Model save
Browse files- README.md +8 -5
- all_results.json +14 -0
- eval_results.json +9 -0
- pytorch_model.bin +1 -1
- train_results.json +8 -0
- trainer_state.json +526 -0
- training_args.bin +1 -1
README.md
CHANGED
|
@@ -17,8 +17,8 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 17 |
|
| 18 |
This model is a fine-tuned version of [openai/whisper-medium](https://huggingface.co/openai/whisper-medium) on the None dataset.
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
-
- Loss: 0.
|
| 21 |
-
- Wer: 0.
|
| 22 |
|
| 23 |
## Model description
|
| 24 |
|
|
@@ -46,14 +46,17 @@ The following hyperparameters were used during training:
|
|
| 46 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 47 |
- lr_scheduler_type: linear
|
| 48 |
- lr_scheduler_warmup_steps: 500
|
| 49 |
-
- training_steps:
|
| 50 |
|
| 51 |
### Training results
|
| 52 |
|
| 53 |
| Training Loss | Epoch | Step | Validation Loss | Wer |
|
| 54 |
|:-------------:|:-----:|:----:|:---------------:|:------:|
|
| 55 |
-
| 0.
|
| 56 |
-
| 0.
|
|
|
|
|
|
|
|
|
|
| 57 |
|
| 58 |
|
| 59 |
### Framework versions
|
|
|
|
| 17 |
|
| 18 |
This model is a fine-tuned version of [openai/whisper-medium](https://huggingface.co/openai/whisper-medium) on the None dataset.
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
+
- Loss: 0.3972
|
| 21 |
+
- Wer: 0.2487
|
| 22 |
|
| 23 |
## Model description
|
| 24 |
|
|
|
|
| 46 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 47 |
- lr_scheduler_type: linear
|
| 48 |
- lr_scheduler_warmup_steps: 500
|
| 49 |
+
- training_steps: 5000
|
| 50 |
|
| 51 |
### Training results
|
| 52 |
|
| 53 |
| Training Loss | Epoch | Step | Validation Loss | Wer |
|
| 54 |
|:-------------:|:-----:|:----:|:---------------:|:------:|
|
| 55 |
+
| 0.0282 | 4.89 | 1000 | 0.2700 | 0.2647 |
|
| 56 |
+
| 0.0025 | 9.78 | 2000 | 0.3434 | 0.2554 |
|
| 57 |
+
| 0.0001 | 14.67 | 3000 | 0.3640 | 0.2471 |
|
| 58 |
+
| 0.0 | 19.56 | 4000 | 0.3902 | 0.2494 |
|
| 59 |
+
| 0.0 | 24.45 | 5000 | 0.3972 | 0.2487 |
|
| 60 |
|
| 61 |
|
| 62 |
### Framework versions
|
all_results.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 9.78,
|
| 3 |
+
"eval_loss": 0.3270670771598816,
|
| 4 |
+
"eval_runtime": 1012.2261,
|
| 5 |
+
"eval_samples": 2894,
|
| 6 |
+
"eval_samples_per_second": 2.859,
|
| 7 |
+
"eval_steps_per_second": 0.179,
|
| 8 |
+
"eval_wer": 0.2435029205113011,
|
| 9 |
+
"train_loss": 0.06386898325150833,
|
| 10 |
+
"train_runtime": 15830.7471,
|
| 11 |
+
"train_samples": 6540,
|
| 12 |
+
"train_samples_per_second": 4.043,
|
| 13 |
+
"train_steps_per_second": 0.126
|
| 14 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 9.78,
|
| 3 |
+
"eval_loss": 0.3270670771598816,
|
| 4 |
+
"eval_runtime": 1012.2261,
|
| 5 |
+
"eval_samples": 2894,
|
| 6 |
+
"eval_samples_per_second": 2.859,
|
| 7 |
+
"eval_steps_per_second": 0.179,
|
| 8 |
+
"eval_wer": 0.2435029205113011
|
| 9 |
+
}
|
pytorch_model.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 3055754969
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8467f4a282e9b110cd68a87ca9bfdf9365e37d6f07bb52c527517b55c463fc9e
|
| 3 |
size 3055754969
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 9.78,
|
| 3 |
+
"train_loss": 0.06386898325150833,
|
| 4 |
+
"train_runtime": 15830.7471,
|
| 5 |
+
"train_samples": 6540,
|
| 6 |
+
"train_samples_per_second": 4.043,
|
| 7 |
+
"train_steps_per_second": 0.126
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,526 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 9.7799511002445,
|
| 5 |
+
"eval_steps": 1000,
|
| 6 |
+
"global_step": 2000,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.12,
|
| 13 |
+
"learning_rate": 5.000000000000001e-07,
|
| 14 |
+
"loss": 0.6666,
|
| 15 |
+
"step": 25
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 0.24,
|
| 19 |
+
"learning_rate": 1.0000000000000002e-06,
|
| 20 |
+
"loss": 0.4618,
|
| 21 |
+
"step": 50
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
"epoch": 0.37,
|
| 25 |
+
"learning_rate": 1.5e-06,
|
| 26 |
+
"loss": 0.4005,
|
| 27 |
+
"step": 75
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 0.49,
|
| 31 |
+
"learning_rate": 2.0000000000000003e-06,
|
| 32 |
+
"loss": 0.2467,
|
| 33 |
+
"step": 100
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 0.61,
|
| 37 |
+
"learning_rate": 2.5e-06,
|
| 38 |
+
"loss": 0.2931,
|
| 39 |
+
"step": 125
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.73,
|
| 43 |
+
"learning_rate": 3e-06,
|
| 44 |
+
"loss": 0.1967,
|
| 45 |
+
"step": 150
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.86,
|
| 49 |
+
"learning_rate": 3.5e-06,
|
| 50 |
+
"loss": 0.253,
|
| 51 |
+
"step": 175
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.98,
|
| 55 |
+
"learning_rate": 4.000000000000001e-06,
|
| 56 |
+
"loss": 0.1654,
|
| 57 |
+
"step": 200
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 1.1,
|
| 61 |
+
"learning_rate": 4.5e-06,
|
| 62 |
+
"loss": 0.1921,
|
| 63 |
+
"step": 225
|
| 64 |
+
},
|
| 65 |
+
{
|
| 66 |
+
"epoch": 1.22,
|
| 67 |
+
"learning_rate": 5e-06,
|
| 68 |
+
"loss": 0.1238,
|
| 69 |
+
"step": 250
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 1.34,
|
| 73 |
+
"learning_rate": 5.500000000000001e-06,
|
| 74 |
+
"loss": 0.1626,
|
| 75 |
+
"step": 275
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 1.47,
|
| 79 |
+
"learning_rate": 6e-06,
|
| 80 |
+
"loss": 0.1193,
|
| 81 |
+
"step": 300
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 1.59,
|
| 85 |
+
"learning_rate": 6.5000000000000004e-06,
|
| 86 |
+
"loss": 0.1468,
|
| 87 |
+
"step": 325
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 1.71,
|
| 91 |
+
"learning_rate": 7e-06,
|
| 92 |
+
"loss": 0.1096,
|
| 93 |
+
"step": 350
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 1.83,
|
| 97 |
+
"learning_rate": 7.500000000000001e-06,
|
| 98 |
+
"loss": 0.1499,
|
| 99 |
+
"step": 375
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 1.96,
|
| 103 |
+
"learning_rate": 8.000000000000001e-06,
|
| 104 |
+
"loss": 0.109,
|
| 105 |
+
"step": 400
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 2.08,
|
| 109 |
+
"learning_rate": 8.5e-06,
|
| 110 |
+
"loss": 0.1019,
|
| 111 |
+
"step": 425
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 2.2,
|
| 115 |
+
"learning_rate": 9e-06,
|
| 116 |
+
"loss": 0.0609,
|
| 117 |
+
"step": 450
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"epoch": 2.32,
|
| 121 |
+
"learning_rate": 9.5e-06,
|
| 122 |
+
"loss": 0.0767,
|
| 123 |
+
"step": 475
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 2.44,
|
| 127 |
+
"learning_rate": 1e-05,
|
| 128 |
+
"loss": 0.0703,
|
| 129 |
+
"step": 500
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 2.57,
|
| 133 |
+
"learning_rate": 9.833333333333333e-06,
|
| 134 |
+
"loss": 0.0874,
|
| 135 |
+
"step": 525
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 2.69,
|
| 139 |
+
"learning_rate": 9.666666666666667e-06,
|
| 140 |
+
"loss": 0.074,
|
| 141 |
+
"step": 550
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 2.81,
|
| 145 |
+
"learning_rate": 9.5e-06,
|
| 146 |
+
"loss": 0.0868,
|
| 147 |
+
"step": 575
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 2.93,
|
| 151 |
+
"learning_rate": 9.333333333333334e-06,
|
| 152 |
+
"loss": 0.0697,
|
| 153 |
+
"step": 600
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 3.06,
|
| 157 |
+
"learning_rate": 9.166666666666666e-06,
|
| 158 |
+
"loss": 0.0552,
|
| 159 |
+
"step": 625
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 3.18,
|
| 163 |
+
"learning_rate": 9e-06,
|
| 164 |
+
"loss": 0.038,
|
| 165 |
+
"step": 650
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 3.3,
|
| 169 |
+
"learning_rate": 8.833333333333334e-06,
|
| 170 |
+
"loss": 0.0427,
|
| 171 |
+
"step": 675
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 3.42,
|
| 175 |
+
"learning_rate": 8.666666666666668e-06,
|
| 176 |
+
"loss": 0.0363,
|
| 177 |
+
"step": 700
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 3.55,
|
| 181 |
+
"learning_rate": 8.5e-06,
|
| 182 |
+
"loss": 0.0422,
|
| 183 |
+
"step": 725
|
| 184 |
+
},
|
| 185 |
+
{
|
| 186 |
+
"epoch": 3.67,
|
| 187 |
+
"learning_rate": 8.333333333333334e-06,
|
| 188 |
+
"loss": 0.0396,
|
| 189 |
+
"step": 750
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 3.79,
|
| 193 |
+
"learning_rate": 8.166666666666668e-06,
|
| 194 |
+
"loss": 0.0465,
|
| 195 |
+
"step": 775
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 3.91,
|
| 199 |
+
"learning_rate": 8.000000000000001e-06,
|
| 200 |
+
"loss": 0.0413,
|
| 201 |
+
"step": 800
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 4.03,
|
| 205 |
+
"learning_rate": 7.833333333333333e-06,
|
| 206 |
+
"loss": 0.0308,
|
| 207 |
+
"step": 825
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 4.16,
|
| 211 |
+
"learning_rate": 7.666666666666667e-06,
|
| 212 |
+
"loss": 0.0211,
|
| 213 |
+
"step": 850
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 4.28,
|
| 217 |
+
"learning_rate": 7.500000000000001e-06,
|
| 218 |
+
"loss": 0.0206,
|
| 219 |
+
"step": 875
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 4.4,
|
| 223 |
+
"learning_rate": 7.333333333333333e-06,
|
| 224 |
+
"loss": 0.0225,
|
| 225 |
+
"step": 900
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 4.52,
|
| 229 |
+
"learning_rate": 7.166666666666667e-06,
|
| 230 |
+
"loss": 0.0206,
|
| 231 |
+
"step": 925
|
| 232 |
+
},
|
| 233 |
+
{
|
| 234 |
+
"epoch": 4.65,
|
| 235 |
+
"learning_rate": 7e-06,
|
| 236 |
+
"loss": 0.0248,
|
| 237 |
+
"step": 950
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 4.77,
|
| 241 |
+
"learning_rate": 6.833333333333334e-06,
|
| 242 |
+
"loss": 0.0228,
|
| 243 |
+
"step": 975
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 4.89,
|
| 247 |
+
"learning_rate": 6.666666666666667e-06,
|
| 248 |
+
"loss": 0.0233,
|
| 249 |
+
"step": 1000
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 4.89,
|
| 253 |
+
"eval_loss": 0.26726391911506653,
|
| 254 |
+
"eval_runtime": 1030.3192,
|
| 255 |
+
"eval_samples_per_second": 2.809,
|
| 256 |
+
"eval_steps_per_second": 0.176,
|
| 257 |
+
"eval_wer": 0.264200457123508,
|
| 258 |
+
"step": 1000
|
| 259 |
+
},
|
| 260 |
+
{
|
| 261 |
+
"epoch": 5.01,
|
| 262 |
+
"learning_rate": 6.5000000000000004e-06,
|
| 263 |
+
"loss": 0.0171,
|
| 264 |
+
"step": 1025
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 5.13,
|
| 268 |
+
"learning_rate": 6.333333333333333e-06,
|
| 269 |
+
"loss": 0.0122,
|
| 270 |
+
"step": 1050
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 5.26,
|
| 274 |
+
"learning_rate": 6.166666666666667e-06,
|
| 275 |
+
"loss": 0.009,
|
| 276 |
+
"step": 1075
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 5.38,
|
| 280 |
+
"learning_rate": 6e-06,
|
| 281 |
+
"loss": 0.0143,
|
| 282 |
+
"step": 1100
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 5.5,
|
| 286 |
+
"learning_rate": 5.833333333333334e-06,
|
| 287 |
+
"loss": 0.0093,
|
| 288 |
+
"step": 1125
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 5.62,
|
| 292 |
+
"learning_rate": 5.666666666666667e-06,
|
| 293 |
+
"loss": 0.0116,
|
| 294 |
+
"step": 1150
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 5.75,
|
| 298 |
+
"learning_rate": 5.500000000000001e-06,
|
| 299 |
+
"loss": 0.01,
|
| 300 |
+
"step": 1175
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"epoch": 5.87,
|
| 304 |
+
"learning_rate": 5.333333333333334e-06,
|
| 305 |
+
"loss": 0.0108,
|
| 306 |
+
"step": 1200
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 5.99,
|
| 310 |
+
"learning_rate": 5.1666666666666675e-06,
|
| 311 |
+
"loss": 0.0061,
|
| 312 |
+
"step": 1225
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 6.11,
|
| 316 |
+
"learning_rate": 5e-06,
|
| 317 |
+
"loss": 0.0072,
|
| 318 |
+
"step": 1250
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 6.23,
|
| 322 |
+
"learning_rate": 4.833333333333333e-06,
|
| 323 |
+
"loss": 0.0031,
|
| 324 |
+
"step": 1275
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 6.36,
|
| 328 |
+
"learning_rate": 4.666666666666667e-06,
|
| 329 |
+
"loss": 0.0058,
|
| 330 |
+
"step": 1300
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 6.48,
|
| 334 |
+
"learning_rate": 4.5e-06,
|
| 335 |
+
"loss": 0.0028,
|
| 336 |
+
"step": 1325
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 6.6,
|
| 340 |
+
"learning_rate": 4.333333333333334e-06,
|
| 341 |
+
"loss": 0.0057,
|
| 342 |
+
"step": 1350
|
| 343 |
+
},
|
| 344 |
+
{
|
| 345 |
+
"epoch": 6.72,
|
| 346 |
+
"learning_rate": 4.166666666666667e-06,
|
| 347 |
+
"loss": 0.0034,
|
| 348 |
+
"step": 1375
|
| 349 |
+
},
|
| 350 |
+
{
|
| 351 |
+
"epoch": 6.85,
|
| 352 |
+
"learning_rate": 4.000000000000001e-06,
|
| 353 |
+
"loss": 0.0052,
|
| 354 |
+
"step": 1400
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 6.97,
|
| 358 |
+
"learning_rate": 3.833333333333334e-06,
|
| 359 |
+
"loss": 0.0023,
|
| 360 |
+
"step": 1425
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 7.09,
|
| 364 |
+
"learning_rate": 3.6666666666666666e-06,
|
| 365 |
+
"loss": 0.0032,
|
| 366 |
+
"step": 1450
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 7.21,
|
| 370 |
+
"learning_rate": 3.5e-06,
|
| 371 |
+
"loss": 0.0011,
|
| 372 |
+
"step": 1475
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"epoch": 7.33,
|
| 376 |
+
"learning_rate": 3.3333333333333333e-06,
|
| 377 |
+
"loss": 0.0017,
|
| 378 |
+
"step": 1500
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"epoch": 7.46,
|
| 382 |
+
"learning_rate": 3.1666666666666667e-06,
|
| 383 |
+
"loss": 0.0012,
|
| 384 |
+
"step": 1525
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 7.58,
|
| 388 |
+
"learning_rate": 3e-06,
|
| 389 |
+
"loss": 0.002,
|
| 390 |
+
"step": 1550
|
| 391 |
+
},
|
| 392 |
+
{
|
| 393 |
+
"epoch": 7.7,
|
| 394 |
+
"learning_rate": 2.8333333333333335e-06,
|
| 395 |
+
"loss": 0.0008,
|
| 396 |
+
"step": 1575
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"epoch": 7.82,
|
| 400 |
+
"learning_rate": 2.666666666666667e-06,
|
| 401 |
+
"loss": 0.0018,
|
| 402 |
+
"step": 1600
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 7.95,
|
| 406 |
+
"learning_rate": 2.5e-06,
|
| 407 |
+
"loss": 0.0017,
|
| 408 |
+
"step": 1625
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 8.07,
|
| 412 |
+
"learning_rate": 2.3333333333333336e-06,
|
| 413 |
+
"loss": 0.0008,
|
| 414 |
+
"step": 1650
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 8.19,
|
| 418 |
+
"learning_rate": 2.166666666666667e-06,
|
| 419 |
+
"loss": 0.0008,
|
| 420 |
+
"step": 1675
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"epoch": 8.31,
|
| 424 |
+
"learning_rate": 2.0000000000000003e-06,
|
| 425 |
+
"loss": 0.0007,
|
| 426 |
+
"step": 1700
|
| 427 |
+
},
|
| 428 |
+
{
|
| 429 |
+
"epoch": 8.44,
|
| 430 |
+
"learning_rate": 1.8333333333333333e-06,
|
| 431 |
+
"loss": 0.0004,
|
| 432 |
+
"step": 1725
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 8.56,
|
| 436 |
+
"learning_rate": 1.6666666666666667e-06,
|
| 437 |
+
"loss": 0.0007,
|
| 438 |
+
"step": 1750
|
| 439 |
+
},
|
| 440 |
+
{
|
| 441 |
+
"epoch": 8.68,
|
| 442 |
+
"learning_rate": 1.5e-06,
|
| 443 |
+
"loss": 0.0004,
|
| 444 |
+
"step": 1775
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 8.8,
|
| 448 |
+
"learning_rate": 1.3333333333333334e-06,
|
| 449 |
+
"loss": 0.0005,
|
| 450 |
+
"step": 1800
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 8.92,
|
| 454 |
+
"learning_rate": 1.1666666666666668e-06,
|
| 455 |
+
"loss": 0.0005,
|
| 456 |
+
"step": 1825
|
| 457 |
+
},
|
| 458 |
+
{
|
| 459 |
+
"epoch": 9.05,
|
| 460 |
+
"learning_rate": 1.0000000000000002e-06,
|
| 461 |
+
"loss": 0.0004,
|
| 462 |
+
"step": 1850
|
| 463 |
+
},
|
| 464 |
+
{
|
| 465 |
+
"epoch": 9.17,
|
| 466 |
+
"learning_rate": 8.333333333333333e-07,
|
| 467 |
+
"loss": 0.0003,
|
| 468 |
+
"step": 1875
|
| 469 |
+
},
|
| 470 |
+
{
|
| 471 |
+
"epoch": 9.29,
|
| 472 |
+
"learning_rate": 6.666666666666667e-07,
|
| 473 |
+
"loss": 0.0004,
|
| 474 |
+
"step": 1900
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 9.41,
|
| 478 |
+
"learning_rate": 5.000000000000001e-07,
|
| 479 |
+
"loss": 0.0003,
|
| 480 |
+
"step": 1925
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 9.54,
|
| 484 |
+
"learning_rate": 3.3333333333333335e-07,
|
| 485 |
+
"loss": 0.0004,
|
| 486 |
+
"step": 1950
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 9.66,
|
| 490 |
+
"learning_rate": 1.6666666666666668e-07,
|
| 491 |
+
"loss": 0.0003,
|
| 492 |
+
"step": 1975
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 9.78,
|
| 496 |
+
"learning_rate": 0.0,
|
| 497 |
+
"loss": 0.0003,
|
| 498 |
+
"step": 2000
|
| 499 |
+
},
|
| 500 |
+
{
|
| 501 |
+
"epoch": 9.78,
|
| 502 |
+
"eval_loss": 0.3270670771598816,
|
| 503 |
+
"eval_runtime": 1014.5775,
|
| 504 |
+
"eval_samples_per_second": 2.852,
|
| 505 |
+
"eval_steps_per_second": 0.178,
|
| 506 |
+
"eval_wer": 0.2435029205113011,
|
| 507 |
+
"step": 2000
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 9.78,
|
| 511 |
+
"step": 2000,
|
| 512 |
+
"total_flos": 6.528197230460928e+19,
|
| 513 |
+
"train_loss": 0.06386898325150833,
|
| 514 |
+
"train_runtime": 15830.7471,
|
| 515 |
+
"train_samples_per_second": 4.043,
|
| 516 |
+
"train_steps_per_second": 0.126
|
| 517 |
+
}
|
| 518 |
+
],
|
| 519 |
+
"logging_steps": 25,
|
| 520 |
+
"max_steps": 2000,
|
| 521 |
+
"num_train_epochs": 10,
|
| 522 |
+
"save_steps": 1000,
|
| 523 |
+
"total_flos": 6.528197230460928e+19,
|
| 524 |
+
"trial_name": null,
|
| 525 |
+
"trial_params": null
|
| 526 |
+
}
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4219
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:eb08232344994b20ec55fad2641a1ee6b0f5f04bbc511c57fee1e6352f8e6a26
|
| 3 |
size 4219
|