Aline Rodrigues
commited on
Commit
•
5b65ce1
1
Parent(s):
65ad224
End of training
Browse files- all_results.json +16 -0
- eval_results.json +10 -0
- train_results.json +9 -0
- trainer_state.json +1129 -0
all_results.json
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 27.0,
|
| 3 |
+
"eval_cer": 0.026817187988041585,
|
| 4 |
+
"eval_loss": 0.13566730916500092,
|
| 5 |
+
"eval_runtime": 6.4693,
|
| 6 |
+
"eval_samples": 341,
|
| 7 |
+
"eval_samples_per_second": 52.711,
|
| 8 |
+
"eval_steps_per_second": 6.647,
|
| 9 |
+
"eval_wer": 0.08438317090563346,
|
| 10 |
+
"total_flos": 1.1392249463344112e+20,
|
| 11 |
+
"train_loss": 0.5354501710375033,
|
| 12 |
+
"train_runtime": 10852.3147,
|
| 13 |
+
"train_samples": 16458,
|
| 14 |
+
"train_samples_per_second": 151.654,
|
| 15 |
+
"train_steps_per_second": 4.736
|
| 16 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 27.0,
|
| 3 |
+
"eval_cer": 0.026817187988041585,
|
| 4 |
+
"eval_loss": 0.13566730916500092,
|
| 5 |
+
"eval_runtime": 6.4693,
|
| 6 |
+
"eval_samples": 341,
|
| 7 |
+
"eval_samples_per_second": 52.711,
|
| 8 |
+
"eval_steps_per_second": 6.647,
|
| 9 |
+
"eval_wer": 0.08438317090563346
|
| 10 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 27.0,
|
| 3 |
+
"total_flos": 1.1392249463344112e+20,
|
| 4 |
+
"train_loss": 0.5354501710375033,
|
| 5 |
+
"train_runtime": 10852.3147,
|
| 6 |
+
"train_samples": 16458,
|
| 7 |
+
"train_samples_per_second": 151.654,
|
| 8 |
+
"train_steps_per_second": 4.736
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1129 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.13566730916500092,
|
| 3 |
+
"best_model_checkpoint": "wav2vec2-large-xlsr-mecita-coraa-portuguese-aug-random-all-03/checkpoint-3601",
|
| 4 |
+
"epoch": 26.99902818270165,
|
| 5 |
+
"global_step": 13891,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.0,
|
| 12 |
+
"learning_rate": 3e-05,
|
| 13 |
+
"loss": 26.0165,
|
| 14 |
+
"step": 1
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.19,
|
| 18 |
+
"learning_rate": 2.9943968871595333e-05,
|
| 19 |
+
"loss": 7.0565,
|
| 20 |
+
"step": 100
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.39,
|
| 24 |
+
"learning_rate": 2.9885603112840468e-05,
|
| 25 |
+
"loss": 3.041,
|
| 26 |
+
"step": 200
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.58,
|
| 30 |
+
"learning_rate": 2.9827237354085602e-05,
|
| 31 |
+
"loss": 2.9395,
|
| 32 |
+
"step": 300
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.78,
|
| 36 |
+
"learning_rate": 2.976887159533074e-05,
|
| 37 |
+
"loss": 2.9123,
|
| 38 |
+
"step": 400
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.97,
|
| 42 |
+
"learning_rate": 2.9710505836575875e-05,
|
| 43 |
+
"loss": 2.883,
|
| 44 |
+
"step": 500
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 1.0,
|
| 48 |
+
"eval_cer": 0.9943331399759047,
|
| 49 |
+
"eval_loss": 2.588413715362549,
|
| 50 |
+
"eval_runtime": 6.3319,
|
| 51 |
+
"eval_samples_per_second": 53.855,
|
| 52 |
+
"eval_steps_per_second": 6.791,
|
| 53 |
+
"eval_wer": 0.9954837176135013,
|
| 54 |
+
"step": 514
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 1.17,
|
| 58 |
+
"learning_rate": 2.9652140077821013e-05,
|
| 59 |
+
"loss": 2.4083,
|
| 60 |
+
"step": 600
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 1.36,
|
| 64 |
+
"learning_rate": 2.9593774319066147e-05,
|
| 65 |
+
"loss": 1.4564,
|
| 66 |
+
"step": 700
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 1.55,
|
| 70 |
+
"learning_rate": 2.9535408560311285e-05,
|
| 71 |
+
"loss": 1.2033,
|
| 72 |
+
"step": 800
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 1.75,
|
| 76 |
+
"learning_rate": 2.947704280155642e-05,
|
| 77 |
+
"loss": 1.0371,
|
| 78 |
+
"step": 900
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"epoch": 1.94,
|
| 82 |
+
"learning_rate": 2.9418677042801558e-05,
|
| 83 |
+
"loss": 0.9481,
|
| 84 |
+
"step": 1000
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 2.0,
|
| 88 |
+
"eval_cer": 0.04359466333496943,
|
| 89 |
+
"eval_loss": 0.2559501826763153,
|
| 90 |
+
"eval_runtime": 6.3303,
|
| 91 |
+
"eval_samples_per_second": 53.868,
|
| 92 |
+
"eval_steps_per_second": 6.793,
|
| 93 |
+
"eval_wer": 0.14737342524364155,
|
| 94 |
+
"step": 1029
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 2.14,
|
| 98 |
+
"learning_rate": 2.9360311284046692e-05,
|
| 99 |
+
"loss": 0.8883,
|
| 100 |
+
"step": 1100
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 2.33,
|
| 104 |
+
"learning_rate": 2.930194552529183e-05,
|
| 105 |
+
"loss": 0.8307,
|
| 106 |
+
"step": 1200
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 2.53,
|
| 110 |
+
"learning_rate": 2.9243579766536965e-05,
|
| 111 |
+
"loss": 0.7553,
|
| 112 |
+
"step": 1300
|
| 113 |
+
},
|
| 114 |
+
{
|
| 115 |
+
"epoch": 2.72,
|
| 116 |
+
"learning_rate": 2.9185214007782103e-05,
|
| 117 |
+
"loss": 0.7774,
|
| 118 |
+
"step": 1400
|
| 119 |
+
},
|
| 120 |
+
{
|
| 121 |
+
"epoch": 2.92,
|
| 122 |
+
"learning_rate": 2.9126848249027238e-05,
|
| 123 |
+
"loss": 0.742,
|
| 124 |
+
"step": 1500
|
| 125 |
+
},
|
| 126 |
+
{
|
| 127 |
+
"epoch": 3.0,
|
| 128 |
+
"eval_cer": 0.03404578108964348,
|
| 129 |
+
"eval_loss": 0.18016588687896729,
|
| 130 |
+
"eval_runtime": 7.4035,
|
| 131 |
+
"eval_samples_per_second": 46.059,
|
| 132 |
+
"eval_steps_per_second": 5.808,
|
| 133 |
+
"eval_wer": 0.10981697171381032,
|
| 134 |
+
"step": 1543
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 3.11,
|
| 138 |
+
"learning_rate": 2.9068482490272376e-05,
|
| 139 |
+
"loss": 0.684,
|
| 140 |
+
"step": 1600
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 3.3,
|
| 144 |
+
"learning_rate": 2.9010116731517514e-05,
|
| 145 |
+
"loss": 0.6992,
|
| 146 |
+
"step": 1700
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 3.5,
|
| 150 |
+
"learning_rate": 2.8951750972762645e-05,
|
| 151 |
+
"loss": 0.6371,
|
| 152 |
+
"step": 1800
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 3.69,
|
| 156 |
+
"learning_rate": 2.8893385214007783e-05,
|
| 157 |
+
"loss": 0.6504,
|
| 158 |
+
"step": 1900
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 3.89,
|
| 162 |
+
"learning_rate": 2.8835019455252917e-05,
|
| 163 |
+
"loss": 0.625,
|
| 164 |
+
"step": 2000
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 4.0,
|
| 168 |
+
"eval_cer": 0.030833073044487083,
|
| 169 |
+
"eval_loss": 0.1589926928281784,
|
| 170 |
+
"eval_runtime": 7.3132,
|
| 171 |
+
"eval_samples_per_second": 46.628,
|
| 172 |
+
"eval_steps_per_second": 5.88,
|
| 173 |
+
"eval_wer": 0.09745661991918232,
|
| 174 |
+
"step": 2058
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 4.08,
|
| 178 |
+
"learning_rate": 2.8776653696498055e-05,
|
| 179 |
+
"loss": 0.641,
|
| 180 |
+
"step": 2100
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"epoch": 4.28,
|
| 184 |
+
"learning_rate": 2.871828793774319e-05,
|
| 185 |
+
"loss": 0.6105,
|
| 186 |
+
"step": 2200
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 4.47,
|
| 190 |
+
"learning_rate": 2.8659922178988328e-05,
|
| 191 |
+
"loss": 0.5456,
|
| 192 |
+
"step": 2300
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 4.66,
|
| 196 |
+
"learning_rate": 2.8601556420233462e-05,
|
| 197 |
+
"loss": 0.611,
|
| 198 |
+
"step": 2400
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 4.86,
|
| 202 |
+
"learning_rate": 2.85431906614786e-05,
|
| 203 |
+
"loss": 0.6001,
|
| 204 |
+
"step": 2500
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 5.0,
|
| 208 |
+
"eval_cer": 0.029226719021908883,
|
| 209 |
+
"eval_loss": 0.14863738417625427,
|
| 210 |
+
"eval_runtime": 6.6768,
|
| 211 |
+
"eval_samples_per_second": 51.072,
|
| 212 |
+
"eval_steps_per_second": 6.44,
|
| 213 |
+
"eval_wer": 0.08866175421915855,
|
| 214 |
+
"step": 2572
|
| 215 |
+
},
|
| 216 |
+
{
|
| 217 |
+
"epoch": 5.05,
|
| 218 |
+
"learning_rate": 2.8484824902723735e-05,
|
| 219 |
+
"loss": 0.5914,
|
| 220 |
+
"step": 2600
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 5.25,
|
| 224 |
+
"learning_rate": 2.8426459143968873e-05,
|
| 225 |
+
"loss": 0.5734,
|
| 226 |
+
"step": 2700
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 5.44,
|
| 230 |
+
"learning_rate": 2.8368093385214007e-05,
|
| 231 |
+
"loss": 0.5532,
|
| 232 |
+
"step": 2800
|
| 233 |
+
},
|
| 234 |
+
{
|
| 235 |
+
"epoch": 5.64,
|
| 236 |
+
"learning_rate": 2.8309727626459145e-05,
|
| 237 |
+
"loss": 0.5305,
|
| 238 |
+
"step": 2900
|
| 239 |
+
},
|
| 240 |
+
{
|
| 241 |
+
"epoch": 5.83,
|
| 242 |
+
"learning_rate": 2.8251361867704283e-05,
|
| 243 |
+
"loss": 0.5208,
|
| 244 |
+
"step": 3000
|
| 245 |
+
},
|
| 246 |
+
{
|
| 247 |
+
"epoch": 6.0,
|
| 248 |
+
"eval_cer": 0.028423542010619786,
|
| 249 |
+
"eval_loss": 0.1423884481191635,
|
| 250 |
+
"eval_runtime": 7.8913,
|
| 251 |
+
"eval_samples_per_second": 43.212,
|
| 252 |
+
"eval_steps_per_second": 5.449,
|
| 253 |
+
"eval_wer": 0.09175184216781554,
|
| 254 |
+
"step": 3087
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 6.03,
|
| 258 |
+
"learning_rate": 2.8192996108949418e-05,
|
| 259 |
+
"loss": 0.5267,
|
| 260 |
+
"step": 3100
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 6.22,
|
| 264 |
+
"learning_rate": 2.8134630350194556e-05,
|
| 265 |
+
"loss": 0.526,
|
| 266 |
+
"step": 3200
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 6.41,
|
| 270 |
+
"learning_rate": 2.807626459143969e-05,
|
| 271 |
+
"loss": 0.5118,
|
| 272 |
+
"step": 3300
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 6.61,
|
| 276 |
+
"learning_rate": 2.8017898832684825e-05,
|
| 277 |
+
"loss": 0.5081,
|
| 278 |
+
"step": 3400
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 6.8,
|
| 282 |
+
"learning_rate": 2.795953307392996e-05,
|
| 283 |
+
"loss": 0.4917,
|
| 284 |
+
"step": 3500
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 7.0,
|
| 288 |
+
"learning_rate": 2.7901167315175098e-05,
|
| 289 |
+
"loss": 0.4857,
|
| 290 |
+
"step": 3600
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 7.0,
|
| 294 |
+
"eval_cer": 0.026817187988041585,
|
| 295 |
+
"eval_loss": 0.13566730916500092,
|
| 296 |
+
"eval_runtime": 6.2287,
|
| 297 |
+
"eval_samples_per_second": 54.747,
|
| 298 |
+
"eval_steps_per_second": 6.904,
|
| 299 |
+
"eval_wer": 0.08438317090563346,
|
| 300 |
+
"step": 3601
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"epoch": 7.19,
|
| 304 |
+
"learning_rate": 2.7842801556420232e-05,
|
| 305 |
+
"loss": 0.4508,
|
| 306 |
+
"step": 3700
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 7.39,
|
| 310 |
+
"learning_rate": 2.778443579766537e-05,
|
| 311 |
+
"loss": 0.462,
|
| 312 |
+
"step": 3800
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 7.58,
|
| 316 |
+
"learning_rate": 2.7726070038910505e-05,
|
| 317 |
+
"loss": 0.4779,
|
| 318 |
+
"step": 3900
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 7.77,
|
| 322 |
+
"learning_rate": 2.7667704280155643e-05,
|
| 323 |
+
"loss": 0.4848,
|
| 324 |
+
"step": 4000
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 7.97,
|
| 328 |
+
"learning_rate": 2.7609338521400777e-05,
|
| 329 |
+
"loss": 0.4458,
|
| 330 |
+
"step": 4100
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 8.0,
|
| 334 |
+
"eval_cer": 0.0316808710008478,
|
| 335 |
+
"eval_loss": 0.13754041492938995,
|
| 336 |
+
"eval_runtime": 8.163,
|
| 337 |
+
"eval_samples_per_second": 41.774,
|
| 338 |
+
"eval_steps_per_second": 5.268,
|
| 339 |
+
"eval_wer": 0.08818635607321132,
|
| 340 |
+
"step": 4116
|
| 341 |
+
},
|
| 342 |
+
{
|
| 343 |
+
"epoch": 8.16,
|
| 344 |
+
"learning_rate": 2.7550972762645915e-05,
|
| 345 |
+
"loss": 0.431,
|
| 346 |
+
"step": 4200
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 8.36,
|
| 350 |
+
"learning_rate": 2.749260700389105e-05,
|
| 351 |
+
"loss": 0.4272,
|
| 352 |
+
"step": 4300
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 8.55,
|
| 356 |
+
"learning_rate": 2.7434241245136188e-05,
|
| 357 |
+
"loss": 0.4333,
|
| 358 |
+
"step": 4400
|
| 359 |
+
},
|
| 360 |
+
{
|
| 361 |
+
"epoch": 8.75,
|
| 362 |
+
"learning_rate": 2.7375875486381326e-05,
|
| 363 |
+
"loss": 0.4415,
|
| 364 |
+
"step": 4500
|
| 365 |
+
},
|
| 366 |
+
{
|
| 367 |
+
"epoch": 8.94,
|
| 368 |
+
"learning_rate": 2.731750972762646e-05,
|
| 369 |
+
"loss": 0.4158,
|
| 370 |
+
"step": 4600
|
| 371 |
+
},
|
| 372 |
+
{
|
| 373 |
+
"epoch": 9.0,
|
| 374 |
+
"eval_cer": 0.0303422426486993,
|
| 375 |
+
"eval_loss": 0.1411372721195221,
|
| 376 |
+
"eval_runtime": 7.463,
|
| 377 |
+
"eval_samples_per_second": 45.692,
|
| 378 |
+
"eval_steps_per_second": 5.762,
|
| 379 |
+
"eval_wer": 0.08390777275968624,
|
| 380 |
+
"step": 4630
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 9.14,
|
| 384 |
+
"learning_rate": 2.72591439688716e-05,
|
| 385 |
+
"loss": 0.4085,
|
| 386 |
+
"step": 4700
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 9.33,
|
| 390 |
+
"learning_rate": 2.7200778210116733e-05,
|
| 391 |
+
"loss": 0.4255,
|
| 392 |
+
"step": 4800
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 9.52,
|
| 396 |
+
"learning_rate": 2.714241245136187e-05,
|
| 397 |
+
"loss": 0.4186,
|
| 398 |
+
"step": 4900
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 9.72,
|
| 402 |
+
"learning_rate": 2.7084046692607002e-05,
|
| 403 |
+
"loss": 0.4326,
|
| 404 |
+
"step": 5000
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 9.91,
|
| 408 |
+
"learning_rate": 2.702568093385214e-05,
|
| 409 |
+
"loss": 0.3915,
|
| 410 |
+
"step": 5100
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 10.0,
|
| 414 |
+
"eval_cer": 0.03190397572620588,
|
| 415 |
+
"eval_loss": 0.14571979641914368,
|
| 416 |
+
"eval_runtime": 6.3221,
|
| 417 |
+
"eval_samples_per_second": 53.938,
|
| 418 |
+
"eval_steps_per_second": 6.802,
|
| 419 |
+
"eval_wer": 0.09151414309484193,
|
| 420 |
+
"step": 5145
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"epoch": 10.11,
|
| 424 |
+
"learning_rate": 2.6967315175097275e-05,
|
| 425 |
+
"loss": 0.3886,
|
| 426 |
+
"step": 5200
|
| 427 |
+
},
|
| 428 |
+
{
|
| 429 |
+
"epoch": 10.3,
|
| 430 |
+
"learning_rate": 2.6908949416342413e-05,
|
| 431 |
+
"loss": 0.3821,
|
| 432 |
+
"step": 5300
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 10.5,
|
| 436 |
+
"learning_rate": 2.6850583657587547e-05,
|
| 437 |
+
"loss": 0.373,
|
| 438 |
+
"step": 5400
|
| 439 |
+
},
|
| 440 |
+
{
|
| 441 |
+
"epoch": 10.69,
|
| 442 |
+
"learning_rate": 2.6792217898832685e-05,
|
| 443 |
+
"loss": 0.385,
|
| 444 |
+
"step": 5500
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 10.88,
|
| 448 |
+
"learning_rate": 2.673385214007782e-05,
|
| 449 |
+
"loss": 0.3898,
|
| 450 |
+
"step": 5600
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 11.0,
|
| 454 |
+
"eval_cer": 0.031011556824773547,
|
| 455 |
+
"eval_loss": 0.14640487730503082,
|
| 456 |
+
"eval_runtime": 7.7134,
|
| 457 |
+
"eval_samples_per_second": 44.209,
|
| 458 |
+
"eval_steps_per_second": 5.575,
|
| 459 |
+
"eval_wer": 0.08699786070834324,
|
| 460 |
+
"step": 5659
|
| 461 |
+
},
|
| 462 |
+
{
|
| 463 |
+
"epoch": 11.08,
|
| 464 |
+
"learning_rate": 2.6675486381322958e-05,
|
| 465 |
+
"loss": 0.3711,
|
| 466 |
+
"step": 5700
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 11.27,
|
| 470 |
+
"learning_rate": 2.6617120622568092e-05,
|
| 471 |
+
"loss": 0.3698,
|
| 472 |
+
"step": 5800
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 11.47,
|
| 476 |
+
"learning_rate": 2.655875486381323e-05,
|
| 477 |
+
"loss": 0.3609,
|
| 478 |
+
"step": 5900
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 11.66,
|
| 482 |
+
"learning_rate": 2.650038910505837e-05,
|
| 483 |
+
"loss": 0.3577,
|
| 484 |
+
"step": 6000
|
| 485 |
+
},
|
| 486 |
+
{
|
| 487 |
+
"epoch": 11.86,
|
| 488 |
+
"learning_rate": 2.6442023346303503e-05,
|
| 489 |
+
"loss": 0.3562,
|
| 490 |
+
"step": 6100
|
| 491 |
+
},
|
| 492 |
+
{
|
| 493 |
+
"epoch": 12.0,
|
| 494 |
+
"eval_cer": 0.0314131453304181,
|
| 495 |
+
"eval_loss": 0.15003812313079834,
|
| 496 |
+
"eval_runtime": 8.0687,
|
| 497 |
+
"eval_samples_per_second": 42.262,
|
| 498 |
+
"eval_steps_per_second": 5.329,
|
| 499 |
+
"eval_wer": 0.08747325885429047,
|
| 500 |
+
"step": 6174
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 12.05,
|
| 504 |
+
"learning_rate": 2.638365758754864e-05,
|
| 505 |
+
"loss": 0.3992,
|
| 506 |
+
"step": 6200
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 12.24,
|
| 510 |
+
"learning_rate": 2.6325291828793776e-05,
|
| 511 |
+
"loss": 0.3677,
|
| 512 |
+
"step": 6300
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 12.44,
|
| 516 |
+
"learning_rate": 2.6266926070038913e-05,
|
| 517 |
+
"loss": 0.345,
|
| 518 |
+
"step": 6400
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 12.63,
|
| 522 |
+
"learning_rate": 2.6208560311284048e-05,
|
| 523 |
+
"loss": 0.3319,
|
| 524 |
+
"step": 6500
|
| 525 |
+
},
|
| 526 |
+
{
|
| 527 |
+
"epoch": 12.83,
|
| 528 |
+
"learning_rate": 2.6150194552529183e-05,
|
| 529 |
+
"loss": 0.3619,
|
| 530 |
+
"step": 6600
|
| 531 |
+
},
|
| 532 |
+
{
|
| 533 |
+
"epoch": 13.0,
|
| 534 |
+
"eval_cer": 0.031279282495203245,
|
| 535 |
+
"eval_loss": 0.15227940678596497,
|
| 536 |
+
"eval_runtime": 6.2463,
|
| 537 |
+
"eval_samples_per_second": 54.592,
|
| 538 |
+
"eval_steps_per_second": 6.884,
|
| 539 |
+
"eval_wer": 0.08771095792726408,
|
| 540 |
+
"step": 6688
|
| 541 |
+
},
|
| 542 |
+
{
|
| 543 |
+
"epoch": 13.02,
|
| 544 |
+
"learning_rate": 2.6091828793774317e-05,
|
| 545 |
+
"loss": 0.3291,
|
| 546 |
+
"step": 6700
|
| 547 |
+
},
|
| 548 |
+
{
|
| 549 |
+
"epoch": 13.22,
|
| 550 |
+
"learning_rate": 2.6033463035019455e-05,
|
| 551 |
+
"loss": 0.3535,
|
| 552 |
+
"step": 6800
|
| 553 |
+
},
|
| 554 |
+
{
|
| 555 |
+
"epoch": 13.41,
|
| 556 |
+
"learning_rate": 2.597509727626459e-05,
|
| 557 |
+
"loss": 0.3331,
|
| 558 |
+
"step": 6900
|
| 559 |
+
},
|
| 560 |
+
{
|
| 561 |
+
"epoch": 13.61,
|
| 562 |
+
"learning_rate": 2.5916731517509728e-05,
|
| 563 |
+
"loss": 0.334,
|
| 564 |
+
"step": 7000
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 13.8,
|
| 568 |
+
"learning_rate": 2.5858365758754862e-05,
|
| 569 |
+
"loss": 0.3435,
|
| 570 |
+
"step": 7100
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 13.99,
|
| 574 |
+
"learning_rate": 2.58e-05,
|
| 575 |
+
"loss": 0.3283,
|
| 576 |
+
"step": 7200
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 14.0,
|
| 580 |
+
"eval_cer": 0.028958993351479185,
|
| 581 |
+
"eval_loss": 0.1473035216331482,
|
| 582 |
+
"eval_runtime": 6.4288,
|
| 583 |
+
"eval_samples_per_second": 53.043,
|
| 584 |
+
"eval_steps_per_second": 6.689,
|
| 585 |
+
"eval_wer": 0.08557166627050154,
|
| 586 |
+
"step": 7203
|
| 587 |
+
},
|
| 588 |
+
{
|
| 589 |
+
"epoch": 14.19,
|
| 590 |
+
"learning_rate": 2.5741634241245138e-05,
|
| 591 |
+
"loss": 0.3169,
|
| 592 |
+
"step": 7300
|
| 593 |
+
},
|
| 594 |
+
{
|
| 595 |
+
"epoch": 14.38,
|
| 596 |
+
"learning_rate": 2.5683268482490273e-05,
|
| 597 |
+
"loss": 0.3146,
|
| 598 |
+
"step": 7400
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 14.58,
|
| 602 |
+
"learning_rate": 2.562490272373541e-05,
|
| 603 |
+
"loss": 0.3201,
|
| 604 |
+
"step": 7500
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 14.77,
|
| 608 |
+
"learning_rate": 2.5566536964980545e-05,
|
| 609 |
+
"loss": 0.3297,
|
| 610 |
+
"step": 7600
|
| 611 |
+
},
|
| 612 |
+
{
|
| 613 |
+
"epoch": 14.97,
|
| 614 |
+
"learning_rate": 2.5508171206225683e-05,
|
| 615 |
+
"loss": 0.3196,
|
| 616 |
+
"step": 7700
|
| 617 |
+
},
|
| 618 |
+
{
|
| 619 |
+
"epoch": 15.0,
|
| 620 |
+
"eval_cer": 0.029896033197983132,
|
| 621 |
+
"eval_loss": 0.14426732063293457,
|
| 622 |
+
"eval_runtime": 6.3433,
|
| 623 |
+
"eval_samples_per_second": 53.757,
|
| 624 |
+
"eval_steps_per_second": 6.779,
|
| 625 |
+
"eval_wer": 0.08438317090563346,
|
| 626 |
+
"step": 7717
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 15.16,
|
| 630 |
+
"learning_rate": 2.5449805447470818e-05,
|
| 631 |
+
"loss": 0.3124,
|
| 632 |
+
"step": 7800
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 15.35,
|
| 636 |
+
"learning_rate": 2.5391439688715956e-05,
|
| 637 |
+
"loss": 0.312,
|
| 638 |
+
"step": 7900
|
| 639 |
+
},
|
| 640 |
+
{
|
| 641 |
+
"epoch": 15.55,
|
| 642 |
+
"learning_rate": 2.533307392996109e-05,
|
| 643 |
+
"loss": 0.3089,
|
| 644 |
+
"step": 8000
|
| 645 |
+
},
|
| 646 |
+
{
|
| 647 |
+
"epoch": 15.74,
|
| 648 |
+
"learning_rate": 2.527470817120623e-05,
|
| 649 |
+
"loss": 0.2985,
|
| 650 |
+
"step": 8100
|
| 651 |
+
},
|
| 652 |
+
{
|
| 653 |
+
"epoch": 15.94,
|
| 654 |
+
"learning_rate": 2.521634241245136e-05,
|
| 655 |
+
"loss": 0.3165,
|
| 656 |
+
"step": 8200
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 16.0,
|
| 660 |
+
"eval_cer": 0.028334300120476552,
|
| 661 |
+
"eval_loss": 0.1413327008485794,
|
| 662 |
+
"eval_runtime": 6.6839,
|
| 663 |
+
"eval_samples_per_second": 51.018,
|
| 664 |
+
"eval_steps_per_second": 6.433,
|
| 665 |
+
"eval_wer": 0.08129308295697647,
|
| 666 |
+
"step": 8232
|
| 667 |
+
},
|
| 668 |
+
{
|
| 669 |
+
"epoch": 16.13,
|
| 670 |
+
"learning_rate": 2.5157976653696498e-05,
|
| 671 |
+
"loss": 0.2891,
|
| 672 |
+
"step": 8300
|
| 673 |
+
},
|
| 674 |
+
{
|
| 675 |
+
"epoch": 16.33,
|
| 676 |
+
"learning_rate": 2.5100194552529184e-05,
|
| 677 |
+
"loss": 0.3053,
|
| 678 |
+
"step": 8400
|
| 679 |
+
},
|
| 680 |
+
{
|
| 681 |
+
"epoch": 16.52,
|
| 682 |
+
"learning_rate": 2.5041828793774322e-05,
|
| 683 |
+
"loss": 0.3003,
|
| 684 |
+
"step": 8500
|
| 685 |
+
},
|
| 686 |
+
{
|
| 687 |
+
"epoch": 16.72,
|
| 688 |
+
"learning_rate": 2.4983463035019457e-05,
|
| 689 |
+
"loss": 0.311,
|
| 690 |
+
"step": 8600
|
| 691 |
+
},
|
| 692 |
+
{
|
| 693 |
+
"epoch": 16.91,
|
| 694 |
+
"learning_rate": 2.4925097276264595e-05,
|
| 695 |
+
"loss": 0.2954,
|
| 696 |
+
"step": 8700
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 17.0,
|
| 700 |
+
"eval_cer": 0.028289679175404935,
|
| 701 |
+
"eval_loss": 0.14512528479099274,
|
| 702 |
+
"eval_runtime": 6.2782,
|
| 703 |
+
"eval_samples_per_second": 54.315,
|
| 704 |
+
"eval_steps_per_second": 6.849,
|
| 705 |
+
"eval_wer": 0.08248157832184455,
|
| 706 |
+
"step": 8746
|
| 707 |
+
},
|
| 708 |
+
{
|
| 709 |
+
"epoch": 17.1,
|
| 710 |
+
"learning_rate": 2.486673151750973e-05,
|
| 711 |
+
"loss": 0.2687,
|
| 712 |
+
"step": 8800
|
| 713 |
+
},
|
| 714 |
+
{
|
| 715 |
+
"epoch": 17.3,
|
| 716 |
+
"learning_rate": 2.4808365758754864e-05,
|
| 717 |
+
"loss": 0.2746,
|
| 718 |
+
"step": 8900
|
| 719 |
+
},
|
| 720 |
+
{
|
| 721 |
+
"epoch": 17.49,
|
| 722 |
+
"learning_rate": 2.475e-05,
|
| 723 |
+
"loss": 0.2857,
|
| 724 |
+
"step": 9000
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 17.69,
|
| 728 |
+
"learning_rate": 2.4691634241245136e-05,
|
| 729 |
+
"loss": 0.2819,
|
| 730 |
+
"step": 9100
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 17.88,
|
| 734 |
+
"learning_rate": 2.463326848249027e-05,
|
| 735 |
+
"loss": 0.293,
|
| 736 |
+
"step": 9200
|
| 737 |
+
},
|
| 738 |
+
{
|
| 739 |
+
"epoch": 18.0,
|
| 740 |
+
"eval_cer": 0.02860202579090625,
|
| 741 |
+
"eval_loss": 0.15394243597984314,
|
| 742 |
+
"eval_runtime": 6.29,
|
| 743 |
+
"eval_samples_per_second": 54.213,
|
| 744 |
+
"eval_steps_per_second": 6.836,
|
| 745 |
+
"eval_wer": 0.08224387924887093,
|
| 746 |
+
"step": 9261
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 18.08,
|
| 750 |
+
"learning_rate": 2.457490272373541e-05,
|
| 751 |
+
"loss": 0.295,
|
| 752 |
+
"step": 9300
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 18.27,
|
| 756 |
+
"learning_rate": 2.4516536964980544e-05,
|
| 757 |
+
"loss": 0.3022,
|
| 758 |
+
"step": 9400
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 18.46,
|
| 762 |
+
"learning_rate": 2.445817120622568e-05,
|
| 763 |
+
"loss": 0.2714,
|
| 764 |
+
"step": 9500
|
| 765 |
+
},
|
| 766 |
+
{
|
| 767 |
+
"epoch": 18.66,
|
| 768 |
+
"learning_rate": 2.4399805447470816e-05,
|
| 769 |
+
"loss": 0.2881,
|
| 770 |
+
"step": 9600
|
| 771 |
+
},
|
| 772 |
+
{
|
| 773 |
+
"epoch": 18.85,
|
| 774 |
+
"learning_rate": 2.4341439688715954e-05,
|
| 775 |
+
"loss": 0.2821,
|
| 776 |
+
"step": 9700
|
| 777 |
+
},
|
| 778 |
+
{
|
| 779 |
+
"epoch": 19.0,
|
| 780 |
+
"eval_cer": 0.029628307527553435,
|
| 781 |
+
"eval_loss": 0.15518580377101898,
|
| 782 |
+
"eval_runtime": 8.2771,
|
| 783 |
+
"eval_samples_per_second": 41.198,
|
| 784 |
+
"eval_steps_per_second": 5.195,
|
| 785 |
+
"eval_wer": 0.08438317090563346,
|
| 786 |
+
"step": 9775
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 19.05,
|
| 790 |
+
"learning_rate": 2.4283073929961092e-05,
|
| 791 |
+
"loss": 0.271,
|
| 792 |
+
"step": 9800
|
| 793 |
+
},
|
| 794 |
+
{
|
| 795 |
+
"epoch": 19.24,
|
| 796 |
+
"learning_rate": 2.4224708171206227e-05,
|
| 797 |
+
"loss": 0.2696,
|
| 798 |
+
"step": 9900
|
| 799 |
+
},
|
| 800 |
+
{
|
| 801 |
+
"epoch": 19.44,
|
| 802 |
+
"learning_rate": 2.4166342412451365e-05,
|
| 803 |
+
"loss": 0.2715,
|
| 804 |
+
"step": 10000
|
| 805 |
+
},
|
| 806 |
+
{
|
| 807 |
+
"epoch": 19.63,
|
| 808 |
+
"learning_rate": 2.41079766536965e-05,
|
| 809 |
+
"loss": 0.2847,
|
| 810 |
+
"step": 10100
|
| 811 |
+
},
|
| 812 |
+
{
|
| 813 |
+
"epoch": 19.83,
|
| 814 |
+
"learning_rate": 2.4049610894941637e-05,
|
| 815 |
+
"loss": 0.2893,
|
| 816 |
+
"step": 10200
|
| 817 |
+
},
|
| 818 |
+
{
|
| 819 |
+
"epoch": 20.0,
|
| 820 |
+
"eval_cer": 0.028468162955691403,
|
| 821 |
+
"eval_loss": 0.1484398990869522,
|
| 822 |
+
"eval_runtime": 6.2768,
|
| 823 |
+
"eval_samples_per_second": 54.327,
|
| 824 |
+
"eval_steps_per_second": 6.851,
|
| 825 |
+
"eval_wer": 0.08200618017589731,
|
| 826 |
+
"step": 10290
|
| 827 |
+
},
|
| 828 |
+
{
|
| 829 |
+
"epoch": 20.02,
|
| 830 |
+
"learning_rate": 2.3991245136186772e-05,
|
| 831 |
+
"loss": 0.2666,
|
| 832 |
+
"step": 10300
|
| 833 |
+
},
|
| 834 |
+
{
|
| 835 |
+
"epoch": 20.21,
|
| 836 |
+
"learning_rate": 2.393287937743191e-05,
|
| 837 |
+
"loss": 0.2589,
|
| 838 |
+
"step": 10400
|
| 839 |
+
},
|
| 840 |
+
{
|
| 841 |
+
"epoch": 20.41,
|
| 842 |
+
"learning_rate": 2.387451361867704e-05,
|
| 843 |
+
"loss": 0.2579,
|
| 844 |
+
"step": 10500
|
| 845 |
+
},
|
| 846 |
+
{
|
| 847 |
+
"epoch": 20.6,
|
| 848 |
+
"learning_rate": 2.381614785992218e-05,
|
| 849 |
+
"loss": 0.2573,
|
| 850 |
+
"step": 10600
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 20.8,
|
| 854 |
+
"learning_rate": 2.3757782101167314e-05,
|
| 855 |
+
"loss": 0.2632,
|
| 856 |
+
"step": 10700
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 20.99,
|
| 860 |
+
"learning_rate": 2.369941634241245e-05,
|
| 861 |
+
"loss": 0.2609,
|
| 862 |
+
"step": 10800
|
| 863 |
+
},
|
| 864 |
+
{
|
| 865 |
+
"epoch": 21.0,
|
| 866 |
+
"eval_cer": 0.03074383115434385,
|
| 867 |
+
"eval_loss": 0.16356056928634644,
|
| 868 |
+
"eval_runtime": 6.6619,
|
| 869 |
+
"eval_samples_per_second": 51.187,
|
| 870 |
+
"eval_steps_per_second": 6.455,
|
| 871 |
+
"eval_wer": 0.08509626812455431,
|
| 872 |
+
"step": 10804
|
| 873 |
+
},
|
| 874 |
+
{
|
| 875 |
+
"epoch": 21.19,
|
| 876 |
+
"learning_rate": 2.3641050583657586e-05,
|
| 877 |
+
"loss": 0.2692,
|
| 878 |
+
"step": 10900
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 21.38,
|
| 882 |
+
"learning_rate": 2.3583268482490276e-05,
|
| 883 |
+
"loss": 0.2576,
|
| 884 |
+
"step": 11000
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 21.57,
|
| 888 |
+
"learning_rate": 2.3524902723735407e-05,
|
| 889 |
+
"loss": 0.2663,
|
| 890 |
+
"step": 11100
|
| 891 |
+
},
|
| 892 |
+
{
|
| 893 |
+
"epoch": 21.77,
|
| 894 |
+
"learning_rate": 2.3466536964980545e-05,
|
| 895 |
+
"loss": 0.2515,
|
| 896 |
+
"step": 11200
|
| 897 |
+
},
|
| 898 |
+
{
|
| 899 |
+
"epoch": 21.96,
|
| 900 |
+
"learning_rate": 2.340817120622568e-05,
|
| 901 |
+
"loss": 0.2526,
|
| 902 |
+
"step": 11300
|
| 903 |
+
},
|
| 904 |
+
{
|
| 905 |
+
"epoch": 22.0,
|
| 906 |
+
"eval_cer": 0.029182098076837266,
|
| 907 |
+
"eval_loss": 0.15198495984077454,
|
| 908 |
+
"eval_runtime": 6.3434,
|
| 909 |
+
"eval_samples_per_second": 53.756,
|
| 910 |
+
"eval_steps_per_second": 6.779,
|
| 911 |
+
"eval_wer": 0.08557166627050154,
|
| 912 |
+
"step": 11319
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 22.16,
|
| 916 |
+
"learning_rate": 2.3349805447470818e-05,
|
| 917 |
+
"loss": 0.2732,
|
| 918 |
+
"step": 11400
|
| 919 |
+
},
|
| 920 |
+
{
|
| 921 |
+
"epoch": 22.35,
|
| 922 |
+
"learning_rate": 2.3291439688715952e-05,
|
| 923 |
+
"loss": 0.2424,
|
| 924 |
+
"step": 11500
|
| 925 |
+
},
|
| 926 |
+
{
|
| 927 |
+
"epoch": 22.55,
|
| 928 |
+
"learning_rate": 2.323307392996109e-05,
|
| 929 |
+
"loss": 0.2586,
|
| 930 |
+
"step": 11600
|
| 931 |
+
},
|
| 932 |
+
{
|
| 933 |
+
"epoch": 22.74,
|
| 934 |
+
"learning_rate": 2.3174708171206225e-05,
|
| 935 |
+
"loss": 0.274,
|
| 936 |
+
"step": 11700
|
| 937 |
+
},
|
| 938 |
+
{
|
| 939 |
+
"epoch": 22.93,
|
| 940 |
+
"learning_rate": 2.3116342412451363e-05,
|
| 941 |
+
"loss": 0.2571,
|
| 942 |
+
"step": 11800
|
| 943 |
+
},
|
| 944 |
+
{
|
| 945 |
+
"epoch": 23.0,
|
| 946 |
+
"eval_cer": 0.029092856186694036,
|
| 947 |
+
"eval_loss": 0.14488764107227325,
|
| 948 |
+
"eval_runtime": 6.7698,
|
| 949 |
+
"eval_samples_per_second": 50.37,
|
| 950 |
+
"eval_steps_per_second": 6.352,
|
| 951 |
+
"eval_wer": 0.08509626812455431,
|
| 952 |
+
"step": 11833
|
| 953 |
+
},
|
| 954 |
+
{
|
| 955 |
+
"epoch": 23.13,
|
| 956 |
+
"learning_rate": 2.3057976653696497e-05,
|
| 957 |
+
"loss": 0.2513,
|
| 958 |
+
"step": 11900
|
| 959 |
+
},
|
| 960 |
+
{
|
| 961 |
+
"epoch": 23.32,
|
| 962 |
+
"learning_rate": 2.2999610894941635e-05,
|
| 963 |
+
"loss": 0.2577,
|
| 964 |
+
"step": 12000
|
| 965 |
+
},
|
| 966 |
+
{
|
| 967 |
+
"epoch": 23.52,
|
| 968 |
+
"learning_rate": 2.294124513618677e-05,
|
| 969 |
+
"loss": 0.2335,
|
| 970 |
+
"step": 12100
|
| 971 |
+
},
|
| 972 |
+
{
|
| 973 |
+
"epoch": 23.71,
|
| 974 |
+
"learning_rate": 2.2882879377431908e-05,
|
| 975 |
+
"loss": 0.2573,
|
| 976 |
+
"step": 12200
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 23.91,
|
| 980 |
+
"learning_rate": 2.2824513618677043e-05,
|
| 981 |
+
"loss": 0.2486,
|
| 982 |
+
"step": 12300
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 24.0,
|
| 986 |
+
"eval_cer": 0.030699210209272233,
|
| 987 |
+
"eval_loss": 0.15738651156425476,
|
| 988 |
+
"eval_runtime": 7.9129,
|
| 989 |
+
"eval_samples_per_second": 43.094,
|
| 990 |
+
"eval_steps_per_second": 5.434,
|
| 991 |
+
"eval_wer": 0.08652246256239601,
|
| 992 |
+
"step": 12348
|
| 993 |
+
},
|
| 994 |
+
{
|
| 995 |
+
"epoch": 24.1,
|
| 996 |
+
"learning_rate": 2.276614785992218e-05,
|
| 997 |
+
"loss": 0.2439,
|
| 998 |
+
"step": 12400
|
| 999 |
+
},
|
| 1000 |
+
{
|
| 1001 |
+
"epoch": 24.3,
|
| 1002 |
+
"learning_rate": 2.270778210116732e-05,
|
| 1003 |
+
"loss": 0.253,
|
| 1004 |
+
"step": 12500
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 24.49,
|
| 1008 |
+
"learning_rate": 2.2649416342412453e-05,
|
| 1009 |
+
"loss": 0.2435,
|
| 1010 |
+
"step": 12600
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 24.68,
|
| 1014 |
+
"learning_rate": 2.2591050583657588e-05,
|
| 1015 |
+
"loss": 0.2378,
|
| 1016 |
+
"step": 12700
|
| 1017 |
+
},
|
| 1018 |
+
{
|
| 1019 |
+
"epoch": 24.88,
|
| 1020 |
+
"learning_rate": 2.2532684824902722e-05,
|
| 1021 |
+
"loss": 0.2501,
|
| 1022 |
+
"step": 12800
|
| 1023 |
+
},
|
| 1024 |
+
{
|
| 1025 |
+
"epoch": 25.0,
|
| 1026 |
+
"eval_cer": 0.0295390656374102,
|
| 1027 |
+
"eval_loss": 0.1490197330713272,
|
| 1028 |
+
"eval_runtime": 6.0948,
|
| 1029 |
+
"eval_samples_per_second": 55.949,
|
| 1030 |
+
"eval_steps_per_second": 7.055,
|
| 1031 |
+
"eval_wer": 0.08557166627050154,
|
| 1032 |
+
"step": 12862
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 25.07,
|
| 1036 |
+
"learning_rate": 2.247431906614786e-05,
|
| 1037 |
+
"loss": 0.2399,
|
| 1038 |
+
"step": 12900
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 25.27,
|
| 1042 |
+
"learning_rate": 2.2415953307392995e-05,
|
| 1043 |
+
"loss": 0.2368,
|
| 1044 |
+
"step": 13000
|
| 1045 |
+
},
|
| 1046 |
+
{
|
| 1047 |
+
"epoch": 25.46,
|
| 1048 |
+
"learning_rate": 2.2357587548638133e-05,
|
| 1049 |
+
"loss": 0.2396,
|
| 1050 |
+
"step": 13100
|
| 1051 |
+
},
|
| 1052 |
+
{
|
| 1053 |
+
"epoch": 25.66,
|
| 1054 |
+
"learning_rate": 2.2299221789883267e-05,
|
| 1055 |
+
"loss": 0.2117,
|
| 1056 |
+
"step": 13200
|
| 1057 |
+
},
|
| 1058 |
+
{
|
| 1059 |
+
"epoch": 25.85,
|
| 1060 |
+
"learning_rate": 2.2240856031128405e-05,
|
| 1061 |
+
"loss": 0.2525,
|
| 1062 |
+
"step": 13300
|
| 1063 |
+
},
|
| 1064 |
+
{
|
| 1065 |
+
"epoch": 26.0,
|
| 1066 |
+
"eval_cer": 0.02940520280219535,
|
| 1067 |
+
"eval_loss": 0.1508348286151886,
|
| 1068 |
+
"eval_runtime": 6.2168,
|
| 1069 |
+
"eval_samples_per_second": 54.851,
|
| 1070 |
+
"eval_steps_per_second": 6.917,
|
| 1071 |
+
"eval_wer": 0.08271927739481816,
|
| 1072 |
+
"step": 13377
|
| 1073 |
+
},
|
| 1074 |
+
{
|
| 1075 |
+
"epoch": 26.04,
|
| 1076 |
+
"learning_rate": 2.218249027237354e-05,
|
| 1077 |
+
"loss": 0.2312,
|
| 1078 |
+
"step": 13400
|
| 1079 |
+
},
|
| 1080 |
+
{
|
| 1081 |
+
"epoch": 26.24,
|
| 1082 |
+
"learning_rate": 2.2124708171206226e-05,
|
| 1083 |
+
"loss": 0.2363,
|
| 1084 |
+
"step": 13500
|
| 1085 |
+
},
|
| 1086 |
+
{
|
| 1087 |
+
"epoch": 26.43,
|
| 1088 |
+
"learning_rate": 2.206634241245136e-05,
|
| 1089 |
+
"loss": 0.2262,
|
| 1090 |
+
"step": 13600
|
| 1091 |
+
},
|
| 1092 |
+
{
|
| 1093 |
+
"epoch": 26.63,
|
| 1094 |
+
"learning_rate": 2.20079766536965e-05,
|
| 1095 |
+
"loss": 0.2415,
|
| 1096 |
+
"step": 13700
|
| 1097 |
+
},
|
| 1098 |
+
{
|
| 1099 |
+
"epoch": 26.82,
|
| 1100 |
+
"learning_rate": 2.1949610894941634e-05,
|
| 1101 |
+
"loss": 0.2452,
|
| 1102 |
+
"step": 13800
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 27.0,
|
| 1106 |
+
"eval_cer": 0.029003614296550802,
|
| 1107 |
+
"eval_loss": 0.15108107030391693,
|
| 1108 |
+
"eval_runtime": 6.0859,
|
| 1109 |
+
"eval_samples_per_second": 56.031,
|
| 1110 |
+
"eval_steps_per_second": 7.065,
|
| 1111 |
+
"eval_wer": 0.08081768481102923,
|
| 1112 |
+
"step": 13891
|
| 1113 |
+
},
|
| 1114 |
+
{
|
| 1115 |
+
"epoch": 27.0,
|
| 1116 |
+
"step": 13891,
|
| 1117 |
+
"total_flos": 1.1392249463344112e+20,
|
| 1118 |
+
"train_loss": 0.5354501710375033,
|
| 1119 |
+
"train_runtime": 10852.3147,
|
| 1120 |
+
"train_samples_per_second": 151.654,
|
| 1121 |
+
"train_steps_per_second": 4.736
|
| 1122 |
+
}
|
| 1123 |
+
],
|
| 1124 |
+
"max_steps": 51400,
|
| 1125 |
+
"num_train_epochs": 100,
|
| 1126 |
+
"total_flos": 1.1392249463344112e+20,
|
| 1127 |
+
"trial_name": null,
|
| 1128 |
+
"trial_params": null
|
| 1129 |
+
}
|