 587 |
-
"epoch":
|
| 588 |
-
"learning_rate":
|
| 589 |
-
"loss": 0.
|
| 590 |
"step": 2350
|
| 591 |
},
|
| 592 |
{
|
| 593 |
-
"epoch":
|
| 594 |
-
"learning_rate":
|
| 595 |
-
"loss": 0.
|
| 596 |
"step": 2375
|
| 597 |
},
|
| 598 |
{
|
| 599 |
-
"epoch":
|
| 600 |
-
"learning_rate":
|
| 601 |
-
"loss": 0.
|
| 602 |
"step": 2400
|
| 603 |
},
|
| 604 |
{
|
| 605 |
-
"epoch":
|
| 606 |
-
"learning_rate":
|
| 607 |
-
"loss": 0.
|
| 608 |
"step": 2425
|
| 609 |
},
|
| 610 |
{
|
| 611 |
-
"epoch":
|
| 612 |
-
"learning_rate":
|
| 613 |
-
"loss": 0.
|
| 614 |
"step": 2450
|
| 615 |
},
|
| 616 |
{
|
| 617 |
-
"epoch":
|
| 618 |
-
"learning_rate":
|
| 619 |
-
"loss": 0.
|
| 620 |
"step": 2475
|
| 621 |
},
|
| 622 |
{
|
| 623 |
-
"epoch":
|
| 624 |
-
"learning_rate":
|
| 625 |
-
"loss": 0.
|
| 626 |
"step": 2500
|
| 627 |
},
|
| 628 |
{
|
| 629 |
-
"epoch":
|
| 630 |
-
"learning_rate":
|
| 631 |
-
"loss": 0.
|
| 632 |
"step": 2525
|
| 633 |
},
|
| 634 |
{
|
| 635 |
-
"epoch":
|
| 636 |
-
"learning_rate":
|
| 637 |
-
"loss": 0.
|
| 638 |
"step": 2550
|
| 639 |
},
|
| 640 |
{
|
| 641 |
-
"epoch":
|
| 642 |
-
"learning_rate":
|
| 643 |
-
"loss": 0.
|
| 644 |
"step": 2575
|
| 645 |
},
|
| 646 |
{
|
| 647 |
-
"epoch":
|
| 648 |
-
"learning_rate":
|
| 649 |
-
"loss": 0.
|
| 650 |
"step": 2600
|
| 651 |
},
|
| 652 |
{
|
| 653 |
-
"epoch":
|
| 654 |
-
"learning_rate":
|
| 655 |
-
"loss": 0.
|
| 656 |
"step": 2625
|
| 657 |
},
|
| 658 |
{
|
| 659 |
-
"epoch":
|
| 660 |
-
"learning_rate":
|
| 661 |
-
"loss": 0.
|
| 662 |
"step": 2650
|
| 663 |
},
|
| 664 |
{
|
| 665 |
-
"epoch":
|
| 666 |
-
"learning_rate":
|
| 667 |
-
"loss": 0.
|
| 668 |
"step": 2675
|
| 669 |
},
|
| 670 |
{
|
| 671 |
-
"epoch":
|
| 672 |
-
"learning_rate":
|
| 673 |
-
"loss": 0.
|
| 674 |
"step": 2700
|
| 675 |
},
|
| 676 |
{
|
| 677 |
-
"epoch":
|
| 678 |
-
"learning_rate":
|
| 679 |
-
"loss": 0.
|
| 680 |
"step": 2725
|
| 681 |
},
|
| 682 |
{
|
| 683 |
-
"epoch":
|
| 684 |
-
"learning_rate":
|
| 685 |
-
"loss": 0.
|
| 686 |
"step": 2750
|
| 687 |
},
|
| 688 |
{
|
| 689 |
-
"epoch":
|
| 690 |
-
"learning_rate":
|
| 691 |
-
"loss": 0.
|
| 692 |
"step": 2775
|
| 693 |
},
|
| 694 |
{
|
| 695 |
-
"epoch":
|
| 696 |
-
"learning_rate":
|
| 697 |
-
"loss": 0.
|
| 698 |
"step": 2800
|
| 699 |
},
|
| 700 |
{
|
| 701 |
-
"epoch":
|
| 702 |
-
"learning_rate":
|
| 703 |
-
"loss": 0.
|
| 704 |
"step": 2825
|
| 705 |
},
|
| 706 |
{
|
| 707 |
-
"epoch":
|
| 708 |
-
"learning_rate":
|
| 709 |
-
"loss": 0.
|
| 710 |
"step": 2850
|
| 711 |
},
|
| 712 |
{
|
| 713 |
-
"epoch":
|
| 714 |
-
"learning_rate":
|
| 715 |
-
"loss": 0.
|
| 716 |
"step": 2875
|
| 717 |
},
|
| 718 |
{
|
| 719 |
-
"epoch":
|
| 720 |
-
"learning_rate":
|
| 721 |
-
"loss": 0.
|
| 722 |
"step": 2900
|
| 723 |
},
|
| 724 |
{
|
| 725 |
-
"epoch":
|
| 726 |
-
"learning_rate":
|
| 727 |
-
"loss": 0.
|
| 728 |
"step": 2925
|
| 729 |
},
|
| 730 |
{
|
| 731 |
-
"epoch":
|
| 732 |
-
"learning_rate":
|
| 733 |
-
"loss": 0.
|
| 734 |
"step": 2950
|
| 735 |
},
|
| 736 |
{
|
| 737 |
-
"epoch":
|
| 738 |
-
"learning_rate":
|
| 739 |
-
"loss": 0.
|
| 740 |
"step": 2975
|
| 741 |
},
|
| 742 |
{
|
| 743 |
-
"epoch":
|
| 744 |
-
"learning_rate":
|
| 745 |
-
"loss": 0.
|
| 746 |
"step": 3000
|
| 747 |
},
|
| 748 |
{
|
| 749 |
-
"epoch":
|
| 750 |
-
"eval_loss": 0.
|
| 751 |
-
"eval_runtime":
|
| 752 |
-
"eval_samples_per_second":
|
| 753 |
-
"eval_steps_per_second": 0.
|
| 754 |
-
"eval_wer":
|
| 755 |
"step": 3000
|
| 756 |
},
|
| 757 |
{
|
| 758 |
-
"epoch":
|
| 759 |
-
"learning_rate":
|
| 760 |
-
"loss": 0.
|
| 761 |
"step": 3025
|
| 762 |
},
|
| 763 |
{
|
| 764 |
-
"epoch":
|
| 765 |
-
"learning_rate":
|
| 766 |
-
"loss": 0.
|
| 767 |
"step": 3050
|
| 768 |
},
|
| 769 |
{
|
| 770 |
-
"epoch":
|
| 771 |
-
"learning_rate":
|
| 772 |
-
"loss": 0.
|
| 773 |
"step": 3075
|
| 774 |
},
|
| 775 |
{
|
| 776 |
-
"epoch":
|
| 777 |
-
"learning_rate":
|
| 778 |
-
"loss": 0.
|
| 779 |
"step": 3100
|
| 780 |
},
|
| 781 |
{
|
| 782 |
-
"epoch":
|
| 783 |
-
"learning_rate":
|
| 784 |
-
"loss": 0.
|
| 785 |
"step": 3125
|
| 786 |
},
|
| 787 |
{
|
| 788 |
-
"epoch":
|
| 789 |
-
"learning_rate":
|
| 790 |
-
"loss": 0.
|
| 791 |
"step": 3150
|
| 792 |
},
|
| 793 |
{
|
| 794 |
-
"epoch":
|
| 795 |
-
"learning_rate":
|
| 796 |
-
"loss": 0.
|
| 797 |
"step": 3175
|
| 798 |
},
|
| 799 |
{
|
| 800 |
-
"epoch":
|
| 801 |
-
"learning_rate":
|
| 802 |
-
"loss": 0.
|
| 803 |
"step": 3200
|
| 804 |
},
|
| 805 |
{
|
| 806 |
-
"epoch":
|
| 807 |
-
"learning_rate":
|
| 808 |
-
"loss": 0.
|
| 809 |
"step": 3225
|
| 810 |
},
|
| 811 |
{
|
| 812 |
-
"epoch":
|
| 813 |
-
"learning_rate":
|
| 814 |
-
"loss": 0.
|
| 815 |
"step": 3250
|
| 816 |
},
|
| 817 |
{
|
| 818 |
-
"epoch":
|
| 819 |
-
"learning_rate":
|
| 820 |
-
"loss": 0.
|
| 821 |
"step": 3275
|
| 822 |
},
|
| 823 |
{
|
| 824 |
-
"epoch": 16
|
| 825 |
-
"learning_rate":
|
| 826 |
-
"loss": 0.
|
| 827 |
"step": 3300
|
| 828 |
},
|
| 829 |
{
|
| 830 |
-
"epoch":
|
| 831 |
-
"learning_rate":
|
| 832 |
-
"loss": 0.
|
| 833 |
"step": 3325
|
| 834 |
},
|
| 835 |
{
|
| 836 |
-
"epoch":
|
| 837 |
-
"learning_rate":
|
| 838 |
-
"loss": 0.
|
| 839 |
"step": 3350
|
| 840 |
},
|
| 841 |
{
|
| 842 |
-
"epoch":
|
| 843 |
-
"learning_rate":
|
| 844 |
-
"loss": 0.
|
| 845 |
"step": 3375
|
| 846 |
},
|
| 847 |
{
|
| 848 |
-
"epoch":
|
| 849 |
-
"learning_rate":
|
| 850 |
-
"loss": 0.
|
| 851 |
"step": 3400
|
| 852 |
},
|
| 853 |
{
|
| 854 |
-
"epoch":
|
| 855 |
-
"learning_rate":
|
| 856 |
-
"loss": 0.
|
| 857 |
"step": 3425
|
| 858 |
},
|
| 859 |
{
|
| 860 |
-
"epoch":
|
| 861 |
-
"learning_rate":
|
| 862 |
-
"loss": 0.
|
| 863 |
"step": 3450
|
| 864 |
},
|
| 865 |
{
|
| 866 |
-
"epoch":
|
| 867 |
-
"learning_rate":
|
| 868 |
-
"loss": 0.
|
| 869 |
"step": 3475
|
| 870 |
},
|
| 871 |
{
|
| 872 |
-
"epoch":
|
| 873 |
-
"learning_rate":
|
| 874 |
-
"loss": 0.
|
| 875 |
"step": 3500
|
| 876 |
},
|
| 877 |
{
|
| 878 |
-
"epoch":
|
| 879 |
-
"learning_rate":
|
| 880 |
-
"loss": 0.
|
| 881 |
"step": 3525
|
| 882 |
},
|
| 883 |
{
|
| 884 |
-
"epoch":
|
| 885 |
-
"learning_rate":
|
| 886 |
-
"loss": 0.
|
| 887 |
"step": 3550
|
| 888 |
},
|
| 889 |
{
|
| 890 |
-
"epoch":
|
| 891 |
-
"learning_rate":
|
| 892 |
-
"loss": 0.
|
| 893 |
"step": 3575
|
| 894 |
},
|
| 895 |
{
|
| 896 |
-
"epoch":
|
| 897 |
-
"learning_rate":
|
| 898 |
-
"loss": 0.
|
| 899 |
"step": 3600
|
| 900 |
},
|
| 901 |
{
|
| 902 |
-
"epoch":
|
| 903 |
-
"learning_rate":
|
| 904 |
-
"loss": 0.
|
| 905 |
"step": 3625
|
| 906 |
},
|
| 907 |
{
|
| 908 |
-
"epoch":
|
| 909 |
-
"learning_rate":
|
| 910 |
-
"loss": 0.
|
| 911 |
"step": 3650
|
| 912 |
},
|
| 913 |
{
|
| 914 |
-
"epoch":
|
| 915 |
-
"learning_rate":
|
| 916 |
-
"loss": 0.
|
| 917 |
"step": 3675
|
| 918 |
},
|
| 919 |
{
|
| 920 |
-
"epoch":
|
| 921 |
-
"learning_rate":
|
| 922 |
-
"loss": 0.
|
| 923 |
"step": 3700
|
| 924 |
},
|
| 925 |
{
|
| 926 |
-
"epoch":
|
| 927 |
-
"learning_rate":
|
| 928 |
-
"loss": 0.
|
| 929 |
"step": 3725
|
| 930 |
},
|
| 931 |
{
|
| 932 |
-
"epoch":
|
| 933 |
-
"learning_rate":
|
| 934 |
-
"loss": 0.
|
| 935 |
"step": 3750
|
| 936 |
},
|
| 937 |
{
|
| 938 |
-
"epoch":
|
| 939 |
-
"learning_rate":
|
| 940 |
-
"loss": 0.
|
| 941 |
"step": 3775
|
| 942 |
},
|
| 943 |
{
|
| 944 |
-
"epoch":
|
| 945 |
-
"learning_rate":
|
| 946 |
-
"loss": 0.
|
| 947 |
"step": 3800
|
| 948 |
},
|
| 949 |
{
|
| 950 |
-
"epoch":
|
| 951 |
-
"learning_rate":
|
| 952 |
-
"loss": 0.
|
| 953 |
"step": 3825
|
| 954 |
},
|
| 955 |
{
|
| 956 |
-
"epoch":
|
| 957 |
-
"learning_rate":
|
| 958 |
-
"loss": 0.
|
| 959 |
"step": 3850
|
| 960 |
},
|
| 961 |
{
|
| 962 |
-
"epoch":
|
| 963 |
-
"learning_rate":
|
| 964 |
-
"loss": 0.
|
| 965 |
"step": 3875
|
| 966 |
},
|
| 967 |
{
|
| 968 |
-
"epoch":
|
| 969 |
-
"learning_rate":
|
| 970 |
-
"loss": 0.
|
| 971 |
"step": 3900
|
| 972 |
},
|
| 973 |
{
|
| 974 |
-
"epoch":
|
| 975 |
-
"learning_rate":
|
| 976 |
-
"loss": 0.
|
| 977 |
"step": 3925
|
| 978 |
},
|
| 979 |
{
|
| 980 |
-
"epoch":
|
| 981 |
-
"learning_rate":
|
| 982 |
-
"loss": 0.
|
| 983 |
"step": 3950
|
| 984 |
},
|
| 985 |
{
|
| 986 |
-
"epoch":
|
| 987 |
-
"learning_rate":
|
| 988 |
-
"loss": 0.
|
| 989 |
"step": 3975
|
| 990 |
},
|
| 991 |
{
|
| 992 |
-
"epoch":
|
| 993 |
-
"learning_rate":
|
| 994 |
-
"loss": 0.
|
| 995 |
"step": 4000
|
| 996 |
},
|
| 997 |
{
|
| 998 |
-
"epoch":
|
| 999 |
-
"eval_loss": 0.
|
| 1000 |
-
"eval_runtime":
|
| 1001 |
-
"eval_samples_per_second":
|
| 1002 |
-
"eval_steps_per_second": 0.
|
| 1003 |
-
"eval_wer":
|
| 1004 |
"step": 4000
|
| 1005 |
},
|
| 1006 |
{
|
| 1007 |
-
"epoch":
|
| 1008 |
-
"learning_rate":
|
| 1009 |
-
"loss": 0.
|
| 1010 |
"step": 4025
|
| 1011 |
},
|
| 1012 |
{
|
| 1013 |
-
"epoch":
|
| 1014 |
-
"learning_rate":
|
| 1015 |
-
"loss": 0.
|
| 1016 |
"step": 4050
|
| 1017 |
},
|
| 1018 |
{
|
| 1019 |
-
"epoch":
|
| 1020 |
-
"learning_rate":
|
| 1021 |
-
"loss": 0.
|
| 1022 |
"step": 4075
|
| 1023 |
},
|
| 1024 |
{
|
| 1025 |
-
"epoch":
|
| 1026 |
-
"learning_rate":
|
| 1027 |
-
"loss": 0.
|
| 1028 |
"step": 4100
|
| 1029 |
},
|
| 1030 |
{
|
| 1031 |
-
"epoch":
|
| 1032 |
-
"learning_rate":
|
| 1033 |
-
"loss": 0.
|
| 1034 |
"step": 4125
|
| 1035 |
},
|
| 1036 |
{
|
| 1037 |
-
"epoch":
|
| 1038 |
-
"learning_rate":
|
| 1039 |
-
"loss": 0.
|
| 1040 |
"step": 4150
|
| 1041 |
},
|
| 1042 |
{
|
| 1043 |
-
"epoch":
|
| 1044 |
-
"learning_rate":
|
| 1045 |
-
"loss": 0.
|
| 1046 |
"step": 4175
|
| 1047 |
},
|
| 1048 |
{
|
| 1049 |
-
"epoch":
|
| 1050 |
-
"learning_rate":
|
| 1051 |
-
"loss": 0.
|
| 1052 |
"step": 4200
|
| 1053 |
},
|
| 1054 |
{
|
| 1055 |
-
"epoch":
|
| 1056 |
-
"learning_rate":
|
| 1057 |
-
"loss": 0.
|
| 1058 |
"step": 4225
|
| 1059 |
},
|
| 1060 |
{
|
| 1061 |
-
"epoch":
|
| 1062 |
-
"learning_rate":
|
| 1063 |
-
"loss": 0.
|
| 1064 |
"step": 4250
|
| 1065 |
},
|
| 1066 |
{
|
| 1067 |
-
"epoch":
|
| 1068 |
-
"learning_rate":
|
| 1069 |
-
"loss": 0.
|
| 1070 |
"step": 4275
|
| 1071 |
},
|
| 1072 |
{
|
| 1073 |
-
"epoch":
|
| 1074 |
-
"learning_rate":
|
| 1075 |
-
"loss": 0.
|
| 1076 |
"step": 4300
|
| 1077 |
},
|
| 1078 |
{
|
| 1079 |
-
"epoch":
|
| 1080 |
-
"learning_rate":
|
| 1081 |
-
"loss": 0.
|
| 1082 |
"step": 4325
|
| 1083 |
},
|
| 1084 |
{
|
| 1085 |
-
"epoch":
|
| 1086 |
-
"learning_rate":
|
| 1087 |
-
"loss": 0.
|
| 1088 |
"step": 4350
|
| 1089 |
},
|
| 1090 |
{
|
| 1091 |
-
"epoch":
|
| 1092 |
-
"learning_rate":
|
| 1093 |
-
"loss": 0.
|
| 1094 |
"step": 4375
|
| 1095 |
},
|
| 1096 |
{
|
| 1097 |
-
"epoch":
|
| 1098 |
-
"learning_rate":
|
| 1099 |
-
"loss": 0.
|
| 1100 |
"step": 4400
|
| 1101 |
},
|
| 1102 |
{
|
| 1103 |
-
"epoch":
|
| 1104 |
-
"learning_rate":
|
| 1105 |
-
"loss": 0.
|
| 1106 |
"step": 4425
|
| 1107 |
},
|
| 1108 |
{
|
| 1109 |
-
"epoch":
|
| 1110 |
-
"learning_rate":
|
| 1111 |
-
"loss": 0.
|
| 1112 |
"step": 4450
|
| 1113 |
},
|
| 1114 |
{
|
| 1115 |
-
"epoch":
|
| 1116 |
-
"learning_rate":
|
| 1117 |
-
"loss": 0.
|
| 1118 |
"step": 4475
|
| 1119 |
},
|
| 1120 |
{
|
| 1121 |
-
"epoch":
|
| 1122 |
-
"learning_rate":
|
| 1123 |
-
"loss": 0.
|
| 1124 |
"step": 4500
|
| 1125 |
},
|
| 1126 |
{
|
| 1127 |
-
"epoch":
|
| 1128 |
-
"learning_rate":
|
| 1129 |
-
"loss": 0.
|
| 1130 |
"step": 4525
|
| 1131 |
},
|
| 1132 |
{
|
| 1133 |
-
"epoch":
|
| 1134 |
-
"learning_rate":
|
| 1135 |
-
"loss": 0.
|
| 1136 |
"step": 4550
|
| 1137 |
},
|
| 1138 |
{
|
| 1139 |
-
"epoch":
|
| 1140 |
-
"learning_rate":
|
| 1141 |
-
"loss": 0.
|
| 1142 |
"step": 4575
|
| 1143 |
},
|
| 1144 |
{
|
| 1145 |
-
"epoch":
|
| 1146 |
-
"learning_rate":
|
| 1147 |
-
"loss": 0.
|
| 1148 |
"step": 4600
|
| 1149 |
},
|
| 1150 |
{
|
| 1151 |
-
"epoch":
|
| 1152 |
-
"learning_rate":
|
| 1153 |
-
"loss": 0.
|
| 1154 |
"step": 4625
|
| 1155 |
},
|
| 1156 |
{
|
| 1157 |
-
"epoch":
|
| 1158 |
-
"learning_rate":
|
| 1159 |
-
"loss": 0.
|
| 1160 |
"step": 4650
|
| 1161 |
},
|
| 1162 |
{
|
| 1163 |
-
"epoch":
|
| 1164 |
-
"learning_rate":
|
| 1165 |
-
"loss": 0.
|
| 1166 |
"step": 4675
|
| 1167 |
},
|
| 1168 |
{
|
| 1169 |
-
"epoch":
|
| 1170 |
-
"learning_rate":
|
| 1171 |
-
"loss": 0.
|
| 1172 |
"step": 4700
|
| 1173 |
},
|
| 1174 |
{
|
| 1175 |
-
"epoch":
|
| 1176 |
-
"learning_rate":
|
| 1177 |
-
"loss": 0.
|
| 1178 |
"step": 4725
|
| 1179 |
},
|
| 1180 |
{
|
| 1181 |
-
"epoch":
|
| 1182 |
-
"learning_rate": 5.
|
| 1183 |
-
"loss": 0.
|
| 1184 |
"step": 4750
|
| 1185 |
},
|
| 1186 |
{
|
| 1187 |
-
"epoch":
|
| 1188 |
-
"learning_rate": 5.
|
| 1189 |
-
"loss": 0.
|
| 1190 |
"step": 4775
|
| 1191 |
},
|
| 1192 |
{
|
| 1193 |
-
"epoch":
|
| 1194 |
-
"learning_rate":
|
| 1195 |
-
"loss": 0.
|
| 1196 |
"step": 4800
|
| 1197 |
},
|
| 1198 |
{
|
| 1199 |
-
"epoch":
|
| 1200 |
-
"learning_rate":
|
| 1201 |
-
"loss": 0.
|
| 1202 |
"step": 4825
|
| 1203 |
},
|
| 1204 |
{
|
| 1205 |
-
"epoch":
|
| 1206 |
-
"learning_rate":
|
| 1207 |
-
"loss": 0.
|
| 1208 |
"step": 4850
|
| 1209 |
},
|
| 1210 |
{
|
| 1211 |
-
"epoch":
|
| 1212 |
-
"learning_rate":
|
| 1213 |
-
"loss": 0.
|
| 1214 |
"step": 4875
|
| 1215 |
},
|
| 1216 |
{
|
| 1217 |
-
"epoch":
|
| 1218 |
-
"learning_rate":
|
| 1219 |
-
"loss": 0.
|
| 1220 |
"step": 4900
|
| 1221 |
},
|
| 1222 |
{
|
| 1223 |
-
"epoch":
|
| 1224 |
-
"learning_rate":
|
| 1225 |
-
"loss": 0.
|
| 1226 |
"step": 4925
|
| 1227 |
},
|
| 1228 |
{
|
| 1229 |
-
"epoch":
|
| 1230 |
-
"learning_rate":
|
| 1231 |
-
"loss": 0.
|
| 1232 |
"step": 4950
|
| 1233 |
},
|
| 1234 |
{
|
| 1235 |
-
"epoch":
|
| 1236 |
-
"learning_rate":
|
| 1237 |
-
"loss": 0.
|
| 1238 |
"step": 4975
|
| 1239 |
},
|
| 1240 |
{
|
| 1241 |
-
"epoch":
|
| 1242 |
-
"learning_rate":
|
| 1243 |
-
"loss": 0.
|
| 1244 |
"step": 5000
|
| 1245 |
},
|
| 1246 |
{
|
| 1247 |
-
"epoch":
|
| 1248 |
-
"eval_loss": 0.
|
| 1249 |
-
"eval_runtime":
|
| 1250 |
-
"eval_samples_per_second":
|
| 1251 |
-
"eval_steps_per_second": 0.
|
| 1252 |
-
"eval_wer":
|
| 1253 |
"step": 5000
|
| 1254 |
},
|
| 1255 |
{
|
| 1256 |
-
"epoch":
|
| 1257 |
-
"
|
| 1258 |
-
"
|
| 1259 |
-
"
|
| 1260 |
-
|
| 1261 |
-
|
| 1262 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1263 |
}
|
| 1264 |
],
|
| 1265 |
-
"max_steps":
|
| 1266 |
-
"num_train_epochs":
|
| 1267 |
-
"total_flos": 1.
|
| 1268 |
"trial_name": null,
|
| 1269 |
"trial_params": null
|
| 1270 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 3.993359771281011,
|
| 3 |
+
"best_model_checkpoint": "./whisper-medium-id/checkpoint-7000",
|
| 4 |
+
"epoch": 3.522367030644593,
|
| 5 |
+
"global_step": 10000,
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
| 9 |
"log_history": [
|
| 10 |
{
|
| 11 |
"epoch": 0.01,
|
| 12 |
+
"learning_rate": 4.8e-08,
|
| 13 |
+
"loss": 0.0892,
|
| 14 |
"step": 25
|
| 15 |
},
|
| 16 |
{
|
| 17 |
+
"epoch": 0.02,
|
| 18 |
+
"learning_rate": 9.8e-08,
|
| 19 |
+
"loss": 0.0727,
|
| 20 |
"step": 50
|
| 21 |
},
|
| 22 |
{
|
| 23 |
+
"epoch": 0.03,
|
| 24 |
+
"learning_rate": 1.4599999999999998e-07,
|
| 25 |
+
"loss": 0.0545,
|
| 26 |
"step": 75
|
| 27 |
},
|
| 28 |
{
|
| 29 |
+
"epoch": 0.04,
|
| 30 |
+
"learning_rate": 1.96e-07,
|
| 31 |
+
"loss": 0.066,
|
| 32 |
"step": 100
|
| 33 |
},
|
| 34 |
{
|
| 35 |
+
"epoch": 0.04,
|
| 36 |
+
"learning_rate": 2.46e-07,
|
| 37 |
+
"loss": 0.0635,
|
| 38 |
"step": 125
|
| 39 |
},
|
| 40 |
{
|
| 41 |
+
"epoch": 0.05,
|
| 42 |
+
"learning_rate": 2.9599999999999995e-07,
|
| 43 |
+
"loss": 0.0499,
|
| 44 |
"step": 150
|
| 45 |
},
|
| 46 |
{
|
| 47 |
+
"epoch": 0.06,
|
| 48 |
+
"learning_rate": 3.4599999999999995e-07,
|
| 49 |
+
"loss": 0.0415,
|
| 50 |
"step": 175
|
| 51 |
},
|
| 52 |
{
|
| 53 |
+
"epoch": 0.07,
|
| 54 |
+
"learning_rate": 3.96e-07,
|
| 55 |
+
"loss": 0.0457,
|
| 56 |
"step": 200
|
| 57 |
},
|
| 58 |
{
|
| 59 |
+
"epoch": 0.08,
|
| 60 |
+
"learning_rate": 4.46e-07,
|
| 61 |
+
"loss": 0.0435,
|
| 62 |
"step": 225
|
| 63 |
},
|
| 64 |
{
|
| 65 |
+
"epoch": 0.09,
|
| 66 |
+
"learning_rate": 4.96e-07,
|
| 67 |
+
"loss": 0.0361,
|
| 68 |
"step": 250
|
| 69 |
},
|
| 70 |
{
|
| 71 |
+
"epoch": 0.1,
|
| 72 |
+
"learning_rate": 5.46e-07,
|
| 73 |
+
"loss": 0.0358,
|
| 74 |
"step": 275
|
| 75 |
},
|
| 76 |
{
|
| 77 |
+
"epoch": 0.11,
|
| 78 |
+
"learning_rate": 5.96e-07,
|
| 79 |
+
"loss": 0.0284,
|
| 80 |
"step": 300
|
| 81 |
},
|
| 82 |
{
|
| 83 |
+
"epoch": 0.11,
|
| 84 |
+
"learning_rate": 6.46e-07,
|
| 85 |
+
"loss": 0.0266,
|
| 86 |
"step": 325
|
| 87 |
},
|
| 88 |
{
|
| 89 |
+
"epoch": 0.12,
|
| 90 |
+
"learning_rate": 6.959999999999999e-07,
|
| 91 |
+
"loss": 0.0342,
|
| 92 |
"step": 350
|
| 93 |
},
|
| 94 |
{
|
| 95 |
+
"epoch": 0.13,
|
| 96 |
+
"learning_rate": 7.459999999999999e-07,
|
| 97 |
+
"loss": 0.0326,
|
| 98 |
"step": 375
|
| 99 |
},
|
| 100 |
{
|
| 101 |
+
"epoch": 0.14,
|
| 102 |
+
"learning_rate": 7.96e-07,
|
| 103 |
+
"loss": 0.032,
|
| 104 |
"step": 400
|
| 105 |
},
|
| 106 |
{
|
| 107 |
+
"epoch": 0.15,
|
| 108 |
+
"learning_rate": 8.459999999999999e-07,
|
| 109 |
+
"loss": 0.0244,
|
| 110 |
"step": 425
|
| 111 |
},
|
| 112 |
{
|
| 113 |
+
"epoch": 0.16,
|
| 114 |
+
"learning_rate": 8.96e-07,
|
| 115 |
+
"loss": 0.024,
|
| 116 |
"step": 450
|
| 117 |
},
|
| 118 |
{
|
| 119 |
+
"epoch": 0.17,
|
| 120 |
+
"learning_rate": 9.459999999999999e-07,
|
| 121 |
+
"loss": 0.0254,
|
| 122 |
"step": 475
|
| 123 |
},
|
| 124 |
{
|
| 125 |
+
"epoch": 0.18,
|
| 126 |
+
"learning_rate": 9.959999999999999e-07,
|
| 127 |
+
"loss": 0.0266,
|
| 128 |
"step": 500
|
| 129 |
},
|
| 130 |
{
|
| 131 |
+
"epoch": 0.18,
|
| 132 |
+
"learning_rate": 9.97578947368421e-07,
|
| 133 |
+
"loss": 0.0229,
|
| 134 |
"step": 525
|
| 135 |
},
|
| 136 |
{
|
| 137 |
+
"epoch": 0.19,
|
| 138 |
+
"learning_rate": 9.949473684210525e-07,
|
| 139 |
+
"loss": 0.0195,
|
| 140 |
"step": 550
|
| 141 |
},
|
| 142 |
{
|
| 143 |
+
"epoch": 0.2,
|
| 144 |
+
"learning_rate": 9.923157894736841e-07,
|
| 145 |
+
"loss": 0.0263,
|
| 146 |
"step": 575
|
| 147 |
},
|
| 148 |
{
|
| 149 |
+
"epoch": 0.21,
|
| 150 |
+
"learning_rate": 9.896842105263157e-07,
|
| 151 |
+
"loss": 0.0185,
|
| 152 |
"step": 600
|
| 153 |
},
|
| 154 |
{
|
| 155 |
+
"epoch": 0.22,
|
| 156 |
+
"learning_rate": 9.870526315789473e-07,
|
| 157 |
+
"loss": 0.0317,
|
| 158 |
"step": 625
|
| 159 |
},
|
| 160 |
{
|
| 161 |
+
"epoch": 0.23,
|
| 162 |
+
"learning_rate": 9.844210526315789e-07,
|
| 163 |
+
"loss": 0.0367,
|
| 164 |
"step": 650
|
| 165 |
},
|
| 166 |
{
|
| 167 |
+
"epoch": 0.24,
|
| 168 |
+
"learning_rate": 9.817894736842105e-07,
|
| 169 |
+
"loss": 0.0162,
|
| 170 |
"step": 675
|
| 171 |
},
|
| 172 |
{
|
| 173 |
+
"epoch": 0.25,
|
| 174 |
+
"learning_rate": 9.79157894736842e-07,
|
| 175 |
+
"loss": 0.019,
|
| 176 |
"step": 700
|
| 177 |
},
|
| 178 |
{
|
| 179 |
+
"epoch": 0.26,
|
| 180 |
+
"learning_rate": 9.765263157894736e-07,
|
| 181 |
+
"loss": 0.021,
|
| 182 |
"step": 725
|
| 183 |
},
|
| 184 |
{
|
| 185 |
+
"epoch": 0.26,
|
| 186 |
+
"learning_rate": 9.738947368421052e-07,
|
| 187 |
+
"loss": 0.0256,
|
| 188 |
"step": 750
|
| 189 |
},
|
| 190 |
{
|
| 191 |
+
"epoch": 0.27,
|
| 192 |
+
"learning_rate": 9.712631578947368e-07,
|
| 193 |
+
"loss": 0.039,
|
| 194 |
"step": 775
|
| 195 |
},
|
| 196 |
{
|
| 197 |
+
"epoch": 0.28,
|
| 198 |
+
"learning_rate": 9.686315789473684e-07,
|
| 199 |
+
"loss": 0.0312,
|
| 200 |
"step": 800
|
| 201 |
},
|
| 202 |
{
|
| 203 |
+
"epoch": 0.29,
|
| 204 |
+
"learning_rate": 9.66e-07,
|
| 205 |
+
"loss": 0.022,
|
| 206 |
"step": 825
|
| 207 |
},
|
| 208 |
{
|
| 209 |
+
"epoch": 0.3,
|
| 210 |
+
"learning_rate": 9.633684210526315e-07,
|
| 211 |
+
"loss": 0.0249,
|
| 212 |
"step": 850
|
| 213 |
},
|
| 214 |
{
|
| 215 |
+
"epoch": 0.31,
|
| 216 |
+
"learning_rate": 9.60736842105263e-07,
|
| 217 |
+
"loss": 0.0206,
|
| 218 |
"step": 875
|
| 219 |
},
|
| 220 |
{
|
| 221 |
+
"epoch": 0.32,
|
| 222 |
+
"learning_rate": 9.581052631578947e-07,
|
| 223 |
+
"loss": 0.0217,
|
| 224 |
"step": 900
|
| 225 |
},
|
| 226 |
{
|
| 227 |
+
"epoch": 0.33,
|
| 228 |
+
"learning_rate": 9.554736842105263e-07,
|
| 229 |
+
"loss": 0.0136,
|
| 230 |
"step": 925
|
| 231 |
},
|
| 232 |
{
|
| 233 |
+
"epoch": 0.33,
|
| 234 |
+
"learning_rate": 9.528421052631579e-07,
|
| 235 |
+
"loss": 0.0311,
|
| 236 |
"step": 950
|
| 237 |
},
|
| 238 |
{
|
| 239 |
+
"epoch": 0.34,
|
| 240 |
+
"learning_rate": 9.502105263157894e-07,
|
| 241 |
+
"loss": 0.0121,
|
| 242 |
"step": 975
|
| 243 |
},
|
| 244 |
{
|
| 245 |
+
"epoch": 0.35,
|
| 246 |
+
"learning_rate": 9.47578947368421e-07,
|
| 247 |
+
"loss": 0.0252,
|
| 248 |
"step": 1000
|
| 249 |
},
|
| 250 |
{
|
| 251 |
+
"epoch": 0.35,
|
| 252 |
+
"eval_loss": 0.07330603152513504,
|
| 253 |
+
"eval_runtime": 2282.5436,
|
| 254 |
+
"eval_samples_per_second": 1.585,
|
| 255 |
+
"eval_steps_per_second": 0.099,
|
| 256 |
+
"eval_wer": 4.482154385317717,
|
| 257 |
"step": 1000
|
| 258 |
},
|
| 259 |
{
|
| 260 |
+
"epoch": 0.36,
|
| 261 |
+
"learning_rate": 9.449473684210526e-07,
|
| 262 |
+
"loss": 0.0262,
|
| 263 |
"step": 1025
|
| 264 |
},
|
| 265 |
{
|
| 266 |
+
"epoch": 0.37,
|
| 267 |
+
"learning_rate": 9.423157894736842e-07,
|
| 268 |
+
"loss": 0.0135,
|
| 269 |
"step": 1050
|
| 270 |
},
|
| 271 |
{
|
| 272 |
+
"epoch": 0.38,
|
| 273 |
+
"learning_rate": 9.396842105263158e-07,
|
| 274 |
+
"loss": 0.0142,
|
| 275 |
"step": 1075
|
| 276 |
},
|
| 277 |
{
|
| 278 |
+
"epoch": 0.39,
|
| 279 |
+
"learning_rate": 9.370526315789473e-07,
|
| 280 |
+
"loss": 0.027,
|
| 281 |
"step": 1100
|
| 282 |
},
|
| 283 |
{
|
| 284 |
+
"epoch": 0.4,
|
| 285 |
+
"learning_rate": 9.344210526315789e-07,
|
| 286 |
+
"loss": 0.0191,
|
| 287 |
"step": 1125
|
| 288 |
},
|
| 289 |
{
|
| 290 |
+
"epoch": 0.41,
|
| 291 |
+
"learning_rate": 9.317894736842105e-07,
|
| 292 |
+
"loss": 0.016,
|
| 293 |
"step": 1150
|
| 294 |
},
|
| 295 |
{
|
| 296 |
+
"epoch": 0.41,
|
| 297 |
+
"learning_rate": 9.29157894736842e-07,
|
| 298 |
+
"loss": 0.0219,
|
| 299 |
"step": 1175
|
| 300 |
},
|
| 301 |
{
|
| 302 |
+
"epoch": 0.42,
|
| 303 |
+
"learning_rate": 9.265263157894736e-07,
|
| 304 |
+
"loss": 0.0339,
|
| 305 |
"step": 1200
|
| 306 |
},
|
| 307 |
{
|
| 308 |
+
"epoch": 0.43,
|
| 309 |
+
"learning_rate": 9.238947368421051e-07,
|
| 310 |
+
"loss": 0.0237,
|
| 311 |
"step": 1225
|
| 312 |
},
|
| 313 |
{
|
| 314 |
+
"epoch": 0.44,
|
| 315 |
+
"learning_rate": 9.212631578947368e-07,
|
| 316 |
+
"loss": 0.0142,
|
| 317 |
"step": 1250
|
| 318 |
},
|
| 319 |
{
|
| 320 |
+
"epoch": 0.45,
|
| 321 |
+
"learning_rate": 9.186315789473684e-07,
|
| 322 |
+
"loss": 0.0212,
|
| 323 |
"step": 1275
|
| 324 |
},
|
| 325 |
{
|
| 326 |
+
"epoch": 0.46,
|
| 327 |
+
"learning_rate": 9.16e-07,
|
| 328 |
+
"loss": 0.015,
|
| 329 |
"step": 1300
|
| 330 |
},
|
| 331 |
{
|
| 332 |
+
"epoch": 0.47,
|
| 333 |
+
"learning_rate": 9.133684210526316e-07,
|
| 334 |
+
"loss": 0.0221,
|
| 335 |
"step": 1325
|
| 336 |
},
|
| 337 |
{
|
| 338 |
+
"epoch": 0.48,
|
| 339 |
+
"learning_rate": 9.107368421052632e-07,
|
| 340 |
+
"loss": 0.0195,
|
| 341 |
"step": 1350
|
| 342 |
},
|
| 343 |
{
|
| 344 |
+
"epoch": 0.48,
|
| 345 |
+
"learning_rate": 9.081052631578947e-07,
|
| 346 |
+
"loss": 0.0279,
|
| 347 |
"step": 1375
|
| 348 |
},
|
| 349 |
{
|
| 350 |
+
"epoch": 0.49,
|
| 351 |
+
"learning_rate": 9.054736842105263e-07,
|
| 352 |
+
"loss": 0.02,
|
| 353 |
"step": 1400
|
| 354 |
},
|
| 355 |
{
|
| 356 |
+
"epoch": 0.5,
|
| 357 |
+
"learning_rate": 9.028421052631578e-07,
|
| 358 |
+
"loss": 0.0216,
|
| 359 |
"step": 1425
|
| 360 |
},
|
| 361 |
{
|
| 362 |
+
"epoch": 0.51,
|
| 363 |
+
"learning_rate": 9.002105263157894e-07,
|
| 364 |
+
"loss": 0.0272,
|
| 365 |
"step": 1450
|
| 366 |
},
|
| 367 |
{
|
| 368 |
+
"epoch": 0.52,
|
| 369 |
+
"learning_rate": 8.97578947368421e-07,
|
| 370 |
+
"loss": 0.0165,
|
| 371 |
"step": 1475
|
| 372 |
},
|
| 373 |
{
|
| 374 |
+
"epoch": 0.53,
|
| 375 |
+
"learning_rate": 8.949473684210525e-07,
|
| 376 |
+
"loss": 0.0328,
|
| 377 |
"step": 1500
|
| 378 |
},
|
| 379 |
{
|
| 380 |
+
"epoch": 0.54,
|
| 381 |
+
"learning_rate": 8.923157894736841e-07,
|
| 382 |
+
"loss": 0.0259,
|
| 383 |
"step": 1525
|
| 384 |
},
|
| 385 |
{
|
| 386 |
+
"epoch": 0.55,
|
| 387 |
+
"learning_rate": 8.896842105263157e-07,
|
| 388 |
+
"loss": 0.0176,
|
| 389 |
"step": 1550
|
| 390 |
},
|
| 391 |
{
|
| 392 |
+
"epoch": 0.55,
|
| 393 |
+
"learning_rate": 8.870526315789473e-07,
|
| 394 |
+
"loss": 0.0194,
|
| 395 |
"step": 1575
|
| 396 |
},
|
| 397 |
{
|
| 398 |
+
"epoch": 0.56,
|
| 399 |
+
"learning_rate": 8.84421052631579e-07,
|
| 400 |
+
"loss": 0.0172,
|
| 401 |
"step": 1600
|
| 402 |
},
|
| 403 |
{
|
| 404 |
+
"epoch": 0.57,
|
| 405 |
+
"learning_rate": 8.817894736842106e-07,
|
| 406 |
+
"loss": 0.0177,
|
| 407 |
"step": 1625
|
| 408 |
},
|
| 409 |
{
|
| 410 |
+
"epoch": 0.58,
|
| 411 |
+
"learning_rate": 8.791578947368421e-07,
|
| 412 |
+
"loss": 0.0293,
|
| 413 |
"step": 1650
|
| 414 |
},
|
| 415 |
{
|
| 416 |
+
"epoch": 0.59,
|
| 417 |
+
"learning_rate": 8.765263157894736e-07,
|
| 418 |
+
"loss": 0.0141,
|
| 419 |
"step": 1675
|
| 420 |
},
|
| 421 |
{
|
| 422 |
+
"epoch": 0.6,
|
| 423 |
+
"learning_rate": 8.738947368421052e-07,
|
| 424 |
+
"loss": 0.025,
|
| 425 |
"step": 1700
|
| 426 |
},
|
| 427 |
{
|
| 428 |
+
"epoch": 0.61,
|
| 429 |
+
"learning_rate": 8.712631578947368e-07,
|
| 430 |
+
"loss": 0.0175,
|
| 431 |
"step": 1725
|
| 432 |
},
|
| 433 |
{
|
| 434 |
+
"epoch": 0.62,
|
| 435 |
+
"learning_rate": 8.686315789473684e-07,
|
| 436 |
+
"loss": 0.0217,
|
| 437 |
"step": 1750
|
| 438 |
},
|
| 439 |
{
|
| 440 |
+
"epoch": 0.63,
|
| 441 |
+
"learning_rate": 8.659999999999999e-07,
|
| 442 |
+
"loss": 0.023,
|
| 443 |
"step": 1775
|
| 444 |
},
|
| 445 |
{
|
| 446 |
+
"epoch": 0.63,
|
| 447 |
+
"learning_rate": 8.633684210526315e-07,
|
| 448 |
+
"loss": 0.0185,
|
| 449 |
"step": 1800
|
| 450 |
},
|
| 451 |
{
|
| 452 |
+
"epoch": 0.64,
|
| 453 |
+
"learning_rate": 8.607368421052631e-07,
|
| 454 |
+
"loss": 0.0255,
|
| 455 |
"step": 1825
|
| 456 |
},
|
| 457 |
{
|
| 458 |
+
"epoch": 0.65,
|
| 459 |
+
"learning_rate": 8.581052631578947e-07,
|
| 460 |
+
"loss": 0.0155,
|
| 461 |
"step": 1850
|
| 462 |
},
|
| 463 |
{
|
| 464 |
+
"epoch": 0.66,
|
| 465 |
+
"learning_rate": 8.554736842105263e-07,
|
| 466 |
+
"loss": 0.026,
|
| 467 |
"step": 1875
|
| 468 |
},
|
| 469 |
{
|
| 470 |
+
"epoch": 0.67,
|
| 471 |
+
"learning_rate": 8.528421052631579e-07,
|
| 472 |
+
"loss": 0.0366,
|
| 473 |
"step": 1900
|
| 474 |
},
|
| 475 |
{
|
| 476 |
+
"epoch": 0.68,
|
| 477 |
+
"learning_rate": 8.502105263157894e-07,
|
| 478 |
+
"loss": 0.0229,
|
| 479 |
"step": 1925
|
| 480 |
},
|
| 481 |
{
|
| 482 |
+
"epoch": 0.69,
|
| 483 |
+
"learning_rate": 8.47578947368421e-07,
|
| 484 |
+
"loss": 0.0241,
|
| 485 |
"step": 1950
|
| 486 |
},
|
| 487 |
{
|
| 488 |
+
"epoch": 0.7,
|
| 489 |
+
"learning_rate": 8.449473684210526e-07,
|
| 490 |
+
"loss": 0.0209,
|
| 491 |
"step": 1975
|
| 492 |
},
|
| 493 |
{
|
| 494 |
+
"epoch": 0.7,
|
| 495 |
+
"learning_rate": 8.423157894736842e-07,
|
| 496 |
+
"loss": 0.0222,
|
| 497 |
"step": 2000
|
| 498 |
},
|
| 499 |
{
|
| 500 |
+
"epoch": 0.7,
|
| 501 |
+
"eval_loss": 0.0697573572397232,
|
| 502 |
+
"eval_runtime": 2277.0169,
|
| 503 |
+
"eval_samples_per_second": 1.589,
|
| 504 |
+
"eval_steps_per_second": 0.1,
|
| 505 |
+
"eval_wer": 4.3392050170616985,
|
| 506 |
"step": 2000
|
| 507 |
},
|
| 508 |
{
|
| 509 |
+
"epoch": 0.71,
|
| 510 |
+
"learning_rate": 8.396842105263158e-07,
|
| 511 |
+
"loss": 0.0333,
|
| 512 |
"step": 2025
|
| 513 |
},
|
| 514 |
{
|
| 515 |
+
"epoch": 0.72,
|
| 516 |
+
"learning_rate": 8.370526315789473e-07,
|
| 517 |
+
"loss": 0.0262,
|
| 518 |
"step": 2050
|
| 519 |
},
|
| 520 |
{
|
| 521 |
+
"epoch": 0.73,
|
| 522 |
+
"learning_rate": 8.344210526315789e-07,
|
| 523 |
+
"loss": 0.0371,
|
| 524 |
"step": 2075
|
| 525 |
},
|
| 526 |
{
|
| 527 |
+
"epoch": 0.74,
|
| 528 |
+
"learning_rate": 8.318947368421052e-07,
|
| 529 |
+
"loss": 0.019,
|
| 530 |
"step": 2100
|
| 531 |
},
|
| 532 |
{
|
| 533 |
+
"epoch": 0.75,
|
| 534 |
+
"learning_rate": 8.292631578947368e-07,
|
| 535 |
+
"loss": 0.0575,
|
| 536 |
"step": 2125
|
| 537 |
},
|
| 538 |
{
|
| 539 |
+
"epoch": 0.76,
|
| 540 |
+
"learning_rate": 8.266315789473684e-07,
|
| 541 |
+
"loss": 0.0192,
|
| 542 |
"step": 2150
|
| 543 |
},
|
| 544 |
{
|
| 545 |
+
"epoch": 0.77,
|
| 546 |
+
"learning_rate": 8.24e-07,
|
| 547 |
+
"loss": 0.0296,
|
| 548 |
"step": 2175
|
| 549 |
},
|
| 550 |
{
|
| 551 |
+
"epoch": 0.77,
|
| 552 |
+
"learning_rate": 8.213684210526314e-07,
|
| 553 |
+
"loss": 0.0234,
|
| 554 |
"step": 2200
|
| 555 |
},
|
| 556 |
{
|
| 557 |
+
"epoch": 0.78,
|
| 558 |
+
"learning_rate": 8.187368421052631e-07,
|
| 559 |
+
"loss": 0.034,
|
| 560 |
"step": 2225
|
| 561 |
},
|
| 562 |
{
|
| 563 |
+
"epoch": 0.79,
|
| 564 |
+
"learning_rate": 8.161052631578947e-07,
|
| 565 |
+
"loss": 0.0389,
|
| 566 |
"step": 2250
|
| 567 |
},
|
| 568 |
{
|
| 569 |
+
"epoch": 0.8,
|
| 570 |
+
"learning_rate": 8.134736842105263e-07,
|
| 571 |
+
"loss": 0.039,
|
| 572 |
"step": 2275
|
| 573 |
},
|
| 574 |
{
|
| 575 |
+
"epoch": 0.81,
|
| 576 |
+
"learning_rate": 8.108421052631579e-07,
|
| 577 |
+
"loss": 0.0264,
|
| 578 |
"step": 2300
|
| 579 |
},
|
| 580 |
{
|
| 581 |
+
"epoch": 0.82,
|
| 582 |
+
"learning_rate": 8.082105263157895e-07,
|
| 583 |
+
"loss": 0.0225,
|
| 584 |
"step": 2325
|
| 585 |
},
|
| 586 |
{
|
| 587 |
+
"epoch": 0.83,
|
| 588 |
+
"learning_rate": 8.05578947368421e-07,
|
| 589 |
+
"loss": 0.0242,
|
| 590 |
"step": 2350
|
| 591 |
},
|
| 592 |
{
|
| 593 |
+
"epoch": 0.84,
|
| 594 |
+
"learning_rate": 8.029473684210526e-07,
|
| 595 |
+
"loss": 0.0416,
|
| 596 |
"step": 2375
|
| 597 |
},
|
| 598 |
{
|
| 599 |
+
"epoch": 0.85,
|
| 600 |
+
"learning_rate": 8.003157894736842e-07,
|
| 601 |
+
"loss": 0.0258,
|
| 602 |
"step": 2400
|
| 603 |
},
|
| 604 |
{
|
| 605 |
+
"epoch": 0.85,
|
| 606 |
+
"learning_rate": 7.976842105263158e-07,
|
| 607 |
+
"loss": 0.0242,
|
| 608 |
"step": 2425
|
| 609 |
},
|
| 610 |
{
|
| 611 |
+
"epoch": 0.86,
|
| 612 |
+
"learning_rate": 7.950526315789473e-07,
|
| 613 |
+
"loss": 0.0241,
|
| 614 |
"step": 2450
|
| 615 |
},
|
| 616 |
{
|
| 617 |
+
"epoch": 0.87,
|
| 618 |
+
"learning_rate": 7.924210526315788e-07,
|
| 619 |
+
"loss": 0.0362,
|
| 620 |
"step": 2475
|
| 621 |
},
|
| 622 |
{
|
| 623 |
+
"epoch": 0.88,
|
| 624 |
+
"learning_rate": 7.897894736842104e-07,
|
| 625 |
+
"loss": 0.0265,
|
| 626 |
"step": 2500
|
| 627 |
},
|
| 628 |
{
|
| 629 |
+
"epoch": 0.89,
|
| 630 |
+
"learning_rate": 7.87157894736842e-07,
|
| 631 |
+
"loss": 0.0421,
|
| 632 |
"step": 2525
|
| 633 |
},
|
| 634 |
{
|
| 635 |
+
"epoch": 0.9,
|
| 636 |
+
"learning_rate": 7.845263157894737e-07,
|
| 637 |
+
"loss": 0.0297,
|
| 638 |
"step": 2550
|
| 639 |
},
|
| 640 |
{
|
| 641 |
+
"epoch": 0.91,
|
| 642 |
+
"learning_rate": 7.818947368421053e-07,
|
| 643 |
+
"loss": 0.0396,
|
| 644 |
"step": 2575
|
| 645 |
},
|
| 646 |
{
|
| 647 |
+
"epoch": 0.92,
|
| 648 |
+
"learning_rate": 7.792631578947369e-07,
|
| 649 |
+
"loss": 0.0432,
|
| 650 |
"step": 2600
|
| 651 |
},
|
| 652 |
{
|
| 653 |
+
"epoch": 0.92,
|
| 654 |
+
"learning_rate": 7.766315789473684e-07,
|
| 655 |
+
"loss": 0.0514,
|
| 656 |
"step": 2625
|
| 657 |
},
|
| 658 |
{
|
| 659 |
+
"epoch": 0.93,
|
| 660 |
+
"learning_rate": 7.74e-07,
|
| 661 |
+
"loss": 0.0371,
|
| 662 |
"step": 2650
|
| 663 |
},
|
| 664 |
{
|
| 665 |
+
"epoch": 0.94,
|
| 666 |
+
"learning_rate": 7.713684210526316e-07,
|
| 667 |
+
"loss": 0.0578,
|
| 668 |
"step": 2675
|
| 669 |
},
|
| 670 |
{
|
| 671 |
+
"epoch": 0.95,
|
| 672 |
+
"learning_rate": 7.687368421052631e-07,
|
| 673 |
+
"loss": 0.0414,
|
| 674 |
"step": 2700
|
| 675 |
},
|
| 676 |
{
|
| 677 |
+
"epoch": 0.96,
|
| 678 |
+
"learning_rate": 7.661052631578947e-07,
|
| 679 |
+
"loss": 0.0499,
|
| 680 |
"step": 2725
|
| 681 |
},
|
| 682 |
{
|
| 683 |
+
"epoch": 0.97,
|
| 684 |
+
"learning_rate": 7.634736842105262e-07,
|
| 685 |
+
"loss": 0.0531,
|
| 686 |
"step": 2750
|
| 687 |
},
|
| 688 |
{
|
| 689 |
+
"epoch": 0.98,
|
| 690 |
+
"learning_rate": 7.608421052631578e-07,
|
| 691 |
+
"loss": 0.0599,
|
| 692 |
"step": 2775
|
| 693 |
},
|
| 694 |
{
|
| 695 |
+
"epoch": 0.99,
|
| 696 |
+
"learning_rate": 7.582105263157894e-07,
|
| 697 |
+
"loss": 0.0482,
|
| 698 |
"step": 2800
|
| 699 |
},
|
| 700 |
{
|
| 701 |
+
"epoch": 1.0,
|
| 702 |
+
"learning_rate": 7.55578947368421e-07,
|
| 703 |
+
"loss": 0.0552,
|
| 704 |
"step": 2825
|
| 705 |
},
|
| 706 |
{
|
| 707 |
+
"epoch": 1.0,
|
| 708 |
+
"learning_rate": 7.529473684210526e-07,
|
| 709 |
+
"loss": 0.0339,
|
| 710 |
"step": 2850
|
| 711 |
},
|
| 712 |
{
|
| 713 |
+
"epoch": 1.01,
|
| 714 |
+
"learning_rate": 7.503157894736842e-07,
|
| 715 |
+
"loss": 0.0302,
|
| 716 |
"step": 2875
|
| 717 |
},
|
| 718 |
{
|
| 719 |
+
"epoch": 1.02,
|
| 720 |
+
"learning_rate": 7.476842105263158e-07,
|
| 721 |
+
"loss": 0.0171,
|
| 722 |
"step": 2900
|
| 723 |
},
|
| 724 |
{
|
| 725 |
+
"epoch": 1.03,
|
| 726 |
+
"learning_rate": 7.450526315789474e-07,
|
| 727 |
+
"loss": 0.0133,
|
| 728 |
"step": 2925
|
| 729 |
},
|
| 730 |
{
|
| 731 |
+
"epoch": 1.04,
|
| 732 |
+
"learning_rate": 7.424210526315789e-07,
|
| 733 |
+
"loss": 0.0234,
|
| 734 |
"step": 2950
|
| 735 |
},
|
| 736 |
{
|
| 737 |
+
"epoch": 1.05,
|
| 738 |
+
"learning_rate": 7.397894736842105e-07,
|
| 739 |
+
"loss": 0.0137,
|
| 740 |
"step": 2975
|
| 741 |
},
|
| 742 |
{
|
| 743 |
+
"epoch": 1.06,
|
| 744 |
+
"learning_rate": 7.371578947368421e-07,
|
| 745 |
+
"loss": 0.0166,
|
| 746 |
"step": 3000
|
| 747 |
},
|
| 748 |
{
|
| 749 |
+
"epoch": 1.06,
|
| 750 |
+
"eval_loss": 0.06964117288589478,
|
| 751 |
+
"eval_runtime": 2279.778,
|
| 752 |
+
"eval_samples_per_second": 1.587,
|
| 753 |
+
"eval_steps_per_second": 0.1,
|
| 754 |
+
"eval_wer": 4.237757078299364,
|
| 755 |
"step": 3000
|
| 756 |
},
|
| 757 |
{
|
| 758 |
+
"epoch": 1.07,
|
| 759 |
+
"learning_rate": 7.345263157894736e-07,
|
| 760 |
+
"loss": 0.0189,
|
| 761 |
"step": 3025
|
| 762 |
},
|
| 763 |
{
|
| 764 |
+
"epoch": 1.07,
|
| 765 |
+
"learning_rate": 7.318947368421052e-07,
|
| 766 |
+
"loss": 0.0154,
|
| 767 |
"step": 3050
|
| 768 |
},
|
| 769 |
{
|
| 770 |
+
"epoch": 1.08,
|
| 771 |
+
"learning_rate": 7.292631578947368e-07,
|
| 772 |
+
"loss": 0.0151,
|
| 773 |
"step": 3075
|
| 774 |
},
|
| 775 |
{
|
| 776 |
+
"epoch": 1.09,
|
| 777 |
+
"learning_rate": 7.266315789473684e-07,
|
| 778 |
+
"loss": 0.0181,
|
| 779 |
"step": 3100
|
| 780 |
},
|
| 781 |
{
|
| 782 |
+
"epoch": 1.1,
|
| 783 |
+
"learning_rate": 7.24e-07,
|
| 784 |
+
"loss": 0.0318,
|
| 785 |
"step": 3125
|
| 786 |
},
|
| 787 |
{
|
| 788 |
+
"epoch": 1.11,
|
| 789 |
+
"learning_rate": 7.213684210526316e-07,
|
| 790 |
+
"loss": 0.0292,
|
| 791 |
"step": 3150
|
| 792 |
},
|
| 793 |
{
|
| 794 |
+
"epoch": 1.12,
|
| 795 |
+
"learning_rate": 7.187368421052631e-07,
|
| 796 |
+
"loss": 0.0204,
|
| 797 |
"step": 3175
|
| 798 |
},
|
| 799 |
{
|
| 800 |
+
"epoch": 1.13,
|
| 801 |
+
"learning_rate": 7.161052631578946e-07,
|
| 802 |
+
"loss": 0.017,
|
| 803 |
"step": 3200
|
| 804 |
},
|
| 805 |
{
|
| 806 |
+
"epoch": 1.14,
|
| 807 |
+
"learning_rate": 7.134736842105263e-07,
|
| 808 |
+
"loss": 0.0086,
|
| 809 |
"step": 3225
|
| 810 |
},
|
| 811 |
{
|
| 812 |
+
"epoch": 1.14,
|
| 813 |
+
"learning_rate": 7.108421052631579e-07,
|
| 814 |
+
"loss": 0.0218,
|
| 815 |
"step": 3250
|
| 816 |
},
|
| 817 |
{
|
| 818 |
+
"epoch": 1.15,
|
| 819 |
+
"learning_rate": 7.082105263157895e-07,
|
| 820 |
+
"loss": 0.0149,
|
| 821 |
"step": 3275
|
| 822 |
},
|
| 823 |
{
|
| 824 |
+
"epoch": 1.16,
|
| 825 |
+
"learning_rate": 7.05578947368421e-07,
|
| 826 |
+
"loss": 0.0201,
|
| 827 |
"step": 3300
|
| 828 |
},
|
| 829 |
{
|
| 830 |
+
"epoch": 1.17,
|
| 831 |
+
"learning_rate": 7.029473684210526e-07,
|
| 832 |
+
"loss": 0.0159,
|
| 833 |
"step": 3325
|
| 834 |
},
|
| 835 |
{
|
| 836 |
+
"epoch": 1.18,
|
| 837 |
+
"learning_rate": 7.003157894736842e-07,
|
| 838 |
+
"loss": 0.0133,
|
| 839 |
"step": 3350
|
| 840 |
},
|
| 841 |
{
|
| 842 |
+
"epoch": 1.19,
|
| 843 |
+
"learning_rate": 6.976842105263158e-07,
|
| 844 |
+
"loss": 0.031,
|
| 845 |
"step": 3375
|
| 846 |
},
|
| 847 |
{
|
| 848 |
+
"epoch": 1.2,
|
| 849 |
+
"learning_rate": 6.950526315789474e-07,
|
| 850 |
+
"loss": 0.0192,
|
| 851 |
"step": 3400
|
| 852 |
},
|
| 853 |
{
|
| 854 |
+
"epoch": 1.21,
|
| 855 |
+
"learning_rate": 6.92421052631579e-07,
|
| 856 |
+
"loss": 0.0174,
|
| 857 |
"step": 3425
|
| 858 |
},
|
| 859 |
{
|
| 860 |
+
"epoch": 1.22,
|
| 861 |
+
"learning_rate": 6.897894736842104e-07,
|
| 862 |
+
"loss": 0.0145,
|
| 863 |
"step": 3450
|
| 864 |
},
|
| 865 |
{
|
| 866 |
+
"epoch": 1.22,
|
| 867 |
+
"learning_rate": 6.87157894736842e-07,
|
| 868 |
+
"loss": 0.0171,
|
| 869 |
"step": 3475
|
| 870 |
},
|
| 871 |
{
|
| 872 |
+
"epoch": 1.23,
|
| 873 |
+
"learning_rate": 6.845263157894736e-07,
|
| 874 |
+
"loss": 0.0182,
|
| 875 |
"step": 3500
|
| 876 |
},
|
| 877 |
{
|
| 878 |
+
"epoch": 1.24,
|
| 879 |
+
"learning_rate": 6.818947368421052e-07,
|
| 880 |
+
"loss": 0.021,
|
| 881 |
"step": 3525
|
| 882 |
},
|
| 883 |
{
|
| 884 |
+
"epoch": 1.25,
|
| 885 |
+
"learning_rate": 6.792631578947368e-07,
|
| 886 |
+
"loss": 0.0182,
|
| 887 |
"step": 3550
|
| 888 |
},
|
| 889 |
{
|
| 890 |
+
"epoch": 1.26,
|
| 891 |
+
"learning_rate": 6.766315789473684e-07,
|
| 892 |
+
"loss": 0.0256,
|
| 893 |
"step": 3575
|
| 894 |
},
|
| 895 |
{
|
| 896 |
+
"epoch": 1.27,
|
| 897 |
+
"learning_rate": 6.74e-07,
|
| 898 |
+
"loss": 0.0192,
|
| 899 |
"step": 3600
|
| 900 |
},
|
| 901 |
{
|
| 902 |
+
"epoch": 1.28,
|
| 903 |
+
"learning_rate": 6.713684210526316e-07,
|
| 904 |
+
"loss": 0.0252,
|
| 905 |
"step": 3625
|
| 906 |
},
|
| 907 |
{
|
| 908 |
+
"epoch": 1.29,
|
| 909 |
+
"learning_rate": 6.687368421052632e-07,
|
| 910 |
+
"loss": 0.0144,
|
| 911 |
"step": 3650
|
| 912 |
},
|
| 913 |
{
|
| 914 |
+
"epoch": 1.29,
|
| 915 |
+
"learning_rate": 6.661052631578948e-07,
|
| 916 |
+
"loss": 0.0233,
|
| 917 |
"step": 3675
|
| 918 |
},
|
| 919 |
{
|
| 920 |
+
"epoch": 1.3,
|
| 921 |
+
"learning_rate": 6.634736842105262e-07,
|
| 922 |
+
"loss": 0.015,
|
| 923 |
"step": 3700
|
| 924 |
},
|
| 925 |
{
|
| 926 |
+
"epoch": 1.31,
|
| 927 |
+
"learning_rate": 6.608421052631578e-07,
|
| 928 |
+
"loss": 0.0279,
|
| 929 |
"step": 3725
|
| 930 |
},
|
| 931 |
{
|
| 932 |
+
"epoch": 1.32,
|
| 933 |
+
"learning_rate": 6.582105263157894e-07,
|
| 934 |
+
"loss": 0.0213,
|
| 935 |
"step": 3750
|
| 936 |
},
|
| 937 |
{
|
| 938 |
+
"epoch": 1.33,
|
| 939 |
+
"learning_rate": 6.55578947368421e-07,
|
| 940 |
+
"loss": 0.021,
|
| 941 |
"step": 3775
|
| 942 |
},
|
| 943 |
{
|
| 944 |
+
"epoch": 1.34,
|
| 945 |
+
"learning_rate": 6.529473684210526e-07,
|
| 946 |
+
"loss": 0.026,
|
| 947 |
"step": 3800
|
| 948 |
},
|
| 949 |
{
|
| 950 |
+
"epoch": 1.35,
|
| 951 |
+
"learning_rate": 6.503157894736842e-07,
|
| 952 |
+
"loss": 0.0366,
|
| 953 |
"step": 3825
|
| 954 |
},
|
| 955 |
{
|
| 956 |
+
"epoch": 1.36,
|
| 957 |
+
"learning_rate": 6.476842105263157e-07,
|
| 958 |
+
"loss": 0.0273,
|
| 959 |
"step": 3850
|
| 960 |
},
|
| 961 |
{
|
| 962 |
+
"epoch": 1.36,
|
| 963 |
+
"learning_rate": 6.450526315789473e-07,
|
| 964 |
+
"loss": 0.0134,
|
| 965 |
"step": 3875
|
| 966 |
},
|
| 967 |
{
|
| 968 |
+
"epoch": 1.37,
|
| 969 |
+
"learning_rate": 6.424210526315789e-07,
|
| 970 |
+
"loss": 0.0193,
|
| 971 |
"step": 3900
|
| 972 |
},
|
| 973 |
{
|
| 974 |
+
"epoch": 1.38,
|
| 975 |
+
"learning_rate": 6.397894736842106e-07,
|
| 976 |
+
"loss": 0.0165,
|
| 977 |
"step": 3925
|
| 978 |
},
|
| 979 |
{
|
| 980 |
+
"epoch": 1.39,
|
| 981 |
+
"learning_rate": 6.371578947368421e-07,
|
| 982 |
+
"loss": 0.0208,
|
| 983 |
"step": 3950
|
| 984 |
},
|
| 985 |
{
|
| 986 |
+
"epoch": 1.4,
|
| 987 |
+
"learning_rate": 6.345263157894736e-07,
|
| 988 |
+
"loss": 0.0167,
|
| 989 |
"step": 3975
|
| 990 |
},
|
| 991 |
{
|
| 992 |
+
"epoch": 1.41,
|
| 993 |
+
"learning_rate": 6.318947368421052e-07,
|
| 994 |
+
"loss": 0.0117,
|
| 995 |
"step": 4000
|
| 996 |
},
|
| 997 |
{
|
| 998 |
+
"epoch": 1.41,
|
| 999 |
+
"eval_loss": 0.0679246336221695,
|
| 1000 |
+
"eval_runtime": 2278.2087,
|
| 1001 |
+
"eval_samples_per_second": 1.588,
|
| 1002 |
+
"eval_steps_per_second": 0.1,
|
| 1003 |
+
"eval_wer": 4.080973900212118,
|
| 1004 |
"step": 4000
|
| 1005 |
},
|
| 1006 |
{
|
| 1007 |
+
"epoch": 1.42,
|
| 1008 |
+
"learning_rate": 6.292631578947368e-07,
|
| 1009 |
+
"loss": 0.0349,
|
| 1010 |
"step": 4025
|
| 1011 |
},
|
| 1012 |
{
|
| 1013 |
+
"epoch": 1.43,
|
| 1014 |
+
"learning_rate": 6.266315789473684e-07,
|
| 1015 |
+
"loss": 0.0173,
|
| 1016 |
"step": 4050
|
| 1017 |
},
|
| 1018 |
{
|
| 1019 |
+
"epoch": 1.44,
|
| 1020 |
+
"learning_rate": 6.24e-07,
|
| 1021 |
+
"loss": 0.0225,
|
| 1022 |
"step": 4075
|
| 1023 |
},
|
| 1024 |
{
|
| 1025 |
+
"epoch": 1.44,
|
| 1026 |
+
"learning_rate": 6.213684210526316e-07,
|
| 1027 |
+
"loss": 0.0156,
|
| 1028 |
"step": 4100
|
| 1029 |
},
|
| 1030 |
{
|
| 1031 |
+
"epoch": 1.45,
|
| 1032 |
+
"learning_rate": 6.187368421052631e-07,
|
| 1033 |
+
"loss": 0.0231,
|
| 1034 |
"step": 4125
|
| 1035 |
},
|
| 1036 |
{
|
| 1037 |
+
"epoch": 1.46,
|
| 1038 |
+
"learning_rate": 6.161052631578947e-07,
|
| 1039 |
+
"loss": 0.0152,
|
| 1040 |
"step": 4150
|
| 1041 |
},
|
| 1042 |
{
|
| 1043 |
+
"epoch": 1.47,
|
| 1044 |
+
"learning_rate": 6.134736842105263e-07,
|
| 1045 |
+
"loss": 0.0209,
|
| 1046 |
"step": 4175
|
| 1047 |
},
|
| 1048 |
{
|
| 1049 |
+
"epoch": 1.48,
|
| 1050 |
+
"learning_rate": 6.108421052631578e-07,
|
| 1051 |
+
"loss": 0.0175,
|
| 1052 |
"step": 4200
|
| 1053 |
},
|
| 1054 |
{
|
| 1055 |
+
"epoch": 1.49,
|
| 1056 |
+
"learning_rate": 6.082105263157894e-07,
|
| 1057 |
+
"loss": 0.0169,
|
| 1058 |
"step": 4225
|
| 1059 |
},
|
| 1060 |
{
|
| 1061 |
+
"epoch": 1.5,
|
| 1062 |
+
"learning_rate": 6.05578947368421e-07,
|
| 1063 |
+
"loss": 0.026,
|
| 1064 |
"step": 4250
|
| 1065 |
},
|
| 1066 |
{
|
| 1067 |
+
"epoch": 1.51,
|
| 1068 |
+
"learning_rate": 6.029473684210526e-07,
|
| 1069 |
+
"loss": 0.0178,
|
| 1070 |
"step": 4275
|
| 1071 |
},
|
| 1072 |
{
|
| 1073 |
+
"epoch": 1.51,
|
| 1074 |
+
"learning_rate": 6.003157894736842e-07,
|
| 1075 |
+
"loss": 0.0084,
|
| 1076 |
"step": 4300
|
| 1077 |
},
|
| 1078 |
{
|
| 1079 |
+
"epoch": 1.52,
|
| 1080 |
+
"learning_rate": 5.976842105263158e-07,
|
| 1081 |
+
"loss": 0.0161,
|
| 1082 |
"step": 4325
|
| 1083 |
},
|
| 1084 |
{
|
| 1085 |
+
"epoch": 1.53,
|
| 1086 |
+
"learning_rate": 5.950526315789474e-07,
|
| 1087 |
+
"loss": 0.0232,
|
| 1088 |
"step": 4350
|
| 1089 |
},
|
| 1090 |
{
|
| 1091 |
+
"epoch": 1.54,
|
| 1092 |
+
"learning_rate": 5.92421052631579e-07,
|
| 1093 |
+
"loss": 0.0153,
|
| 1094 |
"step": 4375
|
| 1095 |
},
|
| 1096 |
{
|
| 1097 |
+
"epoch": 1.55,
|
| 1098 |
+
"learning_rate": 5.897894736842105e-07,
|
| 1099 |
+
"loss": 0.0213,
|
| 1100 |
"step": 4400
|
| 1101 |
},
|
| 1102 |
{
|
| 1103 |
+
"epoch": 1.56,
|
| 1104 |
+
"learning_rate": 5.871578947368421e-07,
|
| 1105 |
+
"loss": 0.0251,
|
| 1106 |
"step": 4425
|
| 1107 |
},
|
| 1108 |
{
|
| 1109 |
+
"epoch": 1.57,
|
| 1110 |
+
"learning_rate": 5.845263157894736e-07,
|
| 1111 |
+
"loss": 0.0121,
|
| 1112 |
"step": 4450
|
| 1113 |
},
|
| 1114 |
{
|
| 1115 |
+
"epoch": 1.58,
|
| 1116 |
+
"learning_rate": 5.818947368421052e-07,
|
| 1117 |
+
"loss": 0.0202,
|
| 1118 |
"step": 4475
|
| 1119 |
},
|
| 1120 |
{
|
| 1121 |
+
"epoch": 1.59,
|
| 1122 |
+
"learning_rate": 5.792631578947368e-07,
|
| 1123 |
+
"loss": 0.0209,
|
| 1124 |
"step": 4500
|
| 1125 |
},
|
| 1126 |
{
|
| 1127 |
+
"epoch": 1.59,
|
| 1128 |
+
"learning_rate": 5.766315789473683e-07,
|
| 1129 |
+
"loss": 0.0211,
|
| 1130 |
"step": 4525
|
| 1131 |
},
|
| 1132 |
{
|
| 1133 |
+
"epoch": 1.6,
|
| 1134 |
+
"learning_rate": 5.739999999999999e-07,
|
| 1135 |
+
"loss": 0.0156,
|
| 1136 |
"step": 4550
|
| 1137 |
},
|
| 1138 |
{
|
| 1139 |
+
"epoch": 1.61,
|
| 1140 |
+
"learning_rate": 5.713684210526315e-07,
|
| 1141 |
+
"loss": 0.0156,
|
| 1142 |
"step": 4575
|
| 1143 |
},
|
| 1144 |
{
|
| 1145 |
+
"epoch": 1.62,
|
| 1146 |
+
"learning_rate": 5.687368421052632e-07,
|
| 1147 |
+
"loss": 0.0175,
|
| 1148 |
"step": 4600
|
| 1149 |
},
|
| 1150 |
{
|
| 1151 |
+
"epoch": 1.63,
|
| 1152 |
+
"learning_rate": 5.661052631578948e-07,
|
| 1153 |
+
"loss": 0.0115,
|
| 1154 |
"step": 4625
|
| 1155 |
},
|
| 1156 |
{
|
| 1157 |
+
"epoch": 1.64,
|
| 1158 |
+
"learning_rate": 5.634736842105264e-07,
|
| 1159 |
+
"loss": 0.0174,
|
| 1160 |
"step": 4650
|
| 1161 |
},
|
| 1162 |
{
|
| 1163 |
+
"epoch": 1.65,
|
| 1164 |
+
"learning_rate": 5.608421052631579e-07,
|
| 1165 |
+
"loss": 0.039,
|
| 1166 |
"step": 4675
|
| 1167 |
},
|
| 1168 |
{
|
| 1169 |
+
"epoch": 1.66,
|
| 1170 |
+
"learning_rate": 5.582105263157894e-07,
|
| 1171 |
+
"loss": 0.0214,
|
| 1172 |
"step": 4700
|
| 1173 |
},
|
| 1174 |
{
|
| 1175 |
+
"epoch": 1.66,
|
| 1176 |
+
"learning_rate": 5.55578947368421e-07,
|
| 1177 |
+
"loss": 0.0234,
|
| 1178 |
"step": 4725
|
| 1179 |
},
|
| 1180 |
{
|
| 1181 |
+
"epoch": 1.67,
|
| 1182 |
+
"learning_rate": 5.529473684210526e-07,
|
| 1183 |
+
"loss": 0.0217,
|
| 1184 |
"step": 4750
|
| 1185 |
},
|
| 1186 |
{
|
| 1187 |
+
"epoch": 1.68,
|
| 1188 |
+
"learning_rate": 5.503157894736842e-07,
|
| 1189 |
+
"loss": 0.0147,
|
| 1190 |
"step": 4775
|
| 1191 |
},
|
| 1192 |
{
|
| 1193 |
+
"epoch": 1.69,
|
| 1194 |
+
"learning_rate": 5.476842105263157e-07,
|
| 1195 |
+
"loss": 0.0124,
|
| 1196 |
"step": 4800
|
| 1197 |
},
|
| 1198 |
{
|
| 1199 |
+
"epoch": 1.7,
|
| 1200 |
+
"learning_rate": 5.450526315789473e-07,
|
| 1201 |
+
"loss": 0.0192,
|
| 1202 |
"step": 4825
|
| 1203 |
},
|
| 1204 |
{
|
| 1205 |
+
"epoch": 1.71,
|
| 1206 |
+
"learning_rate": 5.424210526315789e-07,
|
| 1207 |
+
"loss": 0.0121,
|
| 1208 |
"step": 4850
|
| 1209 |
},
|
| 1210 |
{
|
| 1211 |
+
"epoch": 1.72,
|
| 1212 |
+
"learning_rate": 5.397894736842105e-07,
|
| 1213 |
+
"loss": 0.0133,
|
| 1214 |
"step": 4875
|
| 1215 |
},
|
| 1216 |
{
|
| 1217 |
+
"epoch": 1.73,
|
| 1218 |
+
"learning_rate": 5.371578947368421e-07,
|
| 1219 |
+
"loss": 0.0229,
|
| 1220 |
"step": 4900
|
| 1221 |
},
|
| 1222 |
{
|
| 1223 |
+
"epoch": 1.73,
|
| 1224 |
+
"learning_rate": 5.345263157894738e-07,
|
| 1225 |
+
"loss": 0.0151,
|
| 1226 |
"step": 4925
|
| 1227 |
},
|
| 1228 |
{
|
| 1229 |
+
"epoch": 1.74,
|
| 1230 |
+
"learning_rate": 5.318947368421052e-07,
|
| 1231 |
+
"loss": 0.0151,
|
| 1232 |
"step": 4950
|
| 1233 |
},
|
| 1234 |
{
|
| 1235 |
+
"epoch": 1.75,
|
| 1236 |
+
"learning_rate": 5.292631578947368e-07,
|
| 1237 |
+
"loss": 0.0209,
|
| 1238 |
"step": 4975
|
| 1239 |
},
|
| 1240 |
{
|
| 1241 |
+
"epoch": 1.76,
|
| 1242 |
+
"learning_rate": 5.266315789473684e-07,
|
| 1243 |
+
"loss": 0.0295,
|
| 1244 |
"step": 5000
|
| 1245 |
},
|
| 1246 |
{
|
| 1247 |
+
"epoch": 1.76,
|
| 1248 |
+
"eval_loss": 0.0670861229300499,
|
| 1249 |
+
"eval_runtime": 2277.3949,
|
| 1250 |
+
"eval_samples_per_second": 1.589,
|
| 1251 |
+
"eval_steps_per_second": 0.1,
|
| 1252 |
+
"eval_wer": 4.085585170155861,
|
| 1253 |
"step": 5000
|
| 1254 |
},
|
| 1255 |
{
|
| 1256 |
+
"epoch": 1.77,
|
| 1257 |
+
"learning_rate": 5.24e-07,
|
| 1258 |
+
"loss": 0.0224,
|
| 1259 |
+
"step": 5025
|
| 1260 |
+
},
|
| 1261 |
+
{
|
| 1262 |
+
"epoch": 1.78,
|
| 1263 |
+
"learning_rate": 5.213684210526316e-07,
|
| 1264 |
+
"loss": 0.021,
|
| 1265 |
+
"step": 5050
|
| 1266 |
+
},
|
| 1267 |
+
{
|
| 1268 |
+
"epoch": 1.79,
|
| 1269 |
+
"learning_rate": 5.187368421052631e-07,
|
| 1270 |
+
"loss": 0.0106,
|
| 1271 |
+
"step": 5075
|
| 1272 |
+
},
|
| 1273 |
+
{
|
| 1274 |
+
"epoch": 1.8,
|
| 1275 |
+
"learning_rate": 5.161052631578947e-07,
|
| 1276 |
+
"loss": 0.022,
|
| 1277 |
+
"step": 5100
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 1.81,
|
| 1281 |
+
"learning_rate": 5.134736842105263e-07,
|
| 1282 |
+
"loss": 0.0208,
|
| 1283 |
+
"step": 5125
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 1.81,
|
| 1287 |
+
"learning_rate": 5.108421052631579e-07,
|
| 1288 |
+
"loss": 0.0298,
|
| 1289 |
+
"step": 5150
|
| 1290 |
+
},
|
| 1291 |
+
{
|
| 1292 |
+
"epoch": 1.82,
|
| 1293 |
+
"learning_rate": 5.082105263157895e-07,
|
| 1294 |
+
"loss": 0.0164,
|
| 1295 |
+
"step": 5175
|
| 1296 |
+
},
|
| 1297 |
+
{
|
| 1298 |
+
"epoch": 1.83,
|
| 1299 |
+
"learning_rate": 5.05578947368421e-07,
|
| 1300 |
+
"loss": 0.0212,
|
| 1301 |
+
"step": 5200
|
| 1302 |
+
},
|
| 1303 |
+
{
|
| 1304 |
+
"epoch": 1.84,
|
| 1305 |
+
"learning_rate": 5.029473684210525e-07,
|
| 1306 |
+
"loss": 0.0101,
|
| 1307 |
+
"step": 5225
|
| 1308 |
+
},
|
| 1309 |
+
{
|
| 1310 |
+
"epoch": 1.85,
|
| 1311 |
+
"learning_rate": 5.003157894736841e-07,
|
| 1312 |
+
"loss": 0.0225,
|
| 1313 |
+
"step": 5250
|
| 1314 |
+
},
|
| 1315 |
+
{
|
| 1316 |
+
"epoch": 1.86,
|
| 1317 |
+
"learning_rate": 4.976842105263158e-07,
|
| 1318 |
+
"loss": 0.0135,
|
| 1319 |
+
"step": 5275
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 1.87,
|
| 1323 |
+
"learning_rate": 4.950526315789474e-07,
|
| 1324 |
+
"loss": 0.0343,
|
| 1325 |
+
"step": 5300
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 1.88,
|
| 1329 |
+
"learning_rate": 4.924210526315789e-07,
|
| 1330 |
+
"loss": 0.024,
|
| 1331 |
+
"step": 5325
|
| 1332 |
+
},
|
| 1333 |
+
{
|
| 1334 |
+
"epoch": 1.88,
|
| 1335 |
+
"learning_rate": 4.897894736842105e-07,
|
| 1336 |
+
"loss": 0.022,
|
| 1337 |
+
"step": 5350
|
| 1338 |
+
},
|
| 1339 |
+
{
|
| 1340 |
+
"epoch": 1.89,
|
| 1341 |
+
"learning_rate": 4.871578947368421e-07,
|
| 1342 |
+
"loss": 0.0243,
|
| 1343 |
+
"step": 5375
|
| 1344 |
+
},
|
| 1345 |
+
{
|
| 1346 |
+
"epoch": 1.9,
|
| 1347 |
+
"learning_rate": 4.845263157894737e-07,
|
| 1348 |
+
"loss": 0.0174,
|
| 1349 |
+
"step": 5400
|
| 1350 |
+
},
|
| 1351 |
+
{
|
| 1352 |
+
"epoch": 1.91,
|
| 1353 |
+
"learning_rate": 4.818947368421053e-07,
|
| 1354 |
+
"loss": 0.019,
|
| 1355 |
+
"step": 5425
|
| 1356 |
+
},
|
| 1357 |
+
{
|
| 1358 |
+
"epoch": 1.92,
|
| 1359 |
+
"learning_rate": 4.792631578947368e-07,
|
| 1360 |
+
"loss": 0.02,
|
| 1361 |
+
"step": 5450
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 1.93,
|
| 1365 |
+
"learning_rate": 4.766315789473684e-07,
|
| 1366 |
+
"loss": 0.0265,
|
| 1367 |
+
"step": 5475
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 1.94,
|
| 1371 |
+
"learning_rate": 4.7399999999999993e-07,
|
| 1372 |
+
"loss": 0.0216,
|
| 1373 |
+
"step": 5500
|
| 1374 |
+
},
|
| 1375 |
+
{
|
| 1376 |
+
"epoch": 1.95,
|
| 1377 |
+
"learning_rate": 4.7136842105263156e-07,
|
| 1378 |
+
"loss": 0.013,
|
| 1379 |
+
"step": 5525
|
| 1380 |
+
},
|
| 1381 |
+
{
|
| 1382 |
+
"epoch": 1.95,
|
| 1383 |
+
"learning_rate": 4.6873684210526315e-07,
|
| 1384 |
+
"loss": 0.0207,
|
| 1385 |
+
"step": 5550
|
| 1386 |
+
},
|
| 1387 |
+
{
|
| 1388 |
+
"epoch": 1.96,
|
| 1389 |
+
"learning_rate": 4.6610526315789473e-07,
|
| 1390 |
+
"loss": 0.0264,
|
| 1391 |
+
"step": 5575
|
| 1392 |
+
},
|
| 1393 |
+
{
|
| 1394 |
+
"epoch": 1.97,
|
| 1395 |
+
"learning_rate": 4.634736842105263e-07,
|
| 1396 |
+
"loss": 0.0126,
|
| 1397 |
+
"step": 5600
|
| 1398 |
+
},
|
| 1399 |
+
{
|
| 1400 |
+
"epoch": 1.98,
|
| 1401 |
+
"learning_rate": 4.6084210526315784e-07,
|
| 1402 |
+
"loss": 0.0162,
|
| 1403 |
+
"step": 5625
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 1.99,
|
| 1407 |
+
"learning_rate": 4.582105263157894e-07,
|
| 1408 |
+
"loss": 0.0126,
|
| 1409 |
+
"step": 5650
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 2.0,
|
| 1413 |
+
"learning_rate": 4.55578947368421e-07,
|
| 1414 |
+
"loss": 0.0209,
|
| 1415 |
+
"step": 5675
|
| 1416 |
+
},
|
| 1417 |
+
{
|
| 1418 |
+
"epoch": 2.01,
|
| 1419 |
+
"learning_rate": 4.5294736842105264e-07,
|
| 1420 |
+
"loss": 0.0134,
|
| 1421 |
+
"step": 5700
|
| 1422 |
+
},
|
| 1423 |
+
{
|
| 1424 |
+
"epoch": 2.02,
|
| 1425 |
+
"learning_rate": 4.503157894736842e-07,
|
| 1426 |
+
"loss": 0.0244,
|
| 1427 |
+
"step": 5725
|
| 1428 |
+
},
|
| 1429 |
+
{
|
| 1430 |
+
"epoch": 2.03,
|
| 1431 |
+
"learning_rate": 4.476842105263158e-07,
|
| 1432 |
+
"loss": 0.0066,
|
| 1433 |
+
"step": 5750
|
| 1434 |
+
},
|
| 1435 |
+
{
|
| 1436 |
+
"epoch": 2.03,
|
| 1437 |
+
"learning_rate": 4.4505263157894733e-07,
|
| 1438 |
+
"loss": 0.0155,
|
| 1439 |
+
"step": 5775
|
| 1440 |
+
},
|
| 1441 |
+
{
|
| 1442 |
+
"epoch": 2.04,
|
| 1443 |
+
"learning_rate": 4.424210526315789e-07,
|
| 1444 |
+
"loss": 0.0076,
|
| 1445 |
+
"step": 5800
|
| 1446 |
+
},
|
| 1447 |
+
{
|
| 1448 |
+
"epoch": 2.05,
|
| 1449 |
+
"learning_rate": 4.397894736842105e-07,
|
| 1450 |
+
"loss": 0.0135,
|
| 1451 |
+
"step": 5825
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 2.06,
|
| 1455 |
+
"learning_rate": 4.371578947368421e-07,
|
| 1456 |
+
"loss": 0.0054,
|
| 1457 |
+
"step": 5850
|
| 1458 |
+
},
|
| 1459 |
+
{
|
| 1460 |
+
"epoch": 2.07,
|
| 1461 |
+
"learning_rate": 4.345263157894737e-07,
|
| 1462 |
+
"loss": 0.0186,
|
| 1463 |
+
"step": 5875
|
| 1464 |
+
},
|
| 1465 |
+
{
|
| 1466 |
+
"epoch": 2.08,
|
| 1467 |
+
"learning_rate": 4.3189473684210524e-07,
|
| 1468 |
+
"loss": 0.016,
|
| 1469 |
+
"step": 5900
|
| 1470 |
+
},
|
| 1471 |
+
{
|
| 1472 |
+
"epoch": 2.09,
|
| 1473 |
+
"learning_rate": 4.292631578947368e-07,
|
| 1474 |
+
"loss": 0.0089,
|
| 1475 |
+
"step": 5925
|
| 1476 |
+
},
|
| 1477 |
+
{
|
| 1478 |
+
"epoch": 2.1,
|
| 1479 |
+
"learning_rate": 4.266315789473684e-07,
|
| 1480 |
+
"loss": 0.0065,
|
| 1481 |
+
"step": 5950
|
| 1482 |
+
},
|
| 1483 |
+
{
|
| 1484 |
+
"epoch": 2.1,
|
| 1485 |
+
"learning_rate": 4.24e-07,
|
| 1486 |
+
"loss": 0.0122,
|
| 1487 |
+
"step": 5975
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 2.11,
|
| 1491 |
+
"learning_rate": 4.213684210526315e-07,
|
| 1492 |
+
"loss": 0.0147,
|
| 1493 |
+
"step": 6000
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 2.11,
|
| 1497 |
+
"eval_loss": 0.06900504976511002,
|
| 1498 |
+
"eval_runtime": 2278.2667,
|
| 1499 |
+
"eval_samples_per_second": 1.588,
|
| 1500 |
+
"eval_steps_per_second": 0.1,
|
| 1501 |
+
"eval_wer": 4.0302499308309505,
|
| 1502 |
+
"step": 6000
|
| 1503 |
+
},
|
| 1504 |
+
{
|
| 1505 |
+
"epoch": 2.12,
|
| 1506 |
+
"learning_rate": 4.1873684210526315e-07,
|
| 1507 |
+
"loss": 0.0147,
|
| 1508 |
+
"step": 6025
|
| 1509 |
+
},
|
| 1510 |
+
{
|
| 1511 |
+
"epoch": 2.13,
|
| 1512 |
+
"learning_rate": 4.1610526315789473e-07,
|
| 1513 |
+
"loss": 0.0141,
|
| 1514 |
+
"step": 6050
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 2.14,
|
| 1518 |
+
"learning_rate": 4.134736842105263e-07,
|
| 1519 |
+
"loss": 0.0149,
|
| 1520 |
+
"step": 6075
|
| 1521 |
+
},
|
| 1522 |
+
{
|
| 1523 |
+
"epoch": 2.15,
|
| 1524 |
+
"learning_rate": 4.108421052631579e-07,
|
| 1525 |
+
"loss": 0.01,
|
| 1526 |
+
"step": 6100
|
| 1527 |
+
},
|
| 1528 |
+
{
|
| 1529 |
+
"epoch": 2.16,
|
| 1530 |
+
"learning_rate": 4.083157894736842e-07,
|
| 1531 |
+
"loss": 0.0126,
|
| 1532 |
+
"step": 6125
|
| 1533 |
+
},
|
| 1534 |
+
{
|
| 1535 |
+
"epoch": 2.17,
|
| 1536 |
+
"learning_rate": 4.0568421052631577e-07,
|
| 1537 |
+
"loss": 0.0137,
|
| 1538 |
+
"step": 6150
|
| 1539 |
+
},
|
| 1540 |
+
{
|
| 1541 |
+
"epoch": 2.18,
|
| 1542 |
+
"learning_rate": 4.0305263157894735e-07,
|
| 1543 |
+
"loss": 0.0072,
|
| 1544 |
+
"step": 6175
|
| 1545 |
+
},
|
| 1546 |
+
{
|
| 1547 |
+
"epoch": 2.18,
|
| 1548 |
+
"learning_rate": 4.0042105263157893e-07,
|
| 1549 |
+
"loss": 0.0098,
|
| 1550 |
+
"step": 6200
|
| 1551 |
+
},
|
| 1552 |
+
{
|
| 1553 |
+
"epoch": 2.19,
|
| 1554 |
+
"learning_rate": 3.977894736842105e-07,
|
| 1555 |
+
"loss": 0.0086,
|
| 1556 |
+
"step": 6225
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 2.2,
|
| 1560 |
+
"learning_rate": 3.951578947368421e-07,
|
| 1561 |
+
"loss": 0.0085,
|
| 1562 |
+
"step": 6250
|
| 1563 |
+
},
|
| 1564 |
+
{
|
| 1565 |
+
"epoch": 2.21,
|
| 1566 |
+
"learning_rate": 3.925263157894736e-07,
|
| 1567 |
+
"loss": 0.0184,
|
| 1568 |
+
"step": 6275
|
| 1569 |
+
},
|
| 1570 |
+
{
|
| 1571 |
+
"epoch": 2.22,
|
| 1572 |
+
"learning_rate": 3.898947368421052e-07,
|
| 1573 |
+
"loss": 0.0146,
|
| 1574 |
+
"step": 6300
|
| 1575 |
+
},
|
| 1576 |
+
{
|
| 1577 |
+
"epoch": 2.23,
|
| 1578 |
+
"learning_rate": 3.8726315789473684e-07,
|
| 1579 |
+
"loss": 0.0156,
|
| 1580 |
+
"step": 6325
|
| 1581 |
+
},
|
| 1582 |
+
{
|
| 1583 |
+
"epoch": 2.24,
|
| 1584 |
+
"learning_rate": 3.846315789473684e-07,
|
| 1585 |
+
"loss": 0.0075,
|
| 1586 |
+
"step": 6350
|
| 1587 |
+
},
|
| 1588 |
+
{
|
| 1589 |
+
"epoch": 2.25,
|
| 1590 |
+
"learning_rate": 3.82e-07,
|
| 1591 |
+
"loss": 0.0251,
|
| 1592 |
+
"step": 6375
|
| 1593 |
+
},
|
| 1594 |
+
{
|
| 1595 |
+
"epoch": 2.25,
|
| 1596 |
+
"learning_rate": 3.7936842105263153e-07,
|
| 1597 |
+
"loss": 0.0129,
|
| 1598 |
+
"step": 6400
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 2.26,
|
| 1602 |
+
"learning_rate": 3.767368421052631e-07,
|
| 1603 |
+
"loss": 0.0141,
|
| 1604 |
+
"step": 6425
|
| 1605 |
+
},
|
| 1606 |
+
{
|
| 1607 |
+
"epoch": 2.27,
|
| 1608 |
+
"learning_rate": 3.741052631578947e-07,
|
| 1609 |
+
"loss": 0.0083,
|
| 1610 |
+
"step": 6450
|
| 1611 |
+
},
|
| 1612 |
+
{
|
| 1613 |
+
"epoch": 2.28,
|
| 1614 |
+
"learning_rate": 3.714736842105263e-07,
|
| 1615 |
+
"loss": 0.013,
|
| 1616 |
+
"step": 6475
|
| 1617 |
+
},
|
| 1618 |
+
{
|
| 1619 |
+
"epoch": 2.29,
|
| 1620 |
+
"learning_rate": 3.688421052631579e-07,
|
| 1621 |
+
"loss": 0.0175,
|
| 1622 |
+
"step": 6500
|
| 1623 |
+
},
|
| 1624 |
+
{
|
| 1625 |
+
"epoch": 2.3,
|
| 1626 |
+
"learning_rate": 3.6621052631578944e-07,
|
| 1627 |
+
"loss": 0.0157,
|
| 1628 |
+
"step": 6525
|
| 1629 |
+
},
|
| 1630 |
+
{
|
| 1631 |
+
"epoch": 2.31,
|
| 1632 |
+
"learning_rate": 3.63578947368421e-07,
|
| 1633 |
+
"loss": 0.0163,
|
| 1634 |
+
"step": 6550
|
| 1635 |
+
},
|
| 1636 |
+
{
|
| 1637 |
+
"epoch": 2.32,
|
| 1638 |
+
"learning_rate": 3.609473684210526e-07,
|
| 1639 |
+
"loss": 0.0105,
|
| 1640 |
+
"step": 6575
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 2.32,
|
| 1644 |
+
"learning_rate": 3.583157894736842e-07,
|
| 1645 |
+
"loss": 0.0142,
|
| 1646 |
+
"step": 6600
|
| 1647 |
+
},
|
| 1648 |
+
{
|
| 1649 |
+
"epoch": 2.33,
|
| 1650 |
+
"learning_rate": 3.5568421052631577e-07,
|
| 1651 |
+
"loss": 0.0177,
|
| 1652 |
+
"step": 6625
|
| 1653 |
+
},
|
| 1654 |
+
{
|
| 1655 |
+
"epoch": 2.34,
|
| 1656 |
+
"learning_rate": 3.5305263157894735e-07,
|
| 1657 |
+
"loss": 0.015,
|
| 1658 |
+
"step": 6650
|
| 1659 |
+
},
|
| 1660 |
+
{
|
| 1661 |
+
"epoch": 2.35,
|
| 1662 |
+
"learning_rate": 3.5042105263157894e-07,
|
| 1663 |
+
"loss": 0.0069,
|
| 1664 |
+
"step": 6675
|
| 1665 |
+
},
|
| 1666 |
+
{
|
| 1667 |
+
"epoch": 2.36,
|
| 1668 |
+
"learning_rate": 3.477894736842105e-07,
|
| 1669 |
+
"loss": 0.0178,
|
| 1670 |
+
"step": 6700
|
| 1671 |
+
},
|
| 1672 |
+
{
|
| 1673 |
+
"epoch": 2.37,
|
| 1674 |
+
"learning_rate": 3.451578947368421e-07,
|
| 1675 |
+
"loss": 0.0175,
|
| 1676 |
+
"step": 6725
|
| 1677 |
+
},
|
| 1678 |
+
{
|
| 1679 |
+
"epoch": 2.38,
|
| 1680 |
+
"learning_rate": 3.425263157894737e-07,
|
| 1681 |
+
"loss": 0.0082,
|
| 1682 |
+
"step": 6750
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 2.39,
|
| 1686 |
+
"learning_rate": 3.398947368421052e-07,
|
| 1687 |
+
"loss": 0.0094,
|
| 1688 |
+
"step": 6775
|
| 1689 |
+
},
|
| 1690 |
+
{
|
| 1691 |
+
"epoch": 2.4,
|
| 1692 |
+
"learning_rate": 3.372631578947368e-07,
|
| 1693 |
+
"loss": 0.0114,
|
| 1694 |
+
"step": 6800
|
| 1695 |
+
},
|
| 1696 |
+
{
|
| 1697 |
+
"epoch": 2.4,
|
| 1698 |
+
"learning_rate": 3.346315789473684e-07,
|
| 1699 |
+
"loss": 0.0314,
|
| 1700 |
+
"step": 6825
|
| 1701 |
+
},
|
| 1702 |
+
{
|
| 1703 |
+
"epoch": 2.41,
|
| 1704 |
+
"learning_rate": 3.32e-07,
|
| 1705 |
+
"loss": 0.0127,
|
| 1706 |
+
"step": 6850
|
| 1707 |
+
},
|
| 1708 |
+
{
|
| 1709 |
+
"epoch": 2.42,
|
| 1710 |
+
"learning_rate": 3.293684210526316e-07,
|
| 1711 |
+
"loss": 0.0054,
|
| 1712 |
+
"step": 6875
|
| 1713 |
+
},
|
| 1714 |
+
{
|
| 1715 |
+
"epoch": 2.43,
|
| 1716 |
+
"learning_rate": 3.267368421052631e-07,
|
| 1717 |
+
"loss": 0.0104,
|
| 1718 |
+
"step": 6900
|
| 1719 |
+
},
|
| 1720 |
+
{
|
| 1721 |
+
"epoch": 2.44,
|
| 1722 |
+
"learning_rate": 3.241052631578947e-07,
|
| 1723 |
+
"loss": 0.0121,
|
| 1724 |
+
"step": 6925
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 2.45,
|
| 1728 |
+
"learning_rate": 3.214736842105263e-07,
|
| 1729 |
+
"loss": 0.0136,
|
| 1730 |
+
"step": 6950
|
| 1731 |
+
},
|
| 1732 |
+
{
|
| 1733 |
+
"epoch": 2.46,
|
| 1734 |
+
"learning_rate": 3.1884210526315786e-07,
|
| 1735 |
+
"loss": 0.0141,
|
| 1736 |
+
"step": 6975
|
| 1737 |
+
},
|
| 1738 |
+
{
|
| 1739 |
+
"epoch": 2.47,
|
| 1740 |
+
"learning_rate": 3.162105263157895e-07,
|
| 1741 |
+
"loss": 0.0157,
|
| 1742 |
+
"step": 7000
|
| 1743 |
+
},
|
| 1744 |
+
{
|
| 1745 |
+
"epoch": 2.47,
|
| 1746 |
+
"eval_loss": 0.06976281106472015,
|
| 1747 |
+
"eval_runtime": 2275.6922,
|
| 1748 |
+
"eval_samples_per_second": 1.59,
|
| 1749 |
+
"eval_steps_per_second": 0.1,
|
| 1750 |
+
"eval_wer": 3.993359771281011,
|
| 1751 |
+
"step": 7000
|
| 1752 |
+
},
|
| 1753 |
+
{
|
| 1754 |
+
"epoch": 2.47,
|
| 1755 |
+
"learning_rate": 3.1357894736842103e-07,
|
| 1756 |
+
"loss": 0.0103,
|
| 1757 |
+
"step": 7025
|
| 1758 |
+
},
|
| 1759 |
+
{
|
| 1760 |
+
"epoch": 2.48,
|
| 1761 |
+
"learning_rate": 3.109473684210526e-07,
|
| 1762 |
+
"loss": 0.0102,
|
| 1763 |
+
"step": 7050
|
| 1764 |
+
},
|
| 1765 |
+
{
|
| 1766 |
+
"epoch": 2.49,
|
| 1767 |
+
"learning_rate": 3.083157894736842e-07,
|
| 1768 |
+
"loss": 0.0134,
|
| 1769 |
+
"step": 7075
|
| 1770 |
+
},
|
| 1771 |
+
{
|
| 1772 |
+
"epoch": 2.5,
|
| 1773 |
+
"learning_rate": 3.056842105263158e-07,
|
| 1774 |
+
"loss": 0.008,
|
| 1775 |
+
"step": 7100
|
| 1776 |
+
},
|
| 1777 |
+
{
|
| 1778 |
+
"epoch": 2.51,
|
| 1779 |
+
"learning_rate": 3.0305263157894736e-07,
|
| 1780 |
+
"loss": 0.0148,
|
| 1781 |
+
"step": 7125
|
| 1782 |
+
},
|
| 1783 |
+
{
|
| 1784 |
+
"epoch": 2.52,
|
| 1785 |
+
"learning_rate": 3.004210526315789e-07,
|
| 1786 |
+
"loss": 0.011,
|
| 1787 |
+
"step": 7150
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 2.53,
|
| 1791 |
+
"learning_rate": 2.977894736842105e-07,
|
| 1792 |
+
"loss": 0.0069,
|
| 1793 |
+
"step": 7175
|
| 1794 |
+
},
|
| 1795 |
+
{
|
| 1796 |
+
"epoch": 2.54,
|
| 1797 |
+
"learning_rate": 2.951578947368421e-07,
|
| 1798 |
+
"loss": 0.0098,
|
| 1799 |
+
"step": 7200
|
| 1800 |
+
},
|
| 1801 |
+
{
|
| 1802 |
+
"epoch": 2.54,
|
| 1803 |
+
"learning_rate": 2.925263157894737e-07,
|
| 1804 |
+
"loss": 0.0101,
|
| 1805 |
+
"step": 7225
|
| 1806 |
+
},
|
| 1807 |
+
{
|
| 1808 |
+
"epoch": 2.55,
|
| 1809 |
+
"learning_rate": 2.8989473684210526e-07,
|
| 1810 |
+
"loss": 0.0179,
|
| 1811 |
+
"step": 7250
|
| 1812 |
+
},
|
| 1813 |
+
{
|
| 1814 |
+
"epoch": 2.56,
|
| 1815 |
+
"learning_rate": 2.872631578947368e-07,
|
| 1816 |
+
"loss": 0.0155,
|
| 1817 |
+
"step": 7275
|
| 1818 |
+
},
|
| 1819 |
+
{
|
| 1820 |
+
"epoch": 2.57,
|
| 1821 |
+
"learning_rate": 2.846315789473684e-07,
|
| 1822 |
+
"loss": 0.0183,
|
| 1823 |
+
"step": 7300
|
| 1824 |
+
},
|
| 1825 |
+
{
|
| 1826 |
+
"epoch": 2.58,
|
| 1827 |
+
"learning_rate": 2.8199999999999996e-07,
|
| 1828 |
+
"loss": 0.011,
|
| 1829 |
+
"step": 7325
|
| 1830 |
+
},
|
| 1831 |
+
{
|
| 1832 |
+
"epoch": 2.59,
|
| 1833 |
+
"learning_rate": 2.793684210526316e-07,
|
| 1834 |
+
"loss": 0.0144,
|
| 1835 |
+
"step": 7350
|
| 1836 |
+
},
|
| 1837 |
+
{
|
| 1838 |
+
"epoch": 2.6,
|
| 1839 |
+
"learning_rate": 2.767368421052632e-07,
|
| 1840 |
+
"loss": 0.0194,
|
| 1841 |
+
"step": 7375
|
| 1842 |
+
},
|
| 1843 |
+
{
|
| 1844 |
+
"epoch": 2.61,
|
| 1845 |
+
"learning_rate": 2.741052631578947e-07,
|
| 1846 |
+
"loss": 0.0093,
|
| 1847 |
+
"step": 7400
|
| 1848 |
+
},
|
| 1849 |
+
{
|
| 1850 |
+
"epoch": 2.62,
|
| 1851 |
+
"learning_rate": 2.714736842105263e-07,
|
| 1852 |
+
"loss": 0.0069,
|
| 1853 |
+
"step": 7425
|
| 1854 |
+
},
|
| 1855 |
+
{
|
| 1856 |
+
"epoch": 2.62,
|
| 1857 |
+
"learning_rate": 2.6884210526315787e-07,
|
| 1858 |
+
"loss": 0.0114,
|
| 1859 |
+
"step": 7450
|
| 1860 |
+
},
|
| 1861 |
+
{
|
| 1862 |
+
"epoch": 2.63,
|
| 1863 |
+
"learning_rate": 2.6621052631578945e-07,
|
| 1864 |
+
"loss": 0.0153,
|
| 1865 |
+
"step": 7475
|
| 1866 |
+
},
|
| 1867 |
+
{
|
| 1868 |
+
"epoch": 2.64,
|
| 1869 |
+
"learning_rate": 2.635789473684211e-07,
|
| 1870 |
+
"loss": 0.0081,
|
| 1871 |
+
"step": 7500
|
| 1872 |
+
},
|
| 1873 |
+
{
|
| 1874 |
+
"epoch": 2.65,
|
| 1875 |
+
"learning_rate": 2.609473684210526e-07,
|
| 1876 |
+
"loss": 0.0182,
|
| 1877 |
+
"step": 7525
|
| 1878 |
+
},
|
| 1879 |
+
{
|
| 1880 |
+
"epoch": 2.66,
|
| 1881 |
+
"learning_rate": 2.583157894736842e-07,
|
| 1882 |
+
"loss": 0.0131,
|
| 1883 |
+
"step": 7550
|
| 1884 |
+
},
|
| 1885 |
+
{
|
| 1886 |
+
"epoch": 2.67,
|
| 1887 |
+
"learning_rate": 2.556842105263158e-07,
|
| 1888 |
+
"loss": 0.01,
|
| 1889 |
+
"step": 7575
|
| 1890 |
+
},
|
| 1891 |
+
{
|
| 1892 |
+
"epoch": 2.68,
|
| 1893 |
+
"learning_rate": 2.5305263157894736e-07,
|
| 1894 |
+
"loss": 0.0095,
|
| 1895 |
+
"step": 7600
|
| 1896 |
+
},
|
| 1897 |
+
{
|
| 1898 |
+
"epoch": 2.69,
|
| 1899 |
+
"learning_rate": 2.5042105263157894e-07,
|
| 1900 |
+
"loss": 0.018,
|
| 1901 |
+
"step": 7625
|
| 1902 |
+
},
|
| 1903 |
+
{
|
| 1904 |
+
"epoch": 2.69,
|
| 1905 |
+
"learning_rate": 2.477894736842105e-07,
|
| 1906 |
+
"loss": 0.0369,
|
| 1907 |
+
"step": 7650
|
| 1908 |
+
},
|
| 1909 |
+
{
|
| 1910 |
+
"epoch": 2.7,
|
| 1911 |
+
"learning_rate": 2.451578947368421e-07,
|
| 1912 |
+
"loss": 0.0149,
|
| 1913 |
+
"step": 7675
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 2.71,
|
| 1917 |
+
"learning_rate": 2.425263157894737e-07,
|
| 1918 |
+
"loss": 0.0086,
|
| 1919 |
+
"step": 7700
|
| 1920 |
+
},
|
| 1921 |
+
{
|
| 1922 |
+
"epoch": 2.72,
|
| 1923 |
+
"learning_rate": 2.3989473684210527e-07,
|
| 1924 |
+
"loss": 0.0077,
|
| 1925 |
+
"step": 7725
|
| 1926 |
+
},
|
| 1927 |
+
{
|
| 1928 |
+
"epoch": 2.73,
|
| 1929 |
+
"learning_rate": 2.3726315789473682e-07,
|
| 1930 |
+
"loss": 0.0129,
|
| 1931 |
+
"step": 7750
|
| 1932 |
+
},
|
| 1933 |
+
{
|
| 1934 |
+
"epoch": 2.74,
|
| 1935 |
+
"learning_rate": 2.346315789473684e-07,
|
| 1936 |
+
"loss": 0.0116,
|
| 1937 |
+
"step": 7775
|
| 1938 |
+
},
|
| 1939 |
+
{
|
| 1940 |
+
"epoch": 2.75,
|
| 1941 |
+
"learning_rate": 2.32e-07,
|
| 1942 |
+
"loss": 0.0067,
|
| 1943 |
+
"step": 7800
|
| 1944 |
+
},
|
| 1945 |
+
{
|
| 1946 |
+
"epoch": 2.76,
|
| 1947 |
+
"learning_rate": 2.2936842105263157e-07,
|
| 1948 |
+
"loss": 0.0101,
|
| 1949 |
+
"step": 7825
|
| 1950 |
+
},
|
| 1951 |
+
{
|
| 1952 |
+
"epoch": 2.77,
|
| 1953 |
+
"learning_rate": 2.2673684210526315e-07,
|
| 1954 |
+
"loss": 0.0072,
|
| 1955 |
+
"step": 7850
|
| 1956 |
+
},
|
| 1957 |
+
{
|
| 1958 |
+
"epoch": 2.77,
|
| 1959 |
+
"learning_rate": 2.241052631578947e-07,
|
| 1960 |
+
"loss": 0.0092,
|
| 1961 |
+
"step": 7875
|
| 1962 |
+
},
|
| 1963 |
+
{
|
| 1964 |
+
"epoch": 2.78,
|
| 1965 |
+
"learning_rate": 2.2147368421052631e-07,
|
| 1966 |
+
"loss": 0.011,
|
| 1967 |
+
"step": 7900
|
| 1968 |
+
},
|
| 1969 |
+
{
|
| 1970 |
+
"epoch": 2.79,
|
| 1971 |
+
"learning_rate": 2.188421052631579e-07,
|
| 1972 |
+
"loss": 0.0101,
|
| 1973 |
+
"step": 7925
|
| 1974 |
+
},
|
| 1975 |
+
{
|
| 1976 |
+
"epoch": 2.8,
|
| 1977 |
+
"learning_rate": 2.1621052631578945e-07,
|
| 1978 |
+
"loss": 0.0129,
|
| 1979 |
+
"step": 7950
|
| 1980 |
+
},
|
| 1981 |
+
{
|
| 1982 |
+
"epoch": 2.81,
|
| 1983 |
+
"learning_rate": 2.1357894736842106e-07,
|
| 1984 |
+
"loss": 0.0054,
|
| 1985 |
+
"step": 7975
|
| 1986 |
+
},
|
| 1987 |
+
{
|
| 1988 |
+
"epoch": 2.82,
|
| 1989 |
+
"learning_rate": 2.1094736842105261e-07,
|
| 1990 |
+
"loss": 0.0113,
|
| 1991 |
+
"step": 8000
|
| 1992 |
+
},
|
| 1993 |
+
{
|
| 1994 |
+
"epoch": 2.82,
|
| 1995 |
+
"eval_loss": 0.06976284831762314,
|
| 1996 |
+
"eval_runtime": 2275.4921,
|
| 1997 |
+
"eval_samples_per_second": 1.59,
|
| 1998 |
+
"eval_steps_per_second": 0.1,
|
| 1999 |
+
"eval_wer": 4.0302499308309505,
|
| 2000 |
+
"step": 8000
|
| 2001 |
+
},
|
| 2002 |
+
{
|
| 2003 |
+
"epoch": 2.83,
|
| 2004 |
+
"learning_rate": 2.083157894736842e-07,
|
| 2005 |
+
"loss": 0.0148,
|
| 2006 |
+
"step": 8025
|
| 2007 |
+
},
|
| 2008 |
+
{
|
| 2009 |
+
"epoch": 2.84,
|
| 2010 |
+
"learning_rate": 2.056842105263158e-07,
|
| 2011 |
+
"loss": 0.0144,
|
| 2012 |
+
"step": 8050
|
| 2013 |
+
},
|
| 2014 |
+
{
|
| 2015 |
+
"epoch": 2.84,
|
| 2016 |
+
"learning_rate": 2.0305263157894736e-07,
|
| 2017 |
+
"loss": 0.0232,
|
| 2018 |
+
"step": 8075
|
| 2019 |
+
},
|
| 2020 |
+
{
|
| 2021 |
+
"epoch": 2.85,
|
| 2022 |
+
"learning_rate": 2.0042105263157894e-07,
|
| 2023 |
+
"loss": 0.014,
|
| 2024 |
+
"step": 8100
|
| 2025 |
+
},
|
| 2026 |
+
{
|
| 2027 |
+
"epoch": 2.86,
|
| 2028 |
+
"learning_rate": 1.977894736842105e-07,
|
| 2029 |
+
"loss": 0.0107,
|
| 2030 |
+
"step": 8125
|
| 2031 |
+
},
|
| 2032 |
+
{
|
| 2033 |
+
"epoch": 2.87,
|
| 2034 |
+
"learning_rate": 1.951578947368421e-07,
|
| 2035 |
+
"loss": 0.0109,
|
| 2036 |
+
"step": 8150
|
| 2037 |
+
},
|
| 2038 |
+
{
|
| 2039 |
+
"epoch": 2.88,
|
| 2040 |
+
"learning_rate": 1.925263157894737e-07,
|
| 2041 |
+
"loss": 0.0081,
|
| 2042 |
+
"step": 8175
|
| 2043 |
+
},
|
| 2044 |
+
{
|
| 2045 |
+
"epoch": 2.89,
|
| 2046 |
+
"learning_rate": 1.8999999999999998e-07,
|
| 2047 |
+
"loss": 0.0108,
|
| 2048 |
+
"step": 8200
|
| 2049 |
+
},
|
| 2050 |
+
{
|
| 2051 |
+
"epoch": 2.9,
|
| 2052 |
+
"learning_rate": 1.8736842105263156e-07,
|
| 2053 |
+
"loss": 0.0107,
|
| 2054 |
+
"step": 8225
|
| 2055 |
+
},
|
| 2056 |
+
{
|
| 2057 |
+
"epoch": 2.91,
|
| 2058 |
+
"learning_rate": 1.8473684210526317e-07,
|
| 2059 |
+
"loss": 0.0199,
|
| 2060 |
+
"step": 8250
|
| 2061 |
+
},
|
| 2062 |
+
{
|
| 2063 |
+
"epoch": 2.91,
|
| 2064 |
+
"learning_rate": 1.8210526315789473e-07,
|
| 2065 |
+
"loss": 0.012,
|
| 2066 |
+
"step": 8275
|
| 2067 |
+
},
|
| 2068 |
+
{
|
| 2069 |
+
"epoch": 2.92,
|
| 2070 |
+
"learning_rate": 1.794736842105263e-07,
|
| 2071 |
+
"loss": 0.0073,
|
| 2072 |
+
"step": 8300
|
| 2073 |
+
},
|
| 2074 |
+
{
|
| 2075 |
+
"epoch": 2.93,
|
| 2076 |
+
"learning_rate": 1.7684210526315786e-07,
|
| 2077 |
+
"loss": 0.0156,
|
| 2078 |
+
"step": 8325
|
| 2079 |
+
},
|
| 2080 |
+
{
|
| 2081 |
+
"epoch": 2.94,
|
| 2082 |
+
"learning_rate": 1.7421052631578947e-07,
|
| 2083 |
+
"loss": 0.0091,
|
| 2084 |
+
"step": 8350
|
| 2085 |
+
},
|
| 2086 |
+
{
|
| 2087 |
+
"epoch": 2.95,
|
| 2088 |
+
"learning_rate": 1.7157894736842105e-07,
|
| 2089 |
+
"loss": 0.0218,
|
| 2090 |
+
"step": 8375
|
| 2091 |
+
},
|
| 2092 |
+
{
|
| 2093 |
+
"epoch": 2.96,
|
| 2094 |
+
"learning_rate": 1.689473684210526e-07,
|
| 2095 |
+
"loss": 0.0076,
|
| 2096 |
+
"step": 8400
|
| 2097 |
+
},
|
| 2098 |
+
{
|
| 2099 |
+
"epoch": 2.97,
|
| 2100 |
+
"learning_rate": 1.6631578947368422e-07,
|
| 2101 |
+
"loss": 0.006,
|
| 2102 |
+
"step": 8425
|
| 2103 |
+
},
|
| 2104 |
+
{
|
| 2105 |
+
"epoch": 2.98,
|
| 2106 |
+
"learning_rate": 1.6368421052631577e-07,
|
| 2107 |
+
"loss": 0.0082,
|
| 2108 |
+
"step": 8450
|
| 2109 |
+
},
|
| 2110 |
+
{
|
| 2111 |
+
"epoch": 2.99,
|
| 2112 |
+
"learning_rate": 1.6105263157894735e-07,
|
| 2113 |
+
"loss": 0.0113,
|
| 2114 |
+
"step": 8475
|
| 2115 |
+
},
|
| 2116 |
+
{
|
| 2117 |
+
"epoch": 2.99,
|
| 2118 |
+
"learning_rate": 1.5842105263157894e-07,
|
| 2119 |
+
"loss": 0.0192,
|
| 2120 |
+
"step": 8500
|
| 2121 |
+
},
|
| 2122 |
+
{
|
| 2123 |
+
"epoch": 3.0,
|
| 2124 |
+
"learning_rate": 1.5578947368421052e-07,
|
| 2125 |
+
"loss": 0.0105,
|
| 2126 |
+
"step": 8525
|
| 2127 |
+
},
|
| 2128 |
+
{
|
| 2129 |
+
"epoch": 3.01,
|
| 2130 |
+
"learning_rate": 1.531578947368421e-07,
|
| 2131 |
+
"loss": 0.0052,
|
| 2132 |
+
"step": 8550
|
| 2133 |
+
},
|
| 2134 |
+
{
|
| 2135 |
+
"epoch": 3.02,
|
| 2136 |
+
"learning_rate": 1.5052631578947366e-07,
|
| 2137 |
+
"loss": 0.0047,
|
| 2138 |
+
"step": 8575
|
| 2139 |
+
},
|
| 2140 |
+
{
|
| 2141 |
+
"epoch": 3.03,
|
| 2142 |
+
"learning_rate": 1.4789473684210526e-07,
|
| 2143 |
+
"loss": 0.0106,
|
| 2144 |
+
"step": 8600
|
| 2145 |
+
},
|
| 2146 |
+
{
|
| 2147 |
+
"epoch": 3.04,
|
| 2148 |
+
"learning_rate": 1.4526315789473685e-07,
|
| 2149 |
+
"loss": 0.0073,
|
| 2150 |
+
"step": 8625
|
| 2151 |
+
},
|
| 2152 |
+
{
|
| 2153 |
+
"epoch": 3.05,
|
| 2154 |
+
"learning_rate": 1.426315789473684e-07,
|
| 2155 |
+
"loss": 0.0121,
|
| 2156 |
+
"step": 8650
|
| 2157 |
+
},
|
| 2158 |
+
{
|
| 2159 |
+
"epoch": 3.06,
|
| 2160 |
+
"learning_rate": 1.4e-07,
|
| 2161 |
+
"loss": 0.0076,
|
| 2162 |
+
"step": 8675
|
| 2163 |
+
},
|
| 2164 |
+
{
|
| 2165 |
+
"epoch": 3.06,
|
| 2166 |
+
"learning_rate": 1.3736842105263156e-07,
|
| 2167 |
+
"loss": 0.0045,
|
| 2168 |
+
"step": 8700
|
| 2169 |
+
},
|
| 2170 |
+
{
|
| 2171 |
+
"epoch": 3.07,
|
| 2172 |
+
"learning_rate": 1.3473684210526315e-07,
|
| 2173 |
+
"loss": 0.0061,
|
| 2174 |
+
"step": 8725
|
| 2175 |
+
},
|
| 2176 |
+
{
|
| 2177 |
+
"epoch": 3.08,
|
| 2178 |
+
"learning_rate": 1.3210526315789473e-07,
|
| 2179 |
+
"loss": 0.0088,
|
| 2180 |
+
"step": 8750
|
| 2181 |
+
},
|
| 2182 |
+
{
|
| 2183 |
+
"epoch": 3.09,
|
| 2184 |
+
"learning_rate": 1.294736842105263e-07,
|
| 2185 |
+
"loss": 0.0043,
|
| 2186 |
+
"step": 8775
|
| 2187 |
+
},
|
| 2188 |
+
{
|
| 2189 |
+
"epoch": 3.1,
|
| 2190 |
+
"learning_rate": 1.268421052631579e-07,
|
| 2191 |
+
"loss": 0.0091,
|
| 2192 |
+
"step": 8800
|
| 2193 |
+
},
|
| 2194 |
+
{
|
| 2195 |
+
"epoch": 3.11,
|
| 2196 |
+
"learning_rate": 1.2421052631578947e-07,
|
| 2197 |
+
"loss": 0.0093,
|
| 2198 |
+
"step": 8825
|
| 2199 |
+
},
|
| 2200 |
+
{
|
| 2201 |
+
"epoch": 3.12,
|
| 2202 |
+
"learning_rate": 1.2157894736842106e-07,
|
| 2203 |
+
"loss": 0.008,
|
| 2204 |
+
"step": 8850
|
| 2205 |
+
},
|
| 2206 |
+
{
|
| 2207 |
+
"epoch": 3.13,
|
| 2208 |
+
"learning_rate": 1.1894736842105262e-07,
|
| 2209 |
+
"loss": 0.0068,
|
| 2210 |
+
"step": 8875
|
| 2211 |
+
},
|
| 2212 |
+
{
|
| 2213 |
+
"epoch": 3.13,
|
| 2214 |
+
"learning_rate": 1.163157894736842e-07,
|
| 2215 |
+
"loss": 0.0084,
|
| 2216 |
+
"step": 8900
|
| 2217 |
+
},
|
| 2218 |
+
{
|
| 2219 |
+
"epoch": 3.14,
|
| 2220 |
+
"learning_rate": 1.1368421052631579e-07,
|
| 2221 |
+
"loss": 0.0109,
|
| 2222 |
+
"step": 8925
|
| 2223 |
+
},
|
| 2224 |
+
{
|
| 2225 |
+
"epoch": 3.15,
|
| 2226 |
+
"learning_rate": 1.1105263157894737e-07,
|
| 2227 |
+
"loss": 0.0084,
|
| 2228 |
+
"step": 8950
|
| 2229 |
+
},
|
| 2230 |
+
{
|
| 2231 |
+
"epoch": 3.16,
|
| 2232 |
+
"learning_rate": 1.0842105263157894e-07,
|
| 2233 |
+
"loss": 0.0041,
|
| 2234 |
+
"step": 8975
|
| 2235 |
+
},
|
| 2236 |
+
{
|
| 2237 |
+
"epoch": 3.17,
|
| 2238 |
+
"learning_rate": 1.0578947368421052e-07,
|
| 2239 |
+
"loss": 0.0101,
|
| 2240 |
+
"step": 9000
|
| 2241 |
+
},
|
| 2242 |
+
{
|
| 2243 |
+
"epoch": 3.17,
|
| 2244 |
+
"eval_loss": 0.07073814421892166,
|
| 2245 |
+
"eval_runtime": 2275.7142,
|
| 2246 |
+
"eval_samples_per_second": 1.59,
|
| 2247 |
+
"eval_steps_per_second": 0.1,
|
| 2248 |
+
"eval_wer": 4.057917550493406,
|
| 2249 |
+
"step": 9000
|
| 2250 |
+
},
|
| 2251 |
+
{
|
| 2252 |
+
"epoch": 3.18,
|
| 2253 |
+
"learning_rate": 1.031578947368421e-07,
|
| 2254 |
+
"loss": 0.0138,
|
| 2255 |
+
"step": 9025
|
| 2256 |
+
},
|
| 2257 |
+
{
|
| 2258 |
+
"epoch": 3.19,
|
| 2259 |
+
"learning_rate": 1.0052631578947368e-07,
|
| 2260 |
+
"loss": 0.0146,
|
| 2261 |
+
"step": 9050
|
| 2262 |
+
},
|
| 2263 |
+
{
|
| 2264 |
+
"epoch": 3.2,
|
| 2265 |
+
"learning_rate": 9.789473684210525e-08,
|
| 2266 |
+
"loss": 0.0059,
|
| 2267 |
+
"step": 9075
|
| 2268 |
+
},
|
| 2269 |
+
{
|
| 2270 |
+
"epoch": 3.21,
|
| 2271 |
+
"learning_rate": 9.526315789473683e-08,
|
| 2272 |
+
"loss": 0.0084,
|
| 2273 |
+
"step": 9100
|
| 2274 |
+
},
|
| 2275 |
+
{
|
| 2276 |
+
"epoch": 3.21,
|
| 2277 |
+
"learning_rate": 9.263157894736842e-08,
|
| 2278 |
+
"loss": 0.0091,
|
| 2279 |
+
"step": 9125
|
| 2280 |
+
},
|
| 2281 |
+
{
|
| 2282 |
+
"epoch": 3.22,
|
| 2283 |
+
"learning_rate": 9e-08,
|
| 2284 |
+
"loss": 0.0107,
|
| 2285 |
+
"step": 9150
|
| 2286 |
+
},
|
| 2287 |
+
{
|
| 2288 |
+
"epoch": 3.23,
|
| 2289 |
+
"learning_rate": 8.736842105263158e-08,
|
| 2290 |
+
"loss": 0.0084,
|
| 2291 |
+
"step": 9175
|
| 2292 |
+
},
|
| 2293 |
+
{
|
| 2294 |
+
"epoch": 3.24,
|
| 2295 |
+
"learning_rate": 8.473684210526315e-08,
|
| 2296 |
+
"loss": 0.0116,
|
| 2297 |
+
"step": 9200
|
| 2298 |
+
},
|
| 2299 |
+
{
|
| 2300 |
+
"epoch": 3.25,
|
| 2301 |
+
"learning_rate": 8.210526315789473e-08,
|
| 2302 |
+
"loss": 0.0109,
|
| 2303 |
+
"step": 9225
|
| 2304 |
+
},
|
| 2305 |
+
{
|
| 2306 |
+
"epoch": 3.26,
|
| 2307 |
+
"learning_rate": 7.947368421052631e-08,
|
| 2308 |
+
"loss": 0.0199,
|
| 2309 |
+
"step": 9250
|
| 2310 |
+
},
|
| 2311 |
+
{
|
| 2312 |
+
"epoch": 3.27,
|
| 2313 |
+
"learning_rate": 7.68421052631579e-08,
|
| 2314 |
+
"loss": 0.0061,
|
| 2315 |
+
"step": 9275
|
| 2316 |
+
},
|
| 2317 |
+
{
|
| 2318 |
+
"epoch": 3.28,
|
| 2319 |
+
"learning_rate": 7.421052631578948e-08,
|
| 2320 |
+
"loss": 0.0067,
|
| 2321 |
+
"step": 9300
|
| 2322 |
+
},
|
| 2323 |
+
{
|
| 2324 |
+
"epoch": 3.28,
|
| 2325 |
+
"learning_rate": 7.157894736842105e-08,
|
| 2326 |
+
"loss": 0.0138,
|
| 2327 |
+
"step": 9325
|
| 2328 |
+
},
|
| 2329 |
+
{
|
| 2330 |
+
"epoch": 3.29,
|
| 2331 |
+
"learning_rate": 6.894736842105263e-08,
|
| 2332 |
+
"loss": 0.0106,
|
| 2333 |
+
"step": 9350
|
| 2334 |
+
},
|
| 2335 |
+
{
|
| 2336 |
+
"epoch": 3.3,
|
| 2337 |
+
"learning_rate": 6.631578947368421e-08,
|
| 2338 |
+
"loss": 0.0176,
|
| 2339 |
+
"step": 9375
|
| 2340 |
+
},
|
| 2341 |
+
{
|
| 2342 |
+
"epoch": 3.31,
|
| 2343 |
+
"learning_rate": 6.368421052631579e-08,
|
| 2344 |
+
"loss": 0.0113,
|
| 2345 |
+
"step": 9400
|
| 2346 |
+
},
|
| 2347 |
+
{
|
| 2348 |
+
"epoch": 3.32,
|
| 2349 |
+
"learning_rate": 6.105263157894736e-08,
|
| 2350 |
+
"loss": 0.0082,
|
| 2351 |
+
"step": 9425
|
| 2352 |
+
},
|
| 2353 |
+
{
|
| 2354 |
+
"epoch": 3.33,
|
| 2355 |
+
"learning_rate": 5.842105263157895e-08,
|
| 2356 |
+
"loss": 0.0142,
|
| 2357 |
+
"step": 9450
|
| 2358 |
+
},
|
| 2359 |
+
{
|
| 2360 |
+
"epoch": 3.34,
|
| 2361 |
+
"learning_rate": 5.578947368421052e-08,
|
| 2362 |
+
"loss": 0.0125,
|
| 2363 |
+
"step": 9475
|
| 2364 |
+
},
|
| 2365 |
+
{
|
| 2366 |
+
"epoch": 3.35,
|
| 2367 |
+
"learning_rate": 5.3157894736842105e-08,
|
| 2368 |
+
"loss": 0.0056,
|
| 2369 |
+
"step": 9500
|
| 2370 |
+
},
|
| 2371 |
+
{
|
| 2372 |
+
"epoch": 3.36,
|
| 2373 |
+
"learning_rate": 5.052631578947368e-08,
|
| 2374 |
+
"loss": 0.0058,
|
| 2375 |
+
"step": 9525
|
| 2376 |
+
},
|
| 2377 |
+
{
|
| 2378 |
+
"epoch": 3.36,
|
| 2379 |
+
"learning_rate": 4.789473684210526e-08,
|
| 2380 |
+
"loss": 0.0143,
|
| 2381 |
+
"step": 9550
|
| 2382 |
+
},
|
| 2383 |
+
{
|
| 2384 |
+
"epoch": 3.37,
|
| 2385 |
+
"learning_rate": 4.5263157894736844e-08,
|
| 2386 |
+
"loss": 0.0075,
|
| 2387 |
+
"step": 9575
|
| 2388 |
+
},
|
| 2389 |
+
{
|
| 2390 |
+
"epoch": 3.38,
|
| 2391 |
+
"learning_rate": 4.263157894736842e-08,
|
| 2392 |
+
"loss": 0.0091,
|
| 2393 |
+
"step": 9600
|
| 2394 |
+
},
|
| 2395 |
+
{
|
| 2396 |
+
"epoch": 3.39,
|
| 2397 |
+
"learning_rate": 4e-08,
|
| 2398 |
+
"loss": 0.0166,
|
| 2399 |
+
"step": 9625
|
| 2400 |
+
},
|
| 2401 |
+
{
|
| 2402 |
+
"epoch": 3.4,
|
| 2403 |
+
"learning_rate": 3.7368421052631576e-08,
|
| 2404 |
+
"loss": 0.0059,
|
| 2405 |
+
"step": 9650
|
| 2406 |
+
},
|
| 2407 |
+
{
|
| 2408 |
+
"epoch": 3.41,
|
| 2409 |
+
"learning_rate": 3.473684210526316e-08,
|
| 2410 |
+
"loss": 0.0074,
|
| 2411 |
+
"step": 9675
|
| 2412 |
+
},
|
| 2413 |
+
{
|
| 2414 |
+
"epoch": 3.42,
|
| 2415 |
+
"learning_rate": 3.210526315789473e-08,
|
| 2416 |
+
"loss": 0.0118,
|
| 2417 |
+
"step": 9700
|
| 2418 |
+
},
|
| 2419 |
+
{
|
| 2420 |
+
"epoch": 3.43,
|
| 2421 |
+
"learning_rate": 2.9473684210526315e-08,
|
| 2422 |
+
"loss": 0.0105,
|
| 2423 |
+
"step": 9725
|
| 2424 |
+
},
|
| 2425 |
+
{
|
| 2426 |
+
"epoch": 3.43,
|
| 2427 |
+
"learning_rate": 2.6842105263157894e-08,
|
| 2428 |
+
"loss": 0.0136,
|
| 2429 |
+
"step": 9750
|
| 2430 |
+
},
|
| 2431 |
+
{
|
| 2432 |
+
"epoch": 3.44,
|
| 2433 |
+
"learning_rate": 2.4210526315789472e-08,
|
| 2434 |
+
"loss": 0.0101,
|
| 2435 |
+
"step": 9775
|
| 2436 |
+
},
|
| 2437 |
+
{
|
| 2438 |
+
"epoch": 3.45,
|
| 2439 |
+
"learning_rate": 2.157894736842105e-08,
|
| 2440 |
+
"loss": 0.006,
|
| 2441 |
+
"step": 9800
|
| 2442 |
+
},
|
| 2443 |
+
{
|
| 2444 |
+
"epoch": 3.46,
|
| 2445 |
+
"learning_rate": 1.8947368421052632e-08,
|
| 2446 |
+
"loss": 0.0052,
|
| 2447 |
+
"step": 9825
|
| 2448 |
+
},
|
| 2449 |
+
{
|
| 2450 |
+
"epoch": 3.47,
|
| 2451 |
+
"learning_rate": 1.631578947368421e-08,
|
| 2452 |
+
"loss": 0.0059,
|
| 2453 |
+
"step": 9850
|
| 2454 |
+
},
|
| 2455 |
+
{
|
| 2456 |
+
"epoch": 3.48,
|
| 2457 |
+
"learning_rate": 1.368421052631579e-08,
|
| 2458 |
+
"loss": 0.0129,
|
| 2459 |
+
"step": 9875
|
| 2460 |
+
},
|
| 2461 |
+
{
|
| 2462 |
+
"epoch": 3.49,
|
| 2463 |
+
"learning_rate": 1.1052631578947366e-08,
|
| 2464 |
+
"loss": 0.0106,
|
| 2465 |
+
"step": 9900
|
| 2466 |
+
},
|
| 2467 |
+
{
|
| 2468 |
+
"epoch": 3.5,
|
| 2469 |
+
"learning_rate": 8.421052631578947e-09,
|
| 2470 |
+
"loss": 0.0042,
|
| 2471 |
+
"step": 9925
|
| 2472 |
+
},
|
| 2473 |
+
{
|
| 2474 |
+
"epoch": 3.5,
|
| 2475 |
+
"learning_rate": 5.789473684210526e-09,
|
| 2476 |
+
"loss": 0.0226,
|
| 2477 |
+
"step": 9950
|
| 2478 |
+
},
|
| 2479 |
+
{
|
| 2480 |
+
"epoch": 3.51,
|
| 2481 |
+
"learning_rate": 3.157894736842105e-09,
|
| 2482 |
+
"loss": 0.0079,
|
| 2483 |
+
"step": 9975
|
| 2484 |
+
},
|
| 2485 |
+
{
|
| 2486 |
+
"epoch": 3.52,
|
| 2487 |
+
"learning_rate": 5.263157894736842e-10,
|
| 2488 |
+
"loss": 0.0076,
|
| 2489 |
+
"step": 10000
|
| 2490 |
+
},
|
| 2491 |
+
{
|
| 2492 |
+
"epoch": 3.52,
|
| 2493 |
+
"eval_loss": 0.07079941779375076,
|
| 2494 |
+
"eval_runtime": 2276.0309,
|
| 2495 |
+
"eval_samples_per_second": 1.59,
|
| 2496 |
+
"eval_steps_per_second": 0.1,
|
| 2497 |
+
"eval_wer": 4.062528820437148,
|
| 2498 |
+
"step": 10000
|
| 2499 |
+
},
|
| 2500 |
+
{
|
| 2501 |
+
"epoch": 3.52,
|
| 2502 |
+
"step": 10000,
|
| 2503 |
+
"total_flos": 1.6326922887069696e+20,
|
| 2504 |
+
"train_loss": 0.019241938863694668,
|
| 2505 |
+
"train_runtime": 61111.4423,
|
| 2506 |
+
"train_samples_per_second": 2.618,
|
| 2507 |
+
"train_steps_per_second": 0.164
|
| 2508 |
}
|
| 2509 |
],
|
| 2510 |
+
"max_steps": 10000,
|
| 2511 |
+
"num_train_epochs": 4,
|
| 2512 |
+
"total_flos": 1.6326922887069696e+20,
|
| 2513 |
"trial_name": null,
|
| 2514 |
"trial_params": null
|
| 2515 |
}
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 3515
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0be83aa1bad695541c98846c215c8c3fe1b0157fb9154418a5317e20a528ddfd
|
| 3 |
size 3515
|