End of training
Browse files
all_results.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 42.86,
|
| 3 |
+
"eval_loss": 1.1061279773712158,
|
| 4 |
+
"eval_runtime": 1526.712,
|
| 5 |
+
"eval_samples_per_second": 0.335,
|
| 6 |
+
"eval_steps_per_second": 0.021,
|
| 7 |
+
"eval_wer": 67.83942963422194,
|
| 8 |
+
"train_loss": 0.31906700134277344,
|
| 9 |
+
"train_runtime": 2410.2404,
|
| 10 |
+
"train_samples_per_second": 7.966,
|
| 11 |
+
"train_steps_per_second": 0.124
|
| 12 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 42.86,
|
| 3 |
+
"eval_loss": 1.1061279773712158,
|
| 4 |
+
"eval_runtime": 1526.712,
|
| 5 |
+
"eval_samples_per_second": 0.335,
|
| 6 |
+
"eval_steps_per_second": 0.021,
|
| 7 |
+
"eval_wer": 67.83942963422194
|
| 8 |
+
}
|
runs/Dec14_09-09-02_129-213-22-31/events.out.tfevents.1671013111.129-213-22-31.76517.2
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5ae2972839acc680b0ef08ca60fd111528eb0f6d4b69b1e09084238fa315acc5
|
| 3 |
+
size 358
|
train_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 42.86,
|
| 3 |
+
"train_loss": 0.31906700134277344,
|
| 4 |
+
"train_runtime": 2410.2404,
|
| 5 |
+
"train_samples_per_second": 7.966,
|
| 6 |
+
"train_steps_per_second": 0.124
|
| 7 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,124 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 67.83942963422194,
|
| 3 |
+
"best_model_checkpoint": "./checkpoint-300",
|
| 4 |
+
"epoch": 42.857142857142854,
|
| 5 |
+
"global_step": 300,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 3.57,
|
| 12 |
+
"learning_rate": 2.875862068965517e-07,
|
| 13 |
+
"loss": 2.755,
|
| 14 |
+
"step": 25
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 7.14,
|
| 18 |
+
"learning_rate": 2.6172413793103444e-07,
|
| 19 |
+
"loss": 2.0968,
|
| 20 |
+
"step": 50
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 10.71,
|
| 24 |
+
"learning_rate": 2.3586206896551725e-07,
|
| 25 |
+
"loss": 1.703,
|
| 26 |
+
"step": 75
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 14.29,
|
| 30 |
+
"learning_rate": 2.0999999999999997e-07,
|
| 31 |
+
"loss": 1.4728,
|
| 32 |
+
"step": 100
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 14.29,
|
| 36 |
+
"eval_loss": 1.4428775310516357,
|
| 37 |
+
"eval_runtime": 1406.9013,
|
| 38 |
+
"eval_samples_per_second": 0.364,
|
| 39 |
+
"eval_steps_per_second": 0.023,
|
| 40 |
+
"eval_wer": 83.59423434593924,
|
| 41 |
+
"step": 100
|
| 42 |
+
},
|
| 43 |
+
{
|
| 44 |
+
"epoch": 17.86,
|
| 45 |
+
"learning_rate": 1.8413793103448276e-07,
|
| 46 |
+
"loss": 1.2819,
|
| 47 |
+
"step": 125
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 21.43,
|
| 51 |
+
"learning_rate": 1.5827586206896554e-07,
|
| 52 |
+
"loss": 1.1744,
|
| 53 |
+
"step": 150
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 25.0,
|
| 57 |
+
"learning_rate": 1.3241379310344827e-07,
|
| 58 |
+
"loss": 1.0885,
|
| 59 |
+
"step": 175
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 28.57,
|
| 63 |
+
"learning_rate": 1.0655172413793102e-07,
|
| 64 |
+
"loss": 1.0344,
|
| 65 |
+
"step": 200
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 28.57,
|
| 69 |
+
"eval_loss": 1.158418893814087,
|
| 70 |
+
"eval_runtime": 1566.2048,
|
| 71 |
+
"eval_samples_per_second": 0.327,
|
| 72 |
+
"eval_steps_per_second": 0.02,
|
| 73 |
+
"eval_wer": 69.97055176689399,
|
| 74 |
+
"step": 200
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 32.14,
|
| 78 |
+
"learning_rate": 8.068965517241379e-08,
|
| 79 |
+
"loss": 0.9889,
|
| 80 |
+
"step": 225
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 35.71,
|
| 84 |
+
"learning_rate": 5.482758620689655e-08,
|
| 85 |
+
"loss": 0.9647,
|
| 86 |
+
"step": 250
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 39.29,
|
| 90 |
+
"learning_rate": 2.8965517241379308e-08,
|
| 91 |
+
"loss": 0.9435,
|
| 92 |
+
"step": 275
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 42.86,
|
| 96 |
+
"learning_rate": 3.103448275862069e-09,
|
| 97 |
+
"loss": 0.9318,
|
| 98 |
+
"step": 300
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 42.86,
|
| 102 |
+
"eval_loss": 1.1061279773712158,
|
| 103 |
+
"eval_runtime": 1528.8642,
|
| 104 |
+
"eval_samples_per_second": 0.335,
|
| 105 |
+
"eval_steps_per_second": 0.021,
|
| 106 |
+
"eval_wer": 67.83942963422194,
|
| 107 |
+
"step": 300
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 42.86,
|
| 111 |
+
"step": 300,
|
| 112 |
+
"total_flos": 1.886690232041472e+19,
|
| 113 |
+
"train_loss": 0.31906700134277344,
|
| 114 |
+
"train_runtime": 2410.2404,
|
| 115 |
+
"train_samples_per_second": 7.966,
|
| 116 |
+
"train_steps_per_second": 0.124
|
| 117 |
+
}
|
| 118 |
+
],
|
| 119 |
+
"max_steps": 300,
|
| 120 |
+
"num_train_epochs": 43,
|
| 121 |
+
"total_flos": 1.886690232041472e+19,
|
| 122 |
+
"trial_name": null,
|
| 123 |
+
"trial_params": null
|
| 124 |
+
}
|