End of training
Browse files- README.md +5 -3
- all_results.json +13 -0
- eval_results.json +8 -0
- runs/Dec03_10-24-07_algo-1/events.out.tfevents.1764757902.algo-1.68.1 +3 -0
- train_results.json +8 -0
- trainer_state.json +547 -0
README.md
CHANGED
|
@@ -3,6 +3,8 @@ library_name: transformers
|
|
| 3 |
license: apache-2.0
|
| 4 |
base_model: google/vit-base-patch16-224
|
| 5 |
tags:
|
|
|
|
|
|
|
| 6 |
- generated_from_trainer
|
| 7 |
metrics:
|
| 8 |
- accuracy
|
|
@@ -16,10 +18,10 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 16 |
|
| 17 |
# Validated_Balanced_Raw_Data_model_vit
|
| 18 |
|
| 19 |
-
This model is a fine-tuned version of [google/vit-base-patch16-224](https://huggingface.co/google/vit-base-patch16-224) on
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
-
- Loss: 1.
|
| 22 |
-
- Accuracy: 0.
|
| 23 |
|
| 24 |
## Model description
|
| 25 |
|
|
|
|
| 3 |
license: apache-2.0
|
| 4 |
base_model: google/vit-base-patch16-224
|
| 5 |
tags:
|
| 6 |
+
- image-classification
|
| 7 |
+
- vision
|
| 8 |
- generated_from_trainer
|
| 9 |
metrics:
|
| 10 |
- accuracy
|
|
|
|
| 18 |
|
| 19 |
# Validated_Balanced_Raw_Data_model_vit
|
| 20 |
|
| 21 |
+
This model is a fine-tuned version of [google/vit-base-patch16-224](https://huggingface.co/google/vit-base-patch16-224) on the Logiroad/Validated_Balanced_Raw_Dataset dataset.
|
| 22 |
It achieves the following results on the evaluation set:
|
| 23 |
+
- Loss: 1.2154
|
| 24 |
+
- Accuracy: 0.5094
|
| 25 |
|
| 26 |
## Model description
|
| 27 |
|
all_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 25.0,
|
| 3 |
+
"eval_accuracy": 0.5094339622641509,
|
| 4 |
+
"eval_loss": 1.2154083251953125,
|
| 5 |
+
"eval_runtime": 1.1397,
|
| 6 |
+
"eval_samples_per_second": 186.015,
|
| 7 |
+
"eval_steps_per_second": 23.691,
|
| 8 |
+
"total_flos": 1.2321447211229184e+18,
|
| 9 |
+
"train_loss": 1.1025613689422606,
|
| 10 |
+
"train_runtime": 425.612,
|
| 11 |
+
"train_samples_per_second": 37.358,
|
| 12 |
+
"train_steps_per_second": 4.699
|
| 13 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 25.0,
|
| 3 |
+
"eval_accuracy": 0.5094339622641509,
|
| 4 |
+
"eval_loss": 1.2154083251953125,
|
| 5 |
+
"eval_runtime": 1.1397,
|
| 6 |
+
"eval_samples_per_second": 186.015,
|
| 7 |
+
"eval_steps_per_second": 23.691
|
| 8 |
+
}
|
runs/Dec03_10-24-07_algo-1/events.out.tfevents.1764757902.algo-1.68.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:eed2f286c7c831d033eb178f56d124237c205131273b47845243355024a76cea
|
| 3 |
+
size 411
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 25.0,
|
| 3 |
+
"total_flos": 1.2321447211229184e+18,
|
| 4 |
+
"train_loss": 1.1025613689422606,
|
| 5 |
+
"train_runtime": 425.612,
|
| 6 |
+
"train_samples_per_second": 37.358,
|
| 7 |
+
"train_steps_per_second": 4.699
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,547 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.5094339622641509,
|
| 3 |
+
"best_model_checkpoint": "./Validated_Balanced_Raw_Data_model_vit_outputs/checkpoint-1440",
|
| 4 |
+
"epoch": 25.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 2000,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.625,
|
| 13 |
+
"grad_norm": 10.029894828796387,
|
| 14 |
+
"learning_rate": 1.47e-05,
|
| 15 |
+
"loss": 1.4452,
|
| 16 |
+
"step": 50
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 1.0,
|
| 20 |
+
"eval_accuracy": 0.24056603773584906,
|
| 21 |
+
"eval_loss": 1.4536809921264648,
|
| 22 |
+
"eval_runtime": 1.1349,
|
| 23 |
+
"eval_samples_per_second": 186.798,
|
| 24 |
+
"eval_steps_per_second": 23.79,
|
| 25 |
+
"step": 80
|
| 26 |
+
},
|
| 27 |
+
{
|
| 28 |
+
"epoch": 1.25,
|
| 29 |
+
"grad_norm": 13.045843124389648,
|
| 30 |
+
"learning_rate": 2.97e-05,
|
| 31 |
+
"loss": 1.4021,
|
| 32 |
+
"step": 100
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 1.875,
|
| 36 |
+
"grad_norm": 9.291085243225098,
|
| 37 |
+
"learning_rate": 2.9950795096316707e-05,
|
| 38 |
+
"loss": 1.3534,
|
| 39 |
+
"step": 150
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 2.0,
|
| 43 |
+
"eval_accuracy": 0.35377358490566035,
|
| 44 |
+
"eval_loss": 1.4003523588180542,
|
| 45 |
+
"eval_runtime": 1.1414,
|
| 46 |
+
"eval_samples_per_second": 185.736,
|
| 47 |
+
"eval_steps_per_second": 23.655,
|
| 48 |
+
"step": 160
|
| 49 |
+
},
|
| 50 |
+
{
|
| 51 |
+
"epoch": 2.5,
|
| 52 |
+
"grad_norm": 8.364015579223633,
|
| 53 |
+
"learning_rate": 2.979948160693239e-05,
|
| 54 |
+
"loss": 1.2977,
|
| 55 |
+
"step": 200
|
| 56 |
+
},
|
| 57 |
+
{
|
| 58 |
+
"epoch": 3.0,
|
| 59 |
+
"eval_accuracy": 0.37735849056603776,
|
| 60 |
+
"eval_loss": 1.3416857719421387,
|
| 61 |
+
"eval_runtime": 1.1295,
|
| 62 |
+
"eval_samples_per_second": 187.7,
|
| 63 |
+
"eval_steps_per_second": 23.905,
|
| 64 |
+
"step": 240
|
| 65 |
+
},
|
| 66 |
+
{
|
| 67 |
+
"epoch": 3.125,
|
| 68 |
+
"grad_norm": 10.082795143127441,
|
| 69 |
+
"learning_rate": 2.9547072651695803e-05,
|
| 70 |
+
"loss": 1.2979,
|
| 71 |
+
"step": 250
|
| 72 |
+
},
|
| 73 |
+
{
|
| 74 |
+
"epoch": 3.75,
|
| 75 |
+
"grad_norm": 9.592968940734863,
|
| 76 |
+
"learning_rate": 2.9195292439710517e-05,
|
| 77 |
+
"loss": 1.2604,
|
| 78 |
+
"step": 300
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"epoch": 4.0,
|
| 82 |
+
"eval_accuracy": 0.37735849056603776,
|
| 83 |
+
"eval_loss": 1.3131976127624512,
|
| 84 |
+
"eval_runtime": 1.1222,
|
| 85 |
+
"eval_samples_per_second": 188.91,
|
| 86 |
+
"eval_steps_per_second": 24.059,
|
| 87 |
+
"step": 320
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 4.375,
|
| 91 |
+
"grad_norm": 9.461295127868652,
|
| 92 |
+
"learning_rate": 2.874654398652483e-05,
|
| 93 |
+
"loss": 1.2509,
|
| 94 |
+
"step": 350
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 5.0,
|
| 98 |
+
"grad_norm": 20.441808700561523,
|
| 99 |
+
"learning_rate": 2.8203892699098956e-05,
|
| 100 |
+
"loss": 1.2428,
|
| 101 |
+
"step": 400
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 5.0,
|
| 105 |
+
"eval_accuracy": 0.4009433962264151,
|
| 106 |
+
"eval_loss": 1.2442667484283447,
|
| 107 |
+
"eval_runtime": 1.1294,
|
| 108 |
+
"eval_samples_per_second": 187.712,
|
| 109 |
+
"eval_steps_per_second": 23.907,
|
| 110 |
+
"step": 400
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 5.625,
|
| 114 |
+
"grad_norm": 7.7924604415893555,
|
| 115 |
+
"learning_rate": 2.757104543596717e-05,
|
| 116 |
+
"loss": 1.213,
|
| 117 |
+
"step": 450
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"epoch": 6.0,
|
| 121 |
+
"eval_accuracy": 0.419811320754717,
|
| 122 |
+
"eval_loss": 1.2147694826126099,
|
| 123 |
+
"eval_runtime": 1.1426,
|
| 124 |
+
"eval_samples_per_second": 185.546,
|
| 125 |
+
"eval_steps_per_second": 23.631,
|
| 126 |
+
"step": 480
|
| 127 |
+
},
|
| 128 |
+
{
|
| 129 |
+
"epoch": 6.25,
|
| 130 |
+
"grad_norm": 9.47819709777832,
|
| 131 |
+
"learning_rate": 2.6852325185635357e-05,
|
| 132 |
+
"loss": 1.2163,
|
| 133 |
+
"step": 500
|
| 134 |
+
},
|
| 135 |
+
{
|
| 136 |
+
"epoch": 6.875,
|
| 137 |
+
"grad_norm": 9.0564546585083,
|
| 138 |
+
"learning_rate": 2.605264153618602e-05,
|
| 139 |
+
"loss": 1.1426,
|
| 140 |
+
"step": 550
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 7.0,
|
| 144 |
+
"eval_accuracy": 0.4669811320754717,
|
| 145 |
+
"eval_loss": 1.2095811367034912,
|
| 146 |
+
"eval_runtime": 1.125,
|
| 147 |
+
"eval_samples_per_second": 188.45,
|
| 148 |
+
"eval_steps_per_second": 24.001,
|
| 149 |
+
"step": 560
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 7.5,
|
| 153 |
+
"grad_norm": 6.277278900146484,
|
| 154 |
+
"learning_rate": 2.517745713781345e-05,
|
| 155 |
+
"loss": 1.1657,
|
| 156 |
+
"step": 600
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 8.0,
|
| 160 |
+
"eval_accuracy": 0.4669811320754717,
|
| 161 |
+
"eval_loss": 1.2065520286560059,
|
| 162 |
+
"eval_runtime": 1.1354,
|
| 163 |
+
"eval_samples_per_second": 186.725,
|
| 164 |
+
"eval_steps_per_second": 23.781,
|
| 165 |
+
"step": 640
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 8.125,
|
| 169 |
+
"grad_norm": 9.455373764038086,
|
| 170 |
+
"learning_rate": 2.4252284842435667e-05,
|
| 171 |
+
"loss": 1.1988,
|
| 172 |
+
"step": 650
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"epoch": 8.75,
|
| 176 |
+
"grad_norm": 9.699421882629395,
|
| 177 |
+
"learning_rate": 2.3245704321242494e-05,
|
| 178 |
+
"loss": 1.1249,
|
| 179 |
+
"step": 700
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 9.0,
|
| 183 |
+
"eval_accuracy": 0.4386792452830189,
|
| 184 |
+
"eval_loss": 1.220900297164917,
|
| 185 |
+
"eval_runtime": 1.1295,
|
| 186 |
+
"eval_samples_per_second": 187.686,
|
| 187 |
+
"eval_steps_per_second": 23.903,
|
| 188 |
+
"step": 720
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 9.375,
|
| 192 |
+
"grad_norm": 10.024123191833496,
|
| 193 |
+
"learning_rate": 2.218279727850104e-05,
|
| 194 |
+
"loss": 1.0891,
|
| 195 |
+
"step": 750
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 10.0,
|
| 199 |
+
"grad_norm": 14.003544807434082,
|
| 200 |
+
"learning_rate": 2.107082444708681e-05,
|
| 201 |
+
"loss": 1.1622,
|
| 202 |
+
"step": 800
|
| 203 |
+
},
|
| 204 |
+
{
|
| 205 |
+
"epoch": 10.0,
|
| 206 |
+
"eval_accuracy": 0.4811320754716981,
|
| 207 |
+
"eval_loss": 1.1445928812026978,
|
| 208 |
+
"eval_runtime": 1.1414,
|
| 209 |
+
"eval_samples_per_second": 185.734,
|
| 210 |
+
"eval_steps_per_second": 23.655,
|
| 211 |
+
"step": 800
|
| 212 |
+
},
|
| 213 |
+
{
|
| 214 |
+
"epoch": 10.625,
|
| 215 |
+
"grad_norm": 10.41913890838623,
|
| 216 |
+
"learning_rate": 1.9917381728963962e-05,
|
| 217 |
+
"loss": 1.0625,
|
| 218 |
+
"step": 850
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 11.0,
|
| 222 |
+
"eval_accuracy": 0.4669811320754717,
|
| 223 |
+
"eval_loss": 1.1742380857467651,
|
| 224 |
+
"eval_runtime": 1.1293,
|
| 225 |
+
"eval_samples_per_second": 187.73,
|
| 226 |
+
"eval_steps_per_second": 23.909,
|
| 227 |
+
"step": 880
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 11.25,
|
| 231 |
+
"grad_norm": 9.851762771606445,
|
| 232 |
+
"learning_rate": 1.8730348307472828e-05,
|
| 233 |
+
"loss": 1.043,
|
| 234 |
+
"step": 900
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 11.875,
|
| 238 |
+
"grad_norm": 8.781569480895996,
|
| 239 |
+
"learning_rate": 1.7517832824518216e-05,
|
| 240 |
+
"loss": 1.1157,
|
| 241 |
+
"step": 950
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 12.0,
|
| 245 |
+
"eval_accuracy": 0.44339622641509435,
|
| 246 |
+
"eval_loss": 1.2199500799179077,
|
| 247 |
+
"eval_runtime": 1.1331,
|
| 248 |
+
"eval_samples_per_second": 187.094,
|
| 249 |
+
"eval_steps_per_second": 23.828,
|
| 250 |
+
"step": 960
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 12.5,
|
| 254 |
+
"grad_norm": 10.422264099121094,
|
| 255 |
+
"learning_rate": 1.628811799032326e-05,
|
| 256 |
+
"loss": 1.0807,
|
| 257 |
+
"step": 1000
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 13.0,
|
| 261 |
+
"eval_accuracy": 0.4669811320754717,
|
| 262 |
+
"eval_loss": 1.2117412090301514,
|
| 263 |
+
"eval_runtime": 1.1321,
|
| 264 |
+
"eval_samples_per_second": 187.259,
|
| 265 |
+
"eval_steps_per_second": 23.849,
|
| 266 |
+
"step": 1040
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 13.125,
|
| 270 |
+
"grad_norm": 8.4778470993042,
|
| 271 |
+
"learning_rate": 1.5049604004119928e-05,
|
| 272 |
+
"loss": 1.0891,
|
| 273 |
+
"step": 1050
|
| 274 |
+
},
|
| 275 |
+
{
|
| 276 |
+
"epoch": 13.75,
|
| 277 |
+
"grad_norm": 11.677724838256836,
|
| 278 |
+
"learning_rate": 1.381075117227066e-05,
|
| 279 |
+
"loss": 1.0629,
|
| 280 |
+
"step": 1100
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"epoch": 14.0,
|
| 284 |
+
"eval_accuracy": 0.4811320754716981,
|
| 285 |
+
"eval_loss": 1.2296439409255981,
|
| 286 |
+
"eval_runtime": 1.1234,
|
| 287 |
+
"eval_samples_per_second": 188.721,
|
| 288 |
+
"eval_steps_per_second": 24.035,
|
| 289 |
+
"step": 1120
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 14.375,
|
| 293 |
+
"grad_norm": 11.892010688781738,
|
| 294 |
+
"learning_rate": 1.2580022115797237e-05,
|
| 295 |
+
"loss": 1.0673,
|
| 296 |
+
"step": 1150
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 15.0,
|
| 300 |
+
"grad_norm": 16.539581298828125,
|
| 301 |
+
"learning_rate": 1.1365823962098208e-05,
|
| 302 |
+
"loss": 1.0323,
|
| 303 |
+
"step": 1200
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 15.0,
|
| 307 |
+
"eval_accuracy": 0.49056603773584906,
|
| 308 |
+
"eval_loss": 1.1886956691741943,
|
| 309 |
+
"eval_runtime": 1.1303,
|
| 310 |
+
"eval_samples_per_second": 187.563,
|
| 311 |
+
"eval_steps_per_second": 23.888,
|
| 312 |
+
"step": 1200
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 15.625,
|
| 316 |
+
"grad_norm": 6.841868877410889,
|
| 317 |
+
"learning_rate": 1.0176450915744072e-05,
|
| 318 |
+
"loss": 1.0128,
|
| 319 |
+
"step": 1250
|
| 320 |
+
},
|
| 321 |
+
{
|
| 322 |
+
"epoch": 16.0,
|
| 323 |
+
"eval_accuracy": 0.49528301886792453,
|
| 324 |
+
"eval_loss": 1.2074586153030396,
|
| 325 |
+
"eval_runtime": 1.1321,
|
| 326 |
+
"eval_samples_per_second": 187.267,
|
| 327 |
+
"eval_steps_per_second": 23.85,
|
| 328 |
+
"step": 1280
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"epoch": 16.25,
|
| 332 |
+
"grad_norm": 9.490548133850098,
|
| 333 |
+
"learning_rate": 9.020027600649825e-06,
|
| 334 |
+
"loss": 0.9715,
|
| 335 |
+
"step": 1300
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"epoch": 16.875,
|
| 339 |
+
"grad_norm": 9.569696426391602,
|
| 340 |
+
"learning_rate": 7.904453560655389e-06,
|
| 341 |
+
"loss": 1.0266,
|
| 342 |
+
"step": 1350
|
| 343 |
+
},
|
| 344 |
+
{
|
| 345 |
+
"epoch": 17.0,
|
| 346 |
+
"eval_accuracy": 0.5,
|
| 347 |
+
"eval_loss": 1.2082122564315796,
|
| 348 |
+
"eval_runtime": 1.1449,
|
| 349 |
+
"eval_samples_per_second": 185.172,
|
| 350 |
+
"eval_steps_per_second": 23.583,
|
| 351 |
+
"step": 1360
|
| 352 |
+
},
|
| 353 |
+
{
|
| 354 |
+
"epoch": 17.5,
|
| 355 |
+
"grad_norm": 9.353178977966309,
|
| 356 |
+
"learning_rate": 6.837349297631114e-06,
|
| 357 |
+
"loss": 1.004,
|
| 358 |
+
"step": 1400
|
| 359 |
+
},
|
| 360 |
+
{
|
| 361 |
+
"epoch": 18.0,
|
| 362 |
+
"eval_accuracy": 0.5094339622641509,
|
| 363 |
+
"eval_loss": 1.2154083251953125,
|
| 364 |
+
"eval_runtime": 1.1238,
|
| 365 |
+
"eval_samples_per_second": 188.649,
|
| 366 |
+
"eval_steps_per_second": 24.026,
|
| 367 |
+
"step": 1440
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 18.125,
|
| 371 |
+
"grad_norm": 9.470197677612305,
|
| 372 |
+
"learning_rate": 5.826004215722937e-06,
|
| 373 |
+
"loss": 1.002,
|
| 374 |
+
"step": 1450
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 18.75,
|
| 378 |
+
"grad_norm": 6.417470932006836,
|
| 379 |
+
"learning_rate": 4.87732682733072e-06,
|
| 380 |
+
"loss": 0.9543,
|
| 381 |
+
"step": 1500
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 19.0,
|
| 385 |
+
"eval_accuracy": 0.5047169811320755,
|
| 386 |
+
"eval_loss": 1.2047548294067383,
|
| 387 |
+
"eval_runtime": 1.1355,
|
| 388 |
+
"eval_samples_per_second": 186.702,
|
| 389 |
+
"eval_steps_per_second": 23.778,
|
| 390 |
+
"step": 1520
|
| 391 |
+
},
|
| 392 |
+
{
|
| 393 |
+
"epoch": 19.375,
|
| 394 |
+
"grad_norm": 8.318925857543945,
|
| 395 |
+
"learning_rate": 3.997797560963404e-06,
|
| 396 |
+
"loss": 1.0218,
|
| 397 |
+
"step": 1550
|
| 398 |
+
},
|
| 399 |
+
{
|
| 400 |
+
"epoch": 20.0,
|
| 401 |
+
"grad_norm": 8.360044479370117,
|
| 402 |
+
"learning_rate": 3.193424493341213e-06,
|
| 403 |
+
"loss": 0.9439,
|
| 404 |
+
"step": 1600
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 20.0,
|
| 408 |
+
"eval_accuracy": 0.49056603773584906,
|
| 409 |
+
"eval_loss": 1.2217515707015991,
|
| 410 |
+
"eval_runtime": 1.1305,
|
| 411 |
+
"eval_samples_per_second": 187.525,
|
| 412 |
+
"eval_steps_per_second": 23.883,
|
| 413 |
+
"step": 1600
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 20.625,
|
| 417 |
+
"grad_norm": 12.746959686279297,
|
| 418 |
+
"learning_rate": 2.4697023081395654e-06,
|
| 419 |
+
"loss": 0.9891,
|
| 420 |
+
"step": 1650
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"epoch": 21.0,
|
| 424 |
+
"eval_accuracy": 0.49056603773584906,
|
| 425 |
+
"eval_loss": 1.2135605812072754,
|
| 426 |
+
"eval_runtime": 1.1547,
|
| 427 |
+
"eval_samples_per_second": 183.601,
|
| 428 |
+
"eval_steps_per_second": 23.383,
|
| 429 |
+
"step": 1680
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 21.25,
|
| 433 |
+
"grad_norm": 7.136344909667969,
|
| 434 |
+
"learning_rate": 1.8315747617280382e-06,
|
| 435 |
+
"loss": 0.9586,
|
| 436 |
+
"step": 1700
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 21.875,
|
| 440 |
+
"grad_norm": 8.792285919189453,
|
| 441 |
+
"learning_rate": 1.283400912301434e-06,
|
| 442 |
+
"loss": 0.9801,
|
| 443 |
+
"step": 1750
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 22.0,
|
| 447 |
+
"eval_accuracy": 0.4858490566037736,
|
| 448 |
+
"eval_loss": 1.216606855392456,
|
| 449 |
+
"eval_runtime": 1.133,
|
| 450 |
+
"eval_samples_per_second": 187.117,
|
| 451 |
+
"eval_steps_per_second": 23.831,
|
| 452 |
+
"step": 1760
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 22.5,
|
| 456 |
+
"grad_norm": 8.172861099243164,
|
| 457 |
+
"learning_rate": 8.289253430923128e-07,
|
| 458 |
+
"loss": 0.9632,
|
| 459 |
+
"step": 1800
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 23.0,
|
| 463 |
+
"eval_accuracy": 0.49056603773584906,
|
| 464 |
+
"eval_loss": 1.2148730754852295,
|
| 465 |
+
"eval_runtime": 1.1383,
|
| 466 |
+
"eval_samples_per_second": 186.244,
|
| 467 |
+
"eval_steps_per_second": 23.72,
|
| 468 |
+
"step": 1840
|
| 469 |
+
},
|
| 470 |
+
{
|
| 471 |
+
"epoch": 23.125,
|
| 472 |
+
"grad_norm": 8.669483184814453,
|
| 473 |
+
"learning_rate": 4.7125258307053385e-07,
|
| 474 |
+
"loss": 0.9956,
|
| 475 |
+
"step": 1850
|
| 476 |
+
},
|
| 477 |
+
{
|
| 478 |
+
"epoch": 23.75,
|
| 479 |
+
"grad_norm": 8.101666450500488,
|
| 480 |
+
"learning_rate": 2.1282589986245494e-07,
|
| 481 |
+
"loss": 0.9584,
|
| 482 |
+
"step": 1900
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 24.0,
|
| 486 |
+
"eval_accuracy": 0.49056603773584906,
|
| 487 |
+
"eval_loss": 1.213520884513855,
|
| 488 |
+
"eval_runtime": 1.1315,
|
| 489 |
+
"eval_samples_per_second": 187.37,
|
| 490 |
+
"eval_steps_per_second": 23.863,
|
| 491 |
+
"step": 1920
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 24.375,
|
| 495 |
+
"grad_norm": 25.165281295776367,
|
| 496 |
+
"learning_rate": 5.5410609755597306e-08,
|
| 497 |
+
"loss": 0.948,
|
| 498 |
+
"step": 1950
|
| 499 |
+
},
|
| 500 |
+
{
|
| 501 |
+
"epoch": 25.0,
|
| 502 |
+
"grad_norm": 10.21505069732666,
|
| 503 |
+
"learning_rate": 8.201879839297988e-11,
|
| 504 |
+
"loss": 0.9561,
|
| 505 |
+
"step": 2000
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 25.0,
|
| 509 |
+
"eval_accuracy": 0.49056603773584906,
|
| 510 |
+
"eval_loss": 1.2136294841766357,
|
| 511 |
+
"eval_runtime": 2.3217,
|
| 512 |
+
"eval_samples_per_second": 91.313,
|
| 513 |
+
"eval_steps_per_second": 11.63,
|
| 514 |
+
"step": 2000
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 25.0,
|
| 518 |
+
"step": 2000,
|
| 519 |
+
"total_flos": 1.2321447211229184e+18,
|
| 520 |
+
"train_loss": 1.1025613689422606,
|
| 521 |
+
"train_runtime": 425.612,
|
| 522 |
+
"train_samples_per_second": 37.358,
|
| 523 |
+
"train_steps_per_second": 4.699
|
| 524 |
+
}
|
| 525 |
+
],
|
| 526 |
+
"logging_steps": 50,
|
| 527 |
+
"max_steps": 2000,
|
| 528 |
+
"num_input_tokens_seen": 0,
|
| 529 |
+
"num_train_epochs": 25,
|
| 530 |
+
"save_steps": 500,
|
| 531 |
+
"stateful_callbacks": {
|
| 532 |
+
"TrainerControl": {
|
| 533 |
+
"args": {
|
| 534 |
+
"should_epoch_stop": false,
|
| 535 |
+
"should_evaluate": false,
|
| 536 |
+
"should_log": false,
|
| 537 |
+
"should_save": true,
|
| 538 |
+
"should_training_stop": true
|
| 539 |
+
},
|
| 540 |
+
"attributes": {}
|
| 541 |
+
}
|
| 542 |
+
},
|
| 543 |
+
"total_flos": 1.2321447211229184e+18,
|
| 544 |
+
"train_batch_size": 8,
|
| 545 |
+
"trial_name": null,
|
| 546 |
+
"trial_params": null
|
| 547 |
+
}
|