End of training
Browse files- README.md +16 -14
- all_results.json +26 -0
- eval_results.json +21 -0
- train_results.json +8 -0
- trainer_state.json +738 -0
README.md
CHANGED
|
@@ -3,6 +3,8 @@ library_name: transformers
|
|
| 3 |
license: apache-2.0
|
| 4 |
base_model: ustc-community/dfine-small-coco
|
| 5 |
tags:
|
|
|
|
|
|
|
| 6 |
- generated_from_trainer
|
| 7 |
model-index:
|
| 8 |
- name: red-squirrel-detector
|
|
@@ -14,23 +16,23 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 14 |
|
| 15 |
# red-squirrel-detector
|
| 16 |
|
| 17 |
-
This model is a fine-tuned version of [ustc-community/dfine-small-coco](https://huggingface.co/ustc-community/dfine-small-coco) on
|
| 18 |
It achieves the following results on the evaluation set:
|
| 19 |
-
- Loss: 0.
|
| 20 |
-
- Map: 0.
|
| 21 |
-
- Map 50: 0.
|
| 22 |
-
- Map 75: 0.
|
| 23 |
- Map Small: 0.0
|
| 24 |
-
- Map Medium: 0.
|
| 25 |
-
- Map Large: 0.
|
| 26 |
-
- Mar 1: 0.
|
| 27 |
-
- Mar 10: 0.
|
| 28 |
-
- Mar 100: 0.
|
| 29 |
- Mar Small: 0.0
|
| 30 |
-
- Mar Medium: 0.
|
| 31 |
-
- Mar Large: 0.
|
| 32 |
-
- Map Class 0: 0.
|
| 33 |
-
- Mar 100 Class 0: 0.
|
| 34 |
|
| 35 |
## Model description
|
| 36 |
|
|
|
|
| 3 |
license: apache-2.0
|
| 4 |
base_model: ustc-community/dfine-small-coco
|
| 5 |
tags:
|
| 6 |
+
- object-detection
|
| 7 |
+
- vision
|
| 8 |
- generated_from_trainer
|
| 9 |
model-index:
|
| 10 |
- name: red-squirrel-detector
|
|
|
|
| 16 |
|
| 17 |
# red-squirrel-detector
|
| 18 |
|
| 19 |
+
This model is a fine-tuned version of [ustc-community/dfine-small-coco](https://huggingface.co/ustc-community/dfine-small-coco) on the davanstrien/squirrel-cam-labeled dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
+
- Loss: 0.7587
|
| 22 |
+
- Map: 0.8614
|
| 23 |
+
- Map 50: 0.9087
|
| 24 |
+
- Map 75: 0.8839
|
| 25 |
- Map Small: 0.0
|
| 26 |
+
- Map Medium: 0.63
|
| 27 |
+
- Map Large: 0.8901
|
| 28 |
+
- Mar 1: 0.7528
|
| 29 |
+
- Mar 10: 0.8876
|
| 30 |
+
- Mar 100: 0.9441
|
| 31 |
- Mar Small: 0.0
|
| 32 |
+
- Mar Medium: 0.7769
|
| 33 |
+
- Mar Large: 0.9653
|
| 34 |
+
- Map Class 0: 0.8614
|
| 35 |
+
- Mar 100 Class 0: 0.9441
|
| 36 |
|
| 37 |
## Model description
|
| 38 |
|
all_results.json
ADDED
|
@@ -0,0 +1,26 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 30.0,
|
| 3 |
+
"eval_loss": 0.7586857080459595,
|
| 4 |
+
"eval_map": 0.8614,
|
| 5 |
+
"eval_map_50": 0.9087,
|
| 6 |
+
"eval_map_75": 0.8839,
|
| 7 |
+
"eval_map_class_0": 0.8614,
|
| 8 |
+
"eval_map_large": 0.8901,
|
| 9 |
+
"eval_map_medium": 0.63,
|
| 10 |
+
"eval_map_small": 0.0,
|
| 11 |
+
"eval_mar_1": 0.7528,
|
| 12 |
+
"eval_mar_10": 0.8876,
|
| 13 |
+
"eval_mar_100": 0.9441,
|
| 14 |
+
"eval_mar_100_class_0": 0.9441,
|
| 15 |
+
"eval_mar_large": 0.9653,
|
| 16 |
+
"eval_mar_medium": 0.7769,
|
| 17 |
+
"eval_mar_small": 0.0,
|
| 18 |
+
"eval_runtime": 6.4866,
|
| 19 |
+
"eval_samples_per_second": 19.733,
|
| 20 |
+
"eval_steps_per_second": 2.467,
|
| 21 |
+
"total_flos": 1.6368835442688e+18,
|
| 22 |
+
"train_loss": 11.706639954283997,
|
| 23 |
+
"train_runtime": 1986.2673,
|
| 24 |
+
"train_samples_per_second": 10.95,
|
| 25 |
+
"train_steps_per_second": 1.374
|
| 26 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,21 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 30.0,
|
| 3 |
+
"eval_loss": 0.7586857080459595,
|
| 4 |
+
"eval_map": 0.8614,
|
| 5 |
+
"eval_map_50": 0.9087,
|
| 6 |
+
"eval_map_75": 0.8839,
|
| 7 |
+
"eval_map_class_0": 0.8614,
|
| 8 |
+
"eval_map_large": 0.8901,
|
| 9 |
+
"eval_map_medium": 0.63,
|
| 10 |
+
"eval_map_small": 0.0,
|
| 11 |
+
"eval_mar_1": 0.7528,
|
| 12 |
+
"eval_mar_10": 0.8876,
|
| 13 |
+
"eval_mar_100": 0.9441,
|
| 14 |
+
"eval_mar_100_class_0": 0.9441,
|
| 15 |
+
"eval_mar_large": 0.9653,
|
| 16 |
+
"eval_mar_medium": 0.7769,
|
| 17 |
+
"eval_mar_small": 0.0,
|
| 18 |
+
"eval_runtime": 6.4866,
|
| 19 |
+
"eval_samples_per_second": 19.733,
|
| 20 |
+
"eval_steps_per_second": 2.467
|
| 21 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 30.0,
|
| 3 |
+
"total_flos": 1.6368835442688e+18,
|
| 4 |
+
"train_loss": 11.706639954283997,
|
| 5 |
+
"train_runtime": 1986.2673,
|
| 6 |
+
"train_samples_per_second": 10.95,
|
| 7 |
+
"train_steps_per_second": 1.374
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,738 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": 2184,
|
| 3 |
+
"best_metric": 0.862,
|
| 4 |
+
"best_model_checkpoint": "dfine_squirrel/checkpoint-2184",
|
| 5 |
+
"epoch": 30.0,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 2730,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 1.0,
|
| 14 |
+
"eval_loss": 2.367178440093994,
|
| 15 |
+
"eval_map": 0.1766,
|
| 16 |
+
"eval_map_50": 0.23,
|
| 17 |
+
"eval_map_75": 0.1862,
|
| 18 |
+
"eval_map_class_0": 0.1766,
|
| 19 |
+
"eval_map_large": 0.1989,
|
| 20 |
+
"eval_map_medium": 0.0031,
|
| 21 |
+
"eval_map_small": 0.0,
|
| 22 |
+
"eval_mar_1": 0.3422,
|
| 23 |
+
"eval_mar_10": 0.5248,
|
| 24 |
+
"eval_mar_100": 0.6224,
|
| 25 |
+
"eval_mar_100_class_0": 0.6224,
|
| 26 |
+
"eval_mar_large": 0.651,
|
| 27 |
+
"eval_mar_medium": 0.3462,
|
| 28 |
+
"eval_mar_small": 0.0,
|
| 29 |
+
"eval_runtime": 6.6845,
|
| 30 |
+
"eval_samples_per_second": 19.149,
|
| 31 |
+
"eval_steps_per_second": 2.394,
|
| 32 |
+
"step": 91
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 2.0,
|
| 36 |
+
"eval_loss": 1.0650697946548462,
|
| 37 |
+
"eval_map": 0.7244,
|
| 38 |
+
"eval_map_50": 0.8127,
|
| 39 |
+
"eval_map_75": 0.7856,
|
| 40 |
+
"eval_map_class_0": 0.7244,
|
| 41 |
+
"eval_map_large": 0.7538,
|
| 42 |
+
"eval_map_medium": 0.4261,
|
| 43 |
+
"eval_map_small": 0.0,
|
| 44 |
+
"eval_mar_1": 0.6901,
|
| 45 |
+
"eval_mar_10": 0.7888,
|
| 46 |
+
"eval_mar_100": 0.8596,
|
| 47 |
+
"eval_mar_100_class_0": 0.8596,
|
| 48 |
+
"eval_mar_large": 0.885,
|
| 49 |
+
"eval_mar_medium": 0.6385,
|
| 50 |
+
"eval_mar_small": 0.0,
|
| 51 |
+
"eval_runtime": 6.633,
|
| 52 |
+
"eval_samples_per_second": 19.298,
|
| 53 |
+
"eval_steps_per_second": 2.412,
|
| 54 |
+
"step": 182
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 3.0,
|
| 58 |
+
"eval_loss": 0.8899274468421936,
|
| 59 |
+
"eval_map": 0.7577,
|
| 60 |
+
"eval_map_50": 0.8321,
|
| 61 |
+
"eval_map_75": 0.7881,
|
| 62 |
+
"eval_map_class_0": 0.7577,
|
| 63 |
+
"eval_map_large": 0.7986,
|
| 64 |
+
"eval_map_medium": 0.4732,
|
| 65 |
+
"eval_map_small": 0.0,
|
| 66 |
+
"eval_mar_1": 0.7137,
|
| 67 |
+
"eval_mar_10": 0.8311,
|
| 68 |
+
"eval_mar_100": 0.8863,
|
| 69 |
+
"eval_mar_100_class_0": 0.8863,
|
| 70 |
+
"eval_mar_large": 0.9109,
|
| 71 |
+
"eval_mar_medium": 0.6769,
|
| 72 |
+
"eval_mar_small": 0.0,
|
| 73 |
+
"eval_runtime": 6.5407,
|
| 74 |
+
"eval_samples_per_second": 19.57,
|
| 75 |
+
"eval_steps_per_second": 2.446,
|
| 76 |
+
"step": 273
|
| 77 |
+
},
|
| 78 |
+
{
|
| 79 |
+
"epoch": 4.0,
|
| 80 |
+
"eval_loss": 0.8310188055038452,
|
| 81 |
+
"eval_map": 0.7773,
|
| 82 |
+
"eval_map_50": 0.8432,
|
| 83 |
+
"eval_map_75": 0.8045,
|
| 84 |
+
"eval_map_class_0": 0.7773,
|
| 85 |
+
"eval_map_large": 0.814,
|
| 86 |
+
"eval_map_medium": 0.586,
|
| 87 |
+
"eval_map_small": 0.0,
|
| 88 |
+
"eval_mar_1": 0.7124,
|
| 89 |
+
"eval_mar_10": 0.841,
|
| 90 |
+
"eval_mar_100": 0.8894,
|
| 91 |
+
"eval_mar_100_class_0": 0.8894,
|
| 92 |
+
"eval_mar_large": 0.9122,
|
| 93 |
+
"eval_mar_medium": 0.7,
|
| 94 |
+
"eval_mar_small": 0.0,
|
| 95 |
+
"eval_runtime": 6.599,
|
| 96 |
+
"eval_samples_per_second": 19.397,
|
| 97 |
+
"eval_steps_per_second": 2.425,
|
| 98 |
+
"step": 364
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 5.0,
|
| 102 |
+
"eval_loss": 0.8332623243331909,
|
| 103 |
+
"eval_map": 0.7792,
|
| 104 |
+
"eval_map_50": 0.8513,
|
| 105 |
+
"eval_map_75": 0.8058,
|
| 106 |
+
"eval_map_class_0": 0.7792,
|
| 107 |
+
"eval_map_large": 0.8197,
|
| 108 |
+
"eval_map_medium": 0.6129,
|
| 109 |
+
"eval_map_small": 0.0,
|
| 110 |
+
"eval_mar_1": 0.718,
|
| 111 |
+
"eval_mar_10": 0.8317,
|
| 112 |
+
"eval_mar_100": 0.9031,
|
| 113 |
+
"eval_mar_100_class_0": 0.9031,
|
| 114 |
+
"eval_mar_large": 0.9265,
|
| 115 |
+
"eval_mar_medium": 0.7077,
|
| 116 |
+
"eval_mar_small": 0.0,
|
| 117 |
+
"eval_runtime": 6.61,
|
| 118 |
+
"eval_samples_per_second": 19.365,
|
| 119 |
+
"eval_steps_per_second": 2.421,
|
| 120 |
+
"step": 455
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 5.4945054945054945,
|
| 124 |
+
"grad_norm": 45.534889221191406,
|
| 125 |
+
"learning_rate": 4.086080586080586e-05,
|
| 126 |
+
"loss": 27.38396484375,
|
| 127 |
+
"step": 500
|
| 128 |
+
},
|
| 129 |
+
{
|
| 130 |
+
"epoch": 6.0,
|
| 131 |
+
"eval_loss": 0.8361273407936096,
|
| 132 |
+
"eval_map": 0.8008,
|
| 133 |
+
"eval_map_50": 0.8684,
|
| 134 |
+
"eval_map_75": 0.8296,
|
| 135 |
+
"eval_map_class_0": 0.8008,
|
| 136 |
+
"eval_map_large": 0.8348,
|
| 137 |
+
"eval_map_medium": 0.5788,
|
| 138 |
+
"eval_map_small": 0.0,
|
| 139 |
+
"eval_mar_1": 0.723,
|
| 140 |
+
"eval_mar_10": 0.8609,
|
| 141 |
+
"eval_mar_100": 0.9124,
|
| 142 |
+
"eval_mar_100_class_0": 0.9124,
|
| 143 |
+
"eval_mar_large": 0.9367,
|
| 144 |
+
"eval_mar_medium": 0.7077,
|
| 145 |
+
"eval_mar_small": 0.0,
|
| 146 |
+
"eval_runtime": 6.591,
|
| 147 |
+
"eval_samples_per_second": 19.42,
|
| 148 |
+
"eval_steps_per_second": 2.428,
|
| 149 |
+
"step": 546
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 7.0,
|
| 153 |
+
"eval_loss": 0.8077548146247864,
|
| 154 |
+
"eval_map": 0.8,
|
| 155 |
+
"eval_map_50": 0.8587,
|
| 156 |
+
"eval_map_75": 0.8191,
|
| 157 |
+
"eval_map_class_0": 0.8,
|
| 158 |
+
"eval_map_large": 0.8365,
|
| 159 |
+
"eval_map_medium": 0.5968,
|
| 160 |
+
"eval_map_small": 0.0,
|
| 161 |
+
"eval_mar_1": 0.7224,
|
| 162 |
+
"eval_mar_10": 0.8739,
|
| 163 |
+
"eval_mar_100": 0.9211,
|
| 164 |
+
"eval_mar_100_class_0": 0.9211,
|
| 165 |
+
"eval_mar_large": 0.9442,
|
| 166 |
+
"eval_mar_medium": 0.7308,
|
| 167 |
+
"eval_mar_small": 0.0,
|
| 168 |
+
"eval_runtime": 6.4826,
|
| 169 |
+
"eval_samples_per_second": 19.745,
|
| 170 |
+
"eval_steps_per_second": 2.468,
|
| 171 |
+
"step": 637
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 8.0,
|
| 175 |
+
"eval_loss": 0.8231956958770752,
|
| 176 |
+
"eval_map": 0.8173,
|
| 177 |
+
"eval_map_50": 0.8877,
|
| 178 |
+
"eval_map_75": 0.837,
|
| 179 |
+
"eval_map_class_0": 0.8173,
|
| 180 |
+
"eval_map_large": 0.8476,
|
| 181 |
+
"eval_map_medium": 0.6199,
|
| 182 |
+
"eval_map_small": 0.0,
|
| 183 |
+
"eval_mar_1": 0.7205,
|
| 184 |
+
"eval_mar_10": 0.8677,
|
| 185 |
+
"eval_mar_100": 0.9161,
|
| 186 |
+
"eval_mar_100_class_0": 0.9161,
|
| 187 |
+
"eval_mar_large": 0.9415,
|
| 188 |
+
"eval_mar_medium": 0.7,
|
| 189 |
+
"eval_mar_small": 0.0,
|
| 190 |
+
"eval_runtime": 6.4352,
|
| 191 |
+
"eval_samples_per_second": 19.891,
|
| 192 |
+
"eval_steps_per_second": 2.486,
|
| 193 |
+
"step": 728
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 9.0,
|
| 197 |
+
"eval_loss": 0.7746098041534424,
|
| 198 |
+
"eval_map": 0.8268,
|
| 199 |
+
"eval_map_50": 0.8863,
|
| 200 |
+
"eval_map_75": 0.8541,
|
| 201 |
+
"eval_map_class_0": 0.8268,
|
| 202 |
+
"eval_map_large": 0.8577,
|
| 203 |
+
"eval_map_medium": 0.6029,
|
| 204 |
+
"eval_map_small": 0.0,
|
| 205 |
+
"eval_mar_1": 0.7211,
|
| 206 |
+
"eval_mar_10": 0.8702,
|
| 207 |
+
"eval_mar_100": 0.9224,
|
| 208 |
+
"eval_mar_100_class_0": 0.9224,
|
| 209 |
+
"eval_mar_large": 0.951,
|
| 210 |
+
"eval_mar_medium": 0.6692,
|
| 211 |
+
"eval_mar_small": 0.0,
|
| 212 |
+
"eval_runtime": 6.6084,
|
| 213 |
+
"eval_samples_per_second": 19.369,
|
| 214 |
+
"eval_steps_per_second": 2.421,
|
| 215 |
+
"step": 819
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 10.0,
|
| 219 |
+
"eval_loss": 0.8223000764846802,
|
| 220 |
+
"eval_map": 0.8324,
|
| 221 |
+
"eval_map_50": 0.8875,
|
| 222 |
+
"eval_map_75": 0.8682,
|
| 223 |
+
"eval_map_class_0": 0.8324,
|
| 224 |
+
"eval_map_large": 0.8586,
|
| 225 |
+
"eval_map_medium": 0.6198,
|
| 226 |
+
"eval_map_small": 0.0,
|
| 227 |
+
"eval_mar_1": 0.7311,
|
| 228 |
+
"eval_mar_10": 0.8801,
|
| 229 |
+
"eval_mar_100": 0.9193,
|
| 230 |
+
"eval_mar_100_class_0": 0.9193,
|
| 231 |
+
"eval_mar_large": 0.9456,
|
| 232 |
+
"eval_mar_medium": 0.6923,
|
| 233 |
+
"eval_mar_small": 0.0,
|
| 234 |
+
"eval_runtime": 6.6591,
|
| 235 |
+
"eval_samples_per_second": 19.222,
|
| 236 |
+
"eval_steps_per_second": 2.403,
|
| 237 |
+
"step": 910
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 10.989010989010989,
|
| 241 |
+
"grad_norm": 34.31671142578125,
|
| 242 |
+
"learning_rate": 3.17032967032967e-05,
|
| 243 |
+
"loss": 9.0921865234375,
|
| 244 |
+
"step": 1000
|
| 245 |
+
},
|
| 246 |
+
{
|
| 247 |
+
"epoch": 11.0,
|
| 248 |
+
"eval_loss": 0.7678491473197937,
|
| 249 |
+
"eval_map": 0.8327,
|
| 250 |
+
"eval_map_50": 0.8851,
|
| 251 |
+
"eval_map_75": 0.8689,
|
| 252 |
+
"eval_map_class_0": 0.8327,
|
| 253 |
+
"eval_map_large": 0.8634,
|
| 254 |
+
"eval_map_medium": 0.5991,
|
| 255 |
+
"eval_map_small": 0.0,
|
| 256 |
+
"eval_mar_1": 0.7422,
|
| 257 |
+
"eval_mar_10": 0.8671,
|
| 258 |
+
"eval_mar_100": 0.9255,
|
| 259 |
+
"eval_mar_100_class_0": 0.9255,
|
| 260 |
+
"eval_mar_large": 0.9531,
|
| 261 |
+
"eval_mar_medium": 0.6846,
|
| 262 |
+
"eval_mar_small": 0.0,
|
| 263 |
+
"eval_runtime": 6.6366,
|
| 264 |
+
"eval_samples_per_second": 19.287,
|
| 265 |
+
"eval_steps_per_second": 2.411,
|
| 266 |
+
"step": 1001
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 12.0,
|
| 270 |
+
"eval_loss": 0.7850316166877747,
|
| 271 |
+
"eval_map": 0.8328,
|
| 272 |
+
"eval_map_50": 0.8888,
|
| 273 |
+
"eval_map_75": 0.8545,
|
| 274 |
+
"eval_map_class_0": 0.8328,
|
| 275 |
+
"eval_map_large": 0.8635,
|
| 276 |
+
"eval_map_medium": 0.6134,
|
| 277 |
+
"eval_map_small": 0.0,
|
| 278 |
+
"eval_mar_1": 0.7366,
|
| 279 |
+
"eval_mar_10": 0.8752,
|
| 280 |
+
"eval_mar_100": 0.9335,
|
| 281 |
+
"eval_mar_100_class_0": 0.9335,
|
| 282 |
+
"eval_mar_large": 0.9585,
|
| 283 |
+
"eval_mar_medium": 0.7231,
|
| 284 |
+
"eval_mar_small": 0.0,
|
| 285 |
+
"eval_runtime": 6.866,
|
| 286 |
+
"eval_samples_per_second": 18.643,
|
| 287 |
+
"eval_steps_per_second": 2.33,
|
| 288 |
+
"step": 1092
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 13.0,
|
| 292 |
+
"eval_loss": 0.8330104351043701,
|
| 293 |
+
"eval_map": 0.8294,
|
| 294 |
+
"eval_map_50": 0.8882,
|
| 295 |
+
"eval_map_75": 0.8485,
|
| 296 |
+
"eval_map_class_0": 0.8294,
|
| 297 |
+
"eval_map_large": 0.8585,
|
| 298 |
+
"eval_map_medium": 0.583,
|
| 299 |
+
"eval_map_small": 0.0,
|
| 300 |
+
"eval_mar_1": 0.7416,
|
| 301 |
+
"eval_mar_10": 0.8689,
|
| 302 |
+
"eval_mar_100": 0.9267,
|
| 303 |
+
"eval_mar_100_class_0": 0.9267,
|
| 304 |
+
"eval_mar_large": 0.9503,
|
| 305 |
+
"eval_mar_medium": 0.7308,
|
| 306 |
+
"eval_mar_small": 0.0,
|
| 307 |
+
"eval_runtime": 6.6667,
|
| 308 |
+
"eval_samples_per_second": 19.2,
|
| 309 |
+
"eval_steps_per_second": 2.4,
|
| 310 |
+
"step": 1183
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 14.0,
|
| 314 |
+
"eval_loss": 0.7800494432449341,
|
| 315 |
+
"eval_map": 0.8399,
|
| 316 |
+
"eval_map_50": 0.8977,
|
| 317 |
+
"eval_map_75": 0.8761,
|
| 318 |
+
"eval_map_class_0": 0.8399,
|
| 319 |
+
"eval_map_large": 0.87,
|
| 320 |
+
"eval_map_medium": 0.5989,
|
| 321 |
+
"eval_map_small": 0.0056,
|
| 322 |
+
"eval_mar_1": 0.741,
|
| 323 |
+
"eval_mar_10": 0.8758,
|
| 324 |
+
"eval_mar_100": 0.9348,
|
| 325 |
+
"eval_mar_100_class_0": 0.9348,
|
| 326 |
+
"eval_mar_large": 0.9517,
|
| 327 |
+
"eval_mar_medium": 0.7692,
|
| 328 |
+
"eval_mar_small": 0.6,
|
| 329 |
+
"eval_runtime": 6.5377,
|
| 330 |
+
"eval_samples_per_second": 19.579,
|
| 331 |
+
"eval_steps_per_second": 2.447,
|
| 332 |
+
"step": 1274
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 15.0,
|
| 336 |
+
"eval_loss": 0.7594577074050903,
|
| 337 |
+
"eval_map": 0.8437,
|
| 338 |
+
"eval_map_50": 0.8993,
|
| 339 |
+
"eval_map_75": 0.8771,
|
| 340 |
+
"eval_map_class_0": 0.8437,
|
| 341 |
+
"eval_map_large": 0.8731,
|
| 342 |
+
"eval_map_medium": 0.6092,
|
| 343 |
+
"eval_map_small": 0.0061,
|
| 344 |
+
"eval_mar_1": 0.7509,
|
| 345 |
+
"eval_mar_10": 0.8925,
|
| 346 |
+
"eval_mar_100": 0.9466,
|
| 347 |
+
"eval_mar_100_class_0": 0.9466,
|
| 348 |
+
"eval_mar_large": 0.9633,
|
| 349 |
+
"eval_mar_medium": 0.7846,
|
| 350 |
+
"eval_mar_small": 0.6,
|
| 351 |
+
"eval_runtime": 6.6179,
|
| 352 |
+
"eval_samples_per_second": 19.341,
|
| 353 |
+
"eval_steps_per_second": 2.418,
|
| 354 |
+
"step": 1365
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 16.0,
|
| 358 |
+
"eval_loss": 0.7757935523986816,
|
| 359 |
+
"eval_map": 0.843,
|
| 360 |
+
"eval_map_50": 0.9059,
|
| 361 |
+
"eval_map_75": 0.8657,
|
| 362 |
+
"eval_map_class_0": 0.843,
|
| 363 |
+
"eval_map_large": 0.8713,
|
| 364 |
+
"eval_map_medium": 0.6186,
|
| 365 |
+
"eval_map_small": 0.0,
|
| 366 |
+
"eval_mar_1": 0.7472,
|
| 367 |
+
"eval_mar_10": 0.8783,
|
| 368 |
+
"eval_mar_100": 0.9385,
|
| 369 |
+
"eval_mar_100_class_0": 0.9385,
|
| 370 |
+
"eval_mar_large": 0.9633,
|
| 371 |
+
"eval_mar_medium": 0.7308,
|
| 372 |
+
"eval_mar_small": 0.0,
|
| 373 |
+
"eval_runtime": 6.6266,
|
| 374 |
+
"eval_samples_per_second": 19.316,
|
| 375 |
+
"eval_steps_per_second": 2.415,
|
| 376 |
+
"step": 1456
|
| 377 |
+
},
|
| 378 |
+
{
|
| 379 |
+
"epoch": 16.483516483516482,
|
| 380 |
+
"grad_norm": 38.72512435913086,
|
| 381 |
+
"learning_rate": 2.2545787545787548e-05,
|
| 382 |
+
"loss": 8.323384765625,
|
| 383 |
+
"step": 1500
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 17.0,
|
| 387 |
+
"eval_loss": 0.8078031539916992,
|
| 388 |
+
"eval_map": 0.844,
|
| 389 |
+
"eval_map_50": 0.9017,
|
| 390 |
+
"eval_map_75": 0.8816,
|
| 391 |
+
"eval_map_class_0": 0.844,
|
| 392 |
+
"eval_map_large": 0.8754,
|
| 393 |
+
"eval_map_medium": 0.6091,
|
| 394 |
+
"eval_map_small": 0.009,
|
| 395 |
+
"eval_mar_1": 0.7453,
|
| 396 |
+
"eval_mar_10": 0.8957,
|
| 397 |
+
"eval_mar_100": 0.9366,
|
| 398 |
+
"eval_mar_100_class_0": 0.9366,
|
| 399 |
+
"eval_mar_large": 0.9592,
|
| 400 |
+
"eval_mar_medium": 0.7077,
|
| 401 |
+
"eval_mar_small": 0.6,
|
| 402 |
+
"eval_runtime": 6.5387,
|
| 403 |
+
"eval_samples_per_second": 19.576,
|
| 404 |
+
"eval_steps_per_second": 2.447,
|
| 405 |
+
"step": 1547
|
| 406 |
+
},
|
| 407 |
+
{
|
| 408 |
+
"epoch": 18.0,
|
| 409 |
+
"eval_loss": 0.771867036819458,
|
| 410 |
+
"eval_map": 0.8479,
|
| 411 |
+
"eval_map_50": 0.9069,
|
| 412 |
+
"eval_map_75": 0.8728,
|
| 413 |
+
"eval_map_class_0": 0.8479,
|
| 414 |
+
"eval_map_large": 0.8785,
|
| 415 |
+
"eval_map_medium": 0.6086,
|
| 416 |
+
"eval_map_small": 0.0,
|
| 417 |
+
"eval_mar_1": 0.7522,
|
| 418 |
+
"eval_mar_10": 0.8882,
|
| 419 |
+
"eval_mar_100": 0.9273,
|
| 420 |
+
"eval_mar_100_class_0": 0.9273,
|
| 421 |
+
"eval_mar_large": 0.9531,
|
| 422 |
+
"eval_mar_medium": 0.7077,
|
| 423 |
+
"eval_mar_small": 0.0,
|
| 424 |
+
"eval_runtime": 6.8246,
|
| 425 |
+
"eval_samples_per_second": 18.756,
|
| 426 |
+
"eval_steps_per_second": 2.344,
|
| 427 |
+
"step": 1638
|
| 428 |
+
},
|
| 429 |
+
{
|
| 430 |
+
"epoch": 19.0,
|
| 431 |
+
"eval_loss": 0.756177544593811,
|
| 432 |
+
"eval_map": 0.8552,
|
| 433 |
+
"eval_map_50": 0.9105,
|
| 434 |
+
"eval_map_75": 0.8779,
|
| 435 |
+
"eval_map_class_0": 0.8552,
|
| 436 |
+
"eval_map_large": 0.8853,
|
| 437 |
+
"eval_map_medium": 0.6193,
|
| 438 |
+
"eval_map_small": 0.0,
|
| 439 |
+
"eval_mar_1": 0.7509,
|
| 440 |
+
"eval_mar_10": 0.8882,
|
| 441 |
+
"eval_mar_100": 0.9323,
|
| 442 |
+
"eval_mar_100_class_0": 0.9323,
|
| 443 |
+
"eval_mar_large": 0.9585,
|
| 444 |
+
"eval_mar_medium": 0.7077,
|
| 445 |
+
"eval_mar_small": 0.0,
|
| 446 |
+
"eval_runtime": 6.5483,
|
| 447 |
+
"eval_samples_per_second": 19.547,
|
| 448 |
+
"eval_steps_per_second": 2.443,
|
| 449 |
+
"step": 1729
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 20.0,
|
| 453 |
+
"eval_loss": 0.8165637850761414,
|
| 454 |
+
"eval_map": 0.8478,
|
| 455 |
+
"eval_map_50": 0.9076,
|
| 456 |
+
"eval_map_75": 0.8774,
|
| 457 |
+
"eval_map_class_0": 0.8478,
|
| 458 |
+
"eval_map_large": 0.8765,
|
| 459 |
+
"eval_map_medium": 0.6428,
|
| 460 |
+
"eval_map_small": 0.0,
|
| 461 |
+
"eval_mar_1": 0.7528,
|
| 462 |
+
"eval_mar_10": 0.8988,
|
| 463 |
+
"eval_mar_100": 0.9379,
|
| 464 |
+
"eval_mar_100_class_0": 0.9379,
|
| 465 |
+
"eval_mar_large": 0.9633,
|
| 466 |
+
"eval_mar_medium": 0.7231,
|
| 467 |
+
"eval_mar_small": 0.0,
|
| 468 |
+
"eval_runtime": 6.5543,
|
| 469 |
+
"eval_samples_per_second": 19.529,
|
| 470 |
+
"eval_steps_per_second": 2.441,
|
| 471 |
+
"step": 1820
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 21.0,
|
| 475 |
+
"eval_loss": 0.7615771293640137,
|
| 476 |
+
"eval_map": 0.8519,
|
| 477 |
+
"eval_map_50": 0.9053,
|
| 478 |
+
"eval_map_75": 0.8843,
|
| 479 |
+
"eval_map_class_0": 0.8519,
|
| 480 |
+
"eval_map_large": 0.8818,
|
| 481 |
+
"eval_map_medium": 0.6332,
|
| 482 |
+
"eval_map_small": 0.0,
|
| 483 |
+
"eval_mar_1": 0.7578,
|
| 484 |
+
"eval_mar_10": 0.8919,
|
| 485 |
+
"eval_mar_100": 0.9385,
|
| 486 |
+
"eval_mar_100_class_0": 0.9385,
|
| 487 |
+
"eval_mar_large": 0.9646,
|
| 488 |
+
"eval_mar_medium": 0.7154,
|
| 489 |
+
"eval_mar_small": 0.0,
|
| 490 |
+
"eval_runtime": 6.6955,
|
| 491 |
+
"eval_samples_per_second": 19.117,
|
| 492 |
+
"eval_steps_per_second": 2.39,
|
| 493 |
+
"step": 1911
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 21.978021978021978,
|
| 497 |
+
"grad_norm": 37.53008270263672,
|
| 498 |
+
"learning_rate": 1.3388278388278389e-05,
|
| 499 |
+
"loss": 7.99486328125,
|
| 500 |
+
"step": 2000
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 22.0,
|
| 504 |
+
"eval_loss": 0.7444530129432678,
|
| 505 |
+
"eval_map": 0.857,
|
| 506 |
+
"eval_map_50": 0.9102,
|
| 507 |
+
"eval_map_75": 0.8872,
|
| 508 |
+
"eval_map_class_0": 0.857,
|
| 509 |
+
"eval_map_large": 0.8856,
|
| 510 |
+
"eval_map_medium": 0.6327,
|
| 511 |
+
"eval_map_small": 0.0,
|
| 512 |
+
"eval_mar_1": 0.7497,
|
| 513 |
+
"eval_mar_10": 0.8839,
|
| 514 |
+
"eval_mar_100": 0.9354,
|
| 515 |
+
"eval_mar_100_class_0": 0.9354,
|
| 516 |
+
"eval_mar_large": 0.9605,
|
| 517 |
+
"eval_mar_medium": 0.7231,
|
| 518 |
+
"eval_mar_small": 0.0,
|
| 519 |
+
"eval_runtime": 6.5313,
|
| 520 |
+
"eval_samples_per_second": 19.598,
|
| 521 |
+
"eval_steps_per_second": 2.45,
|
| 522 |
+
"step": 2002
|
| 523 |
+
},
|
| 524 |
+
{
|
| 525 |
+
"epoch": 23.0,
|
| 526 |
+
"eval_loss": 0.7569313645362854,
|
| 527 |
+
"eval_map": 0.85,
|
| 528 |
+
"eval_map_50": 0.9003,
|
| 529 |
+
"eval_map_75": 0.8792,
|
| 530 |
+
"eval_map_class_0": 0.85,
|
| 531 |
+
"eval_map_large": 0.8802,
|
| 532 |
+
"eval_map_medium": 0.6308,
|
| 533 |
+
"eval_map_small": 0.0,
|
| 534 |
+
"eval_mar_1": 0.7516,
|
| 535 |
+
"eval_mar_10": 0.9019,
|
| 536 |
+
"eval_mar_100": 0.9366,
|
| 537 |
+
"eval_mar_100_class_0": 0.9366,
|
| 538 |
+
"eval_mar_large": 0.9626,
|
| 539 |
+
"eval_mar_medium": 0.7154,
|
| 540 |
+
"eval_mar_small": 0.0,
|
| 541 |
+
"eval_runtime": 6.5869,
|
| 542 |
+
"eval_samples_per_second": 19.433,
|
| 543 |
+
"eval_steps_per_second": 2.429,
|
| 544 |
+
"step": 2093
|
| 545 |
+
},
|
| 546 |
+
{
|
| 547 |
+
"epoch": 24.0,
|
| 548 |
+
"eval_loss": 0.755048930644989,
|
| 549 |
+
"eval_map": 0.862,
|
| 550 |
+
"eval_map_50": 0.9093,
|
| 551 |
+
"eval_map_75": 0.8837,
|
| 552 |
+
"eval_map_class_0": 0.862,
|
| 553 |
+
"eval_map_large": 0.8917,
|
| 554 |
+
"eval_map_medium": 0.6277,
|
| 555 |
+
"eval_map_small": 0.0,
|
| 556 |
+
"eval_mar_1": 0.7516,
|
| 557 |
+
"eval_mar_10": 0.8882,
|
| 558 |
+
"eval_mar_100": 0.9441,
|
| 559 |
+
"eval_mar_100_class_0": 0.9441,
|
| 560 |
+
"eval_mar_large": 0.9653,
|
| 561 |
+
"eval_mar_medium": 0.7769,
|
| 562 |
+
"eval_mar_small": 0.0,
|
| 563 |
+
"eval_runtime": 6.6153,
|
| 564 |
+
"eval_samples_per_second": 19.349,
|
| 565 |
+
"eval_steps_per_second": 2.419,
|
| 566 |
+
"step": 2184
|
| 567 |
+
},
|
| 568 |
+
{
|
| 569 |
+
"epoch": 25.0,
|
| 570 |
+
"eval_loss": 0.7434877753257751,
|
| 571 |
+
"eval_map": 0.8561,
|
| 572 |
+
"eval_map_50": 0.9053,
|
| 573 |
+
"eval_map_75": 0.8875,
|
| 574 |
+
"eval_map_class_0": 0.8561,
|
| 575 |
+
"eval_map_large": 0.885,
|
| 576 |
+
"eval_map_medium": 0.6371,
|
| 577 |
+
"eval_map_small": 0.0055,
|
| 578 |
+
"eval_mar_1": 0.7559,
|
| 579 |
+
"eval_mar_10": 0.8981,
|
| 580 |
+
"eval_mar_100": 0.9478,
|
| 581 |
+
"eval_mar_100_class_0": 0.9478,
|
| 582 |
+
"eval_mar_large": 0.9633,
|
| 583 |
+
"eval_mar_medium": 0.7923,
|
| 584 |
+
"eval_mar_small": 0.7,
|
| 585 |
+
"eval_runtime": 6.5533,
|
| 586 |
+
"eval_samples_per_second": 19.532,
|
| 587 |
+
"eval_steps_per_second": 2.442,
|
| 588 |
+
"step": 2275
|
| 589 |
+
},
|
| 590 |
+
{
|
| 591 |
+
"epoch": 26.0,
|
| 592 |
+
"eval_loss": 0.7752219438552856,
|
| 593 |
+
"eval_map": 0.8441,
|
| 594 |
+
"eval_map_50": 0.8961,
|
| 595 |
+
"eval_map_75": 0.8669,
|
| 596 |
+
"eval_map_class_0": 0.8441,
|
| 597 |
+
"eval_map_large": 0.8717,
|
| 598 |
+
"eval_map_medium": 0.6392,
|
| 599 |
+
"eval_map_small": 0.0,
|
| 600 |
+
"eval_mar_1": 0.7553,
|
| 601 |
+
"eval_mar_10": 0.8857,
|
| 602 |
+
"eval_mar_100": 0.9398,
|
| 603 |
+
"eval_mar_100_class_0": 0.9398,
|
| 604 |
+
"eval_mar_large": 0.9592,
|
| 605 |
+
"eval_mar_medium": 0.7923,
|
| 606 |
+
"eval_mar_small": 0.0,
|
| 607 |
+
"eval_runtime": 6.5969,
|
| 608 |
+
"eval_samples_per_second": 19.403,
|
| 609 |
+
"eval_steps_per_second": 2.425,
|
| 610 |
+
"step": 2366
|
| 611 |
+
},
|
| 612 |
+
{
|
| 613 |
+
"epoch": 27.0,
|
| 614 |
+
"eval_loss": 0.7504961490631104,
|
| 615 |
+
"eval_map": 0.8505,
|
| 616 |
+
"eval_map_50": 0.8966,
|
| 617 |
+
"eval_map_75": 0.8745,
|
| 618 |
+
"eval_map_class_0": 0.8505,
|
| 619 |
+
"eval_map_large": 0.88,
|
| 620 |
+
"eval_map_medium": 0.6337,
|
| 621 |
+
"eval_map_small": 0.0,
|
| 622 |
+
"eval_mar_1": 0.7565,
|
| 623 |
+
"eval_mar_10": 0.8857,
|
| 624 |
+
"eval_mar_100": 0.9441,
|
| 625 |
+
"eval_mar_100_class_0": 0.9441,
|
| 626 |
+
"eval_mar_large": 0.9639,
|
| 627 |
+
"eval_mar_medium": 0.7923,
|
| 628 |
+
"eval_mar_small": 0.0,
|
| 629 |
+
"eval_runtime": 6.5131,
|
| 630 |
+
"eval_samples_per_second": 19.653,
|
| 631 |
+
"eval_steps_per_second": 2.457,
|
| 632 |
+
"step": 2457
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 27.47252747252747,
|
| 636 |
+
"grad_norm": 72.2485580444336,
|
| 637 |
+
"learning_rate": 4.230769230769231e-06,
|
| 638 |
+
"loss": 7.68200732421875,
|
| 639 |
+
"step": 2500
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 28.0,
|
| 643 |
+
"eval_loss": 0.7449190616607666,
|
| 644 |
+
"eval_map": 0.8533,
|
| 645 |
+
"eval_map_50": 0.9012,
|
| 646 |
+
"eval_map_75": 0.8801,
|
| 647 |
+
"eval_map_class_0": 0.8533,
|
| 648 |
+
"eval_map_large": 0.8826,
|
| 649 |
+
"eval_map_medium": 0.627,
|
| 650 |
+
"eval_map_small": 0.0,
|
| 651 |
+
"eval_mar_1": 0.7553,
|
| 652 |
+
"eval_mar_10": 0.8901,
|
| 653 |
+
"eval_mar_100": 0.9416,
|
| 654 |
+
"eval_mar_100_class_0": 0.9416,
|
| 655 |
+
"eval_mar_large": 0.9626,
|
| 656 |
+
"eval_mar_medium": 0.7769,
|
| 657 |
+
"eval_mar_small": 0.0,
|
| 658 |
+
"eval_runtime": 6.6251,
|
| 659 |
+
"eval_samples_per_second": 19.32,
|
| 660 |
+
"eval_steps_per_second": 2.415,
|
| 661 |
+
"step": 2548
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 29.0,
|
| 665 |
+
"eval_loss": 0.7330126166343689,
|
| 666 |
+
"eval_map": 0.8577,
|
| 667 |
+
"eval_map_50": 0.9052,
|
| 668 |
+
"eval_map_75": 0.8842,
|
| 669 |
+
"eval_map_class_0": 0.8577,
|
| 670 |
+
"eval_map_large": 0.8874,
|
| 671 |
+
"eval_map_medium": 0.6278,
|
| 672 |
+
"eval_map_small": 0.0076,
|
| 673 |
+
"eval_mar_1": 0.7571,
|
| 674 |
+
"eval_mar_10": 0.8975,
|
| 675 |
+
"eval_mar_100": 0.9497,
|
| 676 |
+
"eval_mar_100_class_0": 0.9497,
|
| 677 |
+
"eval_mar_large": 0.966,
|
| 678 |
+
"eval_mar_medium": 0.7769,
|
| 679 |
+
"eval_mar_small": 0.8,
|
| 680 |
+
"eval_runtime": 6.6188,
|
| 681 |
+
"eval_samples_per_second": 19.339,
|
| 682 |
+
"eval_steps_per_second": 2.417,
|
| 683 |
+
"step": 2639
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 30.0,
|
| 687 |
+
"eval_loss": 0.741165041923523,
|
| 688 |
+
"eval_map": 0.8559,
|
| 689 |
+
"eval_map_50": 0.9016,
|
| 690 |
+
"eval_map_75": 0.8852,
|
| 691 |
+
"eval_map_class_0": 0.8559,
|
| 692 |
+
"eval_map_large": 0.8848,
|
| 693 |
+
"eval_map_medium": 0.6332,
|
| 694 |
+
"eval_map_small": 0.0,
|
| 695 |
+
"eval_mar_1": 0.7571,
|
| 696 |
+
"eval_mar_10": 0.8957,
|
| 697 |
+
"eval_mar_100": 0.9385,
|
| 698 |
+
"eval_mar_100_class_0": 0.9385,
|
| 699 |
+
"eval_mar_large": 0.9639,
|
| 700 |
+
"eval_mar_medium": 0.7231,
|
| 701 |
+
"eval_mar_small": 0.0,
|
| 702 |
+
"eval_runtime": 6.5118,
|
| 703 |
+
"eval_samples_per_second": 19.656,
|
| 704 |
+
"eval_steps_per_second": 2.457,
|
| 705 |
+
"step": 2730
|
| 706 |
+
},
|
| 707 |
+
{
|
| 708 |
+
"epoch": 30.0,
|
| 709 |
+
"step": 2730,
|
| 710 |
+
"total_flos": 1.6368835442688e+18,
|
| 711 |
+
"train_loss": 11.706639954283997,
|
| 712 |
+
"train_runtime": 1986.2673,
|
| 713 |
+
"train_samples_per_second": 10.95,
|
| 714 |
+
"train_steps_per_second": 1.374
|
| 715 |
+
}
|
| 716 |
+
],
|
| 717 |
+
"logging_steps": 500,
|
| 718 |
+
"max_steps": 2730,
|
| 719 |
+
"num_input_tokens_seen": 0,
|
| 720 |
+
"num_train_epochs": 30,
|
| 721 |
+
"save_steps": 500,
|
| 722 |
+
"stateful_callbacks": {
|
| 723 |
+
"TrainerControl": {
|
| 724 |
+
"args": {
|
| 725 |
+
"should_epoch_stop": false,
|
| 726 |
+
"should_evaluate": false,
|
| 727 |
+
"should_log": false,
|
| 728 |
+
"should_save": true,
|
| 729 |
+
"should_training_stop": true
|
| 730 |
+
},
|
| 731 |
+
"attributes": {}
|
| 732 |
+
}
|
| 733 |
+
},
|
| 734 |
+
"total_flos": 1.6368835442688e+18,
|
| 735 |
+
"train_batch_size": 8,
|
| 736 |
+
"trial_name": null,
|
| 737 |
+
"trial_params": null
|
| 738 |
+
}
|