Commit
•
d7d58c1
1
Parent(s):
43a197a
End of training
Browse files- README.md +104 -54
- model.safetensors +1 -1
- trainer_state.json +761 -61
README.md
CHANGED
|
@@ -15,7 +15,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
This model is a fine-tuned version of [microsoft/conditional-detr-resnet-50](https://huggingface.co/microsoft/conditional-detr-resnet-50) on an unknown dataset.
|
| 17 |
It achieves the following results on the evaluation set:
|
| 18 |
-
- Loss: 2.
|
| 19 |
|
| 20 |
## Model description
|
| 21 |
|
|
@@ -40,63 +40,113 @@ The following hyperparameters were used during training:
|
|
| 40 |
- seed: 42
|
| 41 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 42 |
- lr_scheduler_type: linear
|
| 43 |
-
- num_epochs:
|
| 44 |
- mixed_precision_training: Native AMP
|
| 45 |
|
| 46 |
### Training results
|
| 47 |
|
| 48 |
-
| Training Loss | Epoch | Step
|
| 49 |
-
|
| 50 |
-
| 5.4149 | 1.0 | 115
|
| 51 |
-
| 3.9453 | 2.0 | 230
|
| 52 |
-
| 3.7269 | 3.0 | 345
|
| 53 |
-
| 3.5898 | 4.0 | 460
|
| 54 |
-
| 3.486 | 5.0 | 575
|
| 55 |
-
| 3.4073 | 6.0 | 690
|
| 56 |
-
| 3.4181 | 7.0 | 805
|
| 57 |
-
| 3.3603 | 8.0 | 920
|
| 58 |
-
| 3.2777 | 9.0 | 1035
|
| 59 |
-
| 3.2851 | 10.0 | 1150
|
| 60 |
-
| 3.266 | 11.0 | 1265
|
| 61 |
-
| 3.2803 | 12.0 | 1380
|
| 62 |
-
| 3.1733 | 13.0 | 1495
|
| 63 |
-
| 3.2094 | 14.0 | 1610
|
| 64 |
-
| 3.1718 | 15.0 | 1725
|
| 65 |
-
| 3.1303 | 16.0 | 1840
|
| 66 |
-
| 3.0897 | 17.0 | 1955
|
| 67 |
-
| 3.0233 | 18.0 | 2070
|
| 68 |
-
| 3.0156 | 19.0 | 2185
|
| 69 |
-
| 3.0277 | 20.0 | 2300
|
| 70 |
-
| 3.0847 | 21.0 | 2415
|
| 71 |
-
| 3.0333 | 22.0 | 2530
|
| 72 |
-
| 3.0126 | 23.0 | 2645
|
| 73 |
-
| 2.9968 | 24.0 | 2760
|
| 74 |
-
| 3.0295 | 25.0 | 2875
|
| 75 |
-
| 3.0294 | 26.0 | 2990
|
| 76 |
-
| 3.0395 | 27.0 | 3105
|
| 77 |
-
| 3.0445 | 28.0 | 3220
|
| 78 |
-
| 2.9761 | 29.0 | 3335
|
| 79 |
-
| 3.0075 | 30.0 | 3450
|
| 80 |
-
| 3.0198 | 31.0 | 3565
|
| 81 |
-
| 2.9782 | 32.0 | 3680
|
| 82 |
-
| 2.9773 | 33.0 | 3795
|
| 83 |
-
| 2.9528 | 34.0 | 3910
|
| 84 |
-
| 2.9228 | 35.0 | 4025
|
| 85 |
-
| 2.9221 | 36.0 | 4140
|
| 86 |
-
| 2.8933 | 37.0 | 4255
|
| 87 |
-
| 2.8925 | 38.0 | 4370
|
| 88 |
-
| 2.9069 | 39.0 | 4485
|
| 89 |
-
| 2.9246 | 40.0 | 4600
|
| 90 |
-
| 2.9089 | 41.0 | 4715
|
| 91 |
-
| 2.8573 | 42.0 | 4830
|
| 92 |
-
| 2.
|
| 93 |
-
| 2.
|
| 94 |
-
| 2.
|
| 95 |
-
| 2.
|
| 96 |
-
| 2.
|
| 97 |
-
| 2.
|
| 98 |
-
| 2.
|
| 99 |
-
| 2.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 100 |
|
| 101 |
|
| 102 |
### Framework versions
|
|
|
|
| 15 |
|
| 16 |
This model is a fine-tuned version of [microsoft/conditional-detr-resnet-50](https://huggingface.co/microsoft/conditional-detr-resnet-50) on an unknown dataset.
|
| 17 |
It achieves the following results on the evaluation set:
|
| 18 |
+
- Loss: 2.7389
|
| 19 |
|
| 20 |
## Model description
|
| 21 |
|
|
|
|
| 40 |
- seed: 42
|
| 41 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 42 |
- lr_scheduler_type: linear
|
| 43 |
+
- num_epochs: 100
|
| 44 |
- mixed_precision_training: Native AMP
|
| 45 |
|
| 46 |
### Training results
|
| 47 |
|
| 48 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 49 |
+
|:-------------:|:-----:|:-----:|:---------------:|
|
| 50 |
+
| 5.4149 | 1.0 | 115 | 4.3974 |
|
| 51 |
+
| 3.9453 | 2.0 | 230 | 3.6520 |
|
| 52 |
+
| 3.7269 | 3.0 | 345 | 3.7602 |
|
| 53 |
+
| 3.5898 | 4.0 | 460 | 3.5671 |
|
| 54 |
+
| 3.486 | 5.0 | 575 | 3.4912 |
|
| 55 |
+
| 3.4073 | 6.0 | 690 | 3.4095 |
|
| 56 |
+
| 3.4181 | 7.0 | 805 | 3.3183 |
|
| 57 |
+
| 3.3603 | 8.0 | 920 | 3.1111 |
|
| 58 |
+
| 3.2777 | 9.0 | 1035 | 3.1992 |
|
| 59 |
+
| 3.2851 | 10.0 | 1150 | 3.3997 |
|
| 60 |
+
| 3.266 | 11.0 | 1265 | 3.2861 |
|
| 61 |
+
| 3.2803 | 12.0 | 1380 | 3.1813 |
|
| 62 |
+
| 3.1733 | 13.0 | 1495 | 2.9838 |
|
| 63 |
+
| 3.2094 | 14.0 | 1610 | 3.1175 |
|
| 64 |
+
| 3.1718 | 15.0 | 1725 | 3.0064 |
|
| 65 |
+
| 3.1303 | 16.0 | 1840 | 3.0869 |
|
| 66 |
+
| 3.0897 | 17.0 | 1955 | 3.0306 |
|
| 67 |
+
| 3.0233 | 18.0 | 2070 | 2.9479 |
|
| 68 |
+
| 3.0156 | 19.0 | 2185 | 2.9145 |
|
| 69 |
+
| 3.0277 | 20.0 | 2300 | 2.8919 |
|
| 70 |
+
| 3.0847 | 21.0 | 2415 | 2.9321 |
|
| 71 |
+
| 3.0333 | 22.0 | 2530 | 2.9128 |
|
| 72 |
+
| 3.0126 | 23.0 | 2645 | 2.8627 |
|
| 73 |
+
| 2.9968 | 24.0 | 2760 | 3.0186 |
|
| 74 |
+
| 3.0295 | 25.0 | 2875 | 3.0148 |
|
| 75 |
+
| 3.0294 | 26.0 | 2990 | 3.0341 |
|
| 76 |
+
| 3.0395 | 27.0 | 3105 | 2.9997 |
|
| 77 |
+
| 3.0445 | 28.0 | 3220 | 3.0575 |
|
| 78 |
+
| 2.9761 | 29.0 | 3335 | 2.9707 |
|
| 79 |
+
| 3.0075 | 30.0 | 3450 | 2.9392 |
|
| 80 |
+
| 3.0198 | 31.0 | 3565 | 2.9122 |
|
| 81 |
+
| 2.9782 | 32.0 | 3680 | 2.9471 |
|
| 82 |
+
| 2.9773 | 33.0 | 3795 | 3.0306 |
|
| 83 |
+
| 2.9528 | 34.0 | 3910 | 2.8513 |
|
| 84 |
+
| 2.9228 | 35.0 | 4025 | 2.8997 |
|
| 85 |
+
| 2.9221 | 36.0 | 4140 | 2.8646 |
|
| 86 |
+
| 2.8933 | 37.0 | 4255 | 2.8871 |
|
| 87 |
+
| 2.8925 | 38.0 | 4370 | 2.9407 |
|
| 88 |
+
| 2.9069 | 39.0 | 4485 | 2.9625 |
|
| 89 |
+
| 2.9246 | 40.0 | 4600 | 2.9946 |
|
| 90 |
+
| 2.9089 | 41.0 | 4715 | 2.8936 |
|
| 91 |
+
| 2.8573 | 42.0 | 4830 | 2.8272 |
|
| 92 |
+
| 2.8768 | 43.0 | 4945 | 2.9868 |
|
| 93 |
+
| 2.9666 | 44.0 | 5060 | 2.9200 |
|
| 94 |
+
| 2.958 | 45.0 | 5175 | 2.8755 |
|
| 95 |
+
| 2.8923 | 46.0 | 5290 | 2.8518 |
|
| 96 |
+
| 2.9204 | 47.0 | 5405 | 2.9000 |
|
| 97 |
+
| 2.9644 | 48.0 | 5520 | 2.8969 |
|
| 98 |
+
| 2.9011 | 49.0 | 5635 | 2.7918 |
|
| 99 |
+
| 2.9329 | 50.0 | 5750 | 2.9139 |
|
| 100 |
+
| 2.9031 | 51.0 | 5865 | 2.7796 |
|
| 101 |
+
| 2.9029 | 52.0 | 5980 | 2.8025 |
|
| 102 |
+
| 2.9555 | 53.0 | 6095 | 2.9121 |
|
| 103 |
+
| 2.9366 | 54.0 | 6210 | 2.9035 |
|
| 104 |
+
| 2.8871 | 55.0 | 6325 | 2.8759 |
|
| 105 |
+
| 2.863 | 56.0 | 6440 | 2.8540 |
|
| 106 |
+
| 2.8897 | 57.0 | 6555 | 2.8401 |
|
| 107 |
+
| 2.828 | 58.0 | 6670 | 2.8590 |
|
| 108 |
+
| 2.8221 | 59.0 | 6785 | 2.9255 |
|
| 109 |
+
| 2.835 | 60.0 | 6900 | 2.9809 |
|
| 110 |
+
| 2.886 | 61.0 | 7015 | 2.9907 |
|
| 111 |
+
| 2.8227 | 62.0 | 7130 | 2.8283 |
|
| 112 |
+
| 2.7864 | 63.0 | 7245 | 2.8258 |
|
| 113 |
+
| 2.8179 | 64.0 | 7360 | 2.9504 |
|
| 114 |
+
| 2.7944 | 65.0 | 7475 | 2.8042 |
|
| 115 |
+
| 2.7986 | 66.0 | 7590 | 2.8307 |
|
| 116 |
+
| 2.7567 | 67.0 | 7705 | 2.8060 |
|
| 117 |
+
| 2.7552 | 68.0 | 7820 | 2.7994 |
|
| 118 |
+
| 2.7933 | 69.0 | 7935 | 2.8493 |
|
| 119 |
+
| 2.7393 | 70.0 | 8050 | 2.8409 |
|
| 120 |
+
| 2.7357 | 71.0 | 8165 | 2.8086 |
|
| 121 |
+
| 2.7264 | 72.0 | 8280 | 2.7773 |
|
| 122 |
+
| 2.7614 | 73.0 | 8395 | 2.8937 |
|
| 123 |
+
| 2.7279 | 74.0 | 8510 | 2.8887 |
|
| 124 |
+
| 2.745 | 75.0 | 8625 | 2.8274 |
|
| 125 |
+
| 2.7225 | 76.0 | 8740 | 2.7971 |
|
| 126 |
+
| 2.7094 | 77.0 | 8855 | 2.8685 |
|
| 127 |
+
| 2.7306 | 78.0 | 8970 | 2.8482 |
|
| 128 |
+
| 2.6844 | 79.0 | 9085 | 2.7372 |
|
| 129 |
+
| 2.6949 | 80.0 | 9200 | 2.8149 |
|
| 130 |
+
| 2.7342 | 81.0 | 9315 | 2.7647 |
|
| 131 |
+
| 2.6813 | 82.0 | 9430 | 2.7666 |
|
| 132 |
+
| 2.7161 | 83.0 | 9545 | 2.8437 |
|
| 133 |
+
| 2.6953 | 84.0 | 9660 | 2.7895 |
|
| 134 |
+
| 2.6714 | 85.0 | 9775 | 2.7683 |
|
| 135 |
+
| 2.6611 | 86.0 | 9890 | 2.7004 |
|
| 136 |
+
| 2.6714 | 87.0 | 10005 | 2.7183 |
|
| 137 |
+
| 2.6655 | 88.0 | 10120 | 2.7043 |
|
| 138 |
+
| 2.6509 | 89.0 | 10235 | 2.7705 |
|
| 139 |
+
| 2.6266 | 90.0 | 10350 | 2.7152 |
|
| 140 |
+
| 2.6677 | 91.0 | 10465 | 2.7295 |
|
| 141 |
+
| 2.6438 | 92.0 | 10580 | 2.7018 |
|
| 142 |
+
| 2.6267 | 93.0 | 10695 | 2.7063 |
|
| 143 |
+
| 2.6286 | 94.0 | 10810 | 2.7798 |
|
| 144 |
+
| 2.6043 | 95.0 | 10925 | 2.7712 |
|
| 145 |
+
| 2.6188 | 96.0 | 11040 | 2.7614 |
|
| 146 |
+
| 2.6028 | 97.0 | 11155 | 2.7405 |
|
| 147 |
+
| 2.621 | 98.0 | 11270 | 2.7415 |
|
| 148 |
+
| 2.61 | 99.0 | 11385 | 2.7415 |
|
| 149 |
+
| 2.6164 | 100.0 | 11500 | 2.7389 |
|
| 150 |
|
| 151 |
|
| 152 |
### Framework versions
|
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 173870884
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3c1866f4d599d39ba0fe76d52a1a7dcf979852ecd271dca7cd00adff8579d689
|
| 3 |
size 173870884
|
trainer_state.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 2.
|
| 3 |
-
"best_model_checkpoint": "cdetr-mist1-brain-gt-tumors-8ah-6l/checkpoint-
|
| 4 |
-
"epoch":
|
| 5 |
"eval_steps": 500,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
@@ -598,131 +598,831 @@
|
|
| 598 |
},
|
| 599 |
{
|
| 600 |
"epoch": 43.0,
|
| 601 |
-
"learning_rate":
|
| 602 |
-
"loss": 2.
|
| 603 |
"step": 4945
|
| 604 |
},
|
| 605 |
{
|
| 606 |
"epoch": 43.0,
|
| 607 |
-
"eval_loss": 2.
|
| 608 |
-
"eval_runtime": 5.
|
| 609 |
-
"eval_samples_per_second":
|
| 610 |
-
"eval_steps_per_second": 0.
|
| 611 |
"step": 4945
|
| 612 |
},
|
| 613 |
{
|
| 614 |
"epoch": 44.0,
|
| 615 |
-
"learning_rate":
|
| 616 |
-
"loss": 2.
|
| 617 |
"step": 5060
|
| 618 |
},
|
| 619 |
{
|
| 620 |
"epoch": 44.0,
|
| 621 |
-
"eval_loss": 2.
|
| 622 |
-
"eval_runtime": 5.
|
| 623 |
-
"eval_samples_per_second": 7.
|
| 624 |
-
"eval_steps_per_second": 0.
|
| 625 |
"step": 5060
|
| 626 |
},
|
| 627 |
{
|
| 628 |
"epoch": 45.0,
|
| 629 |
-
"learning_rate":
|
| 630 |
-
"loss": 2.
|
| 631 |
"step": 5175
|
| 632 |
},
|
| 633 |
{
|
| 634 |
"epoch": 45.0,
|
| 635 |
-
"eval_loss": 2.
|
| 636 |
-
"eval_runtime": 5.
|
| 637 |
-
"eval_samples_per_second": 7.
|
| 638 |
"eval_steps_per_second": 0.883,
|
| 639 |
"step": 5175
|
| 640 |
},
|
| 641 |
{
|
| 642 |
"epoch": 46.0,
|
| 643 |
-
"learning_rate":
|
| 644 |
-
"loss": 2.
|
| 645 |
"step": 5290
|
| 646 |
},
|
| 647 |
{
|
| 648 |
"epoch": 46.0,
|
| 649 |
-
"eval_loss": 2.
|
| 650 |
-
"eval_runtime": 5.
|
| 651 |
-
"eval_samples_per_second": 7.
|
| 652 |
-
"eval_steps_per_second": 0.
|
| 653 |
"step": 5290
|
| 654 |
},
|
| 655 |
{
|
| 656 |
"epoch": 47.0,
|
| 657 |
-
"learning_rate":
|
| 658 |
-
"loss": 2.
|
| 659 |
"step": 5405
|
| 660 |
},
|
| 661 |
{
|
| 662 |
"epoch": 47.0,
|
| 663 |
-
"eval_loss": 2.
|
| 664 |
-
"eval_runtime": 5.
|
| 665 |
-
"eval_samples_per_second": 7.
|
| 666 |
-
"eval_steps_per_second": 0.
|
| 667 |
"step": 5405
|
| 668 |
},
|
| 669 |
{
|
| 670 |
"epoch": 48.0,
|
| 671 |
-
"learning_rate":
|
| 672 |
-
"loss": 2.
|
| 673 |
"step": 5520
|
| 674 |
},
|
| 675 |
{
|
| 676 |
"epoch": 48.0,
|
| 677 |
-
"eval_loss": 2.
|
| 678 |
-
"eval_runtime": 5.
|
| 679 |
-
"eval_samples_per_second": 7.
|
| 680 |
-
"eval_steps_per_second": 0.
|
| 681 |
"step": 5520
|
| 682 |
},
|
| 683 |
{
|
| 684 |
"epoch": 49.0,
|
| 685 |
-
"learning_rate":
|
| 686 |
-
"loss": 2.
|
| 687 |
"step": 5635
|
| 688 |
},
|
| 689 |
{
|
| 690 |
"epoch": 49.0,
|
| 691 |
-
"eval_loss": 2.
|
| 692 |
-
"eval_runtime": 5.
|
| 693 |
-
"eval_samples_per_second":
|
| 694 |
-
"eval_steps_per_second": 0.
|
| 695 |
"step": 5635
|
| 696 |
},
|
| 697 |
{
|
| 698 |
"epoch": 50.0,
|
| 699 |
-
"learning_rate":
|
| 700 |
-
"loss": 2.
|
| 701 |
"step": 5750
|
| 702 |
},
|
| 703 |
{
|
| 704 |
"epoch": 50.0,
|
| 705 |
-
"eval_loss": 2.
|
| 706 |
-
"eval_runtime": 5.
|
| 707 |
-
"eval_samples_per_second": 7.
|
| 708 |
-
"eval_steps_per_second": 0.
|
| 709 |
"step": 5750
|
| 710 |
},
|
| 711 |
{
|
| 712 |
-
"epoch":
|
| 713 |
-
"
|
| 714 |
-
"
|
| 715 |
-
"
|
| 716 |
-
|
| 717 |
-
|
| 718 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 719 |
}
|
| 720 |
],
|
| 721 |
"logging_steps": 500,
|
| 722 |
-
"max_steps":
|
| 723 |
-
"num_train_epochs":
|
| 724 |
"save_steps": 500,
|
| 725 |
-
"total_flos":
|
| 726 |
"trial_name": null,
|
| 727 |
"trial_params": null
|
| 728 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 2.700421094894409,
|
| 3 |
+
"best_model_checkpoint": "cdetr-mist1-brain-gt-tumors-8ah-6l/checkpoint-9890",
|
| 4 |
+
"epoch": 100.0,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 11500,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
|
|
| 598 |
},
|
| 599 |
{
|
| 600 |
"epoch": 43.0,
|
| 601 |
+
"learning_rate": 5.726956521739131e-06,
|
| 602 |
+
"loss": 2.8768,
|
| 603 |
"step": 4945
|
| 604 |
},
|
| 605 |
{
|
| 606 |
"epoch": 43.0,
|
| 607 |
+
"eval_loss": 2.9868218898773193,
|
| 608 |
+
"eval_runtime": 5.7349,
|
| 609 |
+
"eval_samples_per_second": 6.975,
|
| 610 |
+
"eval_steps_per_second": 0.872,
|
| 611 |
"step": 4945
|
| 612 |
},
|
| 613 |
{
|
| 614 |
"epoch": 44.0,
|
| 615 |
+
"learning_rate": 5.627826086956523e-06,
|
| 616 |
+
"loss": 2.9666,
|
| 617 |
"step": 5060
|
| 618 |
},
|
| 619 |
{
|
| 620 |
"epoch": 44.0,
|
| 621 |
+
"eval_loss": 2.9200377464294434,
|
| 622 |
+
"eval_runtime": 5.6461,
|
| 623 |
+
"eval_samples_per_second": 7.084,
|
| 624 |
+
"eval_steps_per_second": 0.886,
|
| 625 |
"step": 5060
|
| 626 |
},
|
| 627 |
{
|
| 628 |
"epoch": 45.0,
|
| 629 |
+
"learning_rate": 5.527826086956523e-06,
|
| 630 |
+
"loss": 2.958,
|
| 631 |
"step": 5175
|
| 632 |
},
|
| 633 |
{
|
| 634 |
"epoch": 45.0,
|
| 635 |
+
"eval_loss": 2.875474214553833,
|
| 636 |
+
"eval_runtime": 5.664,
|
| 637 |
+
"eval_samples_per_second": 7.062,
|
| 638 |
"eval_steps_per_second": 0.883,
|
| 639 |
"step": 5175
|
| 640 |
},
|
| 641 |
{
|
| 642 |
"epoch": 46.0,
|
| 643 |
+
"learning_rate": 5.427826086956523e-06,
|
| 644 |
+
"loss": 2.8923,
|
| 645 |
"step": 5290
|
| 646 |
},
|
| 647 |
{
|
| 648 |
"epoch": 46.0,
|
| 649 |
+
"eval_loss": 2.851766586303711,
|
| 650 |
+
"eval_runtime": 5.6855,
|
| 651 |
+
"eval_samples_per_second": 7.035,
|
| 652 |
+
"eval_steps_per_second": 0.879,
|
| 653 |
"step": 5290
|
| 654 |
},
|
| 655 |
{
|
| 656 |
"epoch": 47.0,
|
| 657 |
+
"learning_rate": 5.327826086956522e-06,
|
| 658 |
+
"loss": 2.9204,
|
| 659 |
"step": 5405
|
| 660 |
},
|
| 661 |
{
|
| 662 |
"epoch": 47.0,
|
| 663 |
+
"eval_loss": 2.9000306129455566,
|
| 664 |
+
"eval_runtime": 5.6182,
|
| 665 |
+
"eval_samples_per_second": 7.12,
|
| 666 |
+
"eval_steps_per_second": 0.89,
|
| 667 |
"step": 5405
|
| 668 |
},
|
| 669 |
{
|
| 670 |
"epoch": 48.0,
|
| 671 |
+
"learning_rate": 5.227826086956522e-06,
|
| 672 |
+
"loss": 2.9644,
|
| 673 |
"step": 5520
|
| 674 |
},
|
| 675 |
{
|
| 676 |
"epoch": 48.0,
|
| 677 |
+
"eval_loss": 2.8968658447265625,
|
| 678 |
+
"eval_runtime": 5.677,
|
| 679 |
+
"eval_samples_per_second": 7.046,
|
| 680 |
+
"eval_steps_per_second": 0.881,
|
| 681 |
"step": 5520
|
| 682 |
},
|
| 683 |
{
|
| 684 |
"epoch": 49.0,
|
| 685 |
+
"learning_rate": 5.127826086956522e-06,
|
| 686 |
+
"loss": 2.9011,
|
| 687 |
"step": 5635
|
| 688 |
},
|
| 689 |
{
|
| 690 |
"epoch": 49.0,
|
| 691 |
+
"eval_loss": 2.7918035984039307,
|
| 692 |
+
"eval_runtime": 5.6945,
|
| 693 |
+
"eval_samples_per_second": 7.024,
|
| 694 |
+
"eval_steps_per_second": 0.878,
|
| 695 |
"step": 5635
|
| 696 |
},
|
| 697 |
{
|
| 698 |
"epoch": 50.0,
|
| 699 |
+
"learning_rate": 5.028695652173914e-06,
|
| 700 |
+
"loss": 2.9329,
|
| 701 |
"step": 5750
|
| 702 |
},
|
| 703 |
{
|
| 704 |
"epoch": 50.0,
|
| 705 |
+
"eval_loss": 2.9139397144317627,
|
| 706 |
+
"eval_runtime": 5.6814,
|
| 707 |
+
"eval_samples_per_second": 7.041,
|
| 708 |
+
"eval_steps_per_second": 0.88,
|
| 709 |
"step": 5750
|
| 710 |
},
|
| 711 |
{
|
| 712 |
+
"epoch": 51.0,
|
| 713 |
+
"learning_rate": 4.9286956521739135e-06,
|
| 714 |
+
"loss": 2.9031,
|
| 715 |
+
"step": 5865
|
| 716 |
+
},
|
| 717 |
+
{
|
| 718 |
+
"epoch": 51.0,
|
| 719 |
+
"eval_loss": 2.779574155807495,
|
| 720 |
+
"eval_runtime": 5.6982,
|
| 721 |
+
"eval_samples_per_second": 7.02,
|
| 722 |
+
"eval_steps_per_second": 0.877,
|
| 723 |
+
"step": 5865
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 52.0,
|
| 727 |
+
"learning_rate": 4.828695652173914e-06,
|
| 728 |
+
"loss": 2.9029,
|
| 729 |
+
"step": 5980
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 52.0,
|
| 733 |
+
"eval_loss": 2.802475690841675,
|
| 734 |
+
"eval_runtime": 5.6076,
|
| 735 |
+
"eval_samples_per_second": 7.133,
|
| 736 |
+
"eval_steps_per_second": 0.892,
|
| 737 |
+
"step": 5980
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 53.0,
|
| 741 |
+
"learning_rate": 4.728695652173914e-06,
|
| 742 |
+
"loss": 2.9555,
|
| 743 |
+
"step": 6095
|
| 744 |
+
},
|
| 745 |
+
{
|
| 746 |
+
"epoch": 53.0,
|
| 747 |
+
"eval_loss": 2.9120869636535645,
|
| 748 |
+
"eval_runtime": 5.7682,
|
| 749 |
+
"eval_samples_per_second": 6.935,
|
| 750 |
+
"eval_steps_per_second": 0.867,
|
| 751 |
+
"step": 6095
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 54.0,
|
| 755 |
+
"learning_rate": 4.628695652173914e-06,
|
| 756 |
+
"loss": 2.9366,
|
| 757 |
+
"step": 6210
|
| 758 |
+
},
|
| 759 |
+
{
|
| 760 |
+
"epoch": 54.0,
|
| 761 |
+
"eval_loss": 2.9034695625305176,
|
| 762 |
+
"eval_runtime": 5.7147,
|
| 763 |
+
"eval_samples_per_second": 7.0,
|
| 764 |
+
"eval_steps_per_second": 0.875,
|
| 765 |
+
"step": 6210
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 55.0,
|
| 769 |
+
"learning_rate": 4.528695652173913e-06,
|
| 770 |
+
"loss": 2.8871,
|
| 771 |
+
"step": 6325
|
| 772 |
+
},
|
| 773 |
+
{
|
| 774 |
+
"epoch": 55.0,
|
| 775 |
+
"eval_loss": 2.87589168548584,
|
| 776 |
+
"eval_runtime": 5.6683,
|
| 777 |
+
"eval_samples_per_second": 7.057,
|
| 778 |
+
"eval_steps_per_second": 0.882,
|
| 779 |
+
"step": 6325
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 56.0,
|
| 783 |
+
"learning_rate": 4.428695652173913e-06,
|
| 784 |
+
"loss": 2.863,
|
| 785 |
+
"step": 6440
|
| 786 |
+
},
|
| 787 |
+
{
|
| 788 |
+
"epoch": 56.0,
|
| 789 |
+
"eval_loss": 2.8540170192718506,
|
| 790 |
+
"eval_runtime": 5.7794,
|
| 791 |
+
"eval_samples_per_second": 6.921,
|
| 792 |
+
"eval_steps_per_second": 0.865,
|
| 793 |
+
"step": 6440
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 57.0,
|
| 797 |
+
"learning_rate": 4.328695652173913e-06,
|
| 798 |
+
"loss": 2.8897,
|
| 799 |
+
"step": 6555
|
| 800 |
+
},
|
| 801 |
+
{
|
| 802 |
+
"epoch": 57.0,
|
| 803 |
+
"eval_loss": 2.8401310443878174,
|
| 804 |
+
"eval_runtime": 5.6374,
|
| 805 |
+
"eval_samples_per_second": 7.095,
|
| 806 |
+
"eval_steps_per_second": 0.887,
|
| 807 |
+
"step": 6555
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 58.0,
|
| 811 |
+
"learning_rate": 4.228695652173913e-06,
|
| 812 |
+
"loss": 2.828,
|
| 813 |
+
"step": 6670
|
| 814 |
+
},
|
| 815 |
+
{
|
| 816 |
+
"epoch": 58.0,
|
| 817 |
+
"eval_loss": 2.8589885234832764,
|
| 818 |
+
"eval_runtime": 5.697,
|
| 819 |
+
"eval_samples_per_second": 7.021,
|
| 820 |
+
"eval_steps_per_second": 0.878,
|
| 821 |
+
"step": 6670
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 59.0,
|
| 825 |
+
"learning_rate": 4.1286956521739135e-06,
|
| 826 |
+
"loss": 2.8221,
|
| 827 |
+
"step": 6785
|
| 828 |
+
},
|
| 829 |
+
{
|
| 830 |
+
"epoch": 59.0,
|
| 831 |
+
"eval_loss": 2.92549467086792,
|
| 832 |
+
"eval_runtime": 5.7309,
|
| 833 |
+
"eval_samples_per_second": 6.98,
|
| 834 |
+
"eval_steps_per_second": 0.872,
|
| 835 |
+
"step": 6785
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 60.0,
|
| 839 |
+
"learning_rate": 4.028695652173914e-06,
|
| 840 |
+
"loss": 2.835,
|
| 841 |
+
"step": 6900
|
| 842 |
+
},
|
| 843 |
+
{
|
| 844 |
+
"epoch": 60.0,
|
| 845 |
+
"eval_loss": 2.9808831214904785,
|
| 846 |
+
"eval_runtime": 5.6748,
|
| 847 |
+
"eval_samples_per_second": 7.049,
|
| 848 |
+
"eval_steps_per_second": 0.881,
|
| 849 |
+
"step": 6900
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 61.0,
|
| 853 |
+
"learning_rate": 3.928695652173914e-06,
|
| 854 |
+
"loss": 2.886,
|
| 855 |
+
"step": 7015
|
| 856 |
+
},
|
| 857 |
+
{
|
| 858 |
+
"epoch": 61.0,
|
| 859 |
+
"eval_loss": 2.9906742572784424,
|
| 860 |
+
"eval_runtime": 5.7118,
|
| 861 |
+
"eval_samples_per_second": 7.003,
|
| 862 |
+
"eval_steps_per_second": 0.875,
|
| 863 |
+
"step": 7015
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 62.0,
|
| 867 |
+
"learning_rate": 3.828695652173913e-06,
|
| 868 |
+
"loss": 2.8227,
|
| 869 |
+
"step": 7130
|
| 870 |
+
},
|
| 871 |
+
{
|
| 872 |
+
"epoch": 62.0,
|
| 873 |
+
"eval_loss": 2.8283145427703857,
|
| 874 |
+
"eval_runtime": 5.7014,
|
| 875 |
+
"eval_samples_per_second": 7.016,
|
| 876 |
+
"eval_steps_per_second": 0.877,
|
| 877 |
+
"step": 7130
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 63.0,
|
| 881 |
+
"learning_rate": 3.728695652173913e-06,
|
| 882 |
+
"loss": 2.7864,
|
| 883 |
+
"step": 7245
|
| 884 |
+
},
|
| 885 |
+
{
|
| 886 |
+
"epoch": 63.0,
|
| 887 |
+
"eval_loss": 2.8258347511291504,
|
| 888 |
+
"eval_runtime": 5.6903,
|
| 889 |
+
"eval_samples_per_second": 7.029,
|
| 890 |
+
"eval_steps_per_second": 0.879,
|
| 891 |
+
"step": 7245
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 64.0,
|
| 895 |
+
"learning_rate": 3.6286956521739132e-06,
|
| 896 |
+
"loss": 2.8179,
|
| 897 |
+
"step": 7360
|
| 898 |
+
},
|
| 899 |
+
{
|
| 900 |
+
"epoch": 64.0,
|
| 901 |
+
"eval_loss": 2.9504449367523193,
|
| 902 |
+
"eval_runtime": 5.7303,
|
| 903 |
+
"eval_samples_per_second": 6.98,
|
| 904 |
+
"eval_steps_per_second": 0.873,
|
| 905 |
+
"step": 7360
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 65.0,
|
| 909 |
+
"learning_rate": 3.5286956521739133e-06,
|
| 910 |
+
"loss": 2.7944,
|
| 911 |
+
"step": 7475
|
| 912 |
+
},
|
| 913 |
+
{
|
| 914 |
+
"epoch": 65.0,
|
| 915 |
+
"eval_loss": 2.8042430877685547,
|
| 916 |
+
"eval_runtime": 5.7552,
|
| 917 |
+
"eval_samples_per_second": 6.95,
|
| 918 |
+
"eval_steps_per_second": 0.869,
|
| 919 |
+
"step": 7475
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 66.0,
|
| 923 |
+
"learning_rate": 3.4286956521739134e-06,
|
| 924 |
+
"loss": 2.7986,
|
| 925 |
+
"step": 7590
|
| 926 |
+
},
|
| 927 |
+
{
|
| 928 |
+
"epoch": 66.0,
|
| 929 |
+
"eval_loss": 2.8307268619537354,
|
| 930 |
+
"eval_runtime": 5.741,
|
| 931 |
+
"eval_samples_per_second": 6.967,
|
| 932 |
+
"eval_steps_per_second": 0.871,
|
| 933 |
+
"step": 7590
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 67.0,
|
| 937 |
+
"learning_rate": 3.328695652173913e-06,
|
| 938 |
+
"loss": 2.7567,
|
| 939 |
+
"step": 7705
|
| 940 |
+
},
|
| 941 |
+
{
|
| 942 |
+
"epoch": 67.0,
|
| 943 |
+
"eval_loss": 2.805975914001465,
|
| 944 |
+
"eval_runtime": 5.7979,
|
| 945 |
+
"eval_samples_per_second": 6.899,
|
| 946 |
+
"eval_steps_per_second": 0.862,
|
| 947 |
+
"step": 7705
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 68.0,
|
| 951 |
+
"learning_rate": 3.2286956521739132e-06,
|
| 952 |
+
"loss": 2.7552,
|
| 953 |
+
"step": 7820
|
| 954 |
+
},
|
| 955 |
+
{
|
| 956 |
+
"epoch": 68.0,
|
| 957 |
+
"eval_loss": 2.7994372844696045,
|
| 958 |
+
"eval_runtime": 5.7117,
|
| 959 |
+
"eval_samples_per_second": 7.003,
|
| 960 |
+
"eval_steps_per_second": 0.875,
|
| 961 |
+
"step": 7820
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 69.0,
|
| 965 |
+
"learning_rate": 3.1286956521739133e-06,
|
| 966 |
+
"loss": 2.7933,
|
| 967 |
+
"step": 7935
|
| 968 |
+
},
|
| 969 |
+
{
|
| 970 |
+
"epoch": 69.0,
|
| 971 |
+
"eval_loss": 2.849256992340088,
|
| 972 |
+
"eval_runtime": 5.7067,
|
| 973 |
+
"eval_samples_per_second": 7.009,
|
| 974 |
+
"eval_steps_per_second": 0.876,
|
| 975 |
+
"step": 7935
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 70.0,
|
| 979 |
+
"learning_rate": 3.028695652173913e-06,
|
| 980 |
+
"loss": 2.7393,
|
| 981 |
+
"step": 8050
|
| 982 |
+
},
|
| 983 |
+
{
|
| 984 |
+
"epoch": 70.0,
|
| 985 |
+
"eval_loss": 2.8409152030944824,
|
| 986 |
+
"eval_runtime": 5.6797,
|
| 987 |
+
"eval_samples_per_second": 7.043,
|
| 988 |
+
"eval_steps_per_second": 0.88,
|
| 989 |
+
"step": 8050
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 71.0,
|
| 993 |
+
"learning_rate": 2.9286956521739136e-06,
|
| 994 |
+
"loss": 2.7357,
|
| 995 |
+
"step": 8165
|
| 996 |
+
},
|
| 997 |
+
{
|
| 998 |
+
"epoch": 71.0,
|
| 999 |
+
"eval_loss": 2.8086206912994385,
|
| 1000 |
+
"eval_runtime": 5.6798,
|
| 1001 |
+
"eval_samples_per_second": 7.043,
|
| 1002 |
+
"eval_steps_per_second": 0.88,
|
| 1003 |
+
"step": 8165
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 72.0,
|
| 1007 |
+
"learning_rate": 2.8286956521739132e-06,
|
| 1008 |
+
"loss": 2.7264,
|
| 1009 |
+
"step": 8280
|
| 1010 |
+
},
|
| 1011 |
+
{
|
| 1012 |
+
"epoch": 72.0,
|
| 1013 |
+
"eval_loss": 2.7772560119628906,
|
| 1014 |
+
"eval_runtime": 5.7909,
|
| 1015 |
+
"eval_samples_per_second": 6.907,
|
| 1016 |
+
"eval_steps_per_second": 0.863,
|
| 1017 |
+
"step": 8280
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 73.0,
|
| 1021 |
+
"learning_rate": 2.7286956521739134e-06,
|
| 1022 |
+
"loss": 2.7614,
|
| 1023 |
+
"step": 8395
|
| 1024 |
+
},
|
| 1025 |
+
{
|
| 1026 |
+
"epoch": 73.0,
|
| 1027 |
+
"eval_loss": 2.8937366008758545,
|
| 1028 |
+
"eval_runtime": 5.7573,
|
| 1029 |
+
"eval_samples_per_second": 6.948,
|
| 1030 |
+
"eval_steps_per_second": 0.868,
|
| 1031 |
+
"step": 8395
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 74.0,
|
| 1035 |
+
"learning_rate": 2.6295652173913044e-06,
|
| 1036 |
+
"loss": 2.7279,
|
| 1037 |
+
"step": 8510
|
| 1038 |
+
},
|
| 1039 |
+
{
|
| 1040 |
+
"epoch": 74.0,
|
| 1041 |
+
"eval_loss": 2.8887228965759277,
|
| 1042 |
+
"eval_runtime": 5.7093,
|
| 1043 |
+
"eval_samples_per_second": 7.006,
|
| 1044 |
+
"eval_steps_per_second": 0.876,
|
| 1045 |
+
"step": 8510
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 75.0,
|
| 1049 |
+
"learning_rate": 2.5295652173913045e-06,
|
| 1050 |
+
"loss": 2.745,
|
| 1051 |
+
"step": 8625
|
| 1052 |
+
},
|
| 1053 |
+
{
|
| 1054 |
+
"epoch": 75.0,
|
| 1055 |
+
"eval_loss": 2.827376127243042,
|
| 1056 |
+
"eval_runtime": 5.6568,
|
| 1057 |
+
"eval_samples_per_second": 7.071,
|
| 1058 |
+
"eval_steps_per_second": 0.884,
|
| 1059 |
+
"step": 8625
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 76.0,
|
| 1063 |
+
"learning_rate": 2.4295652173913046e-06,
|
| 1064 |
+
"loss": 2.7225,
|
| 1065 |
+
"step": 8740
|
| 1066 |
+
},
|
| 1067 |
+
{
|
| 1068 |
+
"epoch": 76.0,
|
| 1069 |
+
"eval_loss": 2.7970547676086426,
|
| 1070 |
+
"eval_runtime": 5.7165,
|
| 1071 |
+
"eval_samples_per_second": 6.997,
|
| 1072 |
+
"eval_steps_per_second": 0.875,
|
| 1073 |
+
"step": 8740
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 77.0,
|
| 1077 |
+
"learning_rate": 2.3295652173913043e-06,
|
| 1078 |
+
"loss": 2.7094,
|
| 1079 |
+
"step": 8855
|
| 1080 |
+
},
|
| 1081 |
+
{
|
| 1082 |
+
"epoch": 77.0,
|
| 1083 |
+
"eval_loss": 2.868479013442993,
|
| 1084 |
+
"eval_runtime": 5.6813,
|
| 1085 |
+
"eval_samples_per_second": 7.041,
|
| 1086 |
+
"eval_steps_per_second": 0.88,
|
| 1087 |
+
"step": 8855
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 78.0,
|
| 1091 |
+
"learning_rate": 2.229565217391305e-06,
|
| 1092 |
+
"loss": 2.7306,
|
| 1093 |
+
"step": 8970
|
| 1094 |
+
},
|
| 1095 |
+
{
|
| 1096 |
+
"epoch": 78.0,
|
| 1097 |
+
"eval_loss": 2.8482155799865723,
|
| 1098 |
+
"eval_runtime": 5.7403,
|
| 1099 |
+
"eval_samples_per_second": 6.968,
|
| 1100 |
+
"eval_steps_per_second": 0.871,
|
| 1101 |
+
"step": 8970
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 79.0,
|
| 1105 |
+
"learning_rate": 2.1295652173913045e-06,
|
| 1106 |
+
"loss": 2.6844,
|
| 1107 |
+
"step": 9085
|
| 1108 |
+
},
|
| 1109 |
+
{
|
| 1110 |
+
"epoch": 79.0,
|
| 1111 |
+
"eval_loss": 2.7371761798858643,
|
| 1112 |
+
"eval_runtime": 5.7039,
|
| 1113 |
+
"eval_samples_per_second": 7.013,
|
| 1114 |
+
"eval_steps_per_second": 0.877,
|
| 1115 |
+
"step": 9085
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 80.0,
|
| 1119 |
+
"learning_rate": 2.0295652173913046e-06,
|
| 1120 |
+
"loss": 2.6949,
|
| 1121 |
+
"step": 9200
|
| 1122 |
+
},
|
| 1123 |
+
{
|
| 1124 |
+
"epoch": 80.0,
|
| 1125 |
+
"eval_loss": 2.81486177444458,
|
| 1126 |
+
"eval_runtime": 5.7433,
|
| 1127 |
+
"eval_samples_per_second": 6.965,
|
| 1128 |
+
"eval_steps_per_second": 0.871,
|
| 1129 |
+
"step": 9200
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 81.0,
|
| 1133 |
+
"learning_rate": 1.9295652173913047e-06,
|
| 1134 |
+
"loss": 2.7342,
|
| 1135 |
+
"step": 9315
|
| 1136 |
+
},
|
| 1137 |
+
{
|
| 1138 |
+
"epoch": 81.0,
|
| 1139 |
+
"eval_loss": 2.76469087600708,
|
| 1140 |
+
"eval_runtime": 5.6684,
|
| 1141 |
+
"eval_samples_per_second": 7.057,
|
| 1142 |
+
"eval_steps_per_second": 0.882,
|
| 1143 |
+
"step": 9315
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 82.0,
|
| 1147 |
+
"learning_rate": 1.8295652173913044e-06,
|
| 1148 |
+
"loss": 2.6813,
|
| 1149 |
+
"step": 9430
|
| 1150 |
+
},
|
| 1151 |
+
{
|
| 1152 |
+
"epoch": 82.0,
|
| 1153 |
+
"eval_loss": 2.7665936946868896,
|
| 1154 |
+
"eval_runtime": 5.6874,
|
| 1155 |
+
"eval_samples_per_second": 7.033,
|
| 1156 |
+
"eval_steps_per_second": 0.879,
|
| 1157 |
+
"step": 9430
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 83.0,
|
| 1161 |
+
"learning_rate": 1.7295652173913043e-06,
|
| 1162 |
+
"loss": 2.7161,
|
| 1163 |
+
"step": 9545
|
| 1164 |
+
},
|
| 1165 |
+
{
|
| 1166 |
+
"epoch": 83.0,
|
| 1167 |
+
"eval_loss": 2.843693971633911,
|
| 1168 |
+
"eval_runtime": 5.7308,
|
| 1169 |
+
"eval_samples_per_second": 6.98,
|
| 1170 |
+
"eval_steps_per_second": 0.872,
|
| 1171 |
+
"step": 9545
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 84.0,
|
| 1175 |
+
"learning_rate": 1.6295652173913046e-06,
|
| 1176 |
+
"loss": 2.6953,
|
| 1177 |
+
"step": 9660
|
| 1178 |
+
},
|
| 1179 |
+
{
|
| 1180 |
+
"epoch": 84.0,
|
| 1181 |
+
"eval_loss": 2.7895007133483887,
|
| 1182 |
+
"eval_runtime": 5.5926,
|
| 1183 |
+
"eval_samples_per_second": 7.152,
|
| 1184 |
+
"eval_steps_per_second": 0.894,
|
| 1185 |
+
"step": 9660
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 85.0,
|
| 1189 |
+
"learning_rate": 1.5295652173913045e-06,
|
| 1190 |
+
"loss": 2.6714,
|
| 1191 |
+
"step": 9775
|
| 1192 |
+
},
|
| 1193 |
+
{
|
| 1194 |
+
"epoch": 85.0,
|
| 1195 |
+
"eval_loss": 2.768319606781006,
|
| 1196 |
+
"eval_runtime": 5.5781,
|
| 1197 |
+
"eval_samples_per_second": 7.171,
|
| 1198 |
+
"eval_steps_per_second": 0.896,
|
| 1199 |
+
"step": 9775
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 86.0,
|
| 1203 |
+
"learning_rate": 1.4295652173913044e-06,
|
| 1204 |
+
"loss": 2.6611,
|
| 1205 |
+
"step": 9890
|
| 1206 |
+
},
|
| 1207 |
+
{
|
| 1208 |
+
"epoch": 86.0,
|
| 1209 |
+
"eval_loss": 2.700421094894409,
|
| 1210 |
+
"eval_runtime": 5.7156,
|
| 1211 |
+
"eval_samples_per_second": 6.998,
|
| 1212 |
+
"eval_steps_per_second": 0.875,
|
| 1213 |
+
"step": 9890
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 87.0,
|
| 1217 |
+
"learning_rate": 1.3295652173913045e-06,
|
| 1218 |
+
"loss": 2.6714,
|
| 1219 |
+
"step": 10005
|
| 1220 |
+
},
|
| 1221 |
+
{
|
| 1222 |
+
"epoch": 87.0,
|
| 1223 |
+
"eval_loss": 2.7182838916778564,
|
| 1224 |
+
"eval_runtime": 5.6318,
|
| 1225 |
+
"eval_samples_per_second": 7.103,
|
| 1226 |
+
"eval_steps_per_second": 0.888,
|
| 1227 |
+
"step": 10005
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 88.0,
|
| 1231 |
+
"learning_rate": 1.2295652173913044e-06,
|
| 1232 |
+
"loss": 2.6655,
|
| 1233 |
+
"step": 10120
|
| 1234 |
+
},
|
| 1235 |
+
{
|
| 1236 |
+
"epoch": 88.0,
|
| 1237 |
+
"eval_loss": 2.7042617797851562,
|
| 1238 |
+
"eval_runtime": 5.8135,
|
| 1239 |
+
"eval_samples_per_second": 6.881,
|
| 1240 |
+
"eval_steps_per_second": 0.86,
|
| 1241 |
+
"step": 10120
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 89.0,
|
| 1245 |
+
"learning_rate": 1.1295652173913045e-06,
|
| 1246 |
+
"loss": 2.6509,
|
| 1247 |
+
"step": 10235
|
| 1248 |
+
},
|
| 1249 |
+
{
|
| 1250 |
+
"epoch": 89.0,
|
| 1251 |
+
"eval_loss": 2.770498514175415,
|
| 1252 |
+
"eval_runtime": 5.7555,
|
| 1253 |
+
"eval_samples_per_second": 6.95,
|
| 1254 |
+
"eval_steps_per_second": 0.869,
|
| 1255 |
+
"step": 10235
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 90.0,
|
| 1259 |
+
"learning_rate": 1.0295652173913044e-06,
|
| 1260 |
+
"loss": 2.6266,
|
| 1261 |
+
"step": 10350
|
| 1262 |
+
},
|
| 1263 |
+
{
|
| 1264 |
+
"epoch": 90.0,
|
| 1265 |
+
"eval_loss": 2.7151668071746826,
|
| 1266 |
+
"eval_runtime": 5.6476,
|
| 1267 |
+
"eval_samples_per_second": 7.083,
|
| 1268 |
+
"eval_steps_per_second": 0.885,
|
| 1269 |
+
"step": 10350
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 91.0,
|
| 1273 |
+
"learning_rate": 9.295652173913044e-07,
|
| 1274 |
+
"loss": 2.6677,
|
| 1275 |
+
"step": 10465
|
| 1276 |
+
},
|
| 1277 |
+
{
|
| 1278 |
+
"epoch": 91.0,
|
| 1279 |
+
"eval_loss": 2.729531764984131,
|
| 1280 |
+
"eval_runtime": 5.6362,
|
| 1281 |
+
"eval_samples_per_second": 7.097,
|
| 1282 |
+
"eval_steps_per_second": 0.887,
|
| 1283 |
+
"step": 10465
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 92.0,
|
| 1287 |
+
"learning_rate": 8.295652173913043e-07,
|
| 1288 |
+
"loss": 2.6438,
|
| 1289 |
+
"step": 10580
|
| 1290 |
+
},
|
| 1291 |
+
{
|
| 1292 |
+
"epoch": 92.0,
|
| 1293 |
+
"eval_loss": 2.70180344581604,
|
| 1294 |
+
"eval_runtime": 5.6488,
|
| 1295 |
+
"eval_samples_per_second": 7.081,
|
| 1296 |
+
"eval_steps_per_second": 0.885,
|
| 1297 |
+
"step": 10580
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 93.0,
|
| 1301 |
+
"learning_rate": 7.295652173913044e-07,
|
| 1302 |
+
"loss": 2.6267,
|
| 1303 |
+
"step": 10695
|
| 1304 |
+
},
|
| 1305 |
+
{
|
| 1306 |
+
"epoch": 93.0,
|
| 1307 |
+
"eval_loss": 2.7062716484069824,
|
| 1308 |
+
"eval_runtime": 5.5879,
|
| 1309 |
+
"eval_samples_per_second": 7.158,
|
| 1310 |
+
"eval_steps_per_second": 0.895,
|
| 1311 |
+
"step": 10695
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 94.0,
|
| 1315 |
+
"learning_rate": 6.295652173913045e-07,
|
| 1316 |
+
"loss": 2.6286,
|
| 1317 |
+
"step": 10810
|
| 1318 |
+
},
|
| 1319 |
+
{
|
| 1320 |
+
"epoch": 94.0,
|
| 1321 |
+
"eval_loss": 2.779799222946167,
|
| 1322 |
+
"eval_runtime": 5.6552,
|
| 1323 |
+
"eval_samples_per_second": 7.073,
|
| 1324 |
+
"eval_steps_per_second": 0.884,
|
| 1325 |
+
"step": 10810
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 95.0,
|
| 1329 |
+
"learning_rate": 5.295652173913045e-07,
|
| 1330 |
+
"loss": 2.6043,
|
| 1331 |
+
"step": 10925
|
| 1332 |
+
},
|
| 1333 |
+
{
|
| 1334 |
+
"epoch": 95.0,
|
| 1335 |
+
"eval_loss": 2.7712299823760986,
|
| 1336 |
+
"eval_runtime": 5.6131,
|
| 1337 |
+
"eval_samples_per_second": 7.126,
|
| 1338 |
+
"eval_steps_per_second": 0.891,
|
| 1339 |
+
"step": 10925
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 96.0,
|
| 1343 |
+
"learning_rate": 4.295652173913044e-07,
|
| 1344 |
+
"loss": 2.6188,
|
| 1345 |
+
"step": 11040
|
| 1346 |
+
},
|
| 1347 |
+
{
|
| 1348 |
+
"epoch": 96.0,
|
| 1349 |
+
"eval_loss": 2.7614357471466064,
|
| 1350 |
+
"eval_runtime": 5.567,
|
| 1351 |
+
"eval_samples_per_second": 7.185,
|
| 1352 |
+
"eval_steps_per_second": 0.898,
|
| 1353 |
+
"step": 11040
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 97.0,
|
| 1357 |
+
"learning_rate": 3.2956521739130436e-07,
|
| 1358 |
+
"loss": 2.6028,
|
| 1359 |
+
"step": 11155
|
| 1360 |
+
},
|
| 1361 |
+
{
|
| 1362 |
+
"epoch": 97.0,
|
| 1363 |
+
"eval_loss": 2.740495204925537,
|
| 1364 |
+
"eval_runtime": 5.621,
|
| 1365 |
+
"eval_samples_per_second": 7.116,
|
| 1366 |
+
"eval_steps_per_second": 0.89,
|
| 1367 |
+
"step": 11155
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 98.0,
|
| 1371 |
+
"learning_rate": 2.2956521739130436e-07,
|
| 1372 |
+
"loss": 2.621,
|
| 1373 |
+
"step": 11270
|
| 1374 |
+
},
|
| 1375 |
+
{
|
| 1376 |
+
"epoch": 98.0,
|
| 1377 |
+
"eval_loss": 2.741487503051758,
|
| 1378 |
+
"eval_runtime": 5.6116,
|
| 1379 |
+
"eval_samples_per_second": 7.128,
|
| 1380 |
+
"eval_steps_per_second": 0.891,
|
| 1381 |
+
"step": 11270
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 99.0,
|
| 1385 |
+
"learning_rate": 1.2956521739130434e-07,
|
| 1386 |
+
"loss": 2.61,
|
| 1387 |
+
"step": 11385
|
| 1388 |
+
},
|
| 1389 |
+
{
|
| 1390 |
+
"epoch": 99.0,
|
| 1391 |
+
"eval_loss": 2.741542100906372,
|
| 1392 |
+
"eval_runtime": 5.7228,
|
| 1393 |
+
"eval_samples_per_second": 6.99,
|
| 1394 |
+
"eval_steps_per_second": 0.874,
|
| 1395 |
+
"step": 11385
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 100.0,
|
| 1399 |
+
"learning_rate": 2.9565217391304353e-08,
|
| 1400 |
+
"loss": 2.6164,
|
| 1401 |
+
"step": 11500
|
| 1402 |
+
},
|
| 1403 |
+
{
|
| 1404 |
+
"epoch": 100.0,
|
| 1405 |
+
"eval_loss": 2.7389094829559326,
|
| 1406 |
+
"eval_runtime": 5.5344,
|
| 1407 |
+
"eval_samples_per_second": 7.228,
|
| 1408 |
+
"eval_steps_per_second": 0.903,
|
| 1409 |
+
"step": 11500
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 100.0,
|
| 1413 |
+
"step": 11500,
|
| 1414 |
+
"total_flos": 2.295532415232e+19,
|
| 1415 |
+
"train_loss": 1.6043826771611753,
|
| 1416 |
+
"train_runtime": 5669.8878,
|
| 1417 |
+
"train_samples_per_second": 8.113,
|
| 1418 |
+
"train_steps_per_second": 2.028
|
| 1419 |
}
|
| 1420 |
],
|
| 1421 |
"logging_steps": 500,
|
| 1422 |
+
"max_steps": 11500,
|
| 1423 |
+
"num_train_epochs": 100,
|
| 1424 |
"save_steps": 500,
|
| 1425 |
+
"total_flos": 2.295532415232e+19,
|
| 1426 |
"trial_name": null,
|
| 1427 |
"trial_params": null
|
| 1428 |
}
|