Commit
•
58c7e8e
1
Parent(s):
22780cf
End of training
Browse files- all_results.json +21 -21
- eval_pretrained_results.json +5 -5
- pytorch_model.bin +1 -1
- test_finetuned_results.json +6 -6
- test_pretrained_results.json +5 -5
- train_results.json +6 -6
- trainer_state.json +766 -34
- training_args.bin +1 -1
all_results.json
CHANGED
|
@@ -1,33 +1,33 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
"eval_loss": 1.6191972494125366,
|
| 4 |
-
"eval_pretrained_loss": 1.
|
| 5 |
-
"eval_pretrained_runtime":
|
| 6 |
-
"eval_pretrained_samples_per_second":
|
| 7 |
-
"eval_pretrained_steps_per_second": 0.
|
| 8 |
-
"eval_pretrained_wer":
|
| 9 |
"eval_runtime": 56.3363,
|
| 10 |
"eval_samples_per_second": 0.071,
|
| 11 |
"eval_steps_per_second": 0.036,
|
| 12 |
"eval_wer": 153.2258064516129,
|
| 13 |
-
"test_finetuned_loss": 1.
|
| 14 |
-
"test_finetuned_runtime":
|
| 15 |
-
"test_finetuned_samples_per_second":
|
| 16 |
-
"test_finetuned_steps_per_second": 0.
|
| 17 |
-
"test_finetuned_wer":
|
| 18 |
"test_loss": 1.7568330764770508,
|
| 19 |
-
"test_pretrained_loss": 1.
|
| 20 |
-
"test_pretrained_runtime":
|
| 21 |
-
"test_pretrained_samples_per_second":
|
| 22 |
-
"test_pretrained_steps_per_second": 0.
|
| 23 |
-
"test_pretrained_wer":
|
| 24 |
"test_runtime": 37.8582,
|
| 25 |
"test_samples_per_second": 0.106,
|
| 26 |
"test_steps_per_second": 0.053,
|
| 27 |
"test_wer": 138.5964912280702,
|
| 28 |
-
"total_flos":
|
| 29 |
-
"train_loss":
|
| 30 |
-
"train_runtime":
|
| 31 |
-
"train_samples_per_second":
|
| 32 |
-
"train_steps_per_second": 0.
|
| 33 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 2.33,
|
| 3 |
"eval_loss": 1.6191972494125366,
|
| 4 |
+
"eval_pretrained_loss": 1.715580940246582,
|
| 5 |
+
"eval_pretrained_runtime": 131.9468,
|
| 6 |
+
"eval_pretrained_samples_per_second": 2.501,
|
| 7 |
+
"eval_pretrained_steps_per_second": 0.318,
|
| 8 |
+
"eval_pretrained_wer": 264.5217946670924,
|
| 9 |
"eval_runtime": 56.3363,
|
| 10 |
"eval_samples_per_second": 0.071,
|
| 11 |
"eval_steps_per_second": 0.036,
|
| 12 |
"eval_wer": 153.2258064516129,
|
| 13 |
+
"test_finetuned_loss": 1.0018435716629028,
|
| 14 |
+
"test_finetuned_runtime": 238.9736,
|
| 15 |
+
"test_finetuned_samples_per_second": 3.176,
|
| 16 |
+
"test_finetuned_steps_per_second": 0.398,
|
| 17 |
+
"test_finetuned_wer": 173.15939719843325,
|
| 18 |
"test_loss": 1.7568330764770508,
|
| 19 |
+
"test_pretrained_loss": 1.724961757659912,
|
| 20 |
+
"test_pretrained_runtime": 275.9851,
|
| 21 |
+
"test_pretrained_samples_per_second": 2.75,
|
| 22 |
+
"test_pretrained_steps_per_second": 0.344,
|
| 23 |
+
"test_pretrained_wer": 261.9066587001262,
|
| 24 |
"test_runtime": 37.8582,
|
| 25 |
"test_samples_per_second": 0.106,
|
| 26 |
"test_steps_per_second": 0.053,
|
| 27 |
"test_wer": 138.5964912280702,
|
| 28 |
+
"total_flos": 1.7572960198656e+17,
|
| 29 |
+
"train_loss": 0.8751795228038516,
|
| 30 |
+
"train_runtime": 2133.32,
|
| 31 |
+
"train_samples_per_second": 3.36,
|
| 32 |
+
"train_steps_per_second": 0.21
|
| 33 |
}
|
eval_pretrained_results.json
CHANGED
|
@@ -1,7 +1,7 @@
|
|
| 1 |
{
|
| 2 |
-
"eval_pretrained_loss": 1.
|
| 3 |
-
"eval_pretrained_runtime":
|
| 4 |
-
"eval_pretrained_samples_per_second":
|
| 5 |
-
"eval_pretrained_steps_per_second": 0.
|
| 6 |
-
"eval_pretrained_wer":
|
| 7 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"eval_pretrained_loss": 1.715580940246582,
|
| 3 |
+
"eval_pretrained_runtime": 131.9468,
|
| 4 |
+
"eval_pretrained_samples_per_second": 2.501,
|
| 5 |
+
"eval_pretrained_steps_per_second": 0.318,
|
| 6 |
+
"eval_pretrained_wer": 264.5217946670924
|
| 7 |
}
|
pytorch_model.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 151098921
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:68bb0c59266f7c595eaaa4c27443c5b5bde3796bdbf17a2e1ef41723e9e02297
|
| 3 |
size 151098921
|
test_finetuned_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"test_finetuned_loss": 1.
|
| 4 |
-
"test_finetuned_runtime":
|
| 5 |
-
"test_finetuned_samples_per_second":
|
| 6 |
-
"test_finetuned_steps_per_second": 0.
|
| 7 |
-
"test_finetuned_wer":
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 2.33,
|
| 3 |
+
"test_finetuned_loss": 1.0018435716629028,
|
| 4 |
+
"test_finetuned_runtime": 238.9736,
|
| 5 |
+
"test_finetuned_samples_per_second": 3.176,
|
| 6 |
+
"test_finetuned_steps_per_second": 0.398,
|
| 7 |
+
"test_finetuned_wer": 173.15939719843325
|
| 8 |
}
|
test_pretrained_results.json
CHANGED
|
@@ -1,7 +1,7 @@
|
|
| 1 |
{
|
| 2 |
-
"test_pretrained_loss": 1.
|
| 3 |
-
"test_pretrained_runtime":
|
| 4 |
-
"test_pretrained_samples_per_second":
|
| 5 |
-
"test_pretrained_steps_per_second": 0.
|
| 6 |
-
"test_pretrained_wer":
|
| 7 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"test_pretrained_loss": 1.724961757659912,
|
| 3 |
+
"test_pretrained_runtime": 275.9851,
|
| 4 |
+
"test_pretrained_samples_per_second": 2.75,
|
| 5 |
+
"test_pretrained_steps_per_second": 0.344,
|
| 6 |
+
"test_pretrained_wer": 261.9066587001262
|
| 7 |
}
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"total_flos":
|
| 4 |
-
"train_loss":
|
| 5 |
-
"train_runtime":
|
| 6 |
-
"train_samples_per_second":
|
| 7 |
-
"train_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 2.33,
|
| 3 |
+
"total_flos": 1.7572960198656e+17,
|
| 4 |
+
"train_loss": 0.8751795228038516,
|
| 5 |
+
"train_runtime": 2133.32,
|
| 6 |
+
"train_samples_per_second": 3.36,
|
| 7 |
+
"train_steps_per_second": 0.21
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,55 +1,787 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric":
|
| 3 |
-
"best_model_checkpoint": "./whisper-training-blog/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.
|
| 12 |
-
"learning_rate":
|
| 13 |
-
"loss": 1.
|
| 14 |
-
"step":
|
| 15 |
},
|
| 16 |
{
|
| 17 |
-
"epoch": 0.
|
| 18 |
-
"
|
| 19 |
-
"
|
| 20 |
-
"
|
| 21 |
-
"eval_steps_per_second": 0.294,
|
| 22 |
-
"eval_wer": 123.40425531914893,
|
| 23 |
-
"step": 1
|
| 24 |
},
|
| 25 |
{
|
| 26 |
-
"epoch":
|
| 27 |
-
"learning_rate":
|
| 28 |
-
"loss": 1.
|
| 29 |
-
"step":
|
| 30 |
},
|
| 31 |
{
|
| 32 |
-
"epoch":
|
| 33 |
-
"
|
| 34 |
-
"
|
| 35 |
-
"
|
| 36 |
-
|
| 37 |
-
|
| 38 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 39 |
},
|
| 40 |
{
|
| 41 |
"epoch": 1.0,
|
| 42 |
-
"
|
| 43 |
-
"
|
| 44 |
-
"
|
| 45 |
-
|
| 46 |
-
|
| 47 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 48 |
}
|
| 49 |
],
|
| 50 |
-
"max_steps":
|
| 51 |
"num_train_epochs": 9223372036854775807,
|
| 52 |
-
"total_flos":
|
| 53 |
"trial_name": null,
|
| 54 |
"trial_params": null
|
| 55 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 173.47916334025228,
|
| 3 |
+
"best_model_checkpoint": "./whisper-training-blog/checkpoint-396",
|
| 4 |
+
"epoch": 2.330357142857143,
|
| 5 |
+
"global_step": 448,
|
| 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": 1.6666666666666668e-07,
|
| 13 |
+
"loss": 1.7185,
|
| 14 |
+
"step": 4
|
| 15 |
},
|
| 16 |
{
|
| 17 |
+
"epoch": 0.02,
|
| 18 |
+
"learning_rate": 3.3333333333333335e-07,
|
| 19 |
+
"loss": 1.6808,
|
| 20 |
+
"step": 8
|
|
|
|
|
|
|
|
|
|
| 21 |
},
|
| 22 |
{
|
| 23 |
+
"epoch": 0.03,
|
| 24 |
+
"learning_rate": 5.555555555555555e-07,
|
| 25 |
+
"loss": 1.712,
|
| 26 |
+
"step": 12
|
| 27 |
},
|
| 28 |
{
|
| 29 |
+
"epoch": 0.04,
|
| 30 |
+
"learning_rate": 7.777777777777778e-07,
|
| 31 |
+
"loss": 1.543,
|
| 32 |
+
"step": 16
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.04,
|
| 36 |
+
"learning_rate": 1e-06,
|
| 37 |
+
"loss": 1.609,
|
| 38 |
+
"step": 20
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.05,
|
| 42 |
+
"learning_rate": 1.2222222222222223e-06,
|
| 43 |
+
"loss": 1.5657,
|
| 44 |
+
"step": 24
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.06,
|
| 48 |
+
"learning_rate": 1.4444444444444445e-06,
|
| 49 |
+
"loss": 1.608,
|
| 50 |
+
"step": 28
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.07,
|
| 54 |
+
"learning_rate": 1.6666666666666667e-06,
|
| 55 |
+
"loss": 1.5075,
|
| 56 |
+
"step": 32
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.08,
|
| 60 |
+
"learning_rate": 1.8888888888888888e-06,
|
| 61 |
+
"loss": 1.5183,
|
| 62 |
+
"step": 36
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 0.09,
|
| 66 |
+
"learning_rate": 2.1111111111111114e-06,
|
| 67 |
+
"loss": 1.4948,
|
| 68 |
+
"step": 40
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.1,
|
| 72 |
+
"learning_rate": 2.3333333333333336e-06,
|
| 73 |
+
"loss": 1.4112,
|
| 74 |
+
"step": 44
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.1,
|
| 78 |
+
"eval_loss": 1.4919109344482422,
|
| 79 |
+
"eval_runtime": 118.4455,
|
| 80 |
+
"eval_samples_per_second": 2.786,
|
| 81 |
+
"eval_steps_per_second": 0.355,
|
| 82 |
+
"eval_wer": 245.29778061631805,
|
| 83 |
+
"step": 44
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 0.11,
|
| 87 |
+
"learning_rate": 2.5555555555555557e-06,
|
| 88 |
+
"loss": 1.3192,
|
| 89 |
+
"step": 48
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 0.12,
|
| 93 |
+
"learning_rate": 2.7777777777777775e-06,
|
| 94 |
+
"loss": 1.3334,
|
| 95 |
+
"step": 52
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 0.12,
|
| 99 |
+
"learning_rate": 3e-06,
|
| 100 |
+
"loss": 1.2895,
|
| 101 |
+
"step": 56
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.13,
|
| 105 |
+
"learning_rate": 3.2222222222222222e-06,
|
| 106 |
+
"loss": 1.2749,
|
| 107 |
+
"step": 60
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.14,
|
| 111 |
+
"learning_rate": 3.4444444444444444e-06,
|
| 112 |
+
"loss": 1.3856,
|
| 113 |
+
"step": 64
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 0.15,
|
| 117 |
+
"learning_rate": 3.6666666666666666e-06,
|
| 118 |
+
"loss": 1.2078,
|
| 119 |
+
"step": 68
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 0.16,
|
| 123 |
+
"learning_rate": 3.888888888888889e-06,
|
| 124 |
+
"loss": 1.2185,
|
| 125 |
+
"step": 72
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 0.17,
|
| 129 |
+
"learning_rate": 4.111111111111111e-06,
|
| 130 |
+
"loss": 1.1661,
|
| 131 |
+
"step": 76
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 0.18,
|
| 135 |
+
"learning_rate": 4.333333333333333e-06,
|
| 136 |
+
"loss": 1.1351,
|
| 137 |
+
"step": 80
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 0.19,
|
| 141 |
+
"learning_rate": 4.555555555555555e-06,
|
| 142 |
+
"loss": 1.0729,
|
| 143 |
+
"step": 84
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.2,
|
| 147 |
+
"learning_rate": 4.777777777777778e-06,
|
| 148 |
+
"loss": 1.0501,
|
| 149 |
+
"step": 88
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.2,
|
| 153 |
+
"eval_loss": 1.225467324256897,
|
| 154 |
+
"eval_runtime": 119.277,
|
| 155 |
+
"eval_samples_per_second": 2.767,
|
| 156 |
+
"eval_steps_per_second": 0.352,
|
| 157 |
+
"eval_wer": 219.94251955931662,
|
| 158 |
+
"step": 88
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 0.21,
|
| 162 |
+
"learning_rate": 4.9999999999999996e-06,
|
| 163 |
+
"loss": 1.1278,
|
| 164 |
+
"step": 92
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.21,
|
| 168 |
+
"learning_rate": 5.2222222222222226e-06,
|
| 169 |
+
"loss": 1.0563,
|
| 170 |
+
"step": 96
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.22,
|
| 174 |
+
"learning_rate": 5.444444444444445e-06,
|
| 175 |
+
"loss": 0.9929,
|
| 176 |
+
"step": 100
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 0.23,
|
| 180 |
+
"learning_rate": 5.666666666666667e-06,
|
| 181 |
+
"loss": 1.0879,
|
| 182 |
+
"step": 104
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 0.24,
|
| 186 |
+
"learning_rate": 5.888888888888889e-06,
|
| 187 |
+
"loss": 1.0417,
|
| 188 |
+
"step": 108
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 0.25,
|
| 192 |
+
"learning_rate": 6.111111111111111e-06,
|
| 193 |
+
"loss": 0.9337,
|
| 194 |
+
"step": 112
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 0.26,
|
| 198 |
+
"learning_rate": 6.333333333333333e-06,
|
| 199 |
+
"loss": 0.9688,
|
| 200 |
+
"step": 116
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 0.27,
|
| 204 |
+
"learning_rate": 6.555555555555556e-06,
|
| 205 |
+
"loss": 0.9559,
|
| 206 |
+
"step": 120
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.28,
|
| 210 |
+
"learning_rate": 6.777777777777778e-06,
|
| 211 |
+
"loss": 1.0192,
|
| 212 |
+
"step": 124
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.29,
|
| 216 |
+
"learning_rate": 7e-06,
|
| 217 |
+
"loss": 0.8509,
|
| 218 |
+
"step": 128
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 0.29,
|
| 222 |
+
"learning_rate": 7.222222222222222e-06,
|
| 223 |
+
"loss": 0.9033,
|
| 224 |
+
"step": 132
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 0.29,
|
| 228 |
+
"eval_loss": 1.120314359664917,
|
| 229 |
+
"eval_runtime": 111.0003,
|
| 230 |
+
"eval_samples_per_second": 2.973,
|
| 231 |
+
"eval_steps_per_second": 0.378,
|
| 232 |
+
"eval_wer": 205.7799776464953,
|
| 233 |
+
"step": 132
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.3,
|
| 237 |
+
"learning_rate": 7.444444444444444e-06,
|
| 238 |
+
"loss": 0.9171,
|
| 239 |
+
"step": 136
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 0.31,
|
| 243 |
+
"learning_rate": 7.428115015974441e-06,
|
| 244 |
+
"loss": 0.9445,
|
| 245 |
+
"step": 140
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 0.32,
|
| 249 |
+
"learning_rate": 7.332268370607029e-06,
|
| 250 |
+
"loss": 0.8926,
|
| 251 |
+
"step": 144
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 0.33,
|
| 255 |
+
"learning_rate": 7.236421725239617e-06,
|
| 256 |
+
"loss": 0.9975,
|
| 257 |
+
"step": 148
|
| 258 |
},
|
| 259 |
{
|
| 260 |
"epoch": 1.0,
|
| 261 |
+
"learning_rate": 7.140575079872205e-06,
|
| 262 |
+
"loss": 1.239,
|
| 263 |
+
"step": 152
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 1.01,
|
| 267 |
+
"learning_rate": 7.044728434504793e-06,
|
| 268 |
+
"loss": 0.9148,
|
| 269 |
+
"step": 156
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 1.02,
|
| 273 |
+
"learning_rate": 6.948881789137381e-06,
|
| 274 |
+
"loss": 1.0117,
|
| 275 |
+
"step": 160
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 1.03,
|
| 279 |
+
"learning_rate": 6.853035143769968e-06,
|
| 280 |
+
"loss": 0.9081,
|
| 281 |
+
"step": 164
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 1.04,
|
| 285 |
+
"learning_rate": 6.7571884984025565e-06,
|
| 286 |
+
"loss": 0.8434,
|
| 287 |
+
"step": 168
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 1.05,
|
| 291 |
+
"learning_rate": 6.6613418530351436e-06,
|
| 292 |
+
"loss": 0.8672,
|
| 293 |
+
"step": 172
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 1.06,
|
| 297 |
+
"learning_rate": 6.565495207667732e-06,
|
| 298 |
+
"loss": 0.8142,
|
| 299 |
+
"step": 176
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 1.06,
|
| 303 |
+
"eval_loss": 1.0674585103988647,
|
| 304 |
+
"eval_runtime": 102.4925,
|
| 305 |
+
"eval_samples_per_second": 3.22,
|
| 306 |
+
"eval_steps_per_second": 0.41,
|
| 307 |
+
"eval_wer": 192.87881207089254,
|
| 308 |
+
"step": 176
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 1.07,
|
| 312 |
+
"learning_rate": 6.469648562300319e-06,
|
| 313 |
+
"loss": 0.8329,
|
| 314 |
+
"step": 180
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 1.08,
|
| 318 |
+
"learning_rate": 6.373801916932908e-06,
|
| 319 |
+
"loss": 0.8191,
|
| 320 |
+
"step": 184
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 1.08,
|
| 324 |
+
"learning_rate": 6.277955271565495e-06,
|
| 325 |
+
"loss": 0.8659,
|
| 326 |
+
"step": 188
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 1.09,
|
| 330 |
+
"learning_rate": 6.182108626198084e-06,
|
| 331 |
+
"loss": 0.821,
|
| 332 |
+
"step": 192
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 1.1,
|
| 336 |
+
"learning_rate": 6.086261980830671e-06,
|
| 337 |
+
"loss": 0.8518,
|
| 338 |
+
"step": 196
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 1.11,
|
| 342 |
+
"learning_rate": 5.990415335463259e-06,
|
| 343 |
+
"loss": 0.7813,
|
| 344 |
+
"step": 200
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 1.12,
|
| 348 |
+
"learning_rate": 5.894568690095847e-06,
|
| 349 |
+
"loss": 0.7541,
|
| 350 |
+
"step": 204
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 1.13,
|
| 354 |
+
"learning_rate": 5.798722044728435e-06,
|
| 355 |
+
"loss": 0.7938,
|
| 356 |
+
"step": 208
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 1.14,
|
| 360 |
+
"learning_rate": 5.702875399361023e-06,
|
| 361 |
+
"loss": 0.8182,
|
| 362 |
+
"step": 212
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 1.15,
|
| 366 |
+
"learning_rate": 5.607028753993611e-06,
|
| 367 |
+
"loss": 0.7461,
|
| 368 |
+
"step": 216
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 1.16,
|
| 372 |
+
"learning_rate": 5.5111821086261985e-06,
|
| 373 |
+
"loss": 0.8029,
|
| 374 |
+
"step": 220
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 1.16,
|
| 378 |
+
"eval_loss": 1.0393497943878174,
|
| 379 |
+
"eval_runtime": 101.6759,
|
| 380 |
+
"eval_samples_per_second": 3.246,
|
| 381 |
+
"eval_steps_per_second": 0.413,
|
| 382 |
+
"eval_wer": 178.42886795465432,
|
| 383 |
+
"step": 220
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 1.17,
|
| 387 |
+
"learning_rate": 5.415335463258786e-06,
|
| 388 |
+
"loss": 0.754,
|
| 389 |
+
"step": 224
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 1.17,
|
| 393 |
+
"learning_rate": 5.319488817891374e-06,
|
| 394 |
+
"loss": 0.7662,
|
| 395 |
+
"step": 228
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 1.18,
|
| 399 |
+
"learning_rate": 5.223642172523962e-06,
|
| 400 |
+
"loss": 0.7253,
|
| 401 |
+
"step": 232
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 1.19,
|
| 405 |
+
"learning_rate": 5.127795527156549e-06,
|
| 406 |
+
"loss": 0.6944,
|
| 407 |
+
"step": 236
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"epoch": 1.2,
|
| 411 |
+
"learning_rate": 5.031948881789138e-06,
|
| 412 |
+
"loss": 0.7281,
|
| 413 |
+
"step": 240
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 1.21,
|
| 417 |
+
"learning_rate": 4.936102236421725e-06,
|
| 418 |
+
"loss": 0.7841,
|
| 419 |
+
"step": 244
|
| 420 |
+
},
|
| 421 |
+
{
|
| 422 |
+
"epoch": 1.22,
|
| 423 |
+
"learning_rate": 4.840255591054313e-06,
|
| 424 |
+
"loss": 0.7154,
|
| 425 |
+
"step": 248
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 1.23,
|
| 429 |
+
"learning_rate": 4.744408945686901e-06,
|
| 430 |
+
"loss": 0.7077,
|
| 431 |
+
"step": 252
|
| 432 |
+
},
|
| 433 |
+
{
|
| 434 |
+
"epoch": 1.24,
|
| 435 |
+
"learning_rate": 4.648562300319489e-06,
|
| 436 |
+
"loss": 0.7812,
|
| 437 |
+
"step": 256
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 1.25,
|
| 441 |
+
"learning_rate": 4.552715654952077e-06,
|
| 442 |
+
"loss": 0.6817,
|
| 443 |
+
"step": 260
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 1.25,
|
| 447 |
+
"learning_rate": 4.456869009584665e-06,
|
| 448 |
+
"loss": 0.6324,
|
| 449 |
+
"step": 264
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 1.25,
|
| 453 |
+
"eval_loss": 1.0302220582962036,
|
| 454 |
+
"eval_runtime": 114.7389,
|
| 455 |
+
"eval_samples_per_second": 2.876,
|
| 456 |
+
"eval_steps_per_second": 0.366,
|
| 457 |
+
"eval_wer": 216.6054606418649,
|
| 458 |
+
"step": 264
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 1.26,
|
| 462 |
+
"learning_rate": 4.361022364217253e-06,
|
| 463 |
+
"loss": 0.6826,
|
| 464 |
+
"step": 268
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 1.27,
|
| 468 |
+
"learning_rate": 4.26517571884984e-06,
|
| 469 |
+
"loss": 0.7187,
|
| 470 |
+
"step": 272
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 1.28,
|
| 474 |
+
"learning_rate": 4.169329073482428e-06,
|
| 475 |
+
"loss": 0.6699,
|
| 476 |
+
"step": 276
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 1.29,
|
| 480 |
+
"learning_rate": 4.0734824281150155e-06,
|
| 481 |
+
"loss": 0.5677,
|
| 482 |
+
"step": 280
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 1.3,
|
| 486 |
+
"learning_rate": 3.977635782747604e-06,
|
| 487 |
+
"loss": 0.6833,
|
| 488 |
+
"step": 284
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 1.31,
|
| 492 |
+
"learning_rate": 3.881789137380191e-06,
|
| 493 |
+
"loss": 0.6293,
|
| 494 |
+
"step": 288
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 1.32,
|
| 498 |
+
"learning_rate": 3.7859424920127796e-06,
|
| 499 |
+
"loss": 0.672,
|
| 500 |
+
"step": 292
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 1.33,
|
| 504 |
+
"learning_rate": 3.6900958466453675e-06,
|
| 505 |
+
"loss": 0.7121,
|
| 506 |
+
"step": 296
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 1.33,
|
| 510 |
+
"learning_rate": 3.5942492012779555e-06,
|
| 511 |
+
"loss": 0.8766,
|
| 512 |
+
"step": 300
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 2.01,
|
| 516 |
+
"learning_rate": 3.4984025559105434e-06,
|
| 517 |
+
"loss": 0.6726,
|
| 518 |
+
"step": 304
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 2.02,
|
| 522 |
+
"learning_rate": 3.4025559105431313e-06,
|
| 523 |
+
"loss": 0.6971,
|
| 524 |
+
"step": 308
|
| 525 |
+
},
|
| 526 |
+
{
|
| 527 |
+
"epoch": 2.02,
|
| 528 |
+
"eval_loss": 1.0134836435317993,
|
| 529 |
+
"eval_runtime": 101.0358,
|
| 530 |
+
"eval_samples_per_second": 3.266,
|
| 531 |
+
"eval_steps_per_second": 0.416,
|
| 532 |
+
"eval_wer": 179.3709085102986,
|
| 533 |
+
"step": 308
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 2.03,
|
| 537 |
+
"learning_rate": 3.306709265175719e-06,
|
| 538 |
+
"loss": 0.7432,
|
| 539 |
+
"step": 312
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"epoch": 2.04,
|
| 543 |
+
"learning_rate": 3.2108626198083067e-06,
|
| 544 |
+
"loss": 0.6264,
|
| 545 |
+
"step": 316
|
| 546 |
+
},
|
| 547 |
+
{
|
| 548 |
+
"epoch": 2.04,
|
| 549 |
+
"learning_rate": 3.1150159744408946e-06,
|
| 550 |
+
"loss": 0.6604,
|
| 551 |
+
"step": 320
|
| 552 |
+
},
|
| 553 |
+
{
|
| 554 |
+
"epoch": 2.05,
|
| 555 |
+
"learning_rate": 3.0191693290734825e-06,
|
| 556 |
+
"loss": 0.6409,
|
| 557 |
+
"step": 324
|
| 558 |
+
},
|
| 559 |
+
{
|
| 560 |
+
"epoch": 2.06,
|
| 561 |
+
"learning_rate": 2.9233226837060704e-06,
|
| 562 |
+
"loss": 0.6539,
|
| 563 |
+
"step": 328
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 2.07,
|
| 567 |
+
"learning_rate": 2.8274760383386583e-06,
|
| 568 |
+
"loss": 0.611,
|
| 569 |
+
"step": 332
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 2.08,
|
| 573 |
+
"learning_rate": 2.7316293929712462e-06,
|
| 574 |
+
"loss": 0.6828,
|
| 575 |
+
"step": 336
|
| 576 |
+
},
|
| 577 |
+
{
|
| 578 |
+
"epoch": 2.09,
|
| 579 |
+
"learning_rate": 2.635782747603834e-06,
|
| 580 |
+
"loss": 0.6499,
|
| 581 |
+
"step": 340
|
| 582 |
+
},
|
| 583 |
+
{
|
| 584 |
+
"epoch": 2.1,
|
| 585 |
+
"learning_rate": 2.539936102236422e-06,
|
| 586 |
+
"loss": 0.6642,
|
| 587 |
+
"step": 344
|
| 588 |
+
},
|
| 589 |
+
{
|
| 590 |
+
"epoch": 2.11,
|
| 591 |
+
"learning_rate": 2.44408945686901e-06,
|
| 592 |
+
"loss": 0.6147,
|
| 593 |
+
"step": 348
|
| 594 |
+
},
|
| 595 |
+
{
|
| 596 |
+
"epoch": 2.12,
|
| 597 |
+
"learning_rate": 2.3482428115015974e-06,
|
| 598 |
+
"loss": 0.6051,
|
| 599 |
+
"step": 352
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 2.12,
|
| 603 |
+
"eval_loss": 1.0064767599105835,
|
| 604 |
+
"eval_runtime": 107.523,
|
| 605 |
+
"eval_samples_per_second": 3.069,
|
| 606 |
+
"eval_steps_per_second": 0.391,
|
| 607 |
+
"eval_wer": 194.63515886955133,
|
| 608 |
+
"step": 352
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 2.12,
|
| 612 |
+
"learning_rate": 2.2523961661341854e-06,
|
| 613 |
+
"loss": 0.6237,
|
| 614 |
+
"step": 356
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 2.13,
|
| 618 |
+
"learning_rate": 2.1565495207667733e-06,
|
| 619 |
+
"loss": 0.5992,
|
| 620 |
+
"step": 360
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"epoch": 2.14,
|
| 624 |
+
"learning_rate": 2.060702875399361e-06,
|
| 625 |
+
"loss": 0.6777,
|
| 626 |
+
"step": 364
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 2.15,
|
| 630 |
+
"learning_rate": 1.964856230031949e-06,
|
| 631 |
+
"loss": 0.6012,
|
| 632 |
+
"step": 368
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 2.16,
|
| 636 |
+
"learning_rate": 1.8690095846645368e-06,
|
| 637 |
+
"loss": 0.622,
|
| 638 |
+
"step": 372
|
| 639 |
+
},
|
| 640 |
+
{
|
| 641 |
+
"epoch": 2.17,
|
| 642 |
+
"learning_rate": 1.7731629392971245e-06,
|
| 643 |
+
"loss": 0.6288,
|
| 644 |
+
"step": 376
|
| 645 |
+
},
|
| 646 |
+
{
|
| 647 |
+
"epoch": 2.18,
|
| 648 |
+
"learning_rate": 1.6773162939297124e-06,
|
| 649 |
+
"loss": 0.6079,
|
| 650 |
+
"step": 380
|
| 651 |
+
},
|
| 652 |
+
{
|
| 653 |
+
"epoch": 2.19,
|
| 654 |
+
"learning_rate": 1.5814696485623003e-06,
|
| 655 |
+
"loss": 0.5887,
|
| 656 |
+
"step": 384
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 2.2,
|
| 660 |
+
"learning_rate": 1.4856230031948882e-06,
|
| 661 |
+
"loss": 0.5932,
|
| 662 |
+
"step": 388
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 2.21,
|
| 666 |
+
"learning_rate": 1.389776357827476e-06,
|
| 667 |
+
"loss": 0.6373,
|
| 668 |
+
"step": 392
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 2.21,
|
| 672 |
+
"learning_rate": 1.2939297124600638e-06,
|
| 673 |
+
"loss": 0.6048,
|
| 674 |
+
"step": 396
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 2.21,
|
| 678 |
+
"eval_loss": 1.002966284751892,
|
| 679 |
+
"eval_runtime": 97.8102,
|
| 680 |
+
"eval_samples_per_second": 3.374,
|
| 681 |
+
"eval_steps_per_second": 0.429,
|
| 682 |
+
"eval_wer": 173.47916334025228,
|
| 683 |
+
"step": 396
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 2.22,
|
| 687 |
+
"learning_rate": 1.1980830670926517e-06,
|
| 688 |
+
"loss": 0.5774,
|
| 689 |
+
"step": 400
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 2.23,
|
| 693 |
+
"learning_rate": 1.1022364217252397e-06,
|
| 694 |
+
"loss": 0.6325,
|
| 695 |
+
"step": 404
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 2.24,
|
| 699 |
+
"learning_rate": 1.0063897763578274e-06,
|
| 700 |
+
"loss": 0.601,
|
| 701 |
+
"step": 408
|
| 702 |
+
},
|
| 703 |
+
{
|
| 704 |
+
"epoch": 2.25,
|
| 705 |
+
"learning_rate": 9.105431309904153e-07,
|
| 706 |
+
"loss": 0.5618,
|
| 707 |
+
"step": 412
|
| 708 |
+
},
|
| 709 |
+
{
|
| 710 |
+
"epoch": 2.26,
|
| 711 |
+
"learning_rate": 8.146964856230032e-07,
|
| 712 |
+
"loss": 0.5729,
|
| 713 |
+
"step": 416
|
| 714 |
+
},
|
| 715 |
+
{
|
| 716 |
+
"epoch": 2.27,
|
| 717 |
+
"learning_rate": 7.188498402555911e-07,
|
| 718 |
+
"loss": 0.5755,
|
| 719 |
+
"step": 420
|
| 720 |
+
},
|
| 721 |
+
{
|
| 722 |
+
"epoch": 2.28,
|
| 723 |
+
"learning_rate": 6.230031948881789e-07,
|
| 724 |
+
"loss": 0.6207,
|
| 725 |
+
"step": 424
|
| 726 |
+
},
|
| 727 |
+
{
|
| 728 |
+
"epoch": 2.29,
|
| 729 |
+
"learning_rate": 5.271565495207668e-07,
|
| 730 |
+
"loss": 0.5041,
|
| 731 |
+
"step": 428
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 2.29,
|
| 735 |
+
"learning_rate": 4.313099041533546e-07,
|
| 736 |
+
"loss": 0.5544,
|
| 737 |
+
"step": 432
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 2.3,
|
| 741 |
+
"learning_rate": 3.3546325878594247e-07,
|
| 742 |
+
"loss": 0.5608,
|
| 743 |
+
"step": 436
|
| 744 |
+
},
|
| 745 |
+
{
|
| 746 |
+
"epoch": 2.31,
|
| 747 |
+
"learning_rate": 2.3961661341853033e-07,
|
| 748 |
+
"loss": 0.585,
|
| 749 |
+
"step": 440
|
| 750 |
+
},
|
| 751 |
+
{
|
| 752 |
+
"epoch": 2.31,
|
| 753 |
+
"eval_loss": 1.0049320459365845,
|
| 754 |
+
"eval_runtime": 105.5439,
|
| 755 |
+
"eval_samples_per_second": 3.127,
|
| 756 |
+
"eval_steps_per_second": 0.398,
|
| 757 |
+
"eval_wer": 186.6677311192719,
|
| 758 |
+
"step": 440
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 2.32,
|
| 762 |
+
"learning_rate": 1.437699680511182e-07,
|
| 763 |
+
"loss": 0.5486,
|
| 764 |
+
"step": 444
|
| 765 |
+
},
|
| 766 |
+
{
|
| 767 |
+
"epoch": 2.33,
|
| 768 |
+
"learning_rate": 4.792332268370607e-08,
|
| 769 |
+
"loss": 0.6387,
|
| 770 |
+
"step": 448
|
| 771 |
+
},
|
| 772 |
+
{
|
| 773 |
+
"epoch": 2.33,
|
| 774 |
+
"step": 448,
|
| 775 |
+
"total_flos": 1.7572960198656e+17,
|
| 776 |
+
"train_loss": 0.8751795228038516,
|
| 777 |
+
"train_runtime": 2133.32,
|
| 778 |
+
"train_samples_per_second": 3.36,
|
| 779 |
+
"train_steps_per_second": 0.21
|
| 780 |
}
|
| 781 |
],
|
| 782 |
+
"max_steps": 448,
|
| 783 |
"num_train_epochs": 9223372036854775807,
|
| 784 |
+
"total_flos": 1.7572960198656e+17,
|
| 785 |
"trial_name": null,
|
| 786 |
"trial_params": null
|
| 787 |
}
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 3707
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:098eb422614f2b69cada5c370a7aafa45f3e0897c25ca997abbd0fbb3b31385a
|
| 3 |
size 3707
|