End of training
Browse files- README.md +4 -1
- all_results.json +27 -0
- test_results.json +22 -0
- train_results.json +8 -0
- trainer_state.json +642 -0
README.md
CHANGED
|
@@ -3,6 +3,9 @@ library_name: transformers
|
|
| 3 |
license: other
|
| 4 |
base_model: facebook/mask2former-swin-tiny-coco-instance
|
| 5 |
tags:
|
|
|
|
|
|
|
|
|
|
| 6 |
- generated_from_trainer
|
| 7 |
model-index:
|
| 8 |
- name: finetune-instance-segmentation-flowchartseg-mask2former_20epochs_a6000
|
|
@@ -14,7 +17,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 14 |
|
| 15 |
# finetune-instance-segmentation-flowchartseg-mask2former_20epochs_a6000
|
| 16 |
|
| 17 |
-
This model is a fine-tuned version of [facebook/mask2former-swin-tiny-coco-instance](https://huggingface.co/facebook/mask2former-swin-tiny-coco-instance) on
|
| 18 |
It achieves the following results on the evaluation set:
|
| 19 |
- Loss: 6.9523
|
| 20 |
- Map: 0.7683
|
|
|
|
| 3 |
license: other
|
| 4 |
base_model: facebook/mask2former-swin-tiny-coco-instance
|
| 5 |
tags:
|
| 6 |
+
- image-segmentation
|
| 7 |
+
- instance-segmentation
|
| 8 |
+
- vision
|
| 9 |
- generated_from_trainer
|
| 10 |
model-index:
|
| 11 |
- name: finetune-instance-segmentation-flowchartseg-mask2former_20epochs_a6000
|
|
|
|
| 17 |
|
| 18 |
# finetune-instance-segmentation-flowchartseg-mask2former_20epochs_a6000
|
| 19 |
|
| 20 |
+
This model is a fine-tuned version of [facebook/mask2former-swin-tiny-coco-instance](https://huggingface.co/facebook/mask2former-swin-tiny-coco-instance) on the MananSuri27/flowchartseg dataset.
|
| 21 |
It achieves the following results on the evaluation set:
|
| 22 |
- Loss: 6.9523
|
| 23 |
- Map: 0.7683
|
all_results.json
ADDED
|
@@ -0,0 +1,27 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 19.74074074074074,
|
| 3 |
+
"test_classes": 0,
|
| 4 |
+
"test_loss": 6.944503307342529,
|
| 5 |
+
"test_map": 0.7683,
|
| 6 |
+
"test_map_50": 0.9698,
|
| 7 |
+
"test_map_75": 0.9012,
|
| 8 |
+
"test_map_large": 0.9911,
|
| 9 |
+
"test_map_medium": 0.8012,
|
| 10 |
+
"test_map_per_class": 0.7683,
|
| 11 |
+
"test_map_small": 0.6166,
|
| 12 |
+
"test_mar_1": 0.0414,
|
| 13 |
+
"test_mar_10": 0.3998,
|
| 14 |
+
"test_mar_100": 0.8409,
|
| 15 |
+
"test_mar_100_per_class": 0.8409,
|
| 16 |
+
"test_mar_large": 0.9938,
|
| 17 |
+
"test_mar_medium": 0.8711,
|
| 18 |
+
"test_mar_small": 0.6954,
|
| 19 |
+
"test_runtime": 302.5528,
|
| 20 |
+
"test_samples_per_second": 0.436,
|
| 21 |
+
"test_steps_per_second": 0.056,
|
| 22 |
+
"total_flos": 5.235456347568341e+18,
|
| 23 |
+
"train_loss": 8.241098208040805,
|
| 24 |
+
"train_runtime": 50547.6535,
|
| 25 |
+
"train_samples_per_second": 0.47,
|
| 26 |
+
"train_steps_per_second": 0.029
|
| 27 |
+
}
|
test_results.json
ADDED
|
@@ -0,0 +1,22 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 19.74074074074074,
|
| 3 |
+
"test_classes": 0,
|
| 4 |
+
"test_loss": 6.944503307342529,
|
| 5 |
+
"test_map": 0.7683,
|
| 6 |
+
"test_map_50": 0.9698,
|
| 7 |
+
"test_map_75": 0.9012,
|
| 8 |
+
"test_map_large": 0.9911,
|
| 9 |
+
"test_map_medium": 0.8012,
|
| 10 |
+
"test_map_per_class": 0.7683,
|
| 11 |
+
"test_map_small": 0.6166,
|
| 12 |
+
"test_mar_1": 0.0414,
|
| 13 |
+
"test_mar_10": 0.3998,
|
| 14 |
+
"test_mar_100": 0.8409,
|
| 15 |
+
"test_mar_100_per_class": 0.8409,
|
| 16 |
+
"test_mar_large": 0.9938,
|
| 17 |
+
"test_mar_medium": 0.8711,
|
| 18 |
+
"test_mar_small": 0.6954,
|
| 19 |
+
"test_runtime": 302.5528,
|
| 20 |
+
"test_samples_per_second": 0.436,
|
| 21 |
+
"test_steps_per_second": 0.056
|
| 22 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 19.74074074074074,
|
| 3 |
+
"total_flos": 5.235456347568341e+18,
|
| 4 |
+
"train_loss": 8.241098208040805,
|
| 5 |
+
"train_runtime": 50547.6535,
|
| 6 |
+
"train_samples_per_second": 0.47,
|
| 7 |
+
"train_steps_per_second": 0.029
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,642 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 19.74074074074074,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 1480,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 1.0,
|
| 13 |
+
"grad_norm": 52.495697021484375,
|
| 14 |
+
"learning_rate": 9.9464649786582e-06,
|
| 15 |
+
"loss": 21.8993,
|
| 16 |
+
"step": 75
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 1.0,
|
| 20 |
+
"eval_classes": 0,
|
| 21 |
+
"eval_loss": 12.443974494934082,
|
| 22 |
+
"eval_map": 0.6744,
|
| 23 |
+
"eval_map_50": 0.9348,
|
| 24 |
+
"eval_map_75": 0.7871,
|
| 25 |
+
"eval_map_large": 0.9595,
|
| 26 |
+
"eval_map_medium": 0.7115,
|
| 27 |
+
"eval_map_per_class": 0.6744,
|
| 28 |
+
"eval_map_small": 0.4962,
|
| 29 |
+
"eval_mar_1": 0.0394,
|
| 30 |
+
"eval_mar_10": 0.3779,
|
| 31 |
+
"eval_mar_100": 0.7627,
|
| 32 |
+
"eval_mar_100_per_class": 0.7627,
|
| 33 |
+
"eval_mar_large": 0.9741,
|
| 34 |
+
"eval_mar_medium": 0.7977,
|
| 35 |
+
"eval_mar_small": 0.5896,
|
| 36 |
+
"eval_runtime": 301.7158,
|
| 37 |
+
"eval_samples_per_second": 0.437,
|
| 38 |
+
"eval_steps_per_second": 0.056,
|
| 39 |
+
"step": 75
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 2.0,
|
| 43 |
+
"grad_norm": 35.35026550292969,
|
| 44 |
+
"learning_rate": 9.768230005822394e-06,
|
| 45 |
+
"loss": 10.836,
|
| 46 |
+
"step": 150
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 2.0,
|
| 50 |
+
"eval_classes": 0,
|
| 51 |
+
"eval_loss": 10.104583740234375,
|
| 52 |
+
"eval_map": 0.7114,
|
| 53 |
+
"eval_map_50": 0.9553,
|
| 54 |
+
"eval_map_75": 0.8305,
|
| 55 |
+
"eval_map_large": 0.9806,
|
| 56 |
+
"eval_map_medium": 0.7495,
|
| 57 |
+
"eval_map_per_class": 0.7114,
|
| 58 |
+
"eval_map_small": 0.5421,
|
| 59 |
+
"eval_mar_1": 0.0399,
|
| 60 |
+
"eval_mar_10": 0.3853,
|
| 61 |
+
"eval_mar_100": 0.7924,
|
| 62 |
+
"eval_mar_100_per_class": 0.7924,
|
| 63 |
+
"eval_mar_large": 0.9864,
|
| 64 |
+
"eval_mar_medium": 0.8267,
|
| 65 |
+
"eval_mar_small": 0.6239,
|
| 66 |
+
"eval_runtime": 301.7621,
|
| 67 |
+
"eval_samples_per_second": 0.437,
|
| 68 |
+
"eval_steps_per_second": 0.056,
|
| 69 |
+
"step": 150
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 3.0,
|
| 73 |
+
"grad_norm": 77.25425720214844,
|
| 74 |
+
"learning_rate": 9.469397461053838e-06,
|
| 75 |
+
"loss": 9.422,
|
| 76 |
+
"step": 225
|
| 77 |
+
},
|
| 78 |
+
{
|
| 79 |
+
"epoch": 3.0,
|
| 80 |
+
"eval_classes": 0,
|
| 81 |
+
"eval_loss": 9.384869575500488,
|
| 82 |
+
"eval_map": 0.7277,
|
| 83 |
+
"eval_map_50": 0.9599,
|
| 84 |
+
"eval_map_75": 0.8501,
|
| 85 |
+
"eval_map_large": 0.9829,
|
| 86 |
+
"eval_map_medium": 0.7674,
|
| 87 |
+
"eval_map_per_class": 0.7277,
|
| 88 |
+
"eval_map_small": 0.5591,
|
| 89 |
+
"eval_mar_1": 0.0398,
|
| 90 |
+
"eval_mar_10": 0.3882,
|
| 91 |
+
"eval_mar_100": 0.8057,
|
| 92 |
+
"eval_mar_100_per_class": 0.8057,
|
| 93 |
+
"eval_mar_large": 0.9852,
|
| 94 |
+
"eval_mar_medium": 0.8385,
|
| 95 |
+
"eval_mar_small": 0.6454,
|
| 96 |
+
"eval_runtime": 301.8767,
|
| 97 |
+
"eval_samples_per_second": 0.437,
|
| 98 |
+
"eval_steps_per_second": 0.056,
|
| 99 |
+
"step": 225
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 4.0,
|
| 103 |
+
"grad_norm": 25.18535614013672,
|
| 104 |
+
"learning_rate": 9.057525385652877e-06,
|
| 105 |
+
"loss": 8.7626,
|
| 106 |
+
"step": 300
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 4.0,
|
| 110 |
+
"eval_classes": 0,
|
| 111 |
+
"eval_loss": 8.737225532531738,
|
| 112 |
+
"eval_map": 0.7337,
|
| 113 |
+
"eval_map_50": 0.9632,
|
| 114 |
+
"eval_map_75": 0.8564,
|
| 115 |
+
"eval_map_large": 0.9825,
|
| 116 |
+
"eval_map_medium": 0.7713,
|
| 117 |
+
"eval_map_per_class": 0.7337,
|
| 118 |
+
"eval_map_small": 0.5685,
|
| 119 |
+
"eval_mar_1": 0.0401,
|
| 120 |
+
"eval_mar_10": 0.3902,
|
| 121 |
+
"eval_mar_100": 0.8122,
|
| 122 |
+
"eval_mar_100_per_class": 0.8122,
|
| 123 |
+
"eval_mar_large": 0.9889,
|
| 124 |
+
"eval_mar_medium": 0.8441,
|
| 125 |
+
"eval_mar_small": 0.6565,
|
| 126 |
+
"eval_runtime": 301.9674,
|
| 127 |
+
"eval_samples_per_second": 0.437,
|
| 128 |
+
"eval_steps_per_second": 0.056,
|
| 129 |
+
"step": 300
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 5.0,
|
| 133 |
+
"grad_norm": 100.30033111572266,
|
| 134 |
+
"learning_rate": 8.550511739408428e-06,
|
| 135 |
+
"loss": 8.2403,
|
| 136 |
+
"step": 375
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 5.0,
|
| 140 |
+
"eval_classes": 0,
|
| 141 |
+
"eval_loss": 8.39384651184082,
|
| 142 |
+
"eval_map": 0.7416,
|
| 143 |
+
"eval_map_50": 0.9673,
|
| 144 |
+
"eval_map_75": 0.8706,
|
| 145 |
+
"eval_map_large": 0.991,
|
| 146 |
+
"eval_map_medium": 0.7812,
|
| 147 |
+
"eval_map_per_class": 0.7416,
|
| 148 |
+
"eval_map_small": 0.5659,
|
| 149 |
+
"eval_mar_1": 0.0403,
|
| 150 |
+
"eval_mar_10": 0.3914,
|
| 151 |
+
"eval_mar_100": 0.8177,
|
| 152 |
+
"eval_mar_100_per_class": 0.8177,
|
| 153 |
+
"eval_mar_large": 0.9926,
|
| 154 |
+
"eval_mar_medium": 0.8508,
|
| 155 |
+
"eval_mar_small": 0.657,
|
| 156 |
+
"eval_runtime": 301.5787,
|
| 157 |
+
"eval_samples_per_second": 0.438,
|
| 158 |
+
"eval_steps_per_second": 0.056,
|
| 159 |
+
"step": 375
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 6.0,
|
| 163 |
+
"grad_norm": 60.029884338378906,
|
| 164 |
+
"learning_rate": 7.947506126566009e-06,
|
| 165 |
+
"loss": 7.8836,
|
| 166 |
+
"step": 450
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"epoch": 6.0,
|
| 170 |
+
"eval_classes": 0,
|
| 171 |
+
"eval_loss": 7.927948951721191,
|
| 172 |
+
"eval_map": 0.7504,
|
| 173 |
+
"eval_map_50": 0.9692,
|
| 174 |
+
"eval_map_75": 0.8794,
|
| 175 |
+
"eval_map_large": 0.9901,
|
| 176 |
+
"eval_map_medium": 0.7878,
|
| 177 |
+
"eval_map_per_class": 0.7504,
|
| 178 |
+
"eval_map_small": 0.5853,
|
| 179 |
+
"eval_mar_1": 0.0412,
|
| 180 |
+
"eval_mar_10": 0.3964,
|
| 181 |
+
"eval_mar_100": 0.824,
|
| 182 |
+
"eval_mar_100_per_class": 0.824,
|
| 183 |
+
"eval_mar_large": 0.9914,
|
| 184 |
+
"eval_mar_medium": 0.8561,
|
| 185 |
+
"eval_mar_small": 0.6685,
|
| 186 |
+
"eval_runtime": 302.1538,
|
| 187 |
+
"eval_samples_per_second": 0.437,
|
| 188 |
+
"eval_steps_per_second": 0.056,
|
| 189 |
+
"step": 450
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 7.0,
|
| 193 |
+
"grad_norm": 45.95404815673828,
|
| 194 |
+
"learning_rate": 7.2794040652413374e-06,
|
| 195 |
+
"loss": 7.6029,
|
| 196 |
+
"step": 525
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 7.0,
|
| 200 |
+
"eval_classes": 0,
|
| 201 |
+
"eval_loss": 7.830236434936523,
|
| 202 |
+
"eval_map": 0.7562,
|
| 203 |
+
"eval_map_50": 0.9707,
|
| 204 |
+
"eval_map_75": 0.8933,
|
| 205 |
+
"eval_map_large": 0.9909,
|
| 206 |
+
"eval_map_medium": 0.7925,
|
| 207 |
+
"eval_map_per_class": 0.7562,
|
| 208 |
+
"eval_map_small": 0.5965,
|
| 209 |
+
"eval_mar_1": 0.0415,
|
| 210 |
+
"eval_mar_10": 0.3969,
|
| 211 |
+
"eval_mar_100": 0.8294,
|
| 212 |
+
"eval_mar_100_per_class": 0.8294,
|
| 213 |
+
"eval_mar_large": 0.9926,
|
| 214 |
+
"eval_mar_medium": 0.8603,
|
| 215 |
+
"eval_mar_small": 0.68,
|
| 216 |
+
"eval_runtime": 302.0489,
|
| 217 |
+
"eval_samples_per_second": 0.437,
|
| 218 |
+
"eval_steps_per_second": 0.056,
|
| 219 |
+
"step": 525
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 8.0,
|
| 223 |
+
"grad_norm": 35.335750579833984,
|
| 224 |
+
"learning_rate": 6.555175510922047e-06,
|
| 225 |
+
"loss": 7.3823,
|
| 226 |
+
"step": 600
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 8.0,
|
| 230 |
+
"eval_classes": 0,
|
| 231 |
+
"eval_loss": 7.547567844390869,
|
| 232 |
+
"eval_map": 0.756,
|
| 233 |
+
"eval_map_50": 0.9689,
|
| 234 |
+
"eval_map_75": 0.8892,
|
| 235 |
+
"eval_map_large": 0.9908,
|
| 236 |
+
"eval_map_medium": 0.792,
|
| 237 |
+
"eval_map_per_class": 0.756,
|
| 238 |
+
"eval_map_small": 0.5973,
|
| 239 |
+
"eval_mar_1": 0.0412,
|
| 240 |
+
"eval_mar_10": 0.3982,
|
| 241 |
+
"eval_mar_100": 0.8315,
|
| 242 |
+
"eval_mar_100_per_class": 0.8315,
|
| 243 |
+
"eval_mar_large": 0.9926,
|
| 244 |
+
"eval_mar_medium": 0.8626,
|
| 245 |
+
"eval_mar_small": 0.6809,
|
| 246 |
+
"eval_runtime": 302.4595,
|
| 247 |
+
"eval_samples_per_second": 0.436,
|
| 248 |
+
"eval_steps_per_second": 0.056,
|
| 249 |
+
"step": 600
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 9.0,
|
| 253 |
+
"grad_norm": 16.641401290893555,
|
| 254 |
+
"learning_rate": 5.782172325201155e-06,
|
| 255 |
+
"loss": 7.2456,
|
| 256 |
+
"step": 675
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 9.0,
|
| 260 |
+
"eval_classes": 0,
|
| 261 |
+
"eval_loss": 7.461709499359131,
|
| 262 |
+
"eval_map": 0.7611,
|
| 263 |
+
"eval_map_50": 0.969,
|
| 264 |
+
"eval_map_75": 0.8979,
|
| 265 |
+
"eval_map_large": 0.9902,
|
| 266 |
+
"eval_map_medium": 0.795,
|
| 267 |
+
"eval_map_per_class": 0.7611,
|
| 268 |
+
"eval_map_small": 0.6055,
|
| 269 |
+
"eval_mar_1": 0.0411,
|
| 270 |
+
"eval_mar_10": 0.3995,
|
| 271 |
+
"eval_mar_100": 0.8338,
|
| 272 |
+
"eval_mar_100_per_class": 0.8338,
|
| 273 |
+
"eval_mar_large": 0.9926,
|
| 274 |
+
"eval_mar_medium": 0.8642,
|
| 275 |
+
"eval_mar_small": 0.6865,
|
| 276 |
+
"eval_runtime": 302.0761,
|
| 277 |
+
"eval_samples_per_second": 0.437,
|
| 278 |
+
"eval_steps_per_second": 0.056,
|
| 279 |
+
"step": 675
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 10.0,
|
| 283 |
+
"grad_norm": 18.994029998779297,
|
| 284 |
+
"learning_rate": 4.989386519275895e-06,
|
| 285 |
+
"loss": 7.068,
|
| 286 |
+
"step": 750
|
| 287 |
+
},
|
| 288 |
+
{
|
| 289 |
+
"epoch": 10.0,
|
| 290 |
+
"eval_classes": 0,
|
| 291 |
+
"eval_loss": 7.323790550231934,
|
| 292 |
+
"eval_map": 0.7627,
|
| 293 |
+
"eval_map_50": 0.9688,
|
| 294 |
+
"eval_map_75": 0.8988,
|
| 295 |
+
"eval_map_large": 0.9901,
|
| 296 |
+
"eval_map_medium": 0.7964,
|
| 297 |
+
"eval_map_per_class": 0.7627,
|
| 298 |
+
"eval_map_small": 0.6028,
|
| 299 |
+
"eval_mar_1": 0.0413,
|
| 300 |
+
"eval_mar_10": 0.3995,
|
| 301 |
+
"eval_mar_100": 0.8347,
|
| 302 |
+
"eval_mar_100_per_class": 0.8347,
|
| 303 |
+
"eval_mar_large": 0.9938,
|
| 304 |
+
"eval_mar_medium": 0.8656,
|
| 305 |
+
"eval_mar_small": 0.685,
|
| 306 |
+
"eval_runtime": 304.0925,
|
| 307 |
+
"eval_samples_per_second": 0.434,
|
| 308 |
+
"eval_steps_per_second": 0.056,
|
| 309 |
+
"step": 750
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 11.0,
|
| 313 |
+
"grad_norm": 17.86629295349121,
|
| 314 |
+
"learning_rate": 4.196869148389114e-06,
|
| 315 |
+
"loss": 7.0105,
|
| 316 |
+
"step": 825
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 11.0,
|
| 320 |
+
"eval_classes": 0,
|
| 321 |
+
"eval_loss": 7.247533798217773,
|
| 322 |
+
"eval_map": 0.7655,
|
| 323 |
+
"eval_map_50": 0.9703,
|
| 324 |
+
"eval_map_75": 0.8992,
|
| 325 |
+
"eval_map_large": 0.991,
|
| 326 |
+
"eval_map_medium": 0.7982,
|
| 327 |
+
"eval_map_per_class": 0.7655,
|
| 328 |
+
"eval_map_small": 0.6099,
|
| 329 |
+
"eval_mar_1": 0.0409,
|
| 330 |
+
"eval_mar_10": 0.4007,
|
| 331 |
+
"eval_mar_100": 0.8372,
|
| 332 |
+
"eval_mar_100_per_class": 0.8372,
|
| 333 |
+
"eval_mar_large": 0.9938,
|
| 334 |
+
"eval_mar_medium": 0.8679,
|
| 335 |
+
"eval_mar_small": 0.6893,
|
| 336 |
+
"eval_runtime": 302.7002,
|
| 337 |
+
"eval_samples_per_second": 0.436,
|
| 338 |
+
"eval_steps_per_second": 0.056,
|
| 339 |
+
"step": 825
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 12.0,
|
| 343 |
+
"grad_norm": 10.187027931213379,
|
| 344 |
+
"learning_rate": 3.424664478552887e-06,
|
| 345 |
+
"loss": 6.9651,
|
| 346 |
+
"step": 900
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 12.0,
|
| 350 |
+
"eval_classes": 0,
|
| 351 |
+
"eval_loss": 7.234455108642578,
|
| 352 |
+
"eval_map": 0.7679,
|
| 353 |
+
"eval_map_50": 0.9709,
|
| 354 |
+
"eval_map_75": 0.9045,
|
| 355 |
+
"eval_map_large": 0.9899,
|
| 356 |
+
"eval_map_medium": 0.8008,
|
| 357 |
+
"eval_map_per_class": 0.7679,
|
| 358 |
+
"eval_map_small": 0.6147,
|
| 359 |
+
"eval_mar_1": 0.0412,
|
| 360 |
+
"eval_mar_10": 0.3999,
|
| 361 |
+
"eval_mar_100": 0.8393,
|
| 362 |
+
"eval_mar_100_per_class": 0.8393,
|
| 363 |
+
"eval_mar_large": 0.9938,
|
| 364 |
+
"eval_mar_medium": 0.8692,
|
| 365 |
+
"eval_mar_small": 0.6944,
|
| 366 |
+
"eval_runtime": 307.3877,
|
| 367 |
+
"eval_samples_per_second": 0.429,
|
| 368 |
+
"eval_steps_per_second": 0.055,
|
| 369 |
+
"step": 900
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 13.0,
|
| 373 |
+
"grad_norm": 24.03527069091797,
|
| 374 |
+
"learning_rate": 2.6923030290801817e-06,
|
| 375 |
+
"loss": 6.8849,
|
| 376 |
+
"step": 975
|
| 377 |
+
},
|
| 378 |
+
{
|
| 379 |
+
"epoch": 13.0,
|
| 380 |
+
"eval_classes": 0,
|
| 381 |
+
"eval_loss": 7.041961669921875,
|
| 382 |
+
"eval_map": 0.764,
|
| 383 |
+
"eval_map_50": 0.9698,
|
| 384 |
+
"eval_map_75": 0.8996,
|
| 385 |
+
"eval_map_large": 0.9899,
|
| 386 |
+
"eval_map_medium": 0.7977,
|
| 387 |
+
"eval_map_per_class": 0.764,
|
| 388 |
+
"eval_map_small": 0.6091,
|
| 389 |
+
"eval_mar_1": 0.0405,
|
| 390 |
+
"eval_mar_10": 0.3996,
|
| 391 |
+
"eval_mar_100": 0.8377,
|
| 392 |
+
"eval_mar_100_per_class": 0.8377,
|
| 393 |
+
"eval_mar_large": 0.9938,
|
| 394 |
+
"eval_mar_medium": 0.8677,
|
| 395 |
+
"eval_mar_small": 0.692,
|
| 396 |
+
"eval_runtime": 302.0813,
|
| 397 |
+
"eval_samples_per_second": 0.437,
|
| 398 |
+
"eval_steps_per_second": 0.056,
|
| 399 |
+
"step": 975
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 14.0,
|
| 403 |
+
"grad_norm": 25.183696746826172,
|
| 404 |
+
"learning_rate": 2.0183076087440044e-06,
|
| 405 |
+
"loss": 6.8559,
|
| 406 |
+
"step": 1050
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 14.0,
|
| 410 |
+
"eval_classes": 0,
|
| 411 |
+
"eval_loss": 7.071371555328369,
|
| 412 |
+
"eval_map": 0.7643,
|
| 413 |
+
"eval_map_50": 0.9693,
|
| 414 |
+
"eval_map_75": 0.8986,
|
| 415 |
+
"eval_map_large": 0.9899,
|
| 416 |
+
"eval_map_medium": 0.7991,
|
| 417 |
+
"eval_map_per_class": 0.7643,
|
| 418 |
+
"eval_map_small": 0.6084,
|
| 419 |
+
"eval_mar_1": 0.0413,
|
| 420 |
+
"eval_mar_10": 0.3994,
|
| 421 |
+
"eval_mar_100": 0.8387,
|
| 422 |
+
"eval_mar_100_per_class": 0.8387,
|
| 423 |
+
"eval_mar_large": 0.9938,
|
| 424 |
+
"eval_mar_medium": 0.8693,
|
| 425 |
+
"eval_mar_small": 0.6913,
|
| 426 |
+
"eval_runtime": 302.1345,
|
| 427 |
+
"eval_samples_per_second": 0.437,
|
| 428 |
+
"eval_steps_per_second": 0.056,
|
| 429 |
+
"step": 1050
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 15.0,
|
| 433 |
+
"grad_norm": 26.083616256713867,
|
| 434 |
+
"learning_rate": 1.4197248388455693e-06,
|
| 435 |
+
"loss": 6.7735,
|
| 436 |
+
"step": 1125
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 15.0,
|
| 440 |
+
"eval_classes": 0,
|
| 441 |
+
"eval_loss": 6.997360706329346,
|
| 442 |
+
"eval_map": 0.7671,
|
| 443 |
+
"eval_map_50": 0.9702,
|
| 444 |
+
"eval_map_75": 0.9007,
|
| 445 |
+
"eval_map_large": 0.9908,
|
| 446 |
+
"eval_map_medium": 0.8014,
|
| 447 |
+
"eval_map_per_class": 0.7671,
|
| 448 |
+
"eval_map_small": 0.6119,
|
| 449 |
+
"eval_mar_1": 0.0413,
|
| 450 |
+
"eval_mar_10": 0.3998,
|
| 451 |
+
"eval_mar_100": 0.8395,
|
| 452 |
+
"eval_mar_100_per_class": 0.8395,
|
| 453 |
+
"eval_mar_large": 0.9938,
|
| 454 |
+
"eval_mar_medium": 0.8706,
|
| 455 |
+
"eval_mar_small": 0.6902,
|
| 456 |
+
"eval_runtime": 302.6848,
|
| 457 |
+
"eval_samples_per_second": 0.436,
|
| 458 |
+
"eval_steps_per_second": 0.056,
|
| 459 |
+
"step": 1125
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 16.0,
|
| 463 |
+
"grad_norm": 25.00279998779297,
|
| 464 |
+
"learning_rate": 9.116940118607792e-07,
|
| 465 |
+
"loss": 6.7596,
|
| 466 |
+
"step": 1200
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 16.0,
|
| 470 |
+
"eval_classes": 0,
|
| 471 |
+
"eval_loss": 6.9286627769470215,
|
| 472 |
+
"eval_map": 0.7675,
|
| 473 |
+
"eval_map_50": 0.9702,
|
| 474 |
+
"eval_map_75": 0.9041,
|
| 475 |
+
"eval_map_large": 0.991,
|
| 476 |
+
"eval_map_medium": 0.8007,
|
| 477 |
+
"eval_map_per_class": 0.7675,
|
| 478 |
+
"eval_map_small": 0.614,
|
| 479 |
+
"eval_mar_1": 0.0406,
|
| 480 |
+
"eval_mar_10": 0.399,
|
| 481 |
+
"eval_mar_100": 0.8393,
|
| 482 |
+
"eval_mar_100_per_class": 0.8393,
|
| 483 |
+
"eval_mar_large": 0.9938,
|
| 484 |
+
"eval_mar_medium": 0.8696,
|
| 485 |
+
"eval_mar_small": 0.6931,
|
| 486 |
+
"eval_runtime": 302.6539,
|
| 487 |
+
"eval_samples_per_second": 0.436,
|
| 488 |
+
"eval_steps_per_second": 0.056,
|
| 489 |
+
"step": 1200
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 17.0,
|
| 493 |
+
"grad_norm": 29.690765380859375,
|
| 494 |
+
"learning_rate": 5.070641900468149e-07,
|
| 495 |
+
"loss": 6.7191,
|
| 496 |
+
"step": 1275
|
| 497 |
+
},
|
| 498 |
+
{
|
| 499 |
+
"epoch": 17.0,
|
| 500 |
+
"eval_classes": 0,
|
| 501 |
+
"eval_loss": 7.00290584564209,
|
| 502 |
+
"eval_map": 0.7678,
|
| 503 |
+
"eval_map_50": 0.9701,
|
| 504 |
+
"eval_map_75": 0.9039,
|
| 505 |
+
"eval_map_large": 0.991,
|
| 506 |
+
"eval_map_medium": 0.8016,
|
| 507 |
+
"eval_map_per_class": 0.7678,
|
| 508 |
+
"eval_map_small": 0.61,
|
| 509 |
+
"eval_mar_1": 0.0407,
|
| 510 |
+
"eval_mar_10": 0.3991,
|
| 511 |
+
"eval_mar_100": 0.8407,
|
| 512 |
+
"eval_mar_100_per_class": 0.8407,
|
| 513 |
+
"eval_mar_large": 0.9938,
|
| 514 |
+
"eval_mar_medium": 0.8711,
|
| 515 |
+
"eval_mar_small": 0.6943,
|
| 516 |
+
"eval_runtime": 302.683,
|
| 517 |
+
"eval_samples_per_second": 0.436,
|
| 518 |
+
"eval_steps_per_second": 0.056,
|
| 519 |
+
"step": 1275
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 18.0,
|
| 523 |
+
"grad_norm": 10.844990730285645,
|
| 524 |
+
"learning_rate": 2.1606922831058198e-07,
|
| 525 |
+
"loss": 6.6811,
|
| 526 |
+
"step": 1350
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 18.0,
|
| 530 |
+
"eval_classes": 0,
|
| 531 |
+
"eval_loss": 6.940825462341309,
|
| 532 |
+
"eval_map": 0.7678,
|
| 533 |
+
"eval_map_50": 0.97,
|
| 534 |
+
"eval_map_75": 0.901,
|
| 535 |
+
"eval_map_large": 0.991,
|
| 536 |
+
"eval_map_medium": 0.8014,
|
| 537 |
+
"eval_map_per_class": 0.7678,
|
| 538 |
+
"eval_map_small": 0.6121,
|
| 539 |
+
"eval_mar_1": 0.041,
|
| 540 |
+
"eval_mar_10": 0.4008,
|
| 541 |
+
"eval_mar_100": 0.84,
|
| 542 |
+
"eval_mar_100_per_class": 0.84,
|
| 543 |
+
"eval_mar_large": 0.9938,
|
| 544 |
+
"eval_mar_medium": 0.8705,
|
| 545 |
+
"eval_mar_small": 0.6928,
|
| 546 |
+
"eval_runtime": 302.7462,
|
| 547 |
+
"eval_samples_per_second": 0.436,
|
| 548 |
+
"eval_steps_per_second": 0.056,
|
| 549 |
+
"step": 1350
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 19.0,
|
| 553 |
+
"grad_norm": 35.2485237121582,
|
| 554 |
+
"learning_rate": 4.60689406264897e-08,
|
| 555 |
+
"loss": 6.6732,
|
| 556 |
+
"step": 1425
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 19.0,
|
| 560 |
+
"eval_classes": 0,
|
| 561 |
+
"eval_loss": 6.960495471954346,
|
| 562 |
+
"eval_map": 0.7679,
|
| 563 |
+
"eval_map_50": 0.9696,
|
| 564 |
+
"eval_map_75": 0.9012,
|
| 565 |
+
"eval_map_large": 0.9911,
|
| 566 |
+
"eval_map_medium": 0.801,
|
| 567 |
+
"eval_map_per_class": 0.7679,
|
| 568 |
+
"eval_map_small": 0.6142,
|
| 569 |
+
"eval_mar_1": 0.0413,
|
| 570 |
+
"eval_mar_10": 0.4005,
|
| 571 |
+
"eval_mar_100": 0.8405,
|
| 572 |
+
"eval_mar_100_per_class": 0.8405,
|
| 573 |
+
"eval_mar_large": 0.9938,
|
| 574 |
+
"eval_mar_medium": 0.871,
|
| 575 |
+
"eval_mar_small": 0.6935,
|
| 576 |
+
"eval_runtime": 302.8692,
|
| 577 |
+
"eval_samples_per_second": 0.436,
|
| 578 |
+
"eval_steps_per_second": 0.056,
|
| 579 |
+
"step": 1425
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 19.74074074074074,
|
| 583 |
+
"grad_norm": 524.6016845703125,
|
| 584 |
+
"learning_rate": 9.124060015425206e-10,
|
| 585 |
+
"loss": 6.7621,
|
| 586 |
+
"step": 1480
|
| 587 |
+
},
|
| 588 |
+
{
|
| 589 |
+
"epoch": 19.74074074074074,
|
| 590 |
+
"eval_classes": 0,
|
| 591 |
+
"eval_loss": 6.952253341674805,
|
| 592 |
+
"eval_map": 0.7683,
|
| 593 |
+
"eval_map_50": 0.9698,
|
| 594 |
+
"eval_map_75": 0.9012,
|
| 595 |
+
"eval_map_large": 0.9911,
|
| 596 |
+
"eval_map_medium": 0.8012,
|
| 597 |
+
"eval_map_per_class": 0.7683,
|
| 598 |
+
"eval_map_small": 0.6166,
|
| 599 |
+
"eval_mar_1": 0.0414,
|
| 600 |
+
"eval_mar_10": 0.3998,
|
| 601 |
+
"eval_mar_100": 0.8409,
|
| 602 |
+
"eval_mar_100_per_class": 0.8409,
|
| 603 |
+
"eval_mar_large": 0.9938,
|
| 604 |
+
"eval_mar_medium": 0.8711,
|
| 605 |
+
"eval_mar_small": 0.6954,
|
| 606 |
+
"eval_runtime": 304.7146,
|
| 607 |
+
"eval_samples_per_second": 0.433,
|
| 608 |
+
"eval_steps_per_second": 0.056,
|
| 609 |
+
"step": 1480
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 19.74074074074074,
|
| 613 |
+
"step": 1480,
|
| 614 |
+
"total_flos": 5.235456347568341e+18,
|
| 615 |
+
"train_loss": 8.241098208040805,
|
| 616 |
+
"train_runtime": 50547.6535,
|
| 617 |
+
"train_samples_per_second": 0.47,
|
| 618 |
+
"train_steps_per_second": 0.029
|
| 619 |
+
}
|
| 620 |
+
],
|
| 621 |
+
"logging_steps": 500,
|
| 622 |
+
"max_steps": 1480,
|
| 623 |
+
"num_input_tokens_seen": 0,
|
| 624 |
+
"num_train_epochs": 20,
|
| 625 |
+
"save_steps": 500,
|
| 626 |
+
"stateful_callbacks": {
|
| 627 |
+
"TrainerControl": {
|
| 628 |
+
"args": {
|
| 629 |
+
"should_epoch_stop": false,
|
| 630 |
+
"should_evaluate": false,
|
| 631 |
+
"should_log": false,
|
| 632 |
+
"should_save": true,
|
| 633 |
+
"should_training_stop": true
|
| 634 |
+
},
|
| 635 |
+
"attributes": {}
|
| 636 |
+
}
|
| 637 |
+
},
|
| 638 |
+
"total_flos": 5.235456347568341e+18,
|
| 639 |
+
"train_batch_size": 4,
|
| 640 |
+
"trial_name": null,
|
| 641 |
+
"trial_params": null
|
| 642 |
+
}
|