End of training
Browse files
all_results.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 50.0,
|
| 3 |
+
"eval_loss": 0.7895313501358032,
|
| 4 |
+
"eval_runtime": 101.7007,
|
| 5 |
+
"eval_samples": 842,
|
| 6 |
+
"eval_samples_per_second": 8.279,
|
| 7 |
+
"eval_steps_per_second": 1.042,
|
| 8 |
+
"eval_wer": 0.43980671943617605,
|
| 9 |
+
"train_loss": 0.4734274124600994,
|
| 10 |
+
"train_runtime": 45905.6833,
|
| 11 |
+
"train_samples": 3350,
|
| 12 |
+
"train_samples_per_second": 3.649,
|
| 13 |
+
"train_steps_per_second": 0.365
|
| 14 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 50.0,
|
| 3 |
+
"eval_loss": 0.7895313501358032,
|
| 4 |
+
"eval_runtime": 101.7007,
|
| 5 |
+
"eval_samples": 842,
|
| 6 |
+
"eval_samples_per_second": 8.279,
|
| 7 |
+
"eval_steps_per_second": 1.042,
|
| 8 |
+
"eval_wer": 0.43980671943617605
|
| 9 |
+
}
|
runs/Jun30_10-04-05_64df48378bb2/events.out.tfevents.1688166626.64df48378bb2.1419.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ff27a790950cd3b58236c28ec90d046792b81fbb25add0c6019ca4bd97d99bfd
|
| 3 |
+
size 412
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 50.0,
|
| 3 |
+
"train_loss": 0.4734274124600994,
|
| 4 |
+
"train_runtime": 45905.6833,
|
| 5 |
+
"train_samples": 3350,
|
| 6 |
+
"train_samples_per_second": 3.649,
|
| 7 |
+
"train_steps_per_second": 0.365
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,673 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.7895313501358032,
|
| 3 |
+
"best_model_checkpoint": "./wav2vec2-xls-r-300m-paper/checkpoint-3015",
|
| 4 |
+
"epoch": 50.0,
|
| 5 |
+
"global_step": 16750,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 1.0,
|
| 12 |
+
"eval_loss": 3.7156972885131836,
|
| 13 |
+
"eval_runtime": 90.6265,
|
| 14 |
+
"eval_samples_per_second": 9.291,
|
| 15 |
+
"eval_steps_per_second": 1.17,
|
| 16 |
+
"eval_wer": 1.0,
|
| 17 |
+
"step": 335
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 1.49,
|
| 21 |
+
"learning_rate": 4.9770361298224126e-05,
|
| 22 |
+
"loss": 6.2976,
|
| 23 |
+
"step": 500
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 2.0,
|
| 27 |
+
"eval_loss": 3.364386558532715,
|
| 28 |
+
"eval_runtime": 105.7497,
|
| 29 |
+
"eval_samples_per_second": 7.962,
|
| 30 |
+
"eval_steps_per_second": 1.002,
|
| 31 |
+
"eval_wer": 1.0,
|
| 32 |
+
"step": 670
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 2.99,
|
| 36 |
+
"learning_rate": 4.8239436619718316e-05,
|
| 37 |
+
"loss": 3.2342,
|
| 38 |
+
"step": 1000
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 3.0,
|
| 42 |
+
"eval_loss": 2.459735155105591,
|
| 43 |
+
"eval_runtime": 100.2813,
|
| 44 |
+
"eval_samples_per_second": 8.396,
|
| 45 |
+
"eval_steps_per_second": 1.057,
|
| 46 |
+
"eval_wer": 0.9739311504731903,
|
| 47 |
+
"step": 1005
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 4.0,
|
| 51 |
+
"eval_loss": 1.4160194396972656,
|
| 52 |
+
"eval_runtime": 105.5456,
|
| 53 |
+
"eval_samples_per_second": 7.978,
|
| 54 |
+
"eval_steps_per_second": 1.004,
|
| 55 |
+
"eval_wer": 0.7444104809322317,
|
| 56 |
+
"step": 1340
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 4.48,
|
| 60 |
+
"learning_rate": 4.670851194121249e-05,
|
| 61 |
+
"loss": 1.2813,
|
| 62 |
+
"step": 1500
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 5.0,
|
| 66 |
+
"eval_loss": 1.1337928771972656,
|
| 67 |
+
"eval_runtime": 101.0094,
|
| 68 |
+
"eval_samples_per_second": 8.336,
|
| 69 |
+
"eval_steps_per_second": 1.049,
|
| 70 |
+
"eval_wer": 0.6543241186913518,
|
| 71 |
+
"step": 1675
|
| 72 |
+
},
|
| 73 |
+
{
|
| 74 |
+
"epoch": 5.97,
|
| 75 |
+
"learning_rate": 4.5177587262706676e-05,
|
| 76 |
+
"loss": 0.7279,
|
| 77 |
+
"step": 2000
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 6.0,
|
| 81 |
+
"eval_loss": 1.002017617225647,
|
| 82 |
+
"eval_runtime": 97.2569,
|
| 83 |
+
"eval_samples_per_second": 8.657,
|
| 84 |
+
"eval_steps_per_second": 1.09,
|
| 85 |
+
"eval_wer": 0.5855680286183377,
|
| 86 |
+
"step": 2010
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 7.0,
|
| 90 |
+
"eval_loss": 0.8434591293334961,
|
| 91 |
+
"eval_runtime": 101.7349,
|
| 92 |
+
"eval_samples_per_second": 8.276,
|
| 93 |
+
"eval_steps_per_second": 1.042,
|
| 94 |
+
"eval_wer": 0.48233378271954297,
|
| 95 |
+
"step": 2345
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 7.46,
|
| 99 |
+
"learning_rate": 4.364666258420086e-05,
|
| 100 |
+
"loss": 0.5226,
|
| 101 |
+
"step": 2500
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 8.0,
|
| 105 |
+
"eval_loss": 0.875725507736206,
|
| 106 |
+
"eval_runtime": 100.6138,
|
| 107 |
+
"eval_samples_per_second": 8.369,
|
| 108 |
+
"eval_steps_per_second": 1.054,
|
| 109 |
+
"eval_wer": 0.5078286637211847,
|
| 110 |
+
"step": 2680
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 8.96,
|
| 114 |
+
"learning_rate": 4.211573790569504e-05,
|
| 115 |
+
"loss": 0.4218,
|
| 116 |
+
"step": 3000
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 9.0,
|
| 120 |
+
"eval_loss": 0.7895313501358032,
|
| 121 |
+
"eval_runtime": 101.3795,
|
| 122 |
+
"eval_samples_per_second": 8.305,
|
| 123 |
+
"eval_steps_per_second": 1.046,
|
| 124 |
+
"eval_wer": 0.43980671943617605,
|
| 125 |
+
"step": 3015
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 10.0,
|
| 129 |
+
"eval_loss": 0.7991793155670166,
|
| 130 |
+
"eval_runtime": 87.9454,
|
| 131 |
+
"eval_samples_per_second": 9.574,
|
| 132 |
+
"eval_steps_per_second": 1.205,
|
| 133 |
+
"eval_wer": 0.42284127768063323,
|
| 134 |
+
"step": 3350
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 10.45,
|
| 138 |
+
"learning_rate": 4.058481322718922e-05,
|
| 139 |
+
"loss": 0.3421,
|
| 140 |
+
"step": 3500
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 11.0,
|
| 144 |
+
"eval_loss": 0.8118342757225037,
|
| 145 |
+
"eval_runtime": 105.2144,
|
| 146 |
+
"eval_samples_per_second": 8.003,
|
| 147 |
+
"eval_steps_per_second": 1.007,
|
| 148 |
+
"eval_wer": 0.43070331166490916,
|
| 149 |
+
"step": 3685
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 11.94,
|
| 153 |
+
"learning_rate": 3.905388854868341e-05,
|
| 154 |
+
"loss": 0.287,
|
| 155 |
+
"step": 4000
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 12.0,
|
| 159 |
+
"eval_loss": 0.8214530348777771,
|
| 160 |
+
"eval_runtime": 103.7557,
|
| 161 |
+
"eval_samples_per_second": 8.115,
|
| 162 |
+
"eval_steps_per_second": 1.022,
|
| 163 |
+
"eval_wer": 0.4248434934661025,
|
| 164 |
+
"step": 4020
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 13.0,
|
| 168 |
+
"eval_loss": 0.8603241443634033,
|
| 169 |
+
"eval_runtime": 98.5504,
|
| 170 |
+
"eval_samples_per_second": 8.544,
|
| 171 |
+
"eval_steps_per_second": 1.076,
|
| 172 |
+
"eval_wer": 0.40767783013201275,
|
| 173 |
+
"step": 4355
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 13.43,
|
| 177 |
+
"learning_rate": 3.752296387017759e-05,
|
| 178 |
+
"loss": 0.2415,
|
| 179 |
+
"step": 4500
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 14.0,
|
| 183 |
+
"eval_loss": 0.8329221606254578,
|
| 184 |
+
"eval_runtime": 106.1564,
|
| 185 |
+
"eval_samples_per_second": 7.932,
|
| 186 |
+
"eval_steps_per_second": 0.999,
|
| 187 |
+
"eval_wer": 0.3885766915386361,
|
| 188 |
+
"step": 4690
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 14.93,
|
| 192 |
+
"learning_rate": 3.5992039191671776e-05,
|
| 193 |
+
"loss": 0.2132,
|
| 194 |
+
"step": 5000
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 15.0,
|
| 198 |
+
"eval_loss": 0.8728479743003845,
|
| 199 |
+
"eval_runtime": 106.4415,
|
| 200 |
+
"eval_samples_per_second": 7.91,
|
| 201 |
+
"eval_steps_per_second": 0.996,
|
| 202 |
+
"eval_wer": 0.3955310543668326,
|
| 203 |
+
"step": 5025
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 16.0,
|
| 207 |
+
"eval_loss": 0.8741195797920227,
|
| 208 |
+
"eval_runtime": 105.2581,
|
| 209 |
+
"eval_samples_per_second": 7.999,
|
| 210 |
+
"eval_steps_per_second": 1.007,
|
| 211 |
+
"eval_wer": 0.39178023679538687,
|
| 212 |
+
"step": 5360
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 16.42,
|
| 216 |
+
"learning_rate": 3.446111451316595e-05,
|
| 217 |
+
"loss": 0.1857,
|
| 218 |
+
"step": 5500
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 17.0,
|
| 222 |
+
"eval_loss": 0.8632614612579346,
|
| 223 |
+
"eval_runtime": 107.8251,
|
| 224 |
+
"eval_samples_per_second": 7.809,
|
| 225 |
+
"eval_steps_per_second": 0.983,
|
| 226 |
+
"eval_wer": 0.3674733371597902,
|
| 227 |
+
"step": 5695
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 17.91,
|
| 231 |
+
"learning_rate": 3.2930189834660136e-05,
|
| 232 |
+
"loss": 0.1673,
|
| 233 |
+
"step": 6000
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 18.0,
|
| 237 |
+
"eval_loss": 0.8884367346763611,
|
| 238 |
+
"eval_runtime": 96.3592,
|
| 239 |
+
"eval_samples_per_second": 8.738,
|
| 240 |
+
"eval_steps_per_second": 1.1,
|
| 241 |
+
"eval_wer": 0.3803943030286851,
|
| 242 |
+
"step": 6030
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 19.0,
|
| 246 |
+
"eval_loss": 0.9140524864196777,
|
| 247 |
+
"eval_runtime": 99.6527,
|
| 248 |
+
"eval_samples_per_second": 8.449,
|
| 249 |
+
"eval_steps_per_second": 1.064,
|
| 250 |
+
"eval_wer": 0.3679271727378299,
|
| 251 |
+
"step": 6365
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 19.4,
|
| 255 |
+
"learning_rate": 3.139926515615432e-05,
|
| 256 |
+
"loss": 0.1479,
|
| 257 |
+
"step": 6500
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 20.0,
|
| 261 |
+
"eval_loss": 0.9567932486534119,
|
| 262 |
+
"eval_runtime": 99.4352,
|
| 263 |
+
"eval_samples_per_second": 8.468,
|
| 264 |
+
"eval_steps_per_second": 1.066,
|
| 265 |
+
"eval_wer": 0.36047893001588427,
|
| 266 |
+
"step": 6700
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 20.9,
|
| 270 |
+
"learning_rate": 2.98683404776485e-05,
|
| 271 |
+
"loss": 0.1386,
|
| 272 |
+
"step": 7000
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 21.0,
|
| 276 |
+
"eval_loss": 0.9340795874595642,
|
| 277 |
+
"eval_runtime": 103.1923,
|
| 278 |
+
"eval_samples_per_second": 8.16,
|
| 279 |
+
"eval_steps_per_second": 1.027,
|
| 280 |
+
"eval_wer": 0.36301507001081196,
|
| 281 |
+
"step": 7035
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 22.0,
|
| 285 |
+
"eval_loss": 0.9644697308540344,
|
| 286 |
+
"eval_runtime": 98.7834,
|
| 287 |
+
"eval_samples_per_second": 8.524,
|
| 288 |
+
"eval_steps_per_second": 1.073,
|
| 289 |
+
"eval_wer": 0.3536847444505252,
|
| 290 |
+
"step": 7370
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 22.39,
|
| 294 |
+
"learning_rate": 2.8337415799142685e-05,
|
| 295 |
+
"loss": 0.1233,
|
| 296 |
+
"step": 7500
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 23.0,
|
| 300 |
+
"eval_loss": 0.9728893041610718,
|
| 301 |
+
"eval_runtime": 102.9289,
|
| 302 |
+
"eval_samples_per_second": 8.18,
|
| 303 |
+
"eval_steps_per_second": 1.03,
|
| 304 |
+
"eval_wer": 0.35667472002349265,
|
| 305 |
+
"step": 7705
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 23.88,
|
| 309 |
+
"learning_rate": 2.680649112063687e-05,
|
| 310 |
+
"loss": 0.1177,
|
| 311 |
+
"step": 8000
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 24.0,
|
| 315 |
+
"eval_loss": 1.001348614692688,
|
| 316 |
+
"eval_runtime": 106.3343,
|
| 317 |
+
"eval_samples_per_second": 7.918,
|
| 318 |
+
"eval_steps_per_second": 0.997,
|
| 319 |
+
"eval_wer": 0.34538222299344606,
|
| 320 |
+
"step": 8040
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 25.0,
|
| 324 |
+
"eval_loss": 1.0323426723480225,
|
| 325 |
+
"eval_runtime": 102.2816,
|
| 326 |
+
"eval_samples_per_second": 8.232,
|
| 327 |
+
"eval_steps_per_second": 1.036,
|
| 328 |
+
"eval_wer": 0.35967804370169654,
|
| 329 |
+
"step": 8375
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 25.37,
|
| 333 |
+
"learning_rate": 2.527556644213105e-05,
|
| 334 |
+
"loss": 0.1061,
|
| 335 |
+
"step": 8500
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"epoch": 26.0,
|
| 339 |
+
"eval_loss": 1.0269230604171753,
|
| 340 |
+
"eval_runtime": 96.8319,
|
| 341 |
+
"eval_samples_per_second": 8.695,
|
| 342 |
+
"eval_steps_per_second": 1.095,
|
| 343 |
+
"eval_wer": 0.3456491850981753,
|
| 344 |
+
"step": 8710
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 26.87,
|
| 348 |
+
"learning_rate": 2.3744641763625232e-05,
|
| 349 |
+
"loss": 0.1028,
|
| 350 |
+
"step": 9000
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 27.0,
|
| 354 |
+
"eval_loss": 1.0042426586151123,
|
| 355 |
+
"eval_runtime": 94.8369,
|
| 356 |
+
"eval_samples_per_second": 8.878,
|
| 357 |
+
"eval_steps_per_second": 1.118,
|
| 358 |
+
"eval_wer": 0.3424189436309516,
|
| 359 |
+
"step": 9045
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 28.0,
|
| 363 |
+
"eval_loss": 1.0424461364746094,
|
| 364 |
+
"eval_runtime": 93.6728,
|
| 365 |
+
"eval_samples_per_second": 8.989,
|
| 366 |
+
"eval_steps_per_second": 1.132,
|
| 367 |
+
"eval_wer": 0.3394423161632206,
|
| 368 |
+
"step": 9380
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 28.36,
|
| 372 |
+
"learning_rate": 2.2213717085119412e-05,
|
| 373 |
+
"loss": 0.0961,
|
| 374 |
+
"step": 9500
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 29.0,
|
| 378 |
+
"eval_loss": 1.0599919557571411,
|
| 379 |
+
"eval_runtime": 104.5041,
|
| 380 |
+
"eval_samples_per_second": 8.057,
|
| 381 |
+
"eval_steps_per_second": 1.014,
|
| 382 |
+
"eval_wer": 0.3412309622649065,
|
| 383 |
+
"step": 9715
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 29.85,
|
| 387 |
+
"learning_rate": 2.0682792406613595e-05,
|
| 388 |
+
"loss": 0.0949,
|
| 389 |
+
"step": 10000
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 30.0,
|
| 393 |
+
"eval_loss": 1.051209568977356,
|
| 394 |
+
"eval_runtime": 104.5071,
|
| 395 |
+
"eval_samples_per_second": 8.057,
|
| 396 |
+
"eval_steps_per_second": 1.014,
|
| 397 |
+
"eval_wer": 0.33890839195376216,
|
| 398 |
+
"step": 10050
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 31.0,
|
| 402 |
+
"eval_loss": 1.0956796407699585,
|
| 403 |
+
"eval_runtime": 104.4359,
|
| 404 |
+
"eval_samples_per_second": 8.062,
|
| 405 |
+
"eval_steps_per_second": 1.015,
|
| 406 |
+
"eval_wer": 0.3389217400589986,
|
| 407 |
+
"step": 10385
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"epoch": 31.34,
|
| 411 |
+
"learning_rate": 1.9154929577464788e-05,
|
| 412 |
+
"loss": 0.0878,
|
| 413 |
+
"step": 10500
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 32.0,
|
| 417 |
+
"eval_loss": 1.09244704246521,
|
| 418 |
+
"eval_runtime": 106.7914,
|
| 419 |
+
"eval_samples_per_second": 7.885,
|
| 420 |
+
"eval_steps_per_second": 0.993,
|
| 421 |
+
"eval_wer": 0.33107305417995914,
|
| 422 |
+
"step": 10720
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 32.84,
|
| 426 |
+
"learning_rate": 1.7627066748315983e-05,
|
| 427 |
+
"loss": 0.0852,
|
| 428 |
+
"step": 11000
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 33.0,
|
| 432 |
+
"eval_loss": 1.0858749151229858,
|
| 433 |
+
"eval_runtime": 104.2304,
|
| 434 |
+
"eval_samples_per_second": 8.078,
|
| 435 |
+
"eval_steps_per_second": 1.017,
|
| 436 |
+
"eval_wer": 0.3365991697478543,
|
| 437 |
+
"step": 11055
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 34.0,
|
| 441 |
+
"eval_loss": 1.1498078107833862,
|
| 442 |
+
"eval_runtime": 96.0047,
|
| 443 |
+
"eval_samples_per_second": 8.77,
|
| 444 |
+
"eval_steps_per_second": 1.104,
|
| 445 |
+
"eval_wer": 0.3450485203625345,
|
| 446 |
+
"step": 11390
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 34.33,
|
| 450 |
+
"learning_rate": 1.6096142069810167e-05,
|
| 451 |
+
"loss": 0.0837,
|
| 452 |
+
"step": 11500
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 35.0,
|
| 456 |
+
"eval_loss": 1.0844124555587769,
|
| 457 |
+
"eval_runtime": 95.1094,
|
| 458 |
+
"eval_samples_per_second": 8.853,
|
| 459 |
+
"eval_steps_per_second": 1.115,
|
| 460 |
+
"eval_wer": 0.332861700281645,
|
| 461 |
+
"step": 11725
|
| 462 |
+
},
|
| 463 |
+
{
|
| 464 |
+
"epoch": 35.82,
|
| 465 |
+
"learning_rate": 1.4565217391304348e-05,
|
| 466 |
+
"loss": 0.0814,
|
| 467 |
+
"step": 12000
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"epoch": 36.0,
|
| 471 |
+
"eval_loss": 1.1050550937652588,
|
| 472 |
+
"eval_runtime": 105.0526,
|
| 473 |
+
"eval_samples_per_second": 8.015,
|
| 474 |
+
"eval_steps_per_second": 1.009,
|
| 475 |
+
"eval_wer": 0.3321008582831667,
|
| 476 |
+
"step": 12060
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 37.0,
|
| 480 |
+
"eval_loss": 1.0878149271011353,
|
| 481 |
+
"eval_runtime": 97.9103,
|
| 482 |
+
"eval_samples_per_second": 8.6,
|
| 483 |
+
"eval_steps_per_second": 1.083,
|
| 484 |
+
"eval_wer": 0.3310597060747227,
|
| 485 |
+
"step": 12395
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 37.31,
|
| 489 |
+
"learning_rate": 1.3034292712798532e-05,
|
| 490 |
+
"loss": 0.0793,
|
| 491 |
+
"step": 12500
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 38.0,
|
| 495 |
+
"eval_loss": 1.1377496719360352,
|
| 496 |
+
"eval_runtime": 101.9563,
|
| 497 |
+
"eval_samples_per_second": 8.258,
|
| 498 |
+
"eval_steps_per_second": 1.04,
|
| 499 |
+
"eval_wer": 0.32856361039550436,
|
| 500 |
+
"step": 12730
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 38.81,
|
| 504 |
+
"learning_rate": 1.1506429883649724e-05,
|
| 505 |
+
"loss": 0.0759,
|
| 506 |
+
"step": 13000
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 39.0,
|
| 510 |
+
"eval_loss": 1.1136152744293213,
|
| 511 |
+
"eval_runtime": 98.218,
|
| 512 |
+
"eval_samples_per_second": 8.573,
|
| 513 |
+
"eval_steps_per_second": 1.079,
|
| 514 |
+
"eval_wer": 0.3246125712455117,
|
| 515 |
+
"step": 13065
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 40.0,
|
| 519 |
+
"eval_loss": 1.1215593814849854,
|
| 520 |
+
"eval_runtime": 107.5832,
|
| 521 |
+
"eval_samples_per_second": 7.826,
|
| 522 |
+
"eval_steps_per_second": 0.985,
|
| 523 |
+
"eval_wer": 0.32682835671476435,
|
| 524 |
+
"step": 13400
|
| 525 |
+
},
|
| 526 |
+
{
|
| 527 |
+
"epoch": 40.3,
|
| 528 |
+
"learning_rate": 9.975505205143907e-06,
|
| 529 |
+
"loss": 0.0726,
|
| 530 |
+
"step": 13500
|
| 531 |
+
},
|
| 532 |
+
{
|
| 533 |
+
"epoch": 41.0,
|
| 534 |
+
"eval_loss": 1.1300030946731567,
|
| 535 |
+
"eval_runtime": 106.9995,
|
| 536 |
+
"eval_samples_per_second": 7.869,
|
| 537 |
+
"eval_steps_per_second": 0.991,
|
| 538 |
+
"eval_wer": 0.3252666284020983,
|
| 539 |
+
"step": 13735
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"epoch": 41.79,
|
| 543 |
+
"learning_rate": 8.444580526638089e-06,
|
| 544 |
+
"loss": 0.0715,
|
| 545 |
+
"step": 14000
|
| 546 |
+
},
|
| 547 |
+
{
|
| 548 |
+
"epoch": 42.0,
|
| 549 |
+
"eval_loss": 1.1506843566894531,
|
| 550 |
+
"eval_runtime": 93.8834,
|
| 551 |
+
"eval_samples_per_second": 8.969,
|
| 552 |
+
"eval_steps_per_second": 1.129,
|
| 553 |
+
"eval_wer": 0.32622769197912355,
|
| 554 |
+
"step": 14070
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 43.0,
|
| 558 |
+
"eval_loss": 1.1561784744262695,
|
| 559 |
+
"eval_runtime": 104.7418,
|
| 560 |
+
"eval_samples_per_second": 8.039,
|
| 561 |
+
"eval_steps_per_second": 1.012,
|
| 562 |
+
"eval_wer": 0.3275491543975333,
|
| 563 |
+
"step": 14405
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 43.28,
|
| 567 |
+
"learning_rate": 6.913655848132272e-06,
|
| 568 |
+
"loss": 0.0711,
|
| 569 |
+
"step": 14500
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 44.0,
|
| 573 |
+
"eval_loss": 1.1485936641693115,
|
| 574 |
+
"eval_runtime": 104.0283,
|
| 575 |
+
"eval_samples_per_second": 8.094,
|
| 576 |
+
"eval_steps_per_second": 1.019,
|
| 577 |
+
"eval_wer": 0.3218895577772735,
|
| 578 |
+
"step": 14740
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 44.78,
|
| 582 |
+
"learning_rate": 5.382731169626455e-06,
|
| 583 |
+
"loss": 0.0699,
|
| 584 |
+
"step": 15000
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 45.0,
|
| 588 |
+
"eval_loss": 1.1580160856246948,
|
| 589 |
+
"eval_runtime": 98.4301,
|
| 590 |
+
"eval_samples_per_second": 8.554,
|
| 591 |
+
"eval_steps_per_second": 1.077,
|
| 592 |
+
"eval_wer": 0.31940681020329165,
|
| 593 |
+
"step": 15075
|
| 594 |
+
},
|
| 595 |
+
{
|
| 596 |
+
"epoch": 46.0,
|
| 597 |
+
"eval_loss": 1.1580368280410767,
|
| 598 |
+
"eval_runtime": 96.4383,
|
| 599 |
+
"eval_samples_per_second": 8.731,
|
| 600 |
+
"eval_steps_per_second": 1.099,
|
| 601 |
+
"eval_wer": 0.3195269431504198,
|
| 602 |
+
"step": 15410
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 46.27,
|
| 606 |
+
"learning_rate": 3.851806491120637e-06,
|
| 607 |
+
"loss": 0.0667,
|
| 608 |
+
"step": 15500
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 47.0,
|
| 612 |
+
"eval_loss": 1.1504408121109009,
|
| 613 |
+
"eval_runtime": 99.6568,
|
| 614 |
+
"eval_samples_per_second": 8.449,
|
| 615 |
+
"eval_steps_per_second": 1.064,
|
| 616 |
+
"eval_wer": 0.321208804410214,
|
| 617 |
+
"step": 15745
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"epoch": 47.76,
|
| 621 |
+
"learning_rate": 2.3208818126148195e-06,
|
| 622 |
+
"loss": 0.0667,
|
| 623 |
+
"step": 16000
|
| 624 |
+
},
|
| 625 |
+
{
|
| 626 |
+
"epoch": 48.0,
|
| 627 |
+
"eval_loss": 1.1580157279968262,
|
| 628 |
+
"eval_runtime": 110.9917,
|
| 629 |
+
"eval_samples_per_second": 7.586,
|
| 630 |
+
"eval_steps_per_second": 0.955,
|
| 631 |
+
"eval_wer": 0.3202610889384252,
|
| 632 |
+
"step": 16080
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 49.0,
|
| 636 |
+
"eval_loss": 1.1697617769241333,
|
| 637 |
+
"eval_runtime": 107.1691,
|
| 638 |
+
"eval_samples_per_second": 7.857,
|
| 639 |
+
"eval_steps_per_second": 0.989,
|
| 640 |
+
"eval_wer": 0.3192065886247447,
|
| 641 |
+
"step": 16415
|
| 642 |
+
},
|
| 643 |
+
{
|
| 644 |
+
"epoch": 49.25,
|
| 645 |
+
"learning_rate": 7.899571341090019e-07,
|
| 646 |
+
"loss": 0.0664,
|
| 647 |
+
"step": 16500
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 50.0,
|
| 651 |
+
"eval_loss": 1.17439603805542,
|
| 652 |
+
"eval_runtime": 107.0438,
|
| 653 |
+
"eval_samples_per_second": 7.866,
|
| 654 |
+
"eval_steps_per_second": 0.99,
|
| 655 |
+
"eval_wer": 0.3192332848352176,
|
| 656 |
+
"step": 16750
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 50.0,
|
| 660 |
+
"step": 16750,
|
| 661 |
+
"total_flos": 1.4087701795050537e+20,
|
| 662 |
+
"train_loss": 0.4734274124600994,
|
| 663 |
+
"train_runtime": 45905.6833,
|
| 664 |
+
"train_samples_per_second": 3.649,
|
| 665 |
+
"train_steps_per_second": 0.365
|
| 666 |
+
}
|
| 667 |
+
],
|
| 668 |
+
"max_steps": 16750,
|
| 669 |
+
"num_train_epochs": 50,
|
| 670 |
+
"total_flos": 1.4087701795050537e+20,
|
| 671 |
+
"trial_name": null,
|
| 672 |
+
"trial_params": null
|
| 673 |
+
}
|