End of training
Browse files- all_results.json +14 -0
- eval_results.json +9 -0
- pytorch_model.bin +1 -1
- train_results.json +8 -0
- trainer_state.json +940 -0
- wandb/run-20220205_233515-2f29fa6z/files/config.yaml +20 -0
- wandb/run-20220205_233515-2f29fa6z/files/output.log +0 -0
- wandb/run-20220205_233515-2f29fa6z/files/wandb-summary.json +0 -0
- wandb/run-20220205_233515-2f29fa6z/logs/debug-internal.log +0 -0
- wandb/run-20220205_233515-2f29fa6z/run-2f29fa6z.wandb +2 -2
all_results.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 100.0,
|
| 3 |
+
"eval_loss": 0.22779151797294617,
|
| 4 |
+
"eval_runtime": 337.5199,
|
| 5 |
+
"eval_samples": 6463,
|
| 6 |
+
"eval_samples_per_second": 19.149,
|
| 7 |
+
"eval_steps_per_second": 0.267,
|
| 8 |
+
"eval_wer": 0.17871523648578164,
|
| 9 |
+
"train_loss": 0.14547648023750823,
|
| 10 |
+
"train_runtime": 96097.4367,
|
| 11 |
+
"train_samples": 17030,
|
| 12 |
+
"train_samples_per_second": 17.722,
|
| 13 |
+
"train_steps_per_second": 0.123
|
| 14 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 100.0,
|
| 3 |
+
"eval_loss": 0.22779151797294617,
|
| 4 |
+
"eval_runtime": 337.5199,
|
| 5 |
+
"eval_samples": 6463,
|
| 6 |
+
"eval_samples_per_second": 19.149,
|
| 7 |
+
"eval_steps_per_second": 0.267,
|
| 8 |
+
"eval_wer": 0.17871523648578164
|
| 9 |
+
}
|
pytorch_model.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1262058993
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a606acfb7fb479ef93ee010de02a0920109fad8b3b972f6ea1ff6eda4a1e68d6
|
| 3 |
size 1262058993
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 100.0,
|
| 3 |
+
"train_loss": 0.14547648023750823,
|
| 4 |
+
"train_runtime": 96097.4367,
|
| 5 |
+
"train_samples": 17030,
|
| 6 |
+
"train_samples_per_second": 17.722,
|
| 7 |
+
"train_steps_per_second": 0.123
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,940 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 99.9957805907173,
|
| 5 |
+
"global_step": 11800,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.84,
|
| 12 |
+
"learning_rate": 0.00019800000000000002,
|
| 13 |
+
"loss": 5.6106,
|
| 14 |
+
"step": 100
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 1.69,
|
| 18 |
+
"learning_rate": 0.000398,
|
| 19 |
+
"loss": 2.7044,
|
| 20 |
+
"step": 200
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 2.54,
|
| 24 |
+
"learning_rate": 0.000598,
|
| 25 |
+
"loss": 0.8324,
|
| 26 |
+
"step": 300
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 3.39,
|
| 30 |
+
"learning_rate": 0.0007980000000000001,
|
| 31 |
+
"loss": 0.345,
|
| 32 |
+
"step": 400
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 4.24,
|
| 36 |
+
"learning_rate": 0.000998,
|
| 37 |
+
"loss": 0.2548,
|
| 38 |
+
"step": 500
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 4.24,
|
| 42 |
+
"eval_loss": 0.2469930201768875,
|
| 43 |
+
"eval_runtime": 294.3209,
|
| 44 |
+
"eval_samples_per_second": 21.959,
|
| 45 |
+
"eval_steps_per_second": 0.306,
|
| 46 |
+
"eval_wer": 0.36629738582545746,
|
| 47 |
+
"step": 500
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 5.08,
|
| 51 |
+
"learning_rate": 0.0009998106236074637,
|
| 52 |
+
"loss": 0.2203,
|
| 53 |
+
"step": 600
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 5.93,
|
| 57 |
+
"learning_rate": 0.0009992349703990264,
|
| 58 |
+
"loss": 0.1774,
|
| 59 |
+
"step": 700
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 6.78,
|
| 63 |
+
"learning_rate": 0.0009982734659650546,
|
| 64 |
+
"loss": 0.1643,
|
| 65 |
+
"step": 800
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 7.62,
|
| 69 |
+
"learning_rate": 0.0009969268534379132,
|
| 70 |
+
"loss": 0.1516,
|
| 71 |
+
"step": 900
|
| 72 |
+
},
|
| 73 |
+
{
|
| 74 |
+
"epoch": 8.47,
|
| 75 |
+
"learning_rate": 0.00099519617359424,
|
| 76 |
+
"loss": 0.1435,
|
| 77 |
+
"step": 1000
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 8.47,
|
| 81 |
+
"eval_loss": 0.20002000033855438,
|
| 82 |
+
"eval_runtime": 295.0975,
|
| 83 |
+
"eval_samples_per_second": 21.901,
|
| 84 |
+
"eval_steps_per_second": 0.305,
|
| 85 |
+
"eval_wer": 0.2791095533162254,
|
| 86 |
+
"step": 1000
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 9.32,
|
| 90 |
+
"learning_rate": 0.0009930827640505444,
|
| 91 |
+
"loss": 0.1393,
|
| 92 |
+
"step": 1100
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 10.17,
|
| 96 |
+
"learning_rate": 0.0009905882582293853,
|
| 97 |
+
"loss": 0.1323,
|
| 98 |
+
"step": 1200
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 11.02,
|
| 102 |
+
"learning_rate": 0.0009877145840969208,
|
| 103 |
+
"loss": 0.1317,
|
| 104 |
+
"step": 1300
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 11.86,
|
| 108 |
+
"learning_rate": 0.0009844639626728146,
|
| 109 |
+
"loss": 0.1156,
|
| 110 |
+
"step": 1400
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 12.71,
|
| 114 |
+
"learning_rate": 0.000980838906313641,
|
| 115 |
+
"loss": 0.1158,
|
| 116 |
+
"step": 1500
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 12.71,
|
| 120 |
+
"eval_loss": 0.20303700864315033,
|
| 121 |
+
"eval_runtime": 301.1246,
|
| 122 |
+
"eval_samples_per_second": 21.463,
|
| 123 |
+
"eval_steps_per_second": 0.299,
|
| 124 |
+
"eval_wer": 0.26521457929106423,
|
| 125 |
+
"step": 1500
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 13.56,
|
| 129 |
+
"learning_rate": 0.0009768422167711244,
|
| 130 |
+
"loss": 0.1174,
|
| 131 |
+
"step": 1600
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 14.41,
|
| 135 |
+
"learning_rate": 0.0009724769830267044,
|
| 136 |
+
"loss": 0.1118,
|
| 137 |
+
"step": 1700
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 15.25,
|
| 141 |
+
"learning_rate": 0.0009677465789041117,
|
| 142 |
+
"loss": 0.1115,
|
| 143 |
+
"step": 1800
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 16.1,
|
| 147 |
+
"learning_rate": 0.0009626546604617881,
|
| 148 |
+
"loss": 0.1149,
|
| 149 |
+
"step": 1900
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 16.95,
|
| 153 |
+
"learning_rate": 0.0009572051631671743,
|
| 154 |
+
"loss": 0.1094,
|
| 155 |
+
"step": 2000
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 16.95,
|
| 159 |
+
"eval_loss": 0.20957913994789124,
|
| 160 |
+
"eval_runtime": 294.9247,
|
| 161 |
+
"eval_samples_per_second": 21.914,
|
| 162 |
+
"eval_steps_per_second": 0.305,
|
| 163 |
+
"eval_wer": 0.26045774133650457,
|
| 164 |
+
"step": 2000
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 17.79,
|
| 168 |
+
"learning_rate": 0.0009514022988550459,
|
| 169 |
+
"loss": 0.1049,
|
| 170 |
+
"step": 2100
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 18.64,
|
| 174 |
+
"learning_rate": 0.0009452505524722506,
|
| 175 |
+
"loss": 0.1031,
|
| 176 |
+
"step": 2200
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 19.49,
|
| 180 |
+
"learning_rate": 0.0009387546786113578,
|
| 181 |
+
"loss": 0.0991,
|
| 182 |
+
"step": 2300
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 20.34,
|
| 186 |
+
"learning_rate": 0.0009319196978359077,
|
| 187 |
+
"loss": 0.1015,
|
| 188 |
+
"step": 2400
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 21.19,
|
| 192 |
+
"learning_rate": 0.0009247508928000911,
|
| 193 |
+
"loss": 0.1004,
|
| 194 |
+
"step": 2500
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 21.19,
|
| 198 |
+
"eval_loss": 0.21501973271369934,
|
| 199 |
+
"eval_runtime": 295.0309,
|
| 200 |
+
"eval_samples_per_second": 21.906,
|
| 201 |
+
"eval_steps_per_second": 0.305,
|
| 202 |
+
"eval_wer": 0.2477311134756212,
|
| 203 |
+
"step": 2500
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 22.03,
|
| 207 |
+
"learning_rate": 0.0009172538041658665,
|
| 208 |
+
"loss": 0.0949,
|
| 209 |
+
"step": 2600
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 22.88,
|
| 213 |
+
"learning_rate": 0.0009094342263206646,
|
| 214 |
+
"loss": 0.0869,
|
| 215 |
+
"step": 2700
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 23.73,
|
| 219 |
+
"learning_rate": 0.0009012982028989929,
|
| 220 |
+
"loss": 0.0971,
|
| 221 |
+
"step": 2800
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 24.57,
|
| 225 |
+
"learning_rate": 0.000892852022111403,
|
| 226 |
+
"loss": 0.0907,
|
| 227 |
+
"step": 2900
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 25.42,
|
| 231 |
+
"learning_rate": 0.0008841022118844243,
|
| 232 |
+
"loss": 0.0945,
|
| 233 |
+
"step": 3000
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 25.42,
|
| 237 |
+
"eval_loss": 0.20718325674533844,
|
| 238 |
+
"eval_runtime": 295.2378,
|
| 239 |
+
"eval_samples_per_second": 21.891,
|
| 240 |
+
"eval_steps_per_second": 0.305,
|
| 241 |
+
"eval_wer": 0.2369030481316893,
|
| 242 |
+
"step": 3000
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 26.27,
|
| 246 |
+
"learning_rate": 0.0008750555348152298,
|
| 247 |
+
"loss": 0.0884,
|
| 248 |
+
"step": 3100
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 27.12,
|
| 252 |
+
"learning_rate": 0.0008657189829449246,
|
| 253 |
+
"loss": 0.0894,
|
| 254 |
+
"step": 3200
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 27.96,
|
| 258 |
+
"learning_rate": 0.0008560997723545024,
|
| 259 |
+
"loss": 0.0851,
|
| 260 |
+
"step": 3300
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 28.81,
|
| 264 |
+
"learning_rate": 0.0008462053375876455,
|
| 265 |
+
"loss": 0.09,
|
| 266 |
+
"step": 3400
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 29.66,
|
| 270 |
+
"learning_rate": 0.0008360433259046764,
|
| 271 |
+
"loss": 0.0844,
|
| 272 |
+
"step": 3500
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 29.66,
|
| 276 |
+
"eval_loss": 0.1981162428855896,
|
| 277 |
+
"eval_runtime": 296.5443,
|
| 278 |
+
"eval_samples_per_second": 21.794,
|
| 279 |
+
"eval_steps_per_second": 0.303,
|
| 280 |
+
"eval_wer": 0.23275124658363064,
|
| 281 |
+
"step": 3500
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 30.51,
|
| 285 |
+
"learning_rate": 0.0008256215913721052,
|
| 286 |
+
"loss": 0.084,
|
| 287 |
+
"step": 3600
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 31.35,
|
| 291 |
+
"learning_rate": 0.0008149481887923395,
|
| 292 |
+
"loss": 0.081,
|
| 293 |
+
"step": 3700
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 32.2,
|
| 297 |
+
"learning_rate": 0.0008040313674782489,
|
| 298 |
+
"loss": 0.0812,
|
| 299 |
+
"step": 3800
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 33.05,
|
| 303 |
+
"learning_rate": 0.0007928795648773933,
|
| 304 |
+
"loss": 0.0836,
|
| 305 |
+
"step": 3900
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 33.89,
|
| 309 |
+
"learning_rate": 0.0007815014000508474,
|
| 310 |
+
"loss": 0.0877,
|
| 311 |
+
"step": 4000
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 33.89,
|
| 315 |
+
"eval_loss": 0.20411191880702972,
|
| 316 |
+
"eval_runtime": 294.5281,
|
| 317 |
+
"eval_samples_per_second": 21.944,
|
| 318 |
+
"eval_steps_per_second": 0.306,
|
| 319 |
+
"eval_wer": 0.24249441906073313,
|
| 320 |
+
"step": 4000
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 34.74,
|
| 324 |
+
"learning_rate": 0.0007699056670116568,
|
| 325 |
+
"loss": 0.0842,
|
| 326 |
+
"step": 4100
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 35.59,
|
| 330 |
+
"learning_rate": 0.0007581013279280784,
|
| 331 |
+
"loss": 0.0802,
|
| 332 |
+
"step": 4200
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 36.44,
|
| 336 |
+
"learning_rate": 0.0007460975061968531,
|
| 337 |
+
"loss": 0.0812,
|
| 338 |
+
"step": 4300
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 37.29,
|
| 342 |
+
"learning_rate": 0.0007339034793918703,
|
| 343 |
+
"loss": 0.0754,
|
| 344 |
+
"step": 4400
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 38.14,
|
| 348 |
+
"learning_rate": 0.00072152867209367,
|
| 349 |
+
"loss": 0.0741,
|
| 350 |
+
"step": 4500
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 38.14,
|
| 354 |
+
"eval_loss": 0.235269233584404,
|
| 355 |
+
"eval_runtime": 301.8823,
|
| 356 |
+
"eval_samples_per_second": 21.409,
|
| 357 |
+
"eval_steps_per_second": 0.298,
|
| 358 |
+
"eval_wer": 0.24213974254657736,
|
| 359 |
+
"step": 4500
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 38.98,
|
| 363 |
+
"learning_rate": 0.0007089826486053256,
|
| 364 |
+
"loss": 0.0752,
|
| 365 |
+
"step": 4600
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 39.83,
|
| 369 |
+
"learning_rate": 0.0006962751055603378,
|
| 370 |
+
"loss": 0.0725,
|
| 371 |
+
"step": 4700
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 40.68,
|
| 375 |
+
"learning_rate": 0.0006834158644282511,
|
| 376 |
+
"loss": 0.0712,
|
| 377 |
+
"step": 4800
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 41.52,
|
| 381 |
+
"learning_rate": 0.000670414863923788,
|
| 382 |
+
"loss": 0.0673,
|
| 383 |
+
"step": 4900
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 42.37,
|
| 387 |
+
"learning_rate": 0.0006572821523253649,
|
| 388 |
+
"loss": 0.0676,
|
| 389 |
+
"step": 5000
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 42.37,
|
| 393 |
+
"eval_loss": 0.20924021303653717,
|
| 394 |
+
"eval_runtime": 294.7332,
|
| 395 |
+
"eval_samples_per_second": 21.928,
|
| 396 |
+
"eval_steps_per_second": 0.305,
|
| 397 |
+
"eval_wer": 0.22127641818447352,
|
| 398 |
+
"step": 5000
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 43.22,
|
| 402 |
+
"learning_rate": 0.000644027879708929,
|
| 403 |
+
"loss": 0.0691,
|
| 404 |
+
"step": 5100
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 44.07,
|
| 408 |
+
"learning_rate": 0.0006306622901031152,
|
| 409 |
+
"loss": 0.0621,
|
| 410 |
+
"step": 5200
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 44.91,
|
| 414 |
+
"learning_rate": 0.000617195713571792,
|
| 415 |
+
"loss": 0.0648,
|
| 416 |
+
"step": 5300
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 45.76,
|
| 420 |
+
"learning_rate": 0.0006036385582301083,
|
| 421 |
+
"loss": 0.0682,
|
| 422 |
+
"step": 5400
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 46.61,
|
| 426 |
+
"learning_rate": 0.0005900013022002183,
|
| 427 |
+
"loss": 0.0623,
|
| 428 |
+
"step": 5500
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 46.61,
|
| 432 |
+
"eval_loss": 0.22174060344696045,
|
| 433 |
+
"eval_runtime": 298.483,
|
| 434 |
+
"eval_samples_per_second": 21.653,
|
| 435 |
+
"eval_steps_per_second": 0.302,
|
| 436 |
+
"eval_wer": 0.2250318165696522,
|
| 437 |
+
"step": 5500
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 47.46,
|
| 441 |
+
"learning_rate": 0.0005762944855128968,
|
| 442 |
+
"loss": 0.0702,
|
| 443 |
+
"step": 5600
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 48.3,
|
| 447 |
+
"learning_rate": 0.0005625287019613087,
|
| 448 |
+
"loss": 0.0597,
|
| 449 |
+
"step": 5700
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 49.15,
|
| 453 |
+
"learning_rate": 0.0005487145909132243,
|
| 454 |
+
"loss": 0.0557,
|
| 455 |
+
"step": 5800
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 50.0,
|
| 459 |
+
"learning_rate": 0.0005348628290880112,
|
| 460 |
+
"loss": 0.0547,
|
| 461 |
+
"step": 5900
|
| 462 |
+
},
|
| 463 |
+
{
|
| 464 |
+
"epoch": 50.84,
|
| 465 |
+
"learning_rate": 0.0005209841223047587,
|
| 466 |
+
"loss": 0.0574,
|
| 467 |
+
"step": 6000
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"epoch": 50.84,
|
| 471 |
+
"eval_loss": 0.21524770557880402,
|
| 472 |
+
"eval_runtime": 297.0196,
|
| 473 |
+
"eval_samples_per_second": 21.76,
|
| 474 |
+
"eval_steps_per_second": 0.303,
|
| 475 |
+
"eval_wer": 0.2179382862865369,
|
| 476 |
+
"step": 6000
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 51.69,
|
| 480 |
+
"learning_rate": 0.00050708919720791,
|
| 481 |
+
"loss": 0.0571,
|
| 482 |
+
"step": 6100
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 52.54,
|
| 486 |
+
"learning_rate": 0.0004931887929768004,
|
| 487 |
+
"loss": 0.0557,
|
| 488 |
+
"step": 6200
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 53.39,
|
| 492 |
+
"learning_rate": 0.0004792936530255048,
|
| 493 |
+
"loss": 0.0513,
|
| 494 |
+
"step": 6300
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 54.24,
|
| 498 |
+
"learning_rate": 0.00046541451669941626,
|
| 499 |
+
"loss": 0.0542,
|
| 500 |
+
"step": 6400
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 55.08,
|
| 504 |
+
"learning_rate": 0.00045156211097496503,
|
| 505 |
+
"loss": 0.0583,
|
| 506 |
+
"step": 6500
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 55.08,
|
| 510 |
+
"eval_loss": 0.22067895531654358,
|
| 511 |
+
"eval_runtime": 297.4094,
|
| 512 |
+
"eval_samples_per_second": 21.731,
|
| 513 |
+
"eval_steps_per_second": 0.303,
|
| 514 |
+
"eval_wer": 0.21860591266612422,
|
| 515 |
+
"step": 6500
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 55.93,
|
| 519 |
+
"learning_rate": 0.00043774714216890067,
|
| 520 |
+
"loss": 0.0516,
|
| 521 |
+
"step": 6600
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 56.78,
|
| 525 |
+
"learning_rate": 0.000423980287663539,
|
| 526 |
+
"loss": 0.0552,
|
| 527 |
+
"step": 6700
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 57.62,
|
| 531 |
+
"learning_rate": 0.00041027218765437467,
|
| 532 |
+
"loss": 0.0545,
|
| 533 |
+
"step": 6800
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 58.47,
|
| 537 |
+
"learning_rate": 0.0003966334369264325,
|
| 538 |
+
"loss": 0.0497,
|
| 539 |
+
"step": 6900
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"epoch": 59.32,
|
| 543 |
+
"learning_rate": 0.00038307457666571714,
|
| 544 |
+
"loss": 0.0488,
|
| 545 |
+
"step": 7000
|
| 546 |
+
},
|
| 547 |
+
{
|
| 548 |
+
"epoch": 59.32,
|
| 549 |
+
"eval_loss": 0.22246450185775757,
|
| 550 |
+
"eval_runtime": 291.806,
|
| 551 |
+
"eval_samples_per_second": 22.148,
|
| 552 |
+
"eval_steps_per_second": 0.308,
|
| 553 |
+
"eval_wer": 0.21593540714777493,
|
| 554 |
+
"step": 7000
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 60.17,
|
| 558 |
+
"learning_rate": 0.000369606086312089,
|
| 559 |
+
"loss": 0.0467,
|
| 560 |
+
"step": 7100
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 61.02,
|
| 564 |
+
"learning_rate": 0.00035623837545986186,
|
| 565 |
+
"loss": 0.0496,
|
| 566 |
+
"step": 7200
|
| 567 |
+
},
|
| 568 |
+
{
|
| 569 |
+
"epoch": 61.86,
|
| 570 |
+
"learning_rate": 0.0003429817758123833,
|
| 571 |
+
"loss": 0.0477,
|
| 572 |
+
"step": 7300
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 62.71,
|
| 576 |
+
"learning_rate": 0.00032984653319681614,
|
| 577 |
+
"loss": 0.0491,
|
| 578 |
+
"step": 7400
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 63.56,
|
| 582 |
+
"learning_rate": 0.0003168427996452925,
|
| 583 |
+
"loss": 0.0456,
|
| 584 |
+
"step": 7500
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 63.56,
|
| 588 |
+
"eval_loss": 0.22932520508766174,
|
| 589 |
+
"eval_runtime": 293.0052,
|
| 590 |
+
"eval_samples_per_second": 22.058,
|
| 591 |
+
"eval_steps_per_second": 0.307,
|
| 592 |
+
"eval_wer": 0.20310446266508106,
|
| 593 |
+
"step": 7500
|
| 594 |
+
},
|
| 595 |
+
{
|
| 596 |
+
"epoch": 64.41,
|
| 597 |
+
"learning_rate": 0.0003039806255485599,
|
| 598 |
+
"loss": 0.042,
|
| 599 |
+
"step": 7600
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 65.25,
|
| 603 |
+
"learning_rate": 0.0002912699518881855,
|
| 604 |
+
"loss": 0.0431,
|
| 605 |
+
"step": 7700
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 66.1,
|
| 609 |
+
"learning_rate": 0.00027872060255331935,
|
| 610 |
+
"loss": 0.042,
|
| 611 |
+
"step": 7800
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 66.95,
|
| 615 |
+
"learning_rate": 0.0002663422767479589,
|
| 616 |
+
"loss": 0.0422,
|
| 617 |
+
"step": 7900
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"epoch": 67.79,
|
| 621 |
+
"learning_rate": 0.0002541445414945791,
|
| 622 |
+
"loss": 0.041,
|
| 623 |
+
"step": 8000
|
| 624 |
+
},
|
| 625 |
+
{
|
| 626 |
+
"epoch": 67.79,
|
| 627 |
+
"eval_loss": 0.22772204875946045,
|
| 628 |
+
"eval_runtime": 292.1439,
|
| 629 |
+
"eval_samples_per_second": 22.123,
|
| 630 |
+
"eval_steps_per_second": 0.308,
|
| 631 |
+
"eval_wer": 0.20128935344557802,
|
| 632 |
+
"step": 8000
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 68.64,
|
| 636 |
+
"learning_rate": 0.000242136824239924,
|
| 637 |
+
"loss": 0.042,
|
| 638 |
+
"step": 8100
|
| 639 |
+
},
|
| 640 |
+
{
|
| 641 |
+
"epoch": 69.49,
|
| 642 |
+
"learning_rate": 0.00023032840556867502,
|
| 643 |
+
"loss": 0.0404,
|
| 644 |
+
"step": 8200
|
| 645 |
+
},
|
| 646 |
+
{
|
| 647 |
+
"epoch": 70.34,
|
| 648 |
+
"learning_rate": 0.00021872841203062537,
|
| 649 |
+
"loss": 0.0413,
|
| 650 |
+
"step": 8300
|
| 651 |
+
},
|
| 652 |
+
{
|
| 653 |
+
"epoch": 71.19,
|
| 654 |
+
"learning_rate": 0.0002074585300038035,
|
| 655 |
+
"loss": 0.0385,
|
| 656 |
+
"step": 8400
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 72.03,
|
| 660 |
+
"learning_rate": 0.00019629981023833903,
|
| 661 |
+
"loss": 0.0379,
|
| 662 |
+
"step": 8500
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 72.03,
|
| 666 |
+
"eval_loss": 0.22867247462272644,
|
| 667 |
+
"eval_runtime": 297.2969,
|
| 668 |
+
"eval_samples_per_second": 21.739,
|
| 669 |
+
"eval_steps_per_second": 0.303,
|
| 670 |
+
"eval_wer": 0.199077841063195,
|
| 671 |
+
"step": 8500
|
| 672 |
+
},
|
| 673 |
+
{
|
| 674 |
+
"epoch": 72.88,
|
| 675 |
+
"learning_rate": 0.00018537581579734625,
|
| 676 |
+
"loss": 0.037,
|
| 677 |
+
"step": 8600
|
| 678 |
+
},
|
| 679 |
+
{
|
| 680 |
+
"epoch": 73.73,
|
| 681 |
+
"learning_rate": 0.00017469498967238802,
|
| 682 |
+
"loss": 0.0362,
|
| 683 |
+
"step": 8700
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 74.57,
|
| 687 |
+
"learning_rate": 0.00016426558691388015,
|
| 688 |
+
"loss": 0.0348,
|
| 689 |
+
"step": 8800
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 75.42,
|
| 693 |
+
"learning_rate": 0.0001540956682508869,
|
| 694 |
+
"loss": 0.0367,
|
| 695 |
+
"step": 8900
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 76.27,
|
| 699 |
+
"learning_rate": 0.00014419309386110528,
|
| 700 |
+
"loss": 0.0381,
|
| 701 |
+
"step": 9000
|
| 702 |
+
},
|
| 703 |
+
{
|
| 704 |
+
"epoch": 76.27,
|
| 705 |
+
"eval_loss": 0.22330239415168762,
|
| 706 |
+
"eval_runtime": 290.8488,
|
| 707 |
+
"eval_samples_per_second": 22.221,
|
| 708 |
+
"eval_steps_per_second": 0.309,
|
| 709 |
+
"eval_wer": 0.19536416932674053,
|
| 710 |
+
"step": 9000
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 77.12,
|
| 714 |
+
"learning_rate": 0.00013456551729585032,
|
| 715 |
+
"loss": 0.0343,
|
| 716 |
+
"step": 9100
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 77.96,
|
| 720 |
+
"learning_rate": 0.00012522037956474042,
|
| 721 |
+
"loss": 0.033,
|
| 722 |
+
"step": 9200
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"epoch": 78.81,
|
| 726 |
+
"learning_rate": 0.00011616490338465268,
|
| 727 |
+
"loss": 0.032,
|
| 728 |
+
"step": 9300
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 79.66,
|
| 732 |
+
"learning_rate": 0.00010740608759739317,
|
| 733 |
+
"loss": 0.031,
|
| 734 |
+
"step": 9400
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 80.51,
|
| 738 |
+
"learning_rate": 9.895070176039717e-05,
|
| 739 |
+
"loss": 0.0308,
|
| 740 |
+
"step": 9500
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"epoch": 80.51,
|
| 744 |
+
"eval_loss": 0.21947245299816132,
|
| 745 |
+
"eval_runtime": 295.835,
|
| 746 |
+
"eval_samples_per_second": 21.847,
|
| 747 |
+
"eval_steps_per_second": 0.304,
|
| 748 |
+
"eval_wer": 0.18349293776470343,
|
| 749 |
+
"step": 9500
|
| 750 |
+
},
|
| 751 |
+
{
|
| 752 |
+
"epoch": 81.35,
|
| 753 |
+
"learning_rate": 9.080528091463946e-05,
|
| 754 |
+
"loss": 0.0309,
|
| 755 |
+
"step": 9600
|
| 756 |
+
},
|
| 757 |
+
{
|
| 758 |
+
"epoch": 82.2,
|
| 759 |
+
"learning_rate": 8.297612053379883e-05,
|
| 760 |
+
"loss": 0.0302,
|
| 761 |
+
"step": 9700
|
| 762 |
+
},
|
| 763 |
+
{
|
| 764 |
+
"epoch": 83.05,
|
| 765 |
+
"learning_rate": 7.546927165858153e-05,
|
| 766 |
+
"loss": 0.0296,
|
| 767 |
+
"step": 9800
|
| 768 |
+
},
|
| 769 |
+
{
|
| 770 |
+
"epoch": 83.89,
|
| 771 |
+
"learning_rate": 6.82905362199625e-05,
|
| 772 |
+
"loss": 0.0289,
|
| 773 |
+
"step": 9900
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 84.74,
|
| 777 |
+
"learning_rate": 6.14454625549593e-05,
|
| 778 |
+
"loss": 0.0291,
|
| 779 |
+
"step": 10000
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 84.74,
|
| 783 |
+
"eval_loss": 0.22659379243850708,
|
| 784 |
+
"eval_runtime": 301.2419,
|
| 785 |
+
"eval_samples_per_second": 21.455,
|
| 786 |
+
"eval_steps_per_second": 0.299,
|
| 787 |
+
"eval_wer": 0.18251236151968456,
|
| 788 |
+
"step": 10000
|
| 789 |
+
},
|
| 790 |
+
{
|
| 791 |
+
"epoch": 85.59,
|
| 792 |
+
"learning_rate": 5.493934111840604e-05,
|
| 793 |
+
"loss": 0.029,
|
| 794 |
+
"step": 10100
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 86.44,
|
| 798 |
+
"learning_rate": 4.87772003940401e-05,
|
| 799 |
+
"loss": 0.0276,
|
| 800 |
+
"step": 10200
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 87.29,
|
| 804 |
+
"learning_rate": 4.296380300806219e-05,
|
| 805 |
+
"loss": 0.0285,
|
| 806 |
+
"step": 10300
|
| 807 |
+
},
|
| 808 |
+
{
|
| 809 |
+
"epoch": 88.14,
|
| 810 |
+
"learning_rate": 3.750364204817463e-05,
|
| 811 |
+
"loss": 0.0287,
|
| 812 |
+
"step": 10400
|
| 813 |
+
},
|
| 814 |
+
{
|
| 815 |
+
"epoch": 88.98,
|
| 816 |
+
"learning_rate": 3.245018204574429e-05,
|
| 817 |
+
"loss": 0.0266,
|
| 818 |
+
"step": 10500
|
| 819 |
+
},
|
| 820 |
+
{
|
| 821 |
+
"epoch": 88.98,
|
| 822 |
+
"eval_loss": 0.22852544486522675,
|
| 823 |
+
"eval_runtime": 296.9937,
|
| 824 |
+
"eval_samples_per_second": 21.761,
|
| 825 |
+
"eval_steps_per_second": 0.303,
|
| 826 |
+
"eval_wer": 0.1800713525693184,
|
| 827 |
+
"step": 10500
|
| 828 |
+
},
|
| 829 |
+
{
|
| 830 |
+
"epoch": 89.83,
|
| 831 |
+
"learning_rate": 2.7705245284789292e-05,
|
| 832 |
+
"loss": 0.0269,
|
| 833 |
+
"step": 10600
|
| 834 |
+
},
|
| 835 |
+
{
|
| 836 |
+
"epoch": 90.68,
|
| 837 |
+
"learning_rate": 2.3325338060476542e-05,
|
| 838 |
+
"loss": 0.0262,
|
| 839 |
+
"step": 10700
|
| 840 |
+
},
|
| 841 |
+
{
|
| 842 |
+
"epoch": 91.52,
|
| 843 |
+
"learning_rate": 1.9313845537448972e-05,
|
| 844 |
+
"loss": 0.0257,
|
| 845 |
+
"step": 10800
|
| 846 |
+
},
|
| 847 |
+
{
|
| 848 |
+
"epoch": 92.37,
|
| 849 |
+
"learning_rate": 1.567386813814875e-05,
|
| 850 |
+
"loss": 0.0271,
|
| 851 |
+
"step": 10900
|
| 852 |
+
},
|
| 853 |
+
{
|
| 854 |
+
"epoch": 93.22,
|
| 855 |
+
"learning_rate": 1.2408219146547828e-05,
|
| 856 |
+
"loss": 0.0266,
|
| 857 |
+
"step": 11000
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 93.22,
|
| 861 |
+
"eval_loss": 0.229219451546669,
|
| 862 |
+
"eval_runtime": 299.2804,
|
| 863 |
+
"eval_samples_per_second": 21.595,
|
| 864 |
+
"eval_steps_per_second": 0.301,
|
| 865 |
+
"eval_wer": 0.1801130792180426,
|
| 866 |
+
"step": 11000
|
| 867 |
+
},
|
| 868 |
+
{
|
| 869 |
+
"epoch": 94.07,
|
| 870 |
+
"learning_rate": 9.519422533802668e-06,
|
| 871 |
+
"loss": 0.027,
|
| 872 |
+
"step": 11100
|
| 873 |
+
},
|
| 874 |
+
{
|
| 875 |
+
"epoch": 94.91,
|
| 876 |
+
"learning_rate": 7.009711007514796e-06,
|
| 877 |
+
"loss": 0.0266,
|
| 878 |
+
"step": 11200
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 95.76,
|
| 882 |
+
"learning_rate": 4.8810242861045405e-06,
|
| 883 |
+
"loss": 0.0258,
|
| 884 |
+
"step": 11300
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 96.61,
|
| 888 |
+
"learning_rate": 3.1350075996317563e-06,
|
| 889 |
+
"loss": 0.0256,
|
| 890 |
+
"step": 11400
|
| 891 |
+
},
|
| 892 |
+
{
|
| 893 |
+
"epoch": 97.46,
|
| 894 |
+
"learning_rate": 1.773010418222143e-06,
|
| 895 |
+
"loss": 0.0262,
|
| 896 |
+
"step": 11500
|
| 897 |
+
},
|
| 898 |
+
{
|
| 899 |
+
"epoch": 97.46,
|
| 900 |
+
"eval_loss": 0.22784815728664398,
|
| 901 |
+
"eval_runtime": 323.2034,
|
| 902 |
+
"eval_samples_per_second": 19.997,
|
| 903 |
+
"eval_steps_per_second": 0.278,
|
| 904 |
+
"eval_wer": 0.17881955310759218,
|
| 905 |
+
"step": 11500
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 98.3,
|
| 909 |
+
"learning_rate": 7.96085409082159e-07,
|
| 910 |
+
"loss": 0.0267,
|
| 911 |
+
"step": 11600
|
| 912 |
+
},
|
| 913 |
+
{
|
| 914 |
+
"epoch": 99.15,
|
| 915 |
+
"learning_rate": 2.049876229081038e-07,
|
| 916 |
+
"loss": 0.026,
|
| 917 |
+
"step": 11700
|
| 918 |
+
},
|
| 919 |
+
{
|
| 920 |
+
"epoch": 100.0,
|
| 921 |
+
"learning_rate": 1.7391031886804242e-10,
|
| 922 |
+
"loss": 0.0257,
|
| 923 |
+
"step": 11800
|
| 924 |
+
},
|
| 925 |
+
{
|
| 926 |
+
"epoch": 100.0,
|
| 927 |
+
"step": 11800,
|
| 928 |
+
"total_flos": 2.7805901871717807e+20,
|
| 929 |
+
"train_loss": 0.14547648023750823,
|
| 930 |
+
"train_runtime": 96097.4367,
|
| 931 |
+
"train_samples_per_second": 17.722,
|
| 932 |
+
"train_steps_per_second": 0.123
|
| 933 |
+
}
|
| 934 |
+
],
|
| 935 |
+
"max_steps": 11800,
|
| 936 |
+
"num_train_epochs": 100,
|
| 937 |
+
"total_flos": 2.7805901871717807e+20,
|
| 938 |
+
"trial_name": null,
|
| 939 |
+
"trial_params": null
|
| 940 |
+
}
|
wandb/run-20220205_233515-2f29fa6z/files/config.yaml
CHANGED
|
@@ -4802,6 +4802,26 @@ _wandb:
|
|
| 4802 |
5: 1
|
| 4803 |
6:
|
| 4804 |
- 1
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 4805 |
python_version: 3.9.6
|
| 4806 |
start_time: 1644122115
|
| 4807 |
t:
|
|
|
|
| 4802 |
5: 1
|
| 4803 |
6:
|
| 4804 |
- 1
|
| 4805 |
+
- 1: train/train_runtime
|
| 4806 |
+
5: 1
|
| 4807 |
+
6:
|
| 4808 |
+
- 1
|
| 4809 |
+
- 1: train/train_samples_per_second
|
| 4810 |
+
5: 1
|
| 4811 |
+
6:
|
| 4812 |
+
- 1
|
| 4813 |
+
- 1: train/train_steps_per_second
|
| 4814 |
+
5: 1
|
| 4815 |
+
6:
|
| 4816 |
+
- 1
|
| 4817 |
+
- 1: train/total_flos
|
| 4818 |
+
5: 1
|
| 4819 |
+
6:
|
| 4820 |
+
- 1
|
| 4821 |
+
- 1: train/train_loss
|
| 4822 |
+
5: 1
|
| 4823 |
+
6:
|
| 4824 |
+
- 1
|
| 4825 |
python_version: 3.9.6
|
| 4826 |
start_time: 1644122115
|
| 4827 |
t:
|
wandb/run-20220205_233515-2f29fa6z/files/output.log
CHANGED
|
The diff for this file is too large to render.
See raw diff
|
|
|
wandb/run-20220205_233515-2f29fa6z/files/wandb-summary.json
CHANGED
|
The diff for this file is too large to render.
See raw diff
|
|
|
wandb/run-20220205_233515-2f29fa6z/logs/debug-internal.log
CHANGED
|
The diff for this file is too large to render.
See raw diff
|
|
|
wandb/run-20220205_233515-2f29fa6z/run-2f29fa6z.wandb
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:742e2748dedaf1a3dcb8997028a19e4bee59a2093eded406359adc3c8e7fc8ec
|
| 3 |
+
size 85282495
|