Aline Rodrigues
commited on
Commit
•
a589062
1
Parent(s):
e0eefb6
End of training
Browse files- all_results.json +16 -0
- eval_results.json +10 -0
- train_results.json +9 -0
- trainer_state.json +895 -0
all_results.json
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 57.0,
|
| 3 |
+
"eval_cer": 0.026683325152826738,
|
| 4 |
+
"eval_loss": 0.13642320036888123,
|
| 5 |
+
"eval_runtime": 6.0666,
|
| 6 |
+
"eval_samples": 341,
|
| 7 |
+
"eval_samples_per_second": 56.21,
|
| 8 |
+
"eval_steps_per_second": 7.088,
|
| 9 |
+
"eval_wer": 0.08438317090563346,
|
| 10 |
+
"total_flos": 4.0970110980881424e+19,
|
| 11 |
+
"train_loss": 0.6410711561475954,
|
| 12 |
+
"train_runtime": 7770.6868,
|
| 13 |
+
"train_samples": 2743,
|
| 14 |
+
"train_samples_per_second": 35.299,
|
| 15 |
+
"train_steps_per_second": 1.107
|
| 16 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 57.0,
|
| 3 |
+
"eval_cer": 0.026683325152826738,
|
| 4 |
+
"eval_loss": 0.13642320036888123,
|
| 5 |
+
"eval_runtime": 6.0666,
|
| 6 |
+
"eval_samples": 341,
|
| 7 |
+
"eval_samples_per_second": 56.21,
|
| 8 |
+
"eval_steps_per_second": 7.088,
|
| 9 |
+
"eval_wer": 0.08438317090563346
|
| 10 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 57.0,
|
| 3 |
+
"total_flos": 4.0970110980881424e+19,
|
| 4 |
+
"train_loss": 0.6410711561475954,
|
| 5 |
+
"train_runtime": 7770.6868,
|
| 6 |
+
"train_samples": 2743,
|
| 7 |
+
"train_samples_per_second": 35.299,
|
| 8 |
+
"train_steps_per_second": 1.107
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,895 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.13642320036888123,
|
| 3 |
+
"best_model_checkpoint": "wav2vec2-large-xlsr-mecita-coraa-portuguese-random-all-03/checkpoint-3182",
|
| 4 |
+
"epoch": 57.0,
|
| 5 |
+
"global_step": 4902,
|
| 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": 3e-05,
|
| 13 |
+
"loss": 29.1535,
|
| 14 |
+
"step": 1
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 1.0,
|
| 18 |
+
"eval_cer": 1.0,
|
| 19 |
+
"eval_loss": 3.1676151752471924,
|
| 20 |
+
"eval_runtime": 5.9631,
|
| 21 |
+
"eval_samples_per_second": 57.185,
|
| 22 |
+
"eval_steps_per_second": 7.211,
|
| 23 |
+
"eval_wer": 1.0,
|
| 24 |
+
"step": 86
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 1.16,
|
| 28 |
+
"learning_rate": 2.9661627906976745e-05,
|
| 29 |
+
"loss": 7.924,
|
| 30 |
+
"step": 100
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 2.0,
|
| 34 |
+
"eval_cer": 1.0,
|
| 35 |
+
"eval_loss": 2.9288899898529053,
|
| 36 |
+
"eval_runtime": 6.0836,
|
| 37 |
+
"eval_samples_per_second": 56.053,
|
| 38 |
+
"eval_steps_per_second": 7.068,
|
| 39 |
+
"eval_wer": 1.0,
|
| 40 |
+
"step": 172
|
| 41 |
+
},
|
| 42 |
+
{
|
| 43 |
+
"epoch": 2.33,
|
| 44 |
+
"learning_rate": 2.931279069767442e-05,
|
| 45 |
+
"loss": 3.0115,
|
| 46 |
+
"step": 200
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 3.0,
|
| 50 |
+
"eval_cer": 1.0,
|
| 51 |
+
"eval_loss": 2.8972272872924805,
|
| 52 |
+
"eval_runtime": 6.4814,
|
| 53 |
+
"eval_samples_per_second": 52.612,
|
| 54 |
+
"eval_steps_per_second": 6.634,
|
| 55 |
+
"eval_wer": 1.0,
|
| 56 |
+
"step": 258
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 3.49,
|
| 60 |
+
"learning_rate": 2.8963953488372093e-05,
|
| 61 |
+
"loss": 2.9304,
|
| 62 |
+
"step": 300
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 4.0,
|
| 66 |
+
"eval_cer": 1.0,
|
| 67 |
+
"eval_loss": 2.887695550918579,
|
| 68 |
+
"eval_runtime": 6.205,
|
| 69 |
+
"eval_samples_per_second": 54.956,
|
| 70 |
+
"eval_steps_per_second": 6.93,
|
| 71 |
+
"eval_wer": 1.0,
|
| 72 |
+
"step": 344
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 4.65,
|
| 76 |
+
"learning_rate": 2.861511627906977e-05,
|
| 77 |
+
"loss": 2.9073,
|
| 78 |
+
"step": 400
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"epoch": 5.0,
|
| 82 |
+
"eval_cer": 1.0,
|
| 83 |
+
"eval_loss": 2.857363224029541,
|
| 84 |
+
"eval_runtime": 5.8413,
|
| 85 |
+
"eval_samples_per_second": 58.377,
|
| 86 |
+
"eval_steps_per_second": 7.361,
|
| 87 |
+
"eval_wer": 1.0,
|
| 88 |
+
"step": 430
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 5.81,
|
| 92 |
+
"learning_rate": 2.8266279069767444e-05,
|
| 93 |
+
"loss": 2.7919,
|
| 94 |
+
"step": 500
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 6.0,
|
| 98 |
+
"eval_cer": 0.674222480032127,
|
| 99 |
+
"eval_loss": 1.9143861532211304,
|
| 100 |
+
"eval_runtime": 6.2962,
|
| 101 |
+
"eval_samples_per_second": 54.159,
|
| 102 |
+
"eval_steps_per_second": 6.829,
|
| 103 |
+
"eval_wer": 1.0,
|
| 104 |
+
"step": 516
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 6.98,
|
| 108 |
+
"learning_rate": 2.7917441860465118e-05,
|
| 109 |
+
"loss": 1.6061,
|
| 110 |
+
"step": 600
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 7.0,
|
| 114 |
+
"eval_cer": 0.07661416268796573,
|
| 115 |
+
"eval_loss": 0.543914258480072,
|
| 116 |
+
"eval_runtime": 5.8137,
|
| 117 |
+
"eval_samples_per_second": 58.654,
|
| 118 |
+
"eval_steps_per_second": 7.396,
|
| 119 |
+
"eval_wer": 0.2885666745899691,
|
| 120 |
+
"step": 602
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 8.0,
|
| 124 |
+
"eval_cer": 0.055776181339520774,
|
| 125 |
+
"eval_loss": 0.37258943915367126,
|
| 126 |
+
"eval_runtime": 6.0487,
|
| 127 |
+
"eval_samples_per_second": 56.376,
|
| 128 |
+
"eval_steps_per_second": 7.109,
|
| 129 |
+
"eval_wer": 0.19491323983836464,
|
| 130 |
+
"step": 688
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 8.14,
|
| 134 |
+
"learning_rate": 2.7568604651162792e-05,
|
| 135 |
+
"loss": 0.7808,
|
| 136 |
+
"step": 700
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 9.0,
|
| 140 |
+
"eval_cer": 0.05167105439293204,
|
| 141 |
+
"eval_loss": 0.2959502637386322,
|
| 142 |
+
"eval_runtime": 5.8892,
|
| 143 |
+
"eval_samples_per_second": 57.902,
|
| 144 |
+
"eval_steps_per_second": 7.301,
|
| 145 |
+
"eval_wer": 0.18183979082481577,
|
| 146 |
+
"step": 774
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 9.3,
|
| 150 |
+
"learning_rate": 2.7219767441860466e-05,
|
| 151 |
+
"loss": 0.5543,
|
| 152 |
+
"step": 800
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 10.0,
|
| 156 |
+
"eval_cer": 0.04774441122662978,
|
| 157 |
+
"eval_loss": 0.259117066860199,
|
| 158 |
+
"eval_runtime": 5.9755,
|
| 159 |
+
"eval_samples_per_second": 57.066,
|
| 160 |
+
"eval_steps_per_second": 7.196,
|
| 161 |
+
"eval_wer": 0.16876634181126693,
|
| 162 |
+
"step": 860
|
| 163 |
+
},
|
| 164 |
+
{
|
| 165 |
+
"epoch": 10.47,
|
| 166 |
+
"learning_rate": 2.6870930232558143e-05,
|
| 167 |
+
"loss": 0.4721,
|
| 168 |
+
"step": 900
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 11.0,
|
| 172 |
+
"eval_cer": 0.04265762348846549,
|
| 173 |
+
"eval_loss": 0.23672477900981903,
|
| 174 |
+
"eval_runtime": 6.0691,
|
| 175 |
+
"eval_samples_per_second": 56.186,
|
| 176 |
+
"eval_steps_per_second": 7.085,
|
| 177 |
+
"eval_wer": 0.14452103636795816,
|
| 178 |
+
"step": 946
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 11.63,
|
| 182 |
+
"learning_rate": 2.6522093023255817e-05,
|
| 183 |
+
"loss": 0.414,
|
| 184 |
+
"step": 1000
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 12.0,
|
| 188 |
+
"eval_cer": 0.037615456695372806,
|
| 189 |
+
"eval_loss": 0.21672293543815613,
|
| 190 |
+
"eval_runtime": 6.2518,
|
| 191 |
+
"eval_samples_per_second": 54.545,
|
| 192 |
+
"eval_steps_per_second": 6.878,
|
| 193 |
+
"eval_wer": 0.12598050867601615,
|
| 194 |
+
"step": 1032
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 12.79,
|
| 198 |
+
"learning_rate": 2.617325581395349e-05,
|
| 199 |
+
"loss": 0.3819,
|
| 200 |
+
"step": 1100
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 13.0,
|
| 204 |
+
"eval_cer": 0.03502744188121904,
|
| 205 |
+
"eval_loss": 0.1979241967201233,
|
| 206 |
+
"eval_runtime": 5.8651,
|
| 207 |
+
"eval_samples_per_second": 58.141,
|
| 208 |
+
"eval_steps_per_second": 7.332,
|
| 209 |
+
"eval_wer": 0.11504635131922986,
|
| 210 |
+
"step": 1118
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"epoch": 13.95,
|
| 214 |
+
"learning_rate": 2.5824418604651165e-05,
|
| 215 |
+
"loss": 0.3376,
|
| 216 |
+
"step": 1200
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 14.0,
|
| 220 |
+
"eval_cer": 0.03462585337557449,
|
| 221 |
+
"eval_loss": 0.1877206414937973,
|
| 222 |
+
"eval_runtime": 6.3492,
|
| 223 |
+
"eval_samples_per_second": 53.708,
|
| 224 |
+
"eval_steps_per_second": 6.773,
|
| 225 |
+
"eval_wer": 0.11694794390301878,
|
| 226 |
+
"step": 1204
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 15.0,
|
| 230 |
+
"eval_cer": 0.03346570880371246,
|
| 231 |
+
"eval_loss": 0.17655867338180542,
|
| 232 |
+
"eval_runtime": 6.0868,
|
| 233 |
+
"eval_samples_per_second": 56.023,
|
| 234 |
+
"eval_steps_per_second": 7.065,
|
| 235 |
+
"eval_wer": 0.10839077727596863,
|
| 236 |
+
"step": 1290
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 15.12,
|
| 240 |
+
"learning_rate": 2.547558139534884e-05,
|
| 241 |
+
"loss": 0.3199,
|
| 242 |
+
"step": 1300
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 16.0,
|
| 246 |
+
"eval_cer": 0.032260943286778816,
|
| 247 |
+
"eval_loss": 0.175409734249115,
|
| 248 |
+
"eval_runtime": 6.3852,
|
| 249 |
+
"eval_samples_per_second": 53.405,
|
| 250 |
+
"eval_steps_per_second": 6.734,
|
| 251 |
+
"eval_wer": 0.10316139767054909,
|
| 252 |
+
"step": 1376
|
| 253 |
+
},
|
| 254 |
+
{
|
| 255 |
+
"epoch": 16.28,
|
| 256 |
+
"learning_rate": 2.512674418604651e-05,
|
| 257 |
+
"loss": 0.3174,
|
| 258 |
+
"step": 1400
|
| 259 |
+
},
|
| 260 |
+
{
|
| 261 |
+
"epoch": 17.0,
|
| 262 |
+
"eval_cer": 0.031547008165632946,
|
| 263 |
+
"eval_loss": 0.1697303056716919,
|
| 264 |
+
"eval_runtime": 6.098,
|
| 265 |
+
"eval_samples_per_second": 55.92,
|
| 266 |
+
"eval_steps_per_second": 7.051,
|
| 267 |
+
"eval_wer": 0.10173520323270739,
|
| 268 |
+
"step": 1462
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 17.44,
|
| 272 |
+
"learning_rate": 2.4777906976744186e-05,
|
| 273 |
+
"loss": 0.2747,
|
| 274 |
+
"step": 1500
|
| 275 |
+
},
|
| 276 |
+
{
|
| 277 |
+
"epoch": 18.0,
|
| 278 |
+
"eval_cer": 0.030788452099415466,
|
| 279 |
+
"eval_loss": 0.1668318510055542,
|
| 280 |
+
"eval_runtime": 6.0816,
|
| 281 |
+
"eval_samples_per_second": 56.071,
|
| 282 |
+
"eval_steps_per_second": 7.071,
|
| 283 |
+
"eval_wer": 0.09626812455431424,
|
| 284 |
+
"step": 1548
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 18.6,
|
| 288 |
+
"learning_rate": 2.442906976744186e-05,
|
| 289 |
+
"loss": 0.2618,
|
| 290 |
+
"step": 1600
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 19.0,
|
| 294 |
+
"eval_cer": 0.0300745169782696,
|
| 295 |
+
"eval_loss": 0.16257551312446594,
|
| 296 |
+
"eval_runtime": 5.9751,
|
| 297 |
+
"eval_samples_per_second": 57.07,
|
| 298 |
+
"eval_steps_per_second": 7.197,
|
| 299 |
+
"eval_wer": 0.09365343475160447,
|
| 300 |
+
"step": 1634
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"epoch": 19.77,
|
| 304 |
+
"learning_rate": 2.4080232558139534e-05,
|
| 305 |
+
"loss": 0.2557,
|
| 306 |
+
"step": 1700
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 20.0,
|
| 310 |
+
"eval_cer": 0.029851412252911515,
|
| 311 |
+
"eval_loss": 0.15971209108829498,
|
| 312 |
+
"eval_runtime": 6.1983,
|
| 313 |
+
"eval_samples_per_second": 55.015,
|
| 314 |
+
"eval_steps_per_second": 6.937,
|
| 315 |
+
"eval_wer": 0.09460423104349892,
|
| 316 |
+
"step": 1720
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 20.93,
|
| 320 |
+
"learning_rate": 2.3731395348837208e-05,
|
| 321 |
+
"loss": 0.2578,
|
| 322 |
+
"step": 1800
|
| 323 |
+
},
|
| 324 |
+
{
|
| 325 |
+
"epoch": 21.0,
|
| 326 |
+
"eval_cer": 0.029628307527553435,
|
| 327 |
+
"eval_loss": 0.15854881703853607,
|
| 328 |
+
"eval_runtime": 6.6998,
|
| 329 |
+
"eval_samples_per_second": 50.897,
|
| 330 |
+
"eval_steps_per_second": 6.418,
|
| 331 |
+
"eval_wer": 0.09436653197052532,
|
| 332 |
+
"step": 1806
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 22.0,
|
| 336 |
+
"eval_cer": 0.030163758868412834,
|
| 337 |
+
"eval_loss": 0.15489967167377472,
|
| 338 |
+
"eval_runtime": 5.8895,
|
| 339 |
+
"eval_samples_per_second": 57.899,
|
| 340 |
+
"eval_steps_per_second": 7.301,
|
| 341 |
+
"eval_wer": 0.09650582362728785,
|
| 342 |
+
"step": 1892
|
| 343 |
+
},
|
| 344 |
+
{
|
| 345 |
+
"epoch": 22.09,
|
| 346 |
+
"learning_rate": 2.3382558139534882e-05,
|
| 347 |
+
"loss": 0.2288,
|
| 348 |
+
"step": 1900
|
| 349 |
+
},
|
| 350 |
+
{
|
| 351 |
+
"epoch": 23.0,
|
| 352 |
+
"eval_cer": 0.02837892106554817,
|
| 353 |
+
"eval_loss": 0.15013234317302704,
|
| 354 |
+
"eval_runtime": 5.9578,
|
| 355 |
+
"eval_samples_per_second": 57.236,
|
| 356 |
+
"eval_steps_per_second": 7.217,
|
| 357 |
+
"eval_wer": 0.09389113382457809,
|
| 358 |
+
"step": 1978
|
| 359 |
+
},
|
| 360 |
+
{
|
| 361 |
+
"epoch": 23.26,
|
| 362 |
+
"learning_rate": 2.303372093023256e-05,
|
| 363 |
+
"loss": 0.2269,
|
| 364 |
+
"step": 2000
|
| 365 |
+
},
|
| 366 |
+
{
|
| 367 |
+
"epoch": 24.0,
|
| 368 |
+
"eval_cer": 0.029092856186694036,
|
| 369 |
+
"eval_loss": 0.15236006677150726,
|
| 370 |
+
"eval_runtime": 6.4802,
|
| 371 |
+
"eval_samples_per_second": 52.622,
|
| 372 |
+
"eval_steps_per_second": 6.636,
|
| 373 |
+
"eval_wer": 0.09436653197052532,
|
| 374 |
+
"step": 2064
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 24.42,
|
| 378 |
+
"learning_rate": 2.2684883720930233e-05,
|
| 379 |
+
"loss": 0.2327,
|
| 380 |
+
"step": 2100
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 25.0,
|
| 384 |
+
"eval_cer": 0.028111195395118468,
|
| 385 |
+
"eval_loss": 0.14761987328529358,
|
| 386 |
+
"eval_runtime": 5.9425,
|
| 387 |
+
"eval_samples_per_second": 57.383,
|
| 388 |
+
"eval_steps_per_second": 7.236,
|
| 389 |
+
"eval_wer": 0.09032564772997385,
|
| 390 |
+
"step": 2150
|
| 391 |
+
},
|
| 392 |
+
{
|
| 393 |
+
"epoch": 25.58,
|
| 394 |
+
"learning_rate": 2.2336046511627907e-05,
|
| 395 |
+
"loss": 0.2024,
|
| 396 |
+
"step": 2200
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"epoch": 26.0,
|
| 400 |
+
"eval_cer": 0.028423542010619786,
|
| 401 |
+
"eval_loss": 0.14807896316051483,
|
| 402 |
+
"eval_runtime": 6.0178,
|
| 403 |
+
"eval_samples_per_second": 56.665,
|
| 404 |
+
"eval_steps_per_second": 7.145,
|
| 405 |
+
"eval_wer": 0.09032564772997385,
|
| 406 |
+
"step": 2236
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 26.74,
|
| 410 |
+
"learning_rate": 2.198720930232558e-05,
|
| 411 |
+
"loss": 0.2056,
|
| 412 |
+
"step": 2300
|
| 413 |
+
},
|
| 414 |
+
{
|
| 415 |
+
"epoch": 27.0,
|
| 416 |
+
"eval_cer": 0.02837892106554817,
|
| 417 |
+
"eval_loss": 0.1433907002210617,
|
| 418 |
+
"eval_runtime": 6.0685,
|
| 419 |
+
"eval_samples_per_second": 56.192,
|
| 420 |
+
"eval_steps_per_second": 7.086,
|
| 421 |
+
"eval_wer": 0.09246493938673639,
|
| 422 |
+
"step": 2322
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 27.91,
|
| 426 |
+
"learning_rate": 2.1638372093023255e-05,
|
| 427 |
+
"loss": 0.2097,
|
| 428 |
+
"step": 2400
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 28.0,
|
| 432 |
+
"eval_cer": 0.028021953504975234,
|
| 433 |
+
"eval_loss": 0.1468418389558792,
|
| 434 |
+
"eval_runtime": 7.6911,
|
| 435 |
+
"eval_samples_per_second": 44.337,
|
| 436 |
+
"eval_steps_per_second": 5.591,
|
| 437 |
+
"eval_wer": 0.0893748514380794,
|
| 438 |
+
"step": 2408
|
| 439 |
+
},
|
| 440 |
+
{
|
| 441 |
+
"epoch": 29.0,
|
| 442 |
+
"eval_cer": 0.02730801838382937,
|
| 443 |
+
"eval_loss": 0.14347784221172333,
|
| 444 |
+
"eval_runtime": 6.2489,
|
| 445 |
+
"eval_samples_per_second": 54.569,
|
| 446 |
+
"eval_steps_per_second": 6.881,
|
| 447 |
+
"eval_wer": 0.08604706441644877,
|
| 448 |
+
"step": 2494
|
| 449 |
+
},
|
| 450 |
+
{
|
| 451 |
+
"epoch": 29.07,
|
| 452 |
+
"learning_rate": 2.128953488372093e-05,
|
| 453 |
+
"loss": 0.2177,
|
| 454 |
+
"step": 2500
|
| 455 |
+
},
|
| 456 |
+
{
|
| 457 |
+
"epoch": 30.0,
|
| 458 |
+
"eval_cer": 0.028111195395118468,
|
| 459 |
+
"eval_loss": 0.14982478320598602,
|
| 460 |
+
"eval_runtime": 6.7618,
|
| 461 |
+
"eval_samples_per_second": 50.43,
|
| 462 |
+
"eval_steps_per_second": 6.359,
|
| 463 |
+
"eval_wer": 0.08771095792726408,
|
| 464 |
+
"step": 2580
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 30.23,
|
| 468 |
+
"learning_rate": 2.0940697674418606e-05,
|
| 469 |
+
"loss": 0.1935,
|
| 470 |
+
"step": 2600
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 31.0,
|
| 474 |
+
"eval_cer": 0.027798848779617153,
|
| 475 |
+
"eval_loss": 0.14517097175121307,
|
| 476 |
+
"eval_runtime": 5.9019,
|
| 477 |
+
"eval_samples_per_second": 57.778,
|
| 478 |
+
"eval_steps_per_second": 7.286,
|
| 479 |
+
"eval_wer": 0.08913715236510578,
|
| 480 |
+
"step": 2666
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 31.4,
|
| 484 |
+
"learning_rate": 2.059186046511628e-05,
|
| 485 |
+
"loss": 0.1918,
|
| 486 |
+
"step": 2700
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 32.0,
|
| 490 |
+
"eval_cer": 0.027486502164115835,
|
| 491 |
+
"eval_loss": 0.14662335813045502,
|
| 492 |
+
"eval_runtime": 6.9556,
|
| 493 |
+
"eval_samples_per_second": 49.026,
|
| 494 |
+
"eval_steps_per_second": 6.182,
|
| 495 |
+
"eval_wer": 0.08485856905158069,
|
| 496 |
+
"step": 2752
|
| 497 |
+
},
|
| 498 |
+
{
|
| 499 |
+
"epoch": 32.56,
|
| 500 |
+
"learning_rate": 2.0243023255813954e-05,
|
| 501 |
+
"loss": 0.1805,
|
| 502 |
+
"step": 2800
|
| 503 |
+
},
|
| 504 |
+
{
|
| 505 |
+
"epoch": 33.0,
|
| 506 |
+
"eval_cer": 0.028155816340190085,
|
| 507 |
+
"eval_loss": 0.14373721182346344,
|
| 508 |
+
"eval_runtime": 6.2565,
|
| 509 |
+
"eval_samples_per_second": 54.504,
|
| 510 |
+
"eval_steps_per_second": 6.873,
|
| 511 |
+
"eval_wer": 0.08889945329213216,
|
| 512 |
+
"step": 2838
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 33.72,
|
| 516 |
+
"learning_rate": 1.9894186046511628e-05,
|
| 517 |
+
"loss": 0.1805,
|
| 518 |
+
"step": 2900
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 34.0,
|
| 522 |
+
"eval_cer": 0.027441881219044218,
|
| 523 |
+
"eval_loss": 0.14093732833862305,
|
| 524 |
+
"eval_runtime": 6.6898,
|
| 525 |
+
"eval_samples_per_second": 50.973,
|
| 526 |
+
"eval_steps_per_second": 6.428,
|
| 527 |
+
"eval_wer": 0.08699786070834324,
|
| 528 |
+
"step": 2924
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 34.88,
|
| 532 |
+
"learning_rate": 1.9545348837209302e-05,
|
| 533 |
+
"loss": 0.1835,
|
| 534 |
+
"step": 3000
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 35.0,
|
| 538 |
+
"eval_cer": 0.027129534603542903,
|
| 539 |
+
"eval_loss": 0.14223171770572662,
|
| 540 |
+
"eval_runtime": 6.4219,
|
| 541 |
+
"eval_samples_per_second": 53.099,
|
| 542 |
+
"eval_steps_per_second": 6.696,
|
| 543 |
+
"eval_wer": 0.08557166627050154,
|
| 544 |
+
"step": 3010
|
| 545 |
+
},
|
| 546 |
+
{
|
| 547 |
+
"epoch": 36.0,
|
| 548 |
+
"eval_cer": 0.026415599482397037,
|
| 549 |
+
"eval_loss": 0.13766661286354065,
|
| 550 |
+
"eval_runtime": 7.5399,
|
| 551 |
+
"eval_samples_per_second": 45.226,
|
| 552 |
+
"eval_steps_per_second": 5.703,
|
| 553 |
+
"eval_wer": 0.08509626812455431,
|
| 554 |
+
"step": 3096
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 36.05,
|
| 558 |
+
"learning_rate": 1.919651162790698e-05,
|
| 559 |
+
"loss": 0.1787,
|
| 560 |
+
"step": 3100
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 37.0,
|
| 564 |
+
"eval_cer": 0.026683325152826738,
|
| 565 |
+
"eval_loss": 0.13642320036888123,
|
| 566 |
+
"eval_runtime": 6.5226,
|
| 567 |
+
"eval_samples_per_second": 52.28,
|
| 568 |
+
"eval_steps_per_second": 6.592,
|
| 569 |
+
"eval_wer": 0.08438317090563346,
|
| 570 |
+
"step": 3182
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 37.21,
|
| 574 |
+
"learning_rate": 1.8847674418604653e-05,
|
| 575 |
+
"loss": 0.1695,
|
| 576 |
+
"step": 3200
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 38.0,
|
| 580 |
+
"eval_cer": 0.02677256704296997,
|
| 581 |
+
"eval_loss": 0.14176321029663086,
|
| 582 |
+
"eval_runtime": 6.3748,
|
| 583 |
+
"eval_samples_per_second": 53.492,
|
| 584 |
+
"eval_steps_per_second": 6.745,
|
| 585 |
+
"eval_wer": 0.08485856905158069,
|
| 586 |
+
"step": 3268
|
| 587 |
+
},
|
| 588 |
+
{
|
| 589 |
+
"epoch": 38.37,
|
| 590 |
+
"learning_rate": 1.8498837209302327e-05,
|
| 591 |
+
"loss": 0.1775,
|
| 592 |
+
"step": 3300
|
| 593 |
+
},
|
| 594 |
+
{
|
| 595 |
+
"epoch": 39.0,
|
| 596 |
+
"eval_cer": 0.02704029271339967,
|
| 597 |
+
"eval_loss": 0.1400807499885559,
|
| 598 |
+
"eval_runtime": 6.1197,
|
| 599 |
+
"eval_samples_per_second": 55.721,
|
| 600 |
+
"eval_steps_per_second": 7.026,
|
| 601 |
+
"eval_wer": 0.08438317090563346,
|
| 602 |
+
"step": 3354
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 39.53,
|
| 606 |
+
"learning_rate": 1.815e-05,
|
| 607 |
+
"loss": 0.1763,
|
| 608 |
+
"step": 3400
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 40.0,
|
| 612 |
+
"eval_cer": 0.02650484137254027,
|
| 613 |
+
"eval_loss": 0.14022426307201385,
|
| 614 |
+
"eval_runtime": 5.9433,
|
| 615 |
+
"eval_samples_per_second": 57.375,
|
| 616 |
+
"eval_steps_per_second": 7.235,
|
| 617 |
+
"eval_wer": 0.08153078202995008,
|
| 618 |
+
"step": 3440
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 40.7,
|
| 622 |
+
"learning_rate": 1.7801162790697675e-05,
|
| 623 |
+
"loss": 0.1702,
|
| 624 |
+
"step": 3500
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"epoch": 41.0,
|
| 628 |
+
"eval_cer": 0.026415599482397037,
|
| 629 |
+
"eval_loss": 0.14177961647510529,
|
| 630 |
+
"eval_runtime": 6.7266,
|
| 631 |
+
"eval_samples_per_second": 50.694,
|
| 632 |
+
"eval_steps_per_second": 6.393,
|
| 633 |
+
"eval_wer": 0.08295697646779178,
|
| 634 |
+
"step": 3526
|
| 635 |
+
},
|
| 636 |
+
{
|
| 637 |
+
"epoch": 41.86,
|
| 638 |
+
"learning_rate": 1.7452325581395352e-05,
|
| 639 |
+
"loss": 0.1569,
|
| 640 |
+
"step": 3600
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 42.0,
|
| 644 |
+
"eval_cer": 0.02583552719646602,
|
| 645 |
+
"eval_loss": 0.13996584713459015,
|
| 646 |
+
"eval_runtime": 5.888,
|
| 647 |
+
"eval_samples_per_second": 57.914,
|
| 648 |
+
"eval_steps_per_second": 7.303,
|
| 649 |
+
"eval_wer": 0.08248157832184455,
|
| 650 |
+
"step": 3612
|
| 651 |
+
},
|
| 652 |
+
{
|
| 653 |
+
"epoch": 43.0,
|
| 654 |
+
"eval_cer": 0.026192494757038953,
|
| 655 |
+
"eval_loss": 0.14005370438098907,
|
| 656 |
+
"eval_runtime": 5.8324,
|
| 657 |
+
"eval_samples_per_second": 58.466,
|
| 658 |
+
"eval_steps_per_second": 7.373,
|
| 659 |
+
"eval_wer": 0.08153078202995008,
|
| 660 |
+
"step": 3698
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 43.02,
|
| 664 |
+
"learning_rate": 1.7103488372093026e-05,
|
| 665 |
+
"loss": 0.1617,
|
| 666 |
+
"step": 3700
|
| 667 |
+
},
|
| 668 |
+
{
|
| 669 |
+
"epoch": 44.0,
|
| 670 |
+
"eval_cer": 0.026192494757038953,
|
| 671 |
+
"eval_loss": 0.14058516919612885,
|
| 672 |
+
"eval_runtime": 6.1068,
|
| 673 |
+
"eval_samples_per_second": 55.839,
|
| 674 |
+
"eval_steps_per_second": 7.041,
|
| 675 |
+
"eval_wer": 0.07915379130021392,
|
| 676 |
+
"step": 3784
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"epoch": 44.19,
|
| 680 |
+
"learning_rate": 1.67546511627907e-05,
|
| 681 |
+
"loss": 0.1596,
|
| 682 |
+
"step": 3800
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 45.0,
|
| 686 |
+
"eval_cer": 0.02637097853732542,
|
| 687 |
+
"eval_loss": 0.1394573301076889,
|
| 688 |
+
"eval_runtime": 6.5093,
|
| 689 |
+
"eval_samples_per_second": 52.386,
|
| 690 |
+
"eval_steps_per_second": 6.606,
|
| 691 |
+
"eval_wer": 0.0817684811029237,
|
| 692 |
+
"step": 3870
|
| 693 |
+
},
|
| 694 |
+
{
|
| 695 |
+
"epoch": 45.35,
|
| 696 |
+
"learning_rate": 1.6405813953488374e-05,
|
| 697 |
+
"loss": 0.1431,
|
| 698 |
+
"step": 3900
|
| 699 |
+
},
|
| 700 |
+
{
|
| 701 |
+
"epoch": 46.0,
|
| 702 |
+
"eval_cer": 0.026192494757038953,
|
| 703 |
+
"eval_loss": 0.13823828101158142,
|
| 704 |
+
"eval_runtime": 6.0067,
|
| 705 |
+
"eval_samples_per_second": 56.77,
|
| 706 |
+
"eval_steps_per_second": 7.159,
|
| 707 |
+
"eval_wer": 0.08153078202995008,
|
| 708 |
+
"step": 3956
|
| 709 |
+
},
|
| 710 |
+
{
|
| 711 |
+
"epoch": 46.51,
|
| 712 |
+
"learning_rate": 1.6056976744186047e-05,
|
| 713 |
+
"loss": 0.158,
|
| 714 |
+
"step": 4000
|
| 715 |
+
},
|
| 716 |
+
{
|
| 717 |
+
"epoch": 47.0,
|
| 718 |
+
"eval_cer": 0.02650484137254027,
|
| 719 |
+
"eval_loss": 0.13912436366081238,
|
| 720 |
+
"eval_runtime": 6.0425,
|
| 721 |
+
"eval_samples_per_second": 56.433,
|
| 722 |
+
"eval_steps_per_second": 7.116,
|
| 723 |
+
"eval_wer": 0.08129308295697647,
|
| 724 |
+
"step": 4042
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 47.67,
|
| 728 |
+
"learning_rate": 1.570813953488372e-05,
|
| 729 |
+
"loss": 0.1552,
|
| 730 |
+
"step": 4100
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 48.0,
|
| 734 |
+
"eval_cer": 0.026594083262683504,
|
| 735 |
+
"eval_loss": 0.13926950097084045,
|
| 736 |
+
"eval_runtime": 6.5183,
|
| 737 |
+
"eval_samples_per_second": 52.314,
|
| 738 |
+
"eval_steps_per_second": 6.597,
|
| 739 |
+
"eval_wer": 0.08248157832184455,
|
| 740 |
+
"step": 4128
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"epoch": 48.84,
|
| 744 |
+
"learning_rate": 1.5359302325581395e-05,
|
| 745 |
+
"loss": 0.1379,
|
| 746 |
+
"step": 4200
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 49.0,
|
| 750 |
+
"eval_cer": 0.025612422471107937,
|
| 751 |
+
"eval_loss": 0.13708749413490295,
|
| 752 |
+
"eval_runtime": 6.0023,
|
| 753 |
+
"eval_samples_per_second": 56.811,
|
| 754 |
+
"eval_steps_per_second": 7.164,
|
| 755 |
+
"eval_wer": 0.08105538388400285,
|
| 756 |
+
"step": 4214
|
| 757 |
+
},
|
| 758 |
+
{
|
| 759 |
+
"epoch": 50.0,
|
| 760 |
+
"learning_rate": 1.5010465116279071e-05,
|
| 761 |
+
"loss": 0.145,
|
| 762 |
+
"step": 4300
|
| 763 |
+
},
|
| 764 |
+
{
|
| 765 |
+
"epoch": 50.0,
|
| 766 |
+
"eval_cer": 0.02556780152603632,
|
| 767 |
+
"eval_loss": 0.1392103135585785,
|
| 768 |
+
"eval_runtime": 7.2777,
|
| 769 |
+
"eval_samples_per_second": 46.855,
|
| 770 |
+
"eval_steps_per_second": 5.908,
|
| 771 |
+
"eval_wer": 0.0801045875921084,
|
| 772 |
+
"step": 4300
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 51.0,
|
| 776 |
+
"eval_cer": 0.02623711570211057,
|
| 777 |
+
"eval_loss": 0.14161016047000885,
|
| 778 |
+
"eval_runtime": 6.4267,
|
| 779 |
+
"eval_samples_per_second": 53.06,
|
| 780 |
+
"eval_steps_per_second": 6.691,
|
| 781 |
+
"eval_wer": 0.08200618017589731,
|
| 782 |
+
"step": 4386
|
| 783 |
+
},
|
| 784 |
+
{
|
| 785 |
+
"epoch": 51.16,
|
| 786 |
+
"learning_rate": 1.4661627906976745e-05,
|
| 787 |
+
"loss": 0.1647,
|
| 788 |
+
"step": 4400
|
| 789 |
+
},
|
| 790 |
+
{
|
| 791 |
+
"epoch": 52.0,
|
| 792 |
+
"eval_cer": 0.02556780152603632,
|
| 793 |
+
"eval_loss": 0.13915704190731049,
|
| 794 |
+
"eval_runtime": 6.1093,
|
| 795 |
+
"eval_samples_per_second": 55.816,
|
| 796 |
+
"eval_steps_per_second": 7.038,
|
| 797 |
+
"eval_wer": 0.07891609222724032,
|
| 798 |
+
"step": 4472
|
| 799 |
+
},
|
| 800 |
+
{
|
| 801 |
+
"epoch": 52.33,
|
| 802 |
+
"learning_rate": 1.4312790697674419e-05,
|
| 803 |
+
"loss": 0.1493,
|
| 804 |
+
"step": 4500
|
| 805 |
+
},
|
| 806 |
+
{
|
| 807 |
+
"epoch": 53.0,
|
| 808 |
+
"eval_cer": 0.02570166436125117,
|
| 809 |
+
"eval_loss": 0.14251555502414703,
|
| 810 |
+
"eval_runtime": 6.4895,
|
| 811 |
+
"eval_samples_per_second": 52.546,
|
| 812 |
+
"eval_steps_per_second": 6.626,
|
| 813 |
+
"eval_wer": 0.07939149037318755,
|
| 814 |
+
"step": 4558
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 53.49,
|
| 818 |
+
"learning_rate": 1.3963953488372094e-05,
|
| 819 |
+
"loss": 0.1492,
|
| 820 |
+
"step": 4600
|
| 821 |
+
},
|
| 822 |
+
{
|
| 823 |
+
"epoch": 54.0,
|
| 824 |
+
"eval_cer": 0.025746285306322787,
|
| 825 |
+
"eval_loss": 0.1419229656457901,
|
| 826 |
+
"eval_runtime": 5.9077,
|
| 827 |
+
"eval_samples_per_second": 57.721,
|
| 828 |
+
"eval_steps_per_second": 7.279,
|
| 829 |
+
"eval_wer": 0.07962918944616117,
|
| 830 |
+
"step": 4644
|
| 831 |
+
},
|
| 832 |
+
{
|
| 833 |
+
"epoch": 54.65,
|
| 834 |
+
"learning_rate": 1.3615116279069768e-05,
|
| 835 |
+
"loss": 0.139,
|
| 836 |
+
"step": 4700
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 55.0,
|
| 840 |
+
"eval_cer": 0.024987729240105304,
|
| 841 |
+
"eval_loss": 0.13996940851211548,
|
| 842 |
+
"eval_runtime": 5.9605,
|
| 843 |
+
"eval_samples_per_second": 57.21,
|
| 844 |
+
"eval_steps_per_second": 7.214,
|
| 845 |
+
"eval_wer": 0.07582600427858331,
|
| 846 |
+
"step": 4730
|
| 847 |
+
},
|
| 848 |
+
{
|
| 849 |
+
"epoch": 55.81,
|
| 850 |
+
"learning_rate": 1.3266279069767444e-05,
|
| 851 |
+
"loss": 0.1385,
|
| 852 |
+
"step": 4800
|
| 853 |
+
},
|
| 854 |
+
{
|
| 855 |
+
"epoch": 56.0,
|
| 856 |
+
"eval_cer": 0.02534469680067824,
|
| 857 |
+
"eval_loss": 0.1423511654138565,
|
| 858 |
+
"eval_runtime": 6.5323,
|
| 859 |
+
"eval_samples_per_second": 52.203,
|
| 860 |
+
"eval_steps_per_second": 6.583,
|
| 861 |
+
"eval_wer": 0.07915379130021392,
|
| 862 |
+
"step": 4816
|
| 863 |
+
},
|
| 864 |
+
{
|
| 865 |
+
"epoch": 56.98,
|
| 866 |
+
"learning_rate": 1.2917441860465116e-05,
|
| 867 |
+
"loss": 0.128,
|
| 868 |
+
"step": 4900
|
| 869 |
+
},
|
| 870 |
+
{
|
| 871 |
+
"epoch": 57.0,
|
| 872 |
+
"eval_cer": 0.025255454910535005,
|
| 873 |
+
"eval_loss": 0.1403101086616516,
|
| 874 |
+
"eval_runtime": 6.4807,
|
| 875 |
+
"eval_samples_per_second": 52.618,
|
| 876 |
+
"eval_steps_per_second": 6.635,
|
| 877 |
+
"eval_wer": 0.08057998573805562,
|
| 878 |
+
"step": 4902
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 57.0,
|
| 882 |
+
"step": 4902,
|
| 883 |
+
"total_flos": 4.0970110980881424e+19,
|
| 884 |
+
"train_loss": 0.6410711561475954,
|
| 885 |
+
"train_runtime": 7770.6868,
|
| 886 |
+
"train_samples_per_second": 35.299,
|
| 887 |
+
"train_steps_per_second": 1.107
|
| 888 |
+
}
|
| 889 |
+
],
|
| 890 |
+
"max_steps": 8600,
|
| 891 |
+
"num_train_epochs": 100,
|
| 892 |
+
"total_flos": 4.0970110980881424e+19,
|
| 893 |
+
"trial_name": null,
|
| 894 |
+
"trial_params": null
|
| 895 |
+
}
|