Commit
•
a3fad2e
1
Parent(s):
d6e10c7
End of training
Browse files
README.md
CHANGED
|
@@ -3,6 +3,9 @@ library_name: transformers
|
|
| 3 |
license: apache-2.0
|
| 4 |
base_model: c14kevincardenas/beit-large-patch16-384-limb
|
| 5 |
tags:
|
|
|
|
|
|
|
|
|
|
| 6 |
- generated_from_trainer
|
| 7 |
model-index:
|
| 8 |
- name: limbxy_hands
|
|
@@ -14,7 +17,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 14 |
|
| 15 |
# limbxy_hands
|
| 16 |
|
| 17 |
-
This model is a fine-tuned version of [c14kevincardenas/beit-large-patch16-384-limb](https://huggingface.co/c14kevincardenas/beit-large-patch16-384-limb) on
|
| 18 |
It achieves the following results on the evaluation set:
|
| 19 |
- Loss: 0.0048
|
| 20 |
- Rmse: 0.0691
|
|
|
|
| 3 |
license: apache-2.0
|
| 4 |
base_model: c14kevincardenas/beit-large-patch16-384-limb
|
| 5 |
tags:
|
| 6 |
+
- image-regression
|
| 7 |
+
- human-movement
|
| 8 |
+
- vision
|
| 9 |
- generated_from_trainer
|
| 10 |
model-index:
|
| 11 |
- name: limbxy_hands
|
|
|
|
| 17 |
|
| 18 |
# limbxy_hands
|
| 19 |
|
| 20 |
+
This model is a fine-tuned version of [c14kevincardenas/beit-large-patch16-384-limb](https://huggingface.co/c14kevincardenas/beit-large-patch16-384-limb) on the c14kevincardenas/beta_caller_284_limbxy_hands dataset.
|
| 21 |
It achieves the following results on the evaluation set:
|
| 22 |
- Loss: 0.0048
|
| 23 |
- Rmse: 0.0691
|
all_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 20.0,
|
| 3 |
+
"eval_loss": 0.004768455401062965,
|
| 4 |
+
"eval_rmse": 0.0690540000796318,
|
| 5 |
+
"eval_runtime": 5.0406,
|
| 6 |
+
"eval_samples_per_second": 112.684,
|
| 7 |
+
"eval_steps_per_second": 1.785,
|
| 8 |
+
"total_flos": 0.0,
|
| 9 |
+
"train_loss": 0.02494959205973382,
|
| 10 |
+
"train_runtime": 1869.8943,
|
| 11 |
+
"train_samples_per_second": 34.398,
|
| 12 |
+
"train_steps_per_second": 0.545
|
| 13 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 20.0,
|
| 3 |
+
"eval_loss": 0.004768455401062965,
|
| 4 |
+
"eval_rmse": 0.0690540000796318,
|
| 5 |
+
"eval_runtime": 5.0406,
|
| 6 |
+
"eval_samples_per_second": 112.684,
|
| 7 |
+
"eval_steps_per_second": 1.785
|
| 8 |
+
}
|
runs/Oct24_16-30-28_galactica.ad.cirange.net/events.out.tfevents.1729789685.galactica.ad.cirange.net.107524.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a7de17b4b8a0471a5c52933808c3018ea1a921dcb1d7053938747f4d6ab60f42
|
| 3 |
+
size 407
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 20.0,
|
| 3 |
+
"total_flos": 0.0,
|
| 4 |
+
"train_loss": 0.02494959205973382,
|
| 5 |
+
"train_runtime": 1869.8943,
|
| 6 |
+
"train_samples_per_second": 34.398,
|
| 7 |
+
"train_steps_per_second": 0.545
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,502 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.004768455401062965,
|
| 3 |
+
"best_model_checkpoint": "/mnt/ml_drive/kcardenas/limbxy_hands/checkpoint-1020",
|
| 4 |
+
"epoch": 20.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 1020,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.49019607843137253,
|
| 13 |
+
"grad_norm": 556763.0,
|
| 14 |
+
"learning_rate": 5e-06,
|
| 15 |
+
"loss": 0.2302,
|
| 16 |
+
"step": 25
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.9803921568627451,
|
| 20 |
+
"grad_norm": 1358784.625,
|
| 21 |
+
"learning_rate": 1e-05,
|
| 22 |
+
"loss": 0.1563,
|
| 23 |
+
"step": 50
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 1.0,
|
| 27 |
+
"eval_loss": 0.14388924837112427,
|
| 28 |
+
"eval_rmse": 0.37932735681533813,
|
| 29 |
+
"eval_runtime": 4.9465,
|
| 30 |
+
"eval_samples_per_second": 114.83,
|
| 31 |
+
"eval_steps_per_second": 1.819,
|
| 32 |
+
"step": 51
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 1.4705882352941178,
|
| 36 |
+
"grad_norm": 4929205.5,
|
| 37 |
+
"learning_rate": 1.5e-05,
|
| 38 |
+
"loss": 0.1396,
|
| 39 |
+
"step": 75
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 1.9607843137254903,
|
| 43 |
+
"grad_norm": 3301384.0,
|
| 44 |
+
"learning_rate": 2e-05,
|
| 45 |
+
"loss": 0.066,
|
| 46 |
+
"step": 100
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 2.0,
|
| 50 |
+
"eval_loss": 0.06285522878170013,
|
| 51 |
+
"eval_rmse": 0.2507094442844391,
|
| 52 |
+
"eval_runtime": 5.0622,
|
| 53 |
+
"eval_samples_per_second": 112.205,
|
| 54 |
+
"eval_steps_per_second": 1.778,
|
| 55 |
+
"step": 102
|
| 56 |
+
},
|
| 57 |
+
{
|
| 58 |
+
"epoch": 2.450980392156863,
|
| 59 |
+
"grad_norm": 1424321.875,
|
| 60 |
+
"learning_rate": 2.5e-05,
|
| 61 |
+
"loss": 0.0443,
|
| 62 |
+
"step": 125
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 2.9411764705882355,
|
| 66 |
+
"grad_norm": 677306.75,
|
| 67 |
+
"learning_rate": 3e-05,
|
| 68 |
+
"loss": 0.031,
|
| 69 |
+
"step": 150
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 3.0,
|
| 73 |
+
"eval_loss": 0.024024562910199165,
|
| 74 |
+
"eval_rmse": 0.15499858558177948,
|
| 75 |
+
"eval_runtime": 5.0919,
|
| 76 |
+
"eval_samples_per_second": 111.551,
|
| 77 |
+
"eval_steps_per_second": 1.768,
|
| 78 |
+
"step": 153
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"epoch": 3.431372549019608,
|
| 82 |
+
"grad_norm": 1847882.875,
|
| 83 |
+
"learning_rate": 3.5e-05,
|
| 84 |
+
"loss": 0.0227,
|
| 85 |
+
"step": 175
|
| 86 |
+
},
|
| 87 |
+
{
|
| 88 |
+
"epoch": 3.9215686274509802,
|
| 89 |
+
"grad_norm": 2230610.75,
|
| 90 |
+
"learning_rate": 4e-05,
|
| 91 |
+
"loss": 0.0189,
|
| 92 |
+
"step": 200
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 4.0,
|
| 96 |
+
"eval_loss": 0.022217219695448875,
|
| 97 |
+
"eval_rmse": 0.1490544229745865,
|
| 98 |
+
"eval_runtime": 5.1067,
|
| 99 |
+
"eval_samples_per_second": 111.227,
|
| 100 |
+
"eval_steps_per_second": 1.762,
|
| 101 |
+
"step": 204
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 4.411764705882353,
|
| 105 |
+
"grad_norm": 2538623.25,
|
| 106 |
+
"learning_rate": 4.5e-05,
|
| 107 |
+
"loss": 0.0294,
|
| 108 |
+
"step": 225
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 4.901960784313726,
|
| 112 |
+
"grad_norm": 1391400.75,
|
| 113 |
+
"learning_rate": 5e-05,
|
| 114 |
+
"loss": 0.0362,
|
| 115 |
+
"step": 250
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 5.0,
|
| 119 |
+
"eval_loss": 0.07003403455018997,
|
| 120 |
+
"eval_rmse": 0.2646394371986389,
|
| 121 |
+
"eval_runtime": 5.0039,
|
| 122 |
+
"eval_samples_per_second": 113.512,
|
| 123 |
+
"eval_steps_per_second": 1.799,
|
| 124 |
+
"step": 255
|
| 125 |
+
},
|
| 126 |
+
{
|
| 127 |
+
"epoch": 5.392156862745098,
|
| 128 |
+
"grad_norm": 1861511.0,
|
| 129 |
+
"learning_rate": 4.8376623376623384e-05,
|
| 130 |
+
"loss": 0.0243,
|
| 131 |
+
"step": 275
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 5.882352941176471,
|
| 135 |
+
"grad_norm": 591873.25,
|
| 136 |
+
"learning_rate": 4.675324675324675e-05,
|
| 137 |
+
"loss": 0.0106,
|
| 138 |
+
"step": 300
|
| 139 |
+
},
|
| 140 |
+
{
|
| 141 |
+
"epoch": 6.0,
|
| 142 |
+
"eval_loss": 0.011829150840640068,
|
| 143 |
+
"eval_rmse": 0.10876189917325974,
|
| 144 |
+
"eval_runtime": 5.0793,
|
| 145 |
+
"eval_samples_per_second": 111.825,
|
| 146 |
+
"eval_steps_per_second": 1.772,
|
| 147 |
+
"step": 306
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 6.372549019607844,
|
| 151 |
+
"grad_norm": 717852.0625,
|
| 152 |
+
"learning_rate": 4.5129870129870135e-05,
|
| 153 |
+
"loss": 0.011,
|
| 154 |
+
"step": 325
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"epoch": 6.862745098039216,
|
| 158 |
+
"grad_norm": 5022134.0,
|
| 159 |
+
"learning_rate": 4.3506493506493503e-05,
|
| 160 |
+
"loss": 0.0217,
|
| 161 |
+
"step": 350
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 7.0,
|
| 165 |
+
"eval_loss": 0.04193717613816261,
|
| 166 |
+
"eval_rmse": 0.20478568971157074,
|
| 167 |
+
"eval_runtime": 5.0411,
|
| 168 |
+
"eval_samples_per_second": 112.673,
|
| 169 |
+
"eval_steps_per_second": 1.785,
|
| 170 |
+
"step": 357
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 7.352941176470588,
|
| 174 |
+
"grad_norm": 4874459.0,
|
| 175 |
+
"learning_rate": 4.1883116883116886e-05,
|
| 176 |
+
"loss": 0.0272,
|
| 177 |
+
"step": 375
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 7.8431372549019605,
|
| 181 |
+
"grad_norm": 2519686.0,
|
| 182 |
+
"learning_rate": 4.025974025974026e-05,
|
| 183 |
+
"loss": 0.0167,
|
| 184 |
+
"step": 400
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 8.0,
|
| 188 |
+
"eval_loss": 0.01697073131799698,
|
| 189 |
+
"eval_rmse": 0.1302717626094818,
|
| 190 |
+
"eval_runtime": 5.0791,
|
| 191 |
+
"eval_samples_per_second": 111.83,
|
| 192 |
+
"eval_steps_per_second": 1.772,
|
| 193 |
+
"step": 408
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 8.333333333333334,
|
| 197 |
+
"grad_norm": 1922916.0,
|
| 198 |
+
"learning_rate": 3.8636363636363636e-05,
|
| 199 |
+
"loss": 0.0125,
|
| 200 |
+
"step": 425
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 8.823529411764707,
|
| 204 |
+
"grad_norm": 1355185.75,
|
| 205 |
+
"learning_rate": 3.701298701298702e-05,
|
| 206 |
+
"loss": 0.0078,
|
| 207 |
+
"step": 450
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 9.0,
|
| 211 |
+
"eval_loss": 0.011065785773098469,
|
| 212 |
+
"eval_rmse": 0.10519403964281082,
|
| 213 |
+
"eval_runtime": 5.0743,
|
| 214 |
+
"eval_samples_per_second": 111.938,
|
| 215 |
+
"eval_steps_per_second": 1.774,
|
| 216 |
+
"step": 459
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 9.313725490196079,
|
| 220 |
+
"grad_norm": 3503622.25,
|
| 221 |
+
"learning_rate": 3.5389610389610387e-05,
|
| 222 |
+
"loss": 0.0128,
|
| 223 |
+
"step": 475
|
| 224 |
+
},
|
| 225 |
+
{
|
| 226 |
+
"epoch": 9.803921568627452,
|
| 227 |
+
"grad_norm": 2042047.625,
|
| 228 |
+
"learning_rate": 3.376623376623377e-05,
|
| 229 |
+
"loss": 0.0102,
|
| 230 |
+
"step": 500
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 10.0,
|
| 234 |
+
"eval_loss": 0.01578633300960064,
|
| 235 |
+
"eval_rmse": 0.12564367055892944,
|
| 236 |
+
"eval_runtime": 5.0736,
|
| 237 |
+
"eval_samples_per_second": 111.953,
|
| 238 |
+
"eval_steps_per_second": 1.774,
|
| 239 |
+
"step": 510
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 10.294117647058824,
|
| 243 |
+
"grad_norm": 1890998.625,
|
| 244 |
+
"learning_rate": 3.2142857142857144e-05,
|
| 245 |
+
"loss": 0.0067,
|
| 246 |
+
"step": 525
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 10.784313725490197,
|
| 250 |
+
"grad_norm": 1245511.75,
|
| 251 |
+
"learning_rate": 3.051948051948052e-05,
|
| 252 |
+
"loss": 0.009,
|
| 253 |
+
"step": 550
|
| 254 |
+
},
|
| 255 |
+
{
|
| 256 |
+
"epoch": 11.0,
|
| 257 |
+
"eval_loss": 0.006434537936002016,
|
| 258 |
+
"eval_rmse": 0.08021557331085205,
|
| 259 |
+
"eval_runtime": 4.9055,
|
| 260 |
+
"eval_samples_per_second": 115.79,
|
| 261 |
+
"eval_steps_per_second": 1.835,
|
| 262 |
+
"step": 561
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 11.27450980392157,
|
| 266 |
+
"grad_norm": 1712539.625,
|
| 267 |
+
"learning_rate": 2.8896103896103898e-05,
|
| 268 |
+
"loss": 0.0044,
|
| 269 |
+
"step": 575
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 11.764705882352942,
|
| 273 |
+
"grad_norm": 3221122.25,
|
| 274 |
+
"learning_rate": 2.7272727272727273e-05,
|
| 275 |
+
"loss": 0.0067,
|
| 276 |
+
"step": 600
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 12.0,
|
| 280 |
+
"eval_loss": 0.00743119465187192,
|
| 281 |
+
"eval_rmse": 0.08620437234640121,
|
| 282 |
+
"eval_runtime": 5.1361,
|
| 283 |
+
"eval_samples_per_second": 110.589,
|
| 284 |
+
"eval_steps_per_second": 1.752,
|
| 285 |
+
"step": 612
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 12.254901960784313,
|
| 289 |
+
"grad_norm": 862932.25,
|
| 290 |
+
"learning_rate": 2.5649350649350652e-05,
|
| 291 |
+
"loss": 0.0058,
|
| 292 |
+
"step": 625
|
| 293 |
+
},
|
| 294 |
+
{
|
| 295 |
+
"epoch": 12.745098039215687,
|
| 296 |
+
"grad_norm": 1713839.5,
|
| 297 |
+
"learning_rate": 2.4025974025974027e-05,
|
| 298 |
+
"loss": 0.0047,
|
| 299 |
+
"step": 650
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 13.0,
|
| 303 |
+
"eval_loss": 0.0054638562723994255,
|
| 304 |
+
"eval_rmse": 0.07391790300607681,
|
| 305 |
+
"eval_runtime": 5.1487,
|
| 306 |
+
"eval_samples_per_second": 110.319,
|
| 307 |
+
"eval_steps_per_second": 1.748,
|
| 308 |
+
"step": 663
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 13.235294117647058,
|
| 312 |
+
"grad_norm": 1322453.625,
|
| 313 |
+
"learning_rate": 2.2402597402597402e-05,
|
| 314 |
+
"loss": 0.0043,
|
| 315 |
+
"step": 675
|
| 316 |
+
},
|
| 317 |
+
{
|
| 318 |
+
"epoch": 13.72549019607843,
|
| 319 |
+
"grad_norm": 941169.0,
|
| 320 |
+
"learning_rate": 2.077922077922078e-05,
|
| 321 |
+
"loss": 0.0036,
|
| 322 |
+
"step": 700
|
| 323 |
+
},
|
| 324 |
+
{
|
| 325 |
+
"epoch": 14.0,
|
| 326 |
+
"eval_loss": 0.006842494942247868,
|
| 327 |
+
"eval_rmse": 0.08271937817335129,
|
| 328 |
+
"eval_runtime": 4.825,
|
| 329 |
+
"eval_samples_per_second": 117.719,
|
| 330 |
+
"eval_steps_per_second": 1.865,
|
| 331 |
+
"step": 714
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 14.215686274509803,
|
| 335 |
+
"grad_norm": 2176495.5,
|
| 336 |
+
"learning_rate": 1.9155844155844156e-05,
|
| 337 |
+
"loss": 0.0036,
|
| 338 |
+
"step": 725
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 14.705882352941176,
|
| 342 |
+
"grad_norm": 411098.53125,
|
| 343 |
+
"learning_rate": 1.7532467532467535e-05,
|
| 344 |
+
"loss": 0.0072,
|
| 345 |
+
"step": 750
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 15.0,
|
| 349 |
+
"eval_loss": 0.0075486451387405396,
|
| 350 |
+
"eval_rmse": 0.08688294142484665,
|
| 351 |
+
"eval_runtime": 5.1462,
|
| 352 |
+
"eval_samples_per_second": 110.372,
|
| 353 |
+
"eval_steps_per_second": 1.749,
|
| 354 |
+
"step": 765
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 15.196078431372548,
|
| 358 |
+
"grad_norm": 2002094.375,
|
| 359 |
+
"learning_rate": 1.590909090909091e-05,
|
| 360 |
+
"loss": 0.0037,
|
| 361 |
+
"step": 775
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 15.686274509803921,
|
| 365 |
+
"grad_norm": 2112328.75,
|
| 366 |
+
"learning_rate": 1.4285714285714285e-05,
|
| 367 |
+
"loss": 0.006,
|
| 368 |
+
"step": 800
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 16.0,
|
| 372 |
+
"eval_loss": 0.005468303803354502,
|
| 373 |
+
"eval_rmse": 0.073947973549366,
|
| 374 |
+
"eval_runtime": 5.0395,
|
| 375 |
+
"eval_samples_per_second": 112.709,
|
| 376 |
+
"eval_steps_per_second": 1.786,
|
| 377 |
+
"step": 816
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 16.176470588235293,
|
| 381 |
+
"grad_norm": 1033486.875,
|
| 382 |
+
"learning_rate": 1.2662337662337662e-05,
|
| 383 |
+
"loss": 0.0036,
|
| 384 |
+
"step": 825
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 16.666666666666668,
|
| 388 |
+
"grad_norm": 321803.625,
|
| 389 |
+
"learning_rate": 1.103896103896104e-05,
|
| 390 |
+
"loss": 0.0022,
|
| 391 |
+
"step": 850
|
| 392 |
+
},
|
| 393 |
+
{
|
| 394 |
+
"epoch": 17.0,
|
| 395 |
+
"eval_loss": 0.005201099440455437,
|
| 396 |
+
"eval_rmse": 0.07211864739656448,
|
| 397 |
+
"eval_runtime": 5.018,
|
| 398 |
+
"eval_samples_per_second": 113.192,
|
| 399 |
+
"eval_steps_per_second": 1.794,
|
| 400 |
+
"step": 867
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 17.15686274509804,
|
| 404 |
+
"grad_norm": 10877.533203125,
|
| 405 |
+
"learning_rate": 9.415584415584416e-06,
|
| 406 |
+
"loss": 0.0016,
|
| 407 |
+
"step": 875
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"epoch": 17.647058823529413,
|
| 411 |
+
"grad_norm": 1244264.875,
|
| 412 |
+
"learning_rate": 7.792207792207792e-06,
|
| 413 |
+
"loss": 0.0051,
|
| 414 |
+
"step": 900
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 18.0,
|
| 418 |
+
"eval_loss": 0.005439546890556812,
|
| 419 |
+
"eval_rmse": 0.07375328242778778,
|
| 420 |
+
"eval_runtime": 5.1056,
|
| 421 |
+
"eval_samples_per_second": 111.251,
|
| 422 |
+
"eval_steps_per_second": 1.763,
|
| 423 |
+
"step": 918
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 18.137254901960784,
|
| 427 |
+
"grad_norm": 395613.0,
|
| 428 |
+
"learning_rate": 6.168831168831169e-06,
|
| 429 |
+
"loss": 0.0033,
|
| 430 |
+
"step": 925
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 18.627450980392158,
|
| 434 |
+
"grad_norm": 1329285.375,
|
| 435 |
+
"learning_rate": 4.5454545454545455e-06,
|
| 436 |
+
"loss": 0.0023,
|
| 437 |
+
"step": 950
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 19.0,
|
| 441 |
+
"eval_loss": 0.0050988770090043545,
|
| 442 |
+
"eval_rmse": 0.07140642404556274,
|
| 443 |
+
"eval_runtime": 5.0748,
|
| 444 |
+
"eval_samples_per_second": 111.926,
|
| 445 |
+
"eval_steps_per_second": 1.773,
|
| 446 |
+
"step": 969
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 19.11764705882353,
|
| 450 |
+
"grad_norm": 406520.3125,
|
| 451 |
+
"learning_rate": 2.922077922077922e-06,
|
| 452 |
+
"loss": 0.0019,
|
| 453 |
+
"step": 975
|
| 454 |
+
},
|
| 455 |
+
{
|
| 456 |
+
"epoch": 19.607843137254903,
|
| 457 |
+
"grad_norm": 275064.03125,
|
| 458 |
+
"learning_rate": 1.2987012987012988e-06,
|
| 459 |
+
"loss": 0.0016,
|
| 460 |
+
"step": 1000
|
| 461 |
+
},
|
| 462 |
+
{
|
| 463 |
+
"epoch": 20.0,
|
| 464 |
+
"eval_loss": 0.004768455401062965,
|
| 465 |
+
"eval_rmse": 0.0690540000796318,
|
| 466 |
+
"eval_runtime": 5.0305,
|
| 467 |
+
"eval_samples_per_second": 112.912,
|
| 468 |
+
"eval_steps_per_second": 1.789,
|
| 469 |
+
"step": 1020
|
| 470 |
+
},
|
| 471 |
+
{
|
| 472 |
+
"epoch": 20.0,
|
| 473 |
+
"step": 1020,
|
| 474 |
+
"total_flos": 0.0,
|
| 475 |
+
"train_loss": 0.02494959205973382,
|
| 476 |
+
"train_runtime": 1869.8943,
|
| 477 |
+
"train_samples_per_second": 34.398,
|
| 478 |
+
"train_steps_per_second": 0.545
|
| 479 |
+
}
|
| 480 |
+
],
|
| 481 |
+
"logging_steps": 25,
|
| 482 |
+
"max_steps": 1020,
|
| 483 |
+
"num_input_tokens_seen": 0,
|
| 484 |
+
"num_train_epochs": 20,
|
| 485 |
+
"save_steps": 500,
|
| 486 |
+
"stateful_callbacks": {
|
| 487 |
+
"TrainerControl": {
|
| 488 |
+
"args": {
|
| 489 |
+
"should_epoch_stop": false,
|
| 490 |
+
"should_evaluate": false,
|
| 491 |
+
"should_log": false,
|
| 492 |
+
"should_save": true,
|
| 493 |
+
"should_training_stop": true
|
| 494 |
+
},
|
| 495 |
+
"attributes": {}
|
| 496 |
+
}
|
| 497 |
+
},
|
| 498 |
+
"total_flos": 0.0,
|
| 499 |
+
"train_batch_size": 64,
|
| 500 |
+
"trial_name": null,
|
| 501 |
+
"trial_params": null
|
| 502 |
+
}
|