Fin de l'entraînement
Browse files- README.md +3 -3
- all_results.json +10 -10
- eval_results.json +6 -6
- train_results.json +5 -5
- trainer_state.json +511 -37
README.md
CHANGED
|
@@ -14,9 +14,9 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 14 |
|
| 15 |
This model is a fine-tuned version of [](https://huggingface.co/) on an unknown dataset.
|
| 16 |
It achieves the following results on the evaluation set:
|
| 17 |
-
- Loss: 0.
|
| 18 |
-
- Mse: 0.
|
| 19 |
-
- Mae: 0.
|
| 20 |
|
| 21 |
## Model description
|
| 22 |
|
|
|
|
| 14 |
|
| 15 |
This model is a fine-tuned version of [](https://huggingface.co/) on an unknown dataset.
|
| 16 |
It achieves the following results on the evaluation set:
|
| 17 |
+
- Loss: 0.0599
|
| 18 |
+
- Mse: 0.1695
|
| 19 |
+
- Mae: 0.3672
|
| 20 |
|
| 21 |
## Model description
|
| 22 |
|
all_results.json
CHANGED
|
@@ -1,14 +1,14 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_loss": 0.
|
| 4 |
-
"eval_mae": 0.
|
| 5 |
-
"eval_mse": 0.
|
| 6 |
-
"eval_runtime": 2.
|
| 7 |
-
"eval_samples_per_second": 4.
|
| 8 |
"eval_steps_per_second": 0.958,
|
| 9 |
"total_flos": 0.0,
|
| 10 |
-
"train_loss": 0.
|
| 11 |
-
"train_runtime":
|
| 12 |
-
"train_samples_per_second":
|
| 13 |
-
"train_steps_per_second": 0.
|
| 14 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 50.0,
|
| 3 |
+
"eval_loss": 0.05992668867111206,
|
| 4 |
+
"eval_mae": 0.3671800494194031,
|
| 5 |
+
"eval_mse": 0.16954755783081055,
|
| 6 |
+
"eval_runtime": 2.0879,
|
| 7 |
+
"eval_samples_per_second": 4.789,
|
| 8 |
"eval_steps_per_second": 0.958,
|
| 9 |
"total_flos": 0.0,
|
| 10 |
+
"train_loss": 0.08142914772033691,
|
| 11 |
+
"train_runtime": 429.9896,
|
| 12 |
+
"train_samples_per_second": 9.07,
|
| 13 |
+
"train_steps_per_second": 0.233
|
| 14 |
}
|
eval_results.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_loss": 0.
|
| 4 |
-
"eval_mae": 0.
|
| 5 |
-
"eval_mse": 0.
|
| 6 |
-
"eval_runtime": 2.
|
| 7 |
-
"eval_samples_per_second": 4.
|
| 8 |
"eval_steps_per_second": 0.958
|
| 9 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 50.0,
|
| 3 |
+
"eval_loss": 0.05992668867111206,
|
| 4 |
+
"eval_mae": 0.3671800494194031,
|
| 5 |
+
"eval_mse": 0.16954755783081055,
|
| 6 |
+
"eval_runtime": 2.0879,
|
| 7 |
+
"eval_samples_per_second": 4.789,
|
| 8 |
"eval_steps_per_second": 0.958
|
| 9 |
}
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
"total_flos": 0.0,
|
| 4 |
-
"train_loss": 0.
|
| 5 |
-
"train_runtime":
|
| 6 |
-
"train_samples_per_second":
|
| 7 |
-
"train_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 50.0,
|
| 3 |
"total_flos": 0.0,
|
| 4 |
+
"train_loss": 0.08142914772033691,
|
| 5 |
+
"train_runtime": 429.9896,
|
| 6 |
+
"train_samples_per_second": 9.07,
|
| 7 |
+
"train_steps_per_second": 0.233
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,68 +1,542 @@
|
|
| 1 |
{
|
| 2 |
-
"best_global_step":
|
| 3 |
-
"best_metric": 0.
|
| 4 |
-
"best_model_checkpoint": "./vit_focus/checkpoint-
|
| 5 |
-
"epoch":
|
| 6 |
"eval_steps": 500,
|
| 7 |
-
"global_step":
|
| 8 |
"is_hyper_param_search": false,
|
| 9 |
"is_local_process_zero": true,
|
| 10 |
"is_world_process_zero": true,
|
| 11 |
"log_history": [
|
| 12 |
{
|
| 13 |
"epoch": 1.0,
|
| 14 |
-
"eval_loss": 0.
|
| 15 |
-
"eval_mae": 0.
|
| 16 |
-
"eval_mse": 0.
|
| 17 |
-
"eval_runtime": 1.
|
| 18 |
-
"eval_samples_per_second":
|
| 19 |
-
"eval_steps_per_second": 1.
|
| 20 |
"step": 2
|
| 21 |
},
|
| 22 |
{
|
| 23 |
"epoch": 2.0,
|
| 24 |
-
"eval_loss": 0.
|
| 25 |
-
"eval_mae": 0.
|
| 26 |
-
"eval_mse": 0.
|
| 27 |
-
"eval_runtime": 1.
|
| 28 |
-
"eval_samples_per_second":
|
| 29 |
-
"eval_steps_per_second": 1.
|
| 30 |
"step": 4
|
| 31 |
},
|
| 32 |
{
|
| 33 |
"epoch": 3.0,
|
| 34 |
-
"eval_loss": 0.
|
| 35 |
-
"eval_mae": 0.
|
| 36 |
-
"eval_mse": 0.
|
| 37 |
-
"eval_runtime": 1.
|
| 38 |
-
"eval_samples_per_second":
|
| 39 |
-
"eval_steps_per_second": 1.
|
| 40 |
"step": 6
|
| 41 |
},
|
| 42 |
{
|
| 43 |
"epoch": 4.0,
|
| 44 |
-
"eval_loss": 0.
|
| 45 |
-
"eval_mae": 0.
|
| 46 |
-
"eval_mse": 0.
|
| 47 |
-
"eval_runtime": 1.
|
| 48 |
-
"eval_samples_per_second":
|
| 49 |
-
"eval_steps_per_second": 1.
|
| 50 |
"step": 8
|
| 51 |
},
|
| 52 |
{
|
| 53 |
-
"epoch":
|
| 54 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 55 |
"total_flos": 0.0,
|
| 56 |
-
"train_loss": 0.
|
| 57 |
-
"train_runtime":
|
| 58 |
-
"train_samples_per_second":
|
| 59 |
-
"train_steps_per_second": 0.
|
| 60 |
}
|
| 61 |
],
|
| 62 |
"logging_steps": 50,
|
| 63 |
-
"max_steps":
|
| 64 |
"num_input_tokens_seen": 0,
|
| 65 |
-
"num_train_epochs":
|
| 66 |
"save_steps": 500,
|
| 67 |
"stateful_callbacks": {
|
| 68 |
"TrainerControl": {
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_global_step": 58,
|
| 3 |
+
"best_metric": 0.16954755783081055,
|
| 4 |
+
"best_model_checkpoint": "./vit_focus/checkpoint-58",
|
| 5 |
+
"epoch": 50.0,
|
| 6 |
"eval_steps": 500,
|
| 7 |
+
"global_step": 100,
|
| 8 |
"is_hyper_param_search": false,
|
| 9 |
"is_local_process_zero": true,
|
| 10 |
"is_world_process_zero": true,
|
| 11 |
"log_history": [
|
| 12 |
{
|
| 13 |
"epoch": 1.0,
|
| 14 |
+
"eval_loss": 0.12364959716796875,
|
| 15 |
+
"eval_mae": 0.41142117977142334,
|
| 16 |
+
"eval_mse": 0.2097923755645752,
|
| 17 |
+
"eval_runtime": 1.3273,
|
| 18 |
+
"eval_samples_per_second": 7.534,
|
| 19 |
+
"eval_steps_per_second": 1.507,
|
| 20 |
"step": 2
|
| 21 |
},
|
| 22 |
{
|
| 23 |
"epoch": 2.0,
|
| 24 |
+
"eval_loss": 0.1196303591132164,
|
| 25 |
+
"eval_mae": 0.40995553135871887,
|
| 26 |
+
"eval_mse": 0.2083383947610855,
|
| 27 |
+
"eval_runtime": 1.2492,
|
| 28 |
+
"eval_samples_per_second": 8.005,
|
| 29 |
+
"eval_steps_per_second": 1.601,
|
| 30 |
"step": 4
|
| 31 |
},
|
| 32 |
{
|
| 33 |
"epoch": 3.0,
|
| 34 |
+
"eval_loss": 0.11328934133052826,
|
| 35 |
+
"eval_mae": 0.4075067639350891,
|
| 36 |
+
"eval_mse": 0.20581738650798798,
|
| 37 |
+
"eval_runtime": 1.1974,
|
| 38 |
+
"eval_samples_per_second": 8.352,
|
| 39 |
+
"eval_steps_per_second": 1.67,
|
| 40 |
"step": 6
|
| 41 |
},
|
| 42 |
{
|
| 43 |
"epoch": 4.0,
|
| 44 |
+
"eval_loss": 0.10604934394359589,
|
| 45 |
+
"eval_mae": 0.4042162597179413,
|
| 46 |
+
"eval_mse": 0.20277394354343414,
|
| 47 |
+
"eval_runtime": 1.202,
|
| 48 |
+
"eval_samples_per_second": 8.32,
|
| 49 |
+
"eval_steps_per_second": 1.664,
|
| 50 |
"step": 8
|
| 51 |
},
|
| 52 |
{
|
| 53 |
+
"epoch": 5.0,
|
| 54 |
+
"eval_loss": 0.10327418148517609,
|
| 55 |
+
"eval_mae": 0.4024171531200409,
|
| 56 |
+
"eval_mse": 0.20201840996742249,
|
| 57 |
+
"eval_runtime": 1.3089,
|
| 58 |
+
"eval_samples_per_second": 7.64,
|
| 59 |
+
"eval_steps_per_second": 1.528,
|
| 60 |
+
"step": 10
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 6.0,
|
| 64 |
+
"eval_loss": 0.09384196996688843,
|
| 65 |
+
"eval_mae": 0.3968161642551422,
|
| 66 |
+
"eval_mse": 0.19719576835632324,
|
| 67 |
+
"eval_runtime": 1.2505,
|
| 68 |
+
"eval_samples_per_second": 7.997,
|
| 69 |
+
"eval_steps_per_second": 1.599,
|
| 70 |
+
"step": 12
|
| 71 |
+
},
|
| 72 |
+
{
|
| 73 |
+
"epoch": 7.0,
|
| 74 |
+
"eval_loss": 0.10823704302310944,
|
| 75 |
+
"eval_mae": 0.40178966522216797,
|
| 76 |
+
"eval_mse": 0.2021051049232483,
|
| 77 |
+
"eval_runtime": 1.2484,
|
| 78 |
+
"eval_samples_per_second": 8.01,
|
| 79 |
+
"eval_steps_per_second": 1.602,
|
| 80 |
+
"step": 14
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 8.0,
|
| 84 |
+
"eval_loss": 0.0798194482922554,
|
| 85 |
+
"eval_mae": 0.3889862298965454,
|
| 86 |
+
"eval_mse": 0.18965376913547516,
|
| 87 |
+
"eval_runtime": 1.2049,
|
| 88 |
+
"eval_samples_per_second": 8.3,
|
| 89 |
+
"eval_steps_per_second": 1.66,
|
| 90 |
+
"step": 16
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 9.0,
|
| 94 |
+
"eval_loss": 0.10108314454555511,
|
| 95 |
+
"eval_mae": 0.39635446667671204,
|
| 96 |
+
"eval_mse": 0.19831416010856628,
|
| 97 |
+
"eval_runtime": 1.2689,
|
| 98 |
+
"eval_samples_per_second": 7.881,
|
| 99 |
+
"eval_steps_per_second": 1.576,
|
| 100 |
+
"step": 18
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 10.0,
|
| 104 |
+
"eval_loss": 0.08742986619472504,
|
| 105 |
+
"eval_mae": 0.3887980282306671,
|
| 106 |
+
"eval_mse": 0.19129884243011475,
|
| 107 |
+
"eval_runtime": 1.2793,
|
| 108 |
+
"eval_samples_per_second": 7.817,
|
| 109 |
+
"eval_steps_per_second": 1.563,
|
| 110 |
+
"step": 20
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 11.0,
|
| 114 |
+
"eval_loss": 0.06867878139019012,
|
| 115 |
+
"eval_mae": 0.3799101710319519,
|
| 116 |
+
"eval_mse": 0.18114082515239716,
|
| 117 |
+
"eval_runtime": 1.3093,
|
| 118 |
+
"eval_samples_per_second": 7.637,
|
| 119 |
+
"eval_steps_per_second": 1.527,
|
| 120 |
+
"step": 22
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 12.0,
|
| 124 |
+
"eval_loss": 0.08115904778242111,
|
| 125 |
+
"eval_mae": 0.38764962553977966,
|
| 126 |
+
"eval_mse": 0.18922746181488037,
|
| 127 |
+
"eval_runtime": 1.1917,
|
| 128 |
+
"eval_samples_per_second": 8.392,
|
| 129 |
+
"eval_steps_per_second": 1.678,
|
| 130 |
+
"step": 24
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 13.0,
|
| 134 |
+
"eval_loss": 0.07042384147644043,
|
| 135 |
+
"eval_mae": 0.3812001347541809,
|
| 136 |
+
"eval_mse": 0.1831439584493637,
|
| 137 |
+
"eval_runtime": 1.2731,
|
| 138 |
+
"eval_samples_per_second": 7.855,
|
| 139 |
+
"eval_steps_per_second": 1.571,
|
| 140 |
+
"step": 26
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 14.0,
|
| 144 |
+
"eval_loss": 0.057908277958631516,
|
| 145 |
+
"eval_mae": 0.37130409479141235,
|
| 146 |
+
"eval_mse": 0.17441079020500183,
|
| 147 |
+
"eval_runtime": 3.0777,
|
| 148 |
+
"eval_samples_per_second": 3.249,
|
| 149 |
+
"eval_steps_per_second": 0.65,
|
| 150 |
+
"step": 28
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 15.0,
|
| 154 |
+
"eval_loss": 0.06775574386119843,
|
| 155 |
+
"eval_mae": 0.37615475058555603,
|
| 156 |
+
"eval_mse": 0.1790776401758194,
|
| 157 |
+
"eval_runtime": 1.2719,
|
| 158 |
+
"eval_samples_per_second": 7.862,
|
| 159 |
+
"eval_steps_per_second": 1.572,
|
| 160 |
+
"step": 30
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 16.0,
|
| 164 |
+
"eval_loss": 0.08812335133552551,
|
| 165 |
+
"eval_mae": 0.38629305362701416,
|
| 166 |
+
"eval_mse": 0.1883399784564972,
|
| 167 |
+
"eval_runtime": 1.6525,
|
| 168 |
+
"eval_samples_per_second": 6.051,
|
| 169 |
+
"eval_steps_per_second": 1.21,
|
| 170 |
+
"step": 32
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 17.0,
|
| 174 |
+
"eval_loss": 0.09601505100727081,
|
| 175 |
+
"eval_mae": 0.38798943161964417,
|
| 176 |
+
"eval_mse": 0.18960143625736237,
|
| 177 |
+
"eval_runtime": 1.4237,
|
| 178 |
+
"eval_samples_per_second": 7.024,
|
| 179 |
+
"eval_steps_per_second": 1.405,
|
| 180 |
+
"step": 34
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"epoch": 18.0,
|
| 184 |
+
"eval_loss": 0.0696040540933609,
|
| 185 |
+
"eval_mae": 0.3752928674221039,
|
| 186 |
+
"eval_mse": 0.17762857675552368,
|
| 187 |
+
"eval_runtime": 1.2854,
|
| 188 |
+
"eval_samples_per_second": 7.78,
|
| 189 |
+
"eval_steps_per_second": 1.556,
|
| 190 |
+
"step": 36
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 19.0,
|
| 194 |
+
"eval_loss": 0.057551633566617966,
|
| 195 |
+
"eval_mae": 0.3691748082637787,
|
| 196 |
+
"eval_mse": 0.1715679168701172,
|
| 197 |
+
"eval_runtime": 1.4944,
|
| 198 |
+
"eval_samples_per_second": 6.692,
|
| 199 |
+
"eval_steps_per_second": 1.338,
|
| 200 |
+
"step": 38
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 20.0,
|
| 204 |
+
"eval_loss": 0.05853904038667679,
|
| 205 |
+
"eval_mae": 0.3697223663330078,
|
| 206 |
+
"eval_mse": 0.17197829484939575,
|
| 207 |
+
"eval_runtime": 1.2655,
|
| 208 |
+
"eval_samples_per_second": 7.902,
|
| 209 |
+
"eval_steps_per_second": 1.58,
|
| 210 |
+
"step": 40
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"epoch": 21.0,
|
| 214 |
+
"eval_loss": 0.07102437317371368,
|
| 215 |
+
"eval_mae": 0.37734654545783997,
|
| 216 |
+
"eval_mse": 0.1792123019695282,
|
| 217 |
+
"eval_runtime": 1.923,
|
| 218 |
+
"eval_samples_per_second": 5.2,
|
| 219 |
+
"eval_steps_per_second": 1.04,
|
| 220 |
+
"step": 42
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 22.0,
|
| 224 |
+
"eval_loss": 0.08145803213119507,
|
| 225 |
+
"eval_mae": 0.38286036252975464,
|
| 226 |
+
"eval_mse": 0.1842835247516632,
|
| 227 |
+
"eval_runtime": 1.1671,
|
| 228 |
+
"eval_samples_per_second": 8.568,
|
| 229 |
+
"eval_steps_per_second": 1.714,
|
| 230 |
+
"step": 44
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 23.0,
|
| 234 |
+
"eval_loss": 0.06858658790588379,
|
| 235 |
+
"eval_mae": 0.3709622919559479,
|
| 236 |
+
"eval_mse": 0.17373624444007874,
|
| 237 |
+
"eval_runtime": 1.3253,
|
| 238 |
+
"eval_samples_per_second": 7.545,
|
| 239 |
+
"eval_steps_per_second": 1.509,
|
| 240 |
+
"step": 46
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 24.0,
|
| 244 |
+
"eval_loss": 0.06739358603954315,
|
| 245 |
+
"eval_mae": 0.37041887640953064,
|
| 246 |
+
"eval_mse": 0.173396497964859,
|
| 247 |
+
"eval_runtime": 3.1433,
|
| 248 |
+
"eval_samples_per_second": 3.181,
|
| 249 |
+
"eval_steps_per_second": 0.636,
|
| 250 |
+
"step": 48
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 25.0,
|
| 254 |
+
"grad_norm": 0.9524662494659424,
|
| 255 |
+
"learning_rate": 1.53e-05,
|
| 256 |
+
"loss": 0.118,
|
| 257 |
+
"step": 50
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 25.0,
|
| 261 |
+
"eval_loss": 0.07070203125476837,
|
| 262 |
+
"eval_mae": 0.3756571412086487,
|
| 263 |
+
"eval_mse": 0.17759118974208832,
|
| 264 |
+
"eval_runtime": 1.2345,
|
| 265 |
+
"eval_samples_per_second": 8.1,
|
| 266 |
+
"eval_steps_per_second": 1.62,
|
| 267 |
+
"step": 50
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 26.0,
|
| 271 |
+
"eval_loss": 0.07530557364225388,
|
| 272 |
+
"eval_mae": 0.38040465116500854,
|
| 273 |
+
"eval_mse": 0.1816762387752533,
|
| 274 |
+
"eval_runtime": 1.7054,
|
| 275 |
+
"eval_samples_per_second": 5.864,
|
| 276 |
+
"eval_steps_per_second": 1.173,
|
| 277 |
+
"step": 52
|
| 278 |
+
},
|
| 279 |
+
{
|
| 280 |
+
"epoch": 27.0,
|
| 281 |
+
"eval_loss": 0.0708126351237297,
|
| 282 |
+
"eval_mae": 0.37711387872695923,
|
| 283 |
+
"eval_mse": 0.17874492704868317,
|
| 284 |
+
"eval_runtime": 1.1644,
|
| 285 |
+
"eval_samples_per_second": 8.588,
|
| 286 |
+
"eval_steps_per_second": 1.718,
|
| 287 |
+
"step": 54
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 28.0,
|
| 291 |
+
"eval_loss": 0.06371479481458664,
|
| 292 |
+
"eval_mae": 0.3699105381965637,
|
| 293 |
+
"eval_mse": 0.1721055954694748,
|
| 294 |
+
"eval_runtime": 1.5225,
|
| 295 |
+
"eval_samples_per_second": 6.568,
|
| 296 |
+
"eval_steps_per_second": 1.314,
|
| 297 |
+
"step": 56
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 29.0,
|
| 301 |
+
"eval_loss": 0.05992668867111206,
|
| 302 |
+
"eval_mae": 0.3671800494194031,
|
| 303 |
+
"eval_mse": 0.16954755783081055,
|
| 304 |
+
"eval_runtime": 1.3349,
|
| 305 |
+
"eval_samples_per_second": 7.491,
|
| 306 |
+
"eval_steps_per_second": 1.498,
|
| 307 |
+
"step": 58
|
| 308 |
+
},
|
| 309 |
+
{
|
| 310 |
+
"epoch": 30.0,
|
| 311 |
+
"eval_loss": 0.06269104778766632,
|
| 312 |
+
"eval_mae": 0.3710530698299408,
|
| 313 |
+
"eval_mse": 0.17309635877609253,
|
| 314 |
+
"eval_runtime": 1.2258,
|
| 315 |
+
"eval_samples_per_second": 8.158,
|
| 316 |
+
"eval_steps_per_second": 1.632,
|
| 317 |
+
"step": 60
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 31.0,
|
| 321 |
+
"eval_loss": 0.06929778307676315,
|
| 322 |
+
"eval_mae": 0.37701377272605896,
|
| 323 |
+
"eval_mse": 0.1785697340965271,
|
| 324 |
+
"eval_runtime": 1.4115,
|
| 325 |
+
"eval_samples_per_second": 7.085,
|
| 326 |
+
"eval_steps_per_second": 1.417,
|
| 327 |
+
"step": 62
|
| 328 |
+
},
|
| 329 |
+
{
|
| 330 |
+
"epoch": 32.0,
|
| 331 |
+
"eval_loss": 0.07379092276096344,
|
| 332 |
+
"eval_mae": 0.37811335921287537,
|
| 333 |
+
"eval_mse": 0.17990507185459137,
|
| 334 |
+
"eval_runtime": 1.2044,
|
| 335 |
+
"eval_samples_per_second": 8.303,
|
| 336 |
+
"eval_steps_per_second": 1.661,
|
| 337 |
+
"step": 64
|
| 338 |
+
},
|
| 339 |
+
{
|
| 340 |
+
"epoch": 33.0,
|
| 341 |
+
"eval_loss": 0.07304650545120239,
|
| 342 |
+
"eval_mae": 0.3751561939716339,
|
| 343 |
+
"eval_mse": 0.17731547355651855,
|
| 344 |
+
"eval_runtime": 3.3496,
|
| 345 |
+
"eval_samples_per_second": 2.985,
|
| 346 |
+
"eval_steps_per_second": 0.597,
|
| 347 |
+
"step": 66
|
| 348 |
+
},
|
| 349 |
+
{
|
| 350 |
+
"epoch": 34.0,
|
| 351 |
+
"eval_loss": 0.06837408244609833,
|
| 352 |
+
"eval_mae": 0.3710552453994751,
|
| 353 |
+
"eval_mse": 0.17353828251361847,
|
| 354 |
+
"eval_runtime": 1.4629,
|
| 355 |
+
"eval_samples_per_second": 6.836,
|
| 356 |
+
"eval_steps_per_second": 1.367,
|
| 357 |
+
"step": 68
|
| 358 |
+
},
|
| 359 |
+
{
|
| 360 |
+
"epoch": 35.0,
|
| 361 |
+
"eval_loss": 0.06416038423776627,
|
| 362 |
+
"eval_mae": 0.3673146069049835,
|
| 363 |
+
"eval_mse": 0.17018868029117584,
|
| 364 |
+
"eval_runtime": 1.31,
|
| 365 |
+
"eval_samples_per_second": 7.634,
|
| 366 |
+
"eval_steps_per_second": 1.527,
|
| 367 |
+
"step": 70
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 36.0,
|
| 371 |
+
"eval_loss": 0.06411489099264145,
|
| 372 |
+
"eval_mae": 0.36943870782852173,
|
| 373 |
+
"eval_mse": 0.1721121072769165,
|
| 374 |
+
"eval_runtime": 1.5048,
|
| 375 |
+
"eval_samples_per_second": 6.646,
|
| 376 |
+
"eval_steps_per_second": 1.329,
|
| 377 |
+
"step": 72
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 37.0,
|
| 381 |
+
"eval_loss": 0.06872588396072388,
|
| 382 |
+
"eval_mae": 0.373674213886261,
|
| 383 |
+
"eval_mse": 0.17579391598701477,
|
| 384 |
+
"eval_runtime": 1.4802,
|
| 385 |
+
"eval_samples_per_second": 6.756,
|
| 386 |
+
"eval_steps_per_second": 1.351,
|
| 387 |
+
"step": 74
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 38.0,
|
| 391 |
+
"eval_loss": 0.07391238957643509,
|
| 392 |
+
"eval_mae": 0.3772023916244507,
|
| 393 |
+
"eval_mse": 0.17880253493785858,
|
| 394 |
+
"eval_runtime": 1.3549,
|
| 395 |
+
"eval_samples_per_second": 7.381,
|
| 396 |
+
"eval_steps_per_second": 1.476,
|
| 397 |
+
"step": 76
|
| 398 |
+
},
|
| 399 |
+
{
|
| 400 |
+
"epoch": 39.0,
|
| 401 |
+
"eval_loss": 0.07059959322214127,
|
| 402 |
+
"eval_mae": 0.3748144209384918,
|
| 403 |
+
"eval_mse": 0.1765395551919937,
|
| 404 |
+
"eval_runtime": 1.3015,
|
| 405 |
+
"eval_samples_per_second": 7.683,
|
| 406 |
+
"eval_steps_per_second": 1.537,
|
| 407 |
+
"step": 78
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"epoch": 40.0,
|
| 411 |
+
"eval_loss": 0.06648191064596176,
|
| 412 |
+
"eval_mae": 0.37035736441612244,
|
| 413 |
+
"eval_mse": 0.17257149517536163,
|
| 414 |
+
"eval_runtime": 1.4739,
|
| 415 |
+
"eval_samples_per_second": 6.785,
|
| 416 |
+
"eval_steps_per_second": 1.357,
|
| 417 |
+
"step": 80
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 41.0,
|
| 421 |
+
"eval_loss": 0.06419553607702255,
|
| 422 |
+
"eval_mae": 0.36771345138549805,
|
| 423 |
+
"eval_mse": 0.1702636033296585,
|
| 424 |
+
"eval_runtime": 1.3959,
|
| 425 |
+
"eval_samples_per_second": 7.164,
|
| 426 |
+
"eval_steps_per_second": 1.433,
|
| 427 |
+
"step": 82
|
| 428 |
+
},
|
| 429 |
+
{
|
| 430 |
+
"epoch": 42.0,
|
| 431 |
+
"eval_loss": 0.06590329110622406,
|
| 432 |
+
"eval_mae": 0.3694536089897156,
|
| 433 |
+
"eval_mse": 0.17185601592063904,
|
| 434 |
+
"eval_runtime": 1.3544,
|
| 435 |
+
"eval_samples_per_second": 7.383,
|
| 436 |
+
"eval_steps_per_second": 1.477,
|
| 437 |
+
"step": 84
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 43.0,
|
| 441 |
+
"eval_loss": 0.06824100762605667,
|
| 442 |
+
"eval_mae": 0.3721493184566498,
|
| 443 |
+
"eval_mse": 0.1743040680885315,
|
| 444 |
+
"eval_runtime": 1.3016,
|
| 445 |
+
"eval_samples_per_second": 7.683,
|
| 446 |
+
"eval_steps_per_second": 1.537,
|
| 447 |
+
"step": 86
|
| 448 |
+
},
|
| 449 |
+
{
|
| 450 |
+
"epoch": 44.0,
|
| 451 |
+
"eval_loss": 0.07233113795518875,
|
| 452 |
+
"eval_mae": 0.37587830424308777,
|
| 453 |
+
"eval_mse": 0.17762817442417145,
|
| 454 |
+
"eval_runtime": 1.2282,
|
| 455 |
+
"eval_samples_per_second": 8.142,
|
| 456 |
+
"eval_steps_per_second": 1.628,
|
| 457 |
+
"step": 88
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 45.0,
|
| 461 |
+
"eval_loss": 0.0730481892824173,
|
| 462 |
+
"eval_mae": 0.37610477209091187,
|
| 463 |
+
"eval_mse": 0.1778472512960434,
|
| 464 |
+
"eval_runtime": 1.2255,
|
| 465 |
+
"eval_samples_per_second": 8.16,
|
| 466 |
+
"eval_steps_per_second": 1.632,
|
| 467 |
+
"step": 90
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"epoch": 46.0,
|
| 471 |
+
"eval_loss": 0.07376444339752197,
|
| 472 |
+
"eval_mae": 0.37569037079811096,
|
| 473 |
+
"eval_mse": 0.17753194272518158,
|
| 474 |
+
"eval_runtime": 1.2758,
|
| 475 |
+
"eval_samples_per_second": 7.838,
|
| 476 |
+
"eval_steps_per_second": 1.568,
|
| 477 |
+
"step": 92
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 47.0,
|
| 481 |
+
"eval_loss": 0.07429193705320358,
|
| 482 |
+
"eval_mae": 0.3751032054424286,
|
| 483 |
+
"eval_mse": 0.1769881248474121,
|
| 484 |
+
"eval_runtime": 1.1953,
|
| 485 |
+
"eval_samples_per_second": 8.366,
|
| 486 |
+
"eval_steps_per_second": 1.673,
|
| 487 |
+
"step": 94
|
| 488 |
+
},
|
| 489 |
+
{
|
| 490 |
+
"epoch": 48.0,
|
| 491 |
+
"eval_loss": 0.07430405914783478,
|
| 492 |
+
"eval_mae": 0.3747301399707794,
|
| 493 |
+
"eval_mse": 0.17660388350486755,
|
| 494 |
+
"eval_runtime": 1.1671,
|
| 495 |
+
"eval_samples_per_second": 8.568,
|
| 496 |
+
"eval_steps_per_second": 1.714,
|
| 497 |
+
"step": 96
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 49.0,
|
| 501 |
+
"eval_loss": 0.07368143647909164,
|
| 502 |
+
"eval_mae": 0.37456196546554565,
|
| 503 |
+
"eval_mse": 0.17642483115196228,
|
| 504 |
+
"eval_runtime": 1.545,
|
| 505 |
+
"eval_samples_per_second": 6.472,
|
| 506 |
+
"eval_steps_per_second": 1.294,
|
| 507 |
+
"step": 98
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 50.0,
|
| 511 |
+
"grad_norm": 0.7434751987457275,
|
| 512 |
+
"learning_rate": 3.0000000000000004e-07,
|
| 513 |
+
"loss": 0.0449,
|
| 514 |
+
"step": 100
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 50.0,
|
| 518 |
+
"eval_loss": 0.07292107492685318,
|
| 519 |
+
"eval_mae": 0.3741002678871155,
|
| 520 |
+
"eval_mse": 0.17599670588970184,
|
| 521 |
+
"eval_runtime": 1.2607,
|
| 522 |
+
"eval_samples_per_second": 7.932,
|
| 523 |
+
"eval_steps_per_second": 1.586,
|
| 524 |
+
"step": 100
|
| 525 |
+
},
|
| 526 |
+
{
|
| 527 |
+
"epoch": 50.0,
|
| 528 |
+
"step": 100,
|
| 529 |
"total_flos": 0.0,
|
| 530 |
+
"train_loss": 0.08142914772033691,
|
| 531 |
+
"train_runtime": 429.9896,
|
| 532 |
+
"train_samples_per_second": 9.07,
|
| 533 |
+
"train_steps_per_second": 0.233
|
| 534 |
}
|
| 535 |
],
|
| 536 |
"logging_steps": 50,
|
| 537 |
+
"max_steps": 100,
|
| 538 |
"num_input_tokens_seen": 0,
|
| 539 |
+
"num_train_epochs": 100,
|
| 540 |
"save_steps": 500,
|
| 541 |
"stateful_callbacks": {
|
| 542 |
"TrainerControl": {
|