End of training
Browse files- README.md +39 -43
- all_results.json +10 -10
- config.json +8 -4
- eval_results.json +6 -6
- model.safetensors +2 -2
- runs/Dec01_17-09-38_DESKTOP-SKBE9FB/events.out.tfevents.1733094579.DESKTOP-SKBE9FB.18044.0 +3 -0
- runs/Dec01_17-53-11_DESKTOP-SKBE9FB/events.out.tfevents.1733097193.DESKTOP-SKBE9FB.15124.0 +3 -0
- runs/Dec01_18-02-46_DESKTOP-SKBE9FB/events.out.tfevents.1733097768.DESKTOP-SKBE9FB.2668.0 +3 -0
- runs/Dec01_18-10-08_DESKTOP-SKBE9FB/events.out.tfevents.1733098210.DESKTOP-SKBE9FB.2796.0 +3 -0
- runs/Dec01_18-10-08_DESKTOP-SKBE9FB/events.out.tfevents.1733098802.DESKTOP-SKBE9FB.2796.1 +3 -0
- train_results.json +5 -5
- trainer_state.json +309 -441
- training_args.bin +1 -1
README.md
CHANGED
|
@@ -22,7 +22,7 @@ model-index:
|
|
| 22 |
metrics:
|
| 23 |
- name: Accuracy
|
| 24 |
type: accuracy
|
| 25 |
-
value: 0.
|
| 26 |
---
|
| 27 |
|
| 28 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
@@ -32,8 +32,8 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 32 |
|
| 33 |
This model is a fine-tuned version of [microsoft/beit-base-patch16-224](https://huggingface.co/microsoft/beit-base-patch16-224) on the imagefolder dataset.
|
| 34 |
It achieves the following results on the evaluation set:
|
| 35 |
-
- Loss: 0.
|
| 36 |
-
- Accuracy: 0.
|
| 37 |
|
| 38 |
## Model description
|
| 39 |
|
|
@@ -67,46 +67,42 @@ The following hyperparameters were used during training:
|
|
| 67 |
|
| 68 |
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
| 69 |
|:-------------:|:-----:|:----:|:---------------:|:--------:|
|
| 70 |
-
| No log |
|
| 71 |
-
|
|
| 72 |
-
|
|
| 73 |
-
|
|
| 74 |
-
|
|
| 75 |
-
|
|
| 76 |
-
|
|
| 77 |
-
|
|
| 78 |
-
|
|
| 79 |
-
|
|
| 80 |
-
|
|
| 81 |
-
| 0.
|
| 82 |
-
| 0.
|
| 83 |
-
| 0.
|
| 84 |
-
| 0.
|
| 85 |
-
| 0.
|
| 86 |
-
| 0.
|
| 87 |
-
| 0.
|
| 88 |
-
| 0.
|
| 89 |
-
| 0.
|
| 90 |
-
| 0.
|
| 91 |
-
| 0.
|
| 92 |
-
| 0.
|
| 93 |
-
| 0.
|
| 94 |
-
| 0.
|
| 95 |
-
| 0.
|
| 96 |
-
| 0.
|
| 97 |
-
| 0.
|
| 98 |
-
| 0.
|
| 99 |
-
| 0.
|
| 100 |
-
| 0.
|
| 101 |
-
| 0.
|
| 102 |
-
| 0.
|
| 103 |
-
| 0.
|
| 104 |
-
| 0.
|
| 105 |
-
| 0.
|
| 106 |
-
| 0.217 | 37.0 | 296 | 0.5649 | 0.8224 |
|
| 107 |
-
| 0.2111 | 38.0 | 304 | 0.5683 | 0.8131 |
|
| 108 |
-
| 0.2085 | 39.0 | 312 | 0.5398 | 0.8224 |
|
| 109 |
-
| 0.1912 | 40.0 | 320 | 0.5548 | 0.8224 |
|
| 110 |
|
| 111 |
|
| 112 |
### Framework versions
|
|
|
|
| 22 |
metrics:
|
| 23 |
- name: Accuracy
|
| 24 |
type: accuracy
|
| 25 |
+
value: 0.8225806451612904
|
| 26 |
---
|
| 27 |
|
| 28 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
|
| 32 |
|
| 33 |
This model is a fine-tuned version of [microsoft/beit-base-patch16-224](https://huggingface.co/microsoft/beit-base-patch16-224) on the imagefolder dataset.
|
| 34 |
It achieves the following results on the evaluation set:
|
| 35 |
+
- Loss: 0.4801
|
| 36 |
+
- Accuracy: 0.8226
|
| 37 |
|
| 38 |
## Model description
|
| 39 |
|
|
|
|
| 67 |
|
| 68 |
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
| 69 |
|:-------------:|:-----:|:----:|:---------------:|:--------:|
|
| 70 |
+
| No log | 0.89 | 4 | 1.7603 | 0.1452 |
|
| 71 |
+
| No log | 2.0 | 9 | 1.6852 | 0.1452 |
|
| 72 |
+
| 1.7571 | 2.89 | 13 | 1.5655 | 0.1452 |
|
| 73 |
+
| 1.7571 | 4.0 | 18 | 1.3816 | 0.1452 |
|
| 74 |
+
| 1.5255 | 4.89 | 22 | 1.2599 | 0.3226 |
|
| 75 |
+
| 1.5255 | 6.0 | 27 | 1.1534 | 0.4839 |
|
| 76 |
+
| 1.2245 | 6.89 | 31 | 1.0641 | 0.4839 |
|
| 77 |
+
| 1.2245 | 8.0 | 36 | 1.0372 | 0.4355 |
|
| 78 |
+
| 1.0438 | 8.89 | 40 | 0.9988 | 0.4355 |
|
| 79 |
+
| 1.0438 | 10.0 | 45 | 0.9260 | 0.5161 |
|
| 80 |
+
| 1.0438 | 10.89 | 49 | 0.9085 | 0.7097 |
|
| 81 |
+
| 0.9727 | 12.0 | 54 | 0.8433 | 0.7258 |
|
| 82 |
+
| 0.9727 | 12.89 | 58 | 0.7529 | 0.7742 |
|
| 83 |
+
| 0.8469 | 14.0 | 63 | 0.7187 | 0.7581 |
|
| 84 |
+
| 0.8469 | 14.89 | 67 | 0.6806 | 0.7258 |
|
| 85 |
+
| 0.6908 | 16.0 | 72 | 0.6576 | 0.7581 |
|
| 86 |
+
| 0.6908 | 16.89 | 76 | 0.5742 | 0.7903 |
|
| 87 |
+
| 0.6064 | 18.0 | 81 | 0.6447 | 0.7581 |
|
| 88 |
+
| 0.6064 | 18.89 | 85 | 0.5602 | 0.7742 |
|
| 89 |
+
| 0.5303 | 20.0 | 90 | 0.4943 | 0.7903 |
|
| 90 |
+
| 0.5303 | 20.89 | 94 | 0.5304 | 0.7903 |
|
| 91 |
+
| 0.5303 | 22.0 | 99 | 0.4801 | 0.8226 |
|
| 92 |
+
| 0.4903 | 22.89 | 103 | 0.4849 | 0.8226 |
|
| 93 |
+
| 0.4903 | 24.0 | 108 | 0.5710 | 0.7742 |
|
| 94 |
+
| 0.4261 | 24.89 | 112 | 0.4803 | 0.7903 |
|
| 95 |
+
| 0.4261 | 26.0 | 117 | 0.5671 | 0.7258 |
|
| 96 |
+
| 0.4122 | 26.89 | 121 | 0.4585 | 0.8065 |
|
| 97 |
+
| 0.4122 | 28.0 | 126 | 0.5910 | 0.7097 |
|
| 98 |
+
| 0.3739 | 28.89 | 130 | 0.5821 | 0.7581 |
|
| 99 |
+
| 0.3739 | 30.0 | 135 | 0.5329 | 0.7742 |
|
| 100 |
+
| 0.3739 | 30.89 | 139 | 0.4423 | 0.8226 |
|
| 101 |
+
| 0.3896 | 32.0 | 144 | 0.4716 | 0.7581 |
|
| 102 |
+
| 0.3896 | 32.89 | 148 | 0.4786 | 0.7903 |
|
| 103 |
+
| 0.3472 | 34.0 | 153 | 0.4538 | 0.7903 |
|
| 104 |
+
| 0.3472 | 34.89 | 157 | 0.4553 | 0.7903 |
|
| 105 |
+
| 0.3349 | 35.56 | 160 | 0.4528 | 0.7903 |
|
|
|
|
|
|
|
|
|
|
|
|
|
| 106 |
|
| 107 |
|
| 108 |
### Framework versions
|
all_results.json
CHANGED
|
@@ -1,12 +1,12 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss": 0.
|
| 5 |
-
"eval_runtime": 2.
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second":
|
| 8 |
-
"train_loss": 0.
|
| 9 |
-
"train_runtime":
|
| 10 |
-
"train_samples_per_second":
|
| 11 |
-
"train_steps_per_second": 0.
|
| 12 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 35.56,
|
| 3 |
+
"eval_accuracy": 0.8225806451612904,
|
| 4 |
+
"eval_loss": 0.48010584712028503,
|
| 5 |
+
"eval_runtime": 2.5596,
|
| 6 |
+
"eval_samples_per_second": 24.223,
|
| 7 |
+
"eval_steps_per_second": 1.563,
|
| 8 |
+
"train_loss": 0.7482577681541442,
|
| 9 |
+
"train_runtime": 589.7762,
|
| 10 |
+
"train_samples_per_second": 19.533,
|
| 11 |
+
"train_steps_per_second": 0.271
|
| 12 |
}
|
config.json
CHANGED
|
@@ -14,15 +14,19 @@
|
|
| 14 |
"hidden_dropout_prob": 0.0,
|
| 15 |
"hidden_size": 768,
|
| 16 |
"id2label": {
|
| 17 |
-
"0": "
|
| 18 |
-
"1": "
|
|
|
|
|
|
|
| 19 |
},
|
| 20 |
"image_size": 224,
|
| 21 |
"initializer_range": 0.02,
|
| 22 |
"intermediate_size": 3072,
|
| 23 |
"label2id": {
|
| 24 |
-
"
|
| 25 |
-
"
|
|
|
|
|
|
|
| 26 |
},
|
| 27 |
"layer_norm_eps": 1e-12,
|
| 28 |
"layer_scale_init_value": 0.1,
|
|
|
|
| 14 |
"hidden_dropout_prob": 0.0,
|
| 15 |
"hidden_size": 768,
|
| 16 |
"id2label": {
|
| 17 |
+
"0": "active",
|
| 18 |
+
"1": "active-inactive",
|
| 19 |
+
"2": "healthy",
|
| 20 |
+
"3": "inactive"
|
| 21 |
},
|
| 22 |
"image_size": 224,
|
| 23 |
"initializer_range": 0.02,
|
| 24 |
"intermediate_size": 3072,
|
| 25 |
"label2id": {
|
| 26 |
+
"active": 0,
|
| 27 |
+
"active-inactive": 1,
|
| 28 |
+
"healthy": 2,
|
| 29 |
+
"inactive": 3
|
| 30 |
},
|
| 31 |
"layer_norm_eps": 1e-12,
|
| 32 |
"layer_scale_init_value": 0.1,
|
eval_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss": 0.
|
| 5 |
-
"eval_runtime": 2.
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second":
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 35.56,
|
| 3 |
+
"eval_accuracy": 0.8225806451612904,
|
| 4 |
+
"eval_loss": 0.48010584712028503,
|
| 5 |
+
"eval_runtime": 2.5596,
|
| 6 |
+
"eval_samples_per_second": 24.223,
|
| 7 |
+
"eval_steps_per_second": 1.563
|
| 8 |
}
|
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d2e0354dd3875d2a95e10cb59e09081044efc892a545ff863b0aaa18b051efa6
|
| 3 |
+
size 343086480
|
runs/Dec01_17-09-38_DESKTOP-SKBE9FB/events.out.tfevents.1733094579.DESKTOP-SKBE9FB.18044.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c90a088556944c18e49c91b9047a211812a0ae629235b3b9577acedb47130633
|
| 3 |
+
size 8146
|
runs/Dec01_17-53-11_DESKTOP-SKBE9FB/events.out.tfevents.1733097193.DESKTOP-SKBE9FB.15124.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3129a7fb381f5a5b5536bb13a15331b4a1d56416073bcd6536b696503b67e472
|
| 3 |
+
size 17651
|
runs/Dec01_18-02-46_DESKTOP-SKBE9FB/events.out.tfevents.1733097768.DESKTOP-SKBE9FB.2668.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:bb3c6fd85e0cd85df68369e4dbfb4a09ba174302b615b80037cd450200187de3
|
| 3 |
+
size 14469
|
runs/Dec01_18-10-08_DESKTOP-SKBE9FB/events.out.tfevents.1733098210.DESKTOP-SKBE9FB.2796.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8a9e884b466dce56ecf87f1ddb8372d84339ad4da748a8b8c73aea615a584ad6
|
| 3 |
+
size 19611
|
runs/Dec01_18-10-08_DESKTOP-SKBE9FB/events.out.tfevents.1733098802.DESKTOP-SKBE9FB.2796.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fc1f54208f94aab9d58c8a1f6921b72cd15fb82aa5c029bf45826f5593eb5cab
|
| 3 |
+
size 411
|
train_results.json
CHANGED
|
@@ -1,7 +1,7 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"train_loss": 0.
|
| 4 |
-
"train_runtime":
|
| 5 |
-
"train_samples_per_second":
|
| 6 |
-
"train_steps_per_second": 0.
|
| 7 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 35.56,
|
| 3 |
+
"train_loss": 0.7482577681541442,
|
| 4 |
+
"train_runtime": 589.7762,
|
| 5 |
+
"train_samples_per_second": 19.533,
|
| 6 |
+
"train_steps_per_second": 0.271
|
| 7 |
}
|
trainer_state.json
CHANGED
|
@@ -1,581 +1,449 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 0.
|
| 3 |
-
"best_model_checkpoint": "beit-base-patch16-224-OT\\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,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
-
"epoch":
|
| 13 |
-
"eval_accuracy": 0.
|
| 14 |
-
"eval_loss":
|
| 15 |
-
"eval_runtime": 2.
|
| 16 |
-
"eval_samples_per_second":
|
| 17 |
-
"eval_steps_per_second":
|
| 18 |
-
"step":
|
| 19 |
-
},
|
| 20 |
-
{
|
| 21 |
-
"epoch": 1.25,
|
| 22 |
-
"learning_rate": 1.5625e-05,
|
| 23 |
-
"loss": 0.692,
|
| 24 |
-
"step": 10
|
| 25 |
},
|
| 26 |
{
|
| 27 |
"epoch": 2.0,
|
| 28 |
-
"eval_accuracy": 0.
|
| 29 |
-
"eval_loss":
|
| 30 |
-
"eval_runtime":
|
| 31 |
-
"eval_samples_per_second":
|
| 32 |
-
"eval_steps_per_second":
|
| 33 |
-
"step":
|
| 34 |
},
|
| 35 |
{
|
| 36 |
-
"epoch": 2.
|
| 37 |
"learning_rate": 3.125e-05,
|
| 38 |
-
"loss":
|
| 39 |
-
"step":
|
| 40 |
-
},
|
| 41 |
-
{
|
| 42 |
-
"epoch": 3.0,
|
| 43 |
-
"eval_accuracy": 0.5887850467289719,
|
| 44 |
-
"eval_loss": 0.6668981313705444,
|
| 45 |
-
"eval_runtime": 2.0656,
|
| 46 |
-
"eval_samples_per_second": 51.802,
|
| 47 |
-
"eval_steps_per_second": 3.389,
|
| 48 |
-
"step": 24
|
| 49 |
},
|
| 50 |
{
|
| 51 |
-
"epoch":
|
| 52 |
-
"
|
| 53 |
-
"
|
| 54 |
-
"
|
|
|
|
|
|
|
|
|
|
| 55 |
},
|
| 56 |
{
|
| 57 |
"epoch": 4.0,
|
| 58 |
-
"eval_accuracy": 0.
|
| 59 |
-
"eval_loss":
|
| 60 |
-
"eval_runtime": 2.
|
| 61 |
-
"eval_samples_per_second":
|
| 62 |
-
"eval_steps_per_second":
|
| 63 |
-
"step":
|
| 64 |
},
|
| 65 |
{
|
| 66 |
-
"epoch":
|
| 67 |
"learning_rate": 4.8611111111111115e-05,
|
| 68 |
-
"loss":
|
| 69 |
-
"step":
|
| 70 |
},
|
| 71 |
{
|
| 72 |
-
"epoch":
|
| 73 |
-
"eval_accuracy": 0.
|
| 74 |
-
"eval_loss":
|
| 75 |
-
"eval_runtime":
|
| 76 |
-
"eval_samples_per_second":
|
| 77 |
-
"eval_steps_per_second":
|
| 78 |
-
"step":
|
| 79 |
},
|
| 80 |
{
|
| 81 |
"epoch": 6.0,
|
| 82 |
-
"eval_accuracy": 0.
|
| 83 |
-
"eval_loss":
|
| 84 |
-
"eval_runtime": 2.
|
| 85 |
-
"eval_samples_per_second":
|
| 86 |
-
"eval_steps_per_second":
|
| 87 |
-
"step":
|
| 88 |
-
},
|
| 89 |
-
{
|
| 90 |
-
"epoch": 6.25,
|
| 91 |
-
"learning_rate": 4.6875e-05,
|
| 92 |
-
"loss": 0.6341,
|
| 93 |
-
"step": 50
|
| 94 |
},
|
| 95 |
{
|
| 96 |
-
"epoch":
|
| 97 |
-
"eval_accuracy": 0.6542056074766355,
|
| 98 |
-
"eval_loss": 0.6234968900680542,
|
| 99 |
-
"eval_runtime": 1.9955,
|
| 100 |
-
"eval_samples_per_second": 53.622,
|
| 101 |
-
"eval_steps_per_second": 3.508,
|
| 102 |
-
"step": 56
|
| 103 |
-
},
|
| 104 |
-
{
|
| 105 |
-
"epoch": 7.5,
|
| 106 |
"learning_rate": 4.5138888888888894e-05,
|
| 107 |
-
"loss":
|
| 108 |
-
"step":
|
| 109 |
},
|
| 110 |
{
|
| 111 |
-
"epoch":
|
| 112 |
-
"eval_accuracy": 0.
|
| 113 |
-
"eval_loss":
|
| 114 |
-
"eval_runtime": 2.
|
| 115 |
-
"eval_samples_per_second":
|
| 116 |
-
"eval_steps_per_second":
|
| 117 |
-
"step":
|
| 118 |
},
|
| 119 |
{
|
| 120 |
-
"epoch": 8.
|
| 121 |
-
"
|
| 122 |
-
"
|
| 123 |
-
"
|
|
|
|
|
|
|
|
|
|
| 124 |
},
|
| 125 |
{
|
| 126 |
-
"epoch":
|
| 127 |
-
"
|
| 128 |
-
"
|
| 129 |
-
"
|
| 130 |
-
"eval_samples_per_second": 53.235,
|
| 131 |
-
"eval_steps_per_second": 3.483,
|
| 132 |
-
"step": 72
|
| 133 |
},
|
| 134 |
{
|
| 135 |
-
"epoch":
|
| 136 |
-
"
|
| 137 |
-
"
|
| 138 |
-
"
|
|
|
|
|
|
|
|
|
|
| 139 |
},
|
| 140 |
{
|
| 141 |
"epoch": 10.0,
|
| 142 |
-
"eval_accuracy": 0.
|
| 143 |
-
"eval_loss": 0.
|
| 144 |
-
"eval_runtime": 2.
|
| 145 |
-
"eval_samples_per_second":
|
| 146 |
-
"eval_steps_per_second":
|
| 147 |
-
"step":
|
| 148 |
},
|
| 149 |
{
|
| 150 |
-
"epoch":
|
| 151 |
-
"eval_accuracy": 0.
|
| 152 |
-
"eval_loss": 0.
|
| 153 |
-
"eval_runtime":
|
| 154 |
-
"eval_samples_per_second":
|
| 155 |
-
"eval_steps_per_second":
|
| 156 |
-
"step":
|
| 157 |
},
|
| 158 |
{
|
| 159 |
-
"epoch": 11.
|
| 160 |
-
"learning_rate": 3.
|
| 161 |
-
"loss": 0.
|
| 162 |
-
"step":
|
| 163 |
},
|
| 164 |
{
|
| 165 |
"epoch": 12.0,
|
| 166 |
-
"eval_accuracy": 0.
|
| 167 |
-
"eval_loss": 0.
|
| 168 |
-
"eval_runtime": 2.
|
| 169 |
-
"eval_samples_per_second":
|
| 170 |
-
"eval_steps_per_second":
|
| 171 |
-
"step":
|
| 172 |
-
},
|
| 173 |
-
{
|
| 174 |
-
"epoch": 12.5,
|
| 175 |
-
"learning_rate": 3.8194444444444444e-05,
|
| 176 |
-
"loss": 0.5336,
|
| 177 |
-
"step": 100
|
| 178 |
},
|
| 179 |
{
|
| 180 |
-
"epoch":
|
| 181 |
-
"eval_accuracy": 0.
|
| 182 |
-
"eval_loss": 0.
|
| 183 |
-
"eval_runtime":
|
| 184 |
-
"eval_samples_per_second":
|
| 185 |
-
"eval_steps_per_second":
|
| 186 |
-
"step":
|
| 187 |
},
|
| 188 |
{
|
| 189 |
-
"epoch": 13.
|
| 190 |
-
"learning_rate": 3.
|
| 191 |
-
"loss": 0.
|
| 192 |
-
"step":
|
| 193 |
},
|
| 194 |
{
|
| 195 |
"epoch": 14.0,
|
| 196 |
-
"eval_accuracy": 0.
|
| 197 |
-
"eval_loss": 0.
|
| 198 |
-
"eval_runtime": 2.
|
| 199 |
-
"eval_samples_per_second":
|
| 200 |
-
"eval_steps_per_second":
|
| 201 |
-
"step":
|
| 202 |
},
|
| 203 |
{
|
| 204 |
-
"epoch":
|
| 205 |
-
"
|
| 206 |
-
"
|
| 207 |
-
"
|
|
|
|
|
|
|
|
|
|
| 208 |
},
|
| 209 |
{
|
| 210 |
-
"epoch": 15.
|
| 211 |
-
"
|
| 212 |
-
"
|
| 213 |
-
"
|
| 214 |
-
"eval_samples_per_second": 53.541,
|
| 215 |
-
"eval_steps_per_second": 3.503,
|
| 216 |
-
"step": 120
|
| 217 |
},
|
| 218 |
{
|
| 219 |
"epoch": 16.0,
|
| 220 |
-
"eval_accuracy": 0.
|
| 221 |
-
"eval_loss": 0.
|
| 222 |
-
"eval_runtime":
|
| 223 |
-
"eval_samples_per_second":
|
| 224 |
-
"eval_steps_per_second":
|
| 225 |
-
"step":
|
| 226 |
-
},
|
| 227 |
-
{
|
| 228 |
-
"epoch": 16.25,
|
| 229 |
-
"learning_rate": 3.2986111111111115e-05,
|
| 230 |
-
"loss": 0.544,
|
| 231 |
-
"step": 130
|
| 232 |
},
|
| 233 |
{
|
| 234 |
-
"epoch":
|
| 235 |
-
"eval_accuracy": 0.
|
| 236 |
-
"eval_loss": 0.
|
| 237 |
-
"eval_runtime":
|
| 238 |
-
"eval_samples_per_second":
|
| 239 |
-
"eval_steps_per_second":
|
| 240 |
-
"step":
|
| 241 |
},
|
| 242 |
{
|
| 243 |
-
"epoch": 17.
|
| 244 |
-
"learning_rate":
|
| 245 |
-
"loss": 0.
|
| 246 |
-
"step":
|
| 247 |
},
|
| 248 |
{
|
| 249 |
"epoch": 18.0,
|
| 250 |
-
"eval_accuracy": 0.
|
| 251 |
-
"eval_loss": 0.
|
| 252 |
-
"eval_runtime":
|
| 253 |
-
"eval_samples_per_second":
|
| 254 |
-
"eval_steps_per_second":
|
| 255 |
-
"step":
|
| 256 |
-
},
|
| 257 |
-
{
|
| 258 |
-
"epoch": 18.75,
|
| 259 |
-
"learning_rate": 2.951388888888889e-05,
|
| 260 |
-
"loss": 0.4082,
|
| 261 |
-
"step": 150
|
| 262 |
},
|
| 263 |
{
|
| 264 |
-
"epoch":
|
| 265 |
-
"eval_accuracy": 0.
|
| 266 |
-
"eval_loss": 0.
|
| 267 |
-
"eval_runtime":
|
| 268 |
-
"eval_samples_per_second":
|
| 269 |
-
"eval_steps_per_second":
|
| 270 |
-
"step":
|
| 271 |
},
|
| 272 |
{
|
| 273 |
"epoch": 20.0,
|
| 274 |
-
"learning_rate": 2.
|
| 275 |
-
"loss": 0.
|
| 276 |
-
"step":
|
| 277 |
},
|
| 278 |
{
|
| 279 |
"epoch": 20.0,
|
| 280 |
-
"eval_accuracy": 0.
|
| 281 |
-
"eval_loss": 0.
|
| 282 |
-
"eval_runtime":
|
| 283 |
-
"eval_samples_per_second":
|
| 284 |
-
"eval_steps_per_second":
|
| 285 |
-
"step":
|
| 286 |
-
},
|
| 287 |
-
{
|
| 288 |
-
"epoch": 21.0,
|
| 289 |
-
"eval_accuracy": 0.7757009345794392,
|
| 290 |
-
"eval_loss": 0.5392723083496094,
|
| 291 |
-
"eval_runtime": 2.117,
|
| 292 |
-
"eval_samples_per_second": 50.542,
|
| 293 |
-
"eval_steps_per_second": 3.306,
|
| 294 |
-
"step": 168
|
| 295 |
},
|
| 296 |
{
|
| 297 |
-
"epoch":
|
| 298 |
-
"
|
| 299 |
-
"
|
| 300 |
-
"
|
|
|
|
|
|
|
|
|
|
| 301 |
},
|
| 302 |
{
|
| 303 |
"epoch": 22.0,
|
| 304 |
-
"eval_accuracy": 0.
|
| 305 |
-
"eval_loss": 0.
|
| 306 |
-
"eval_runtime":
|
| 307 |
-
"eval_samples_per_second":
|
| 308 |
-
"eval_steps_per_second":
|
| 309 |
-
"step":
|
| 310 |
},
|
| 311 |
{
|
| 312 |
-
"epoch": 22.
|
| 313 |
-
"learning_rate": 2.
|
| 314 |
-
"loss": 0.
|
| 315 |
-
"step":
|
| 316 |
-
},
|
| 317 |
-
{
|
| 318 |
-
"epoch": 23.0,
|
| 319 |
-
"eval_accuracy": 0.7383177570093458,
|
| 320 |
-
"eval_loss": 0.5568514466285706,
|
| 321 |
-
"eval_runtime": 2.1005,
|
| 322 |
-
"eval_samples_per_second": 50.94,
|
| 323 |
-
"eval_steps_per_second": 3.333,
|
| 324 |
-
"step": 184
|
| 325 |
},
|
| 326 |
{
|
| 327 |
-
"epoch":
|
| 328 |
-
"
|
| 329 |
-
"
|
| 330 |
-
"
|
|
|
|
|
|
|
|
|
|
| 331 |
},
|
| 332 |
{
|
| 333 |
"epoch": 24.0,
|
| 334 |
-
"eval_accuracy": 0.
|
| 335 |
-
"eval_loss": 0.
|
| 336 |
-
"eval_runtime":
|
| 337 |
-
"eval_samples_per_second":
|
| 338 |
-
"eval_steps_per_second":
|
| 339 |
-
"step":
|
| 340 |
},
|
| 341 |
{
|
| 342 |
-
"epoch":
|
| 343 |
-
"learning_rate":
|
| 344 |
-
"loss": 0.
|
| 345 |
-
"step":
|
| 346 |
},
|
| 347 |
{
|
| 348 |
-
"epoch":
|
| 349 |
-
"eval_accuracy": 0.
|
| 350 |
-
"eval_loss": 0.
|
| 351 |
-
"eval_runtime": 2.
|
| 352 |
-
"eval_samples_per_second":
|
| 353 |
-
"eval_steps_per_second":
|
| 354 |
-
"step":
|
| 355 |
},
|
| 356 |
{
|
| 357 |
"epoch": 26.0,
|
| 358 |
-
"eval_accuracy": 0.
|
| 359 |
-
"eval_loss": 0.
|
| 360 |
-
"eval_runtime": 2.
|
| 361 |
-
"eval_samples_per_second":
|
| 362 |
-
"eval_steps_per_second":
|
| 363 |
-
"step":
|
| 364 |
},
|
| 365 |
{
|
| 366 |
-
"epoch": 26.
|
| 367 |
-
"learning_rate": 1.
|
| 368 |
-
"loss": 0.
|
| 369 |
-
"step":
|
| 370 |
-
},
|
| 371 |
-
{
|
| 372 |
-
"epoch": 27.0,
|
| 373 |
-
"eval_accuracy": 0.7663551401869159,
|
| 374 |
-
"eval_loss": 0.6095036864280701,
|
| 375 |
-
"eval_runtime": 1.9725,
|
| 376 |
-
"eval_samples_per_second": 54.247,
|
| 377 |
-
"eval_steps_per_second": 3.549,
|
| 378 |
-
"step": 216
|
| 379 |
},
|
| 380 |
{
|
| 381 |
-
"epoch":
|
| 382 |
-
"
|
| 383 |
-
"
|
| 384 |
-
"
|
|
|
|
|
|
|
|
|
|
| 385 |
},
|
| 386 |
{
|
| 387 |
"epoch": 28.0,
|
| 388 |
-
"eval_accuracy": 0.
|
| 389 |
-
"eval_loss": 0.
|
| 390 |
-
"eval_runtime": 2.
|
| 391 |
-
"eval_samples_per_second":
|
| 392 |
-
"eval_steps_per_second":
|
| 393 |
-
"step":
|
| 394 |
-
},
|
| 395 |
-
{
|
| 396 |
-
"epoch": 28.75,
|
| 397 |
-
"learning_rate": 1.5625e-05,
|
| 398 |
-
"loss": 0.2332,
|
| 399 |
-
"step": 230
|
| 400 |
},
|
| 401 |
{
|
| 402 |
-
"epoch":
|
| 403 |
-
"
|
| 404 |
-
"
|
| 405 |
-
"
|
| 406 |
-
"eval_samples_per_second": 54.412,
|
| 407 |
-
"eval_steps_per_second": 3.56,
|
| 408 |
-
"step": 232
|
| 409 |
},
|
| 410 |
{
|
| 411 |
-
"epoch":
|
| 412 |
-
"
|
| 413 |
-
"
|
| 414 |
-
"
|
|
|
|
|
|
|
|
|
|
| 415 |
},
|
| 416 |
{
|
| 417 |
"epoch": 30.0,
|
| 418 |
-
"eval_accuracy": 0.
|
| 419 |
-
"eval_loss": 0.
|
| 420 |
-
"eval_runtime":
|
| 421 |
-
"eval_samples_per_second":
|
| 422 |
-
"eval_steps_per_second":
|
| 423 |
-
"step":
|
| 424 |
},
|
| 425 |
{
|
| 426 |
-
"epoch":
|
| 427 |
-
"eval_accuracy": 0.
|
| 428 |
-
"eval_loss": 0.
|
| 429 |
-
"eval_runtime":
|
| 430 |
-
"eval_samples_per_second":
|
| 431 |
-
"eval_steps_per_second":
|
| 432 |
-
"step":
|
| 433 |
},
|
| 434 |
{
|
| 435 |
-
"epoch": 31.
|
| 436 |
-
"learning_rate":
|
| 437 |
-
"loss": 0.
|
| 438 |
-
"step":
|
| 439 |
},
|
| 440 |
{
|
| 441 |
"epoch": 32.0,
|
| 442 |
-
"eval_accuracy": 0.
|
| 443 |
-
"eval_loss": 0.
|
| 444 |
-
"eval_runtime": 2.
|
| 445 |
-
"eval_samples_per_second":
|
| 446 |
-
"eval_steps_per_second":
|
| 447 |
-
"step":
|
| 448 |
-
},
|
| 449 |
-
{
|
| 450 |
-
"epoch": 32.5,
|
| 451 |
-
"learning_rate": 1.0416666666666668e-05,
|
| 452 |
-
"loss": 0.2127,
|
| 453 |
-
"step": 260
|
| 454 |
},
|
| 455 |
{
|
| 456 |
-
"epoch":
|
| 457 |
-
"eval_accuracy": 0.
|
| 458 |
-
"eval_loss": 0.
|
| 459 |
-
"eval_runtime":
|
| 460 |
-
"eval_samples_per_second":
|
| 461 |
-
"eval_steps_per_second":
|
| 462 |
-
"step":
|
| 463 |
},
|
| 464 |
{
|
| 465 |
-
"epoch": 33.
|
| 466 |
-
"learning_rate":
|
| 467 |
-
"loss": 0.
|
| 468 |
-
"step":
|
| 469 |
},
|
| 470 |
{
|
| 471 |
"epoch": 34.0,
|
| 472 |
-
"eval_accuracy": 0.
|
| 473 |
-
"eval_loss": 0.
|
| 474 |
-
"eval_runtime":
|
| 475 |
-
"eval_samples_per_second":
|
| 476 |
-
"eval_steps_per_second":
|
| 477 |
-
"step":
|
| 478 |
-
},
|
| 479 |
-
{
|
| 480 |
-
"epoch": 35.0,
|
| 481 |
-
"learning_rate": 6.944444444444445e-06,
|
| 482 |
-
"loss": 0.2618,
|
| 483 |
-
"step": 280
|
| 484 |
-
},
|
| 485 |
-
{
|
| 486 |
-
"epoch": 35.0,
|
| 487 |
-
"eval_accuracy": 0.8037383177570093,
|
| 488 |
-
"eval_loss": 0.622107982635498,
|
| 489 |
-
"eval_runtime": 2.0025,
|
| 490 |
-
"eval_samples_per_second": 53.434,
|
| 491 |
-
"eval_steps_per_second": 3.496,
|
| 492 |
-
"step": 280
|
| 493 |
-
},
|
| 494 |
-
{
|
| 495 |
-
"epoch": 36.0,
|
| 496 |
-
"eval_accuracy": 0.8317757009345794,
|
| 497 |
-
"eval_loss": 0.5089600682258606,
|
| 498 |
-
"eval_runtime": 1.9755,
|
| 499 |
-
"eval_samples_per_second": 54.164,
|
| 500 |
-
"eval_steps_per_second": 3.543,
|
| 501 |
-
"step": 288
|
| 502 |
-
},
|
| 503 |
-
{
|
| 504 |
-
"epoch": 36.25,
|
| 505 |
-
"learning_rate": 5.208333333333334e-06,
|
| 506 |
-
"loss": 0.217,
|
| 507 |
-
"step": 290
|
| 508 |
-
},
|
| 509 |
-
{
|
| 510 |
-
"epoch": 37.0,
|
| 511 |
-
"eval_accuracy": 0.822429906542056,
|
| 512 |
-
"eval_loss": 0.5649047493934631,
|
| 513 |
-
"eval_runtime": 1.991,
|
| 514 |
-
"eval_samples_per_second": 53.743,
|
| 515 |
-
"eval_steps_per_second": 3.516,
|
| 516 |
-
"step": 296
|
| 517 |
},
|
| 518 |
{
|
| 519 |
-
"epoch":
|
| 520 |
-
"
|
| 521 |
-
"
|
| 522 |
-
"
|
| 523 |
-
|
| 524 |
-
|
| 525 |
-
"
|
| 526 |
-
"eval_accuracy": 0.8130841121495327,
|
| 527 |
-
"eval_loss": 0.568317174911499,
|
| 528 |
-
"eval_runtime": 2.0365,
|
| 529 |
-
"eval_samples_per_second": 52.542,
|
| 530 |
-
"eval_steps_per_second": 3.437,
|
| 531 |
-
"step": 304
|
| 532 |
},
|
| 533 |
{
|
| 534 |
-
"epoch":
|
| 535 |
-
"learning_rate":
|
| 536 |
-
"loss": 0.
|
| 537 |
-
"step":
|
| 538 |
},
|
| 539 |
{
|
| 540 |
-
"epoch":
|
| 541 |
-
"eval_accuracy": 0.
|
| 542 |
-
"eval_loss": 0.
|
| 543 |
-
"eval_runtime":
|
| 544 |
-
"eval_samples_per_second":
|
| 545 |
-
"eval_steps_per_second":
|
| 546 |
-
"step":
|
| 547 |
},
|
| 548 |
{
|
| 549 |
-
"epoch":
|
| 550 |
-
"
|
| 551 |
-
"
|
| 552 |
-
"
|
| 553 |
-
|
| 554 |
-
|
| 555 |
-
"
|
| 556 |
-
"eval_accuracy": 0.822429906542056,
|
| 557 |
-
"eval_loss": 0.5548034310340881,
|
| 558 |
-
"eval_runtime": 1.978,
|
| 559 |
-
"eval_samples_per_second": 54.096,
|
| 560 |
-
"eval_steps_per_second": 3.539,
|
| 561 |
-
"step": 320
|
| 562 |
-
},
|
| 563 |
-
{
|
| 564 |
-
"epoch": 40.0,
|
| 565 |
-
"step": 320,
|
| 566 |
-
"total_flos": 1.5429806632629043e+18,
|
| 567 |
-
"train_loss": 0.4049976162612438,
|
| 568 |
-
"train_runtime": 701.3243,
|
| 569 |
-
"train_samples_per_second": 28.403,
|
| 570 |
-
"train_steps_per_second": 0.456
|
| 571 |
}
|
| 572 |
],
|
| 573 |
"logging_steps": 10,
|
| 574 |
-
"max_steps":
|
| 575 |
"num_input_tokens_seen": 0,
|
| 576 |
"num_train_epochs": 40,
|
| 577 |
"save_steps": 500,
|
| 578 |
-
"total_flos":
|
| 579 |
"train_batch_size": 16,
|
| 580 |
"trial_name": null,
|
| 581 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 0.8225806451612904,
|
| 3 |
+
"best_model_checkpoint": "beit-base-patch16-224-OT\\checkpoint-99",
|
| 4 |
+
"epoch": 35.55555555555556,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 160,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
+
"epoch": 0.89,
|
| 13 |
+
"eval_accuracy": 0.14516129032258066,
|
| 14 |
+
"eval_loss": 1.7602994441986084,
|
| 15 |
+
"eval_runtime": 2.148,
|
| 16 |
+
"eval_samples_per_second": 28.864,
|
| 17 |
+
"eval_steps_per_second": 1.862,
|
| 18 |
+
"step": 4
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 19 |
},
|
| 20 |
{
|
| 21 |
"epoch": 2.0,
|
| 22 |
+
"eval_accuracy": 0.14516129032258066,
|
| 23 |
+
"eval_loss": 1.685233473777771,
|
| 24 |
+
"eval_runtime": 2.219,
|
| 25 |
+
"eval_samples_per_second": 27.94,
|
| 26 |
+
"eval_steps_per_second": 1.803,
|
| 27 |
+
"step": 9
|
| 28 |
},
|
| 29 |
{
|
| 30 |
+
"epoch": 2.22,
|
| 31 |
"learning_rate": 3.125e-05,
|
| 32 |
+
"loss": 1.7571,
|
| 33 |
+
"step": 10
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 34 |
},
|
| 35 |
{
|
| 36 |
+
"epoch": 2.89,
|
| 37 |
+
"eval_accuracy": 0.14516129032258066,
|
| 38 |
+
"eval_loss": 1.5655227899551392,
|
| 39 |
+
"eval_runtime": 2.222,
|
| 40 |
+
"eval_samples_per_second": 27.903,
|
| 41 |
+
"eval_steps_per_second": 1.8,
|
| 42 |
+
"step": 13
|
| 43 |
},
|
| 44 |
{
|
| 45 |
"epoch": 4.0,
|
| 46 |
+
"eval_accuracy": 0.14516129032258066,
|
| 47 |
+
"eval_loss": 1.3815577030181885,
|
| 48 |
+
"eval_runtime": 2.3696,
|
| 49 |
+
"eval_samples_per_second": 26.165,
|
| 50 |
+
"eval_steps_per_second": 1.688,
|
| 51 |
+
"step": 18
|
| 52 |
},
|
| 53 |
{
|
| 54 |
+
"epoch": 4.44,
|
| 55 |
"learning_rate": 4.8611111111111115e-05,
|
| 56 |
+
"loss": 1.5255,
|
| 57 |
+
"step": 20
|
| 58 |
},
|
| 59 |
{
|
| 60 |
+
"epoch": 4.89,
|
| 61 |
+
"eval_accuracy": 0.3225806451612903,
|
| 62 |
+
"eval_loss": 1.2598901987075806,
|
| 63 |
+
"eval_runtime": 2.178,
|
| 64 |
+
"eval_samples_per_second": 28.466,
|
| 65 |
+
"eval_steps_per_second": 1.837,
|
| 66 |
+
"step": 22
|
| 67 |
},
|
| 68 |
{
|
| 69 |
"epoch": 6.0,
|
| 70 |
+
"eval_accuracy": 0.4838709677419355,
|
| 71 |
+
"eval_loss": 1.153410792350769,
|
| 72 |
+
"eval_runtime": 2.227,
|
| 73 |
+
"eval_samples_per_second": 27.84,
|
| 74 |
+
"eval_steps_per_second": 1.796,
|
| 75 |
+
"step": 27
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 76 |
},
|
| 77 |
{
|
| 78 |
+
"epoch": 6.67,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 79 |
"learning_rate": 4.5138888888888894e-05,
|
| 80 |
+
"loss": 1.2245,
|
| 81 |
+
"step": 30
|
| 82 |
},
|
| 83 |
{
|
| 84 |
+
"epoch": 6.89,
|
| 85 |
+
"eval_accuracy": 0.4838709677419355,
|
| 86 |
+
"eval_loss": 1.0641188621520996,
|
| 87 |
+
"eval_runtime": 2.203,
|
| 88 |
+
"eval_samples_per_second": 28.143,
|
| 89 |
+
"eval_steps_per_second": 1.816,
|
| 90 |
+
"step": 31
|
| 91 |
},
|
| 92 |
{
|
| 93 |
+
"epoch": 8.0,
|
| 94 |
+
"eval_accuracy": 0.43548387096774194,
|
| 95 |
+
"eval_loss": 1.0371758937835693,
|
| 96 |
+
"eval_runtime": 2.2355,
|
| 97 |
+
"eval_samples_per_second": 27.734,
|
| 98 |
+
"eval_steps_per_second": 1.789,
|
| 99 |
+
"step": 36
|
| 100 |
},
|
| 101 |
{
|
| 102 |
+
"epoch": 8.89,
|
| 103 |
+
"learning_rate": 4.166666666666667e-05,
|
| 104 |
+
"loss": 1.0438,
|
| 105 |
+
"step": 40
|
|
|
|
|
|
|
|
|
|
| 106 |
},
|
| 107 |
{
|
| 108 |
+
"epoch": 8.89,
|
| 109 |
+
"eval_accuracy": 0.43548387096774194,
|
| 110 |
+
"eval_loss": 0.9987961649894714,
|
| 111 |
+
"eval_runtime": 2.3086,
|
| 112 |
+
"eval_samples_per_second": 26.857,
|
| 113 |
+
"eval_steps_per_second": 1.733,
|
| 114 |
+
"step": 40
|
| 115 |
},
|
| 116 |
{
|
| 117 |
"epoch": 10.0,
|
| 118 |
+
"eval_accuracy": 0.5161290322580645,
|
| 119 |
+
"eval_loss": 0.9259945154190063,
|
| 120 |
+
"eval_runtime": 2.2905,
|
| 121 |
+
"eval_samples_per_second": 27.068,
|
| 122 |
+
"eval_steps_per_second": 1.746,
|
| 123 |
+
"step": 45
|
| 124 |
},
|
| 125 |
{
|
| 126 |
+
"epoch": 10.89,
|
| 127 |
+
"eval_accuracy": 0.7096774193548387,
|
| 128 |
+
"eval_loss": 0.9084866046905518,
|
| 129 |
+
"eval_runtime": 2.2545,
|
| 130 |
+
"eval_samples_per_second": 27.5,
|
| 131 |
+
"eval_steps_per_second": 1.774,
|
| 132 |
+
"step": 49
|
| 133 |
},
|
| 134 |
{
|
| 135 |
+
"epoch": 11.11,
|
| 136 |
+
"learning_rate": 3.8194444444444444e-05,
|
| 137 |
+
"loss": 0.9727,
|
| 138 |
+
"step": 50
|
| 139 |
},
|
| 140 |
{
|
| 141 |
"epoch": 12.0,
|
| 142 |
+
"eval_accuracy": 0.7258064516129032,
|
| 143 |
+
"eval_loss": 0.843325674533844,
|
| 144 |
+
"eval_runtime": 2.1865,
|
| 145 |
+
"eval_samples_per_second": 28.355,
|
| 146 |
+
"eval_steps_per_second": 1.829,
|
| 147 |
+
"step": 54
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 148 |
},
|
| 149 |
{
|
| 150 |
+
"epoch": 12.89,
|
| 151 |
+
"eval_accuracy": 0.7741935483870968,
|
| 152 |
+
"eval_loss": 0.7529016137123108,
|
| 153 |
+
"eval_runtime": 2.1995,
|
| 154 |
+
"eval_samples_per_second": 28.188,
|
| 155 |
+
"eval_steps_per_second": 1.819,
|
| 156 |
+
"step": 58
|
| 157 |
},
|
| 158 |
{
|
| 159 |
+
"epoch": 13.33,
|
| 160 |
+
"learning_rate": 3.472222222222222e-05,
|
| 161 |
+
"loss": 0.8469,
|
| 162 |
+
"step": 60
|
| 163 |
},
|
| 164 |
{
|
| 165 |
"epoch": 14.0,
|
| 166 |
+
"eval_accuracy": 0.7580645161290323,
|
| 167 |
+
"eval_loss": 0.7186572551727295,
|
| 168 |
+
"eval_runtime": 2.2225,
|
| 169 |
+
"eval_samples_per_second": 27.896,
|
| 170 |
+
"eval_steps_per_second": 1.8,
|
| 171 |
+
"step": 63
|
| 172 |
},
|
| 173 |
{
|
| 174 |
+
"epoch": 14.89,
|
| 175 |
+
"eval_accuracy": 0.7258064516129032,
|
| 176 |
+
"eval_loss": 0.6805762648582458,
|
| 177 |
+
"eval_runtime": 2.1855,
|
| 178 |
+
"eval_samples_per_second": 28.369,
|
| 179 |
+
"eval_steps_per_second": 1.83,
|
| 180 |
+
"step": 67
|
| 181 |
},
|
| 182 |
{
|
| 183 |
+
"epoch": 15.56,
|
| 184 |
+
"learning_rate": 3.125e-05,
|
| 185 |
+
"loss": 0.6908,
|
| 186 |
+
"step": 70
|
|
|
|
|
|
|
|
|
|
| 187 |
},
|
| 188 |
{
|
| 189 |
"epoch": 16.0,
|
| 190 |
+
"eval_accuracy": 0.7580645161290323,
|
| 191 |
+
"eval_loss": 0.6575707197189331,
|
| 192 |
+
"eval_runtime": 2.2315,
|
| 193 |
+
"eval_samples_per_second": 27.784,
|
| 194 |
+
"eval_steps_per_second": 1.792,
|
| 195 |
+
"step": 72
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 196 |
},
|
| 197 |
{
|
| 198 |
+
"epoch": 16.89,
|
| 199 |
+
"eval_accuracy": 0.7903225806451613,
|
| 200 |
+
"eval_loss": 0.5742202997207642,
|
| 201 |
+
"eval_runtime": 2.261,
|
| 202 |
+
"eval_samples_per_second": 27.421,
|
| 203 |
+
"eval_steps_per_second": 1.769,
|
| 204 |
+
"step": 76
|
| 205 |
},
|
| 206 |
{
|
| 207 |
+
"epoch": 17.78,
|
| 208 |
+
"learning_rate": 2.777777777777778e-05,
|
| 209 |
+
"loss": 0.6064,
|
| 210 |
+
"step": 80
|
| 211 |
},
|
| 212 |
{
|
| 213 |
"epoch": 18.0,
|
| 214 |
+
"eval_accuracy": 0.7580645161290323,
|
| 215 |
+
"eval_loss": 0.6446634531021118,
|
| 216 |
+
"eval_runtime": 2.3311,
|
| 217 |
+
"eval_samples_per_second": 26.597,
|
| 218 |
+
"eval_steps_per_second": 1.716,
|
| 219 |
+
"step": 81
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 220 |
},
|
| 221 |
{
|
| 222 |
+
"epoch": 18.89,
|
| 223 |
+
"eval_accuracy": 0.7741935483870968,
|
| 224 |
+
"eval_loss": 0.5602142810821533,
|
| 225 |
+
"eval_runtime": 2.1895,
|
| 226 |
+
"eval_samples_per_second": 28.317,
|
| 227 |
+
"eval_steps_per_second": 1.827,
|
| 228 |
+
"step": 85
|
| 229 |
},
|
| 230 |
{
|
| 231 |
"epoch": 20.0,
|
| 232 |
+
"learning_rate": 2.4305555555555558e-05,
|
| 233 |
+
"loss": 0.5303,
|
| 234 |
+
"step": 90
|
| 235 |
},
|
| 236 |
{
|
| 237 |
"epoch": 20.0,
|
| 238 |
+
"eval_accuracy": 0.7903225806451613,
|
| 239 |
+
"eval_loss": 0.49427932500839233,
|
| 240 |
+
"eval_runtime": 2.2745,
|
| 241 |
+
"eval_samples_per_second": 27.258,
|
| 242 |
+
"eval_steps_per_second": 1.759,
|
| 243 |
+
"step": 90
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 244 |
},
|
| 245 |
{
|
| 246 |
+
"epoch": 20.89,
|
| 247 |
+
"eval_accuracy": 0.7903225806451613,
|
| 248 |
+
"eval_loss": 0.530381441116333,
|
| 249 |
+
"eval_runtime": 2.142,
|
| 250 |
+
"eval_samples_per_second": 28.945,
|
| 251 |
+
"eval_steps_per_second": 1.867,
|
| 252 |
+
"step": 94
|
| 253 |
},
|
| 254 |
{
|
| 255 |
"epoch": 22.0,
|
| 256 |
+
"eval_accuracy": 0.8225806451612904,
|
| 257 |
+
"eval_loss": 0.48010584712028503,
|
| 258 |
+
"eval_runtime": 2.3616,
|
| 259 |
+
"eval_samples_per_second": 26.254,
|
| 260 |
+
"eval_steps_per_second": 1.694,
|
| 261 |
+
"step": 99
|
| 262 |
},
|
| 263 |
{
|
| 264 |
+
"epoch": 22.22,
|
| 265 |
+
"learning_rate": 2.0833333333333336e-05,
|
| 266 |
+
"loss": 0.4903,
|
| 267 |
+
"step": 100
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 268 |
},
|
| 269 |
{
|
| 270 |
+
"epoch": 22.89,
|
| 271 |
+
"eval_accuracy": 0.8225806451612904,
|
| 272 |
+
"eval_loss": 0.4848884344100952,
|
| 273 |
+
"eval_runtime": 2.16,
|
| 274 |
+
"eval_samples_per_second": 28.703,
|
| 275 |
+
"eval_steps_per_second": 1.852,
|
| 276 |
+
"step": 103
|
| 277 |
},
|
| 278 |
{
|
| 279 |
"epoch": 24.0,
|
| 280 |
+
"eval_accuracy": 0.7741935483870968,
|
| 281 |
+
"eval_loss": 0.5709980726242065,
|
| 282 |
+
"eval_runtime": 2.1685,
|
| 283 |
+
"eval_samples_per_second": 28.591,
|
| 284 |
+
"eval_steps_per_second": 1.845,
|
| 285 |
+
"step": 108
|
| 286 |
},
|
| 287 |
{
|
| 288 |
+
"epoch": 24.44,
|
| 289 |
+
"learning_rate": 1.736111111111111e-05,
|
| 290 |
+
"loss": 0.4261,
|
| 291 |
+
"step": 110
|
| 292 |
},
|
| 293 |
{
|
| 294 |
+
"epoch": 24.89,
|
| 295 |
+
"eval_accuracy": 0.7903225806451613,
|
| 296 |
+
"eval_loss": 0.4803168475627899,
|
| 297 |
+
"eval_runtime": 2.2145,
|
| 298 |
+
"eval_samples_per_second": 27.997,
|
| 299 |
+
"eval_steps_per_second": 1.806,
|
| 300 |
+
"step": 112
|
| 301 |
},
|
| 302 |
{
|
| 303 |
"epoch": 26.0,
|
| 304 |
+
"eval_accuracy": 0.7258064516129032,
|
| 305 |
+
"eval_loss": 0.5670634508132935,
|
| 306 |
+
"eval_runtime": 2.1993,
|
| 307 |
+
"eval_samples_per_second": 28.191,
|
| 308 |
+
"eval_steps_per_second": 1.819,
|
| 309 |
+
"step": 117
|
| 310 |
},
|
| 311 |
{
|
| 312 |
+
"epoch": 26.67,
|
| 313 |
+
"learning_rate": 1.388888888888889e-05,
|
| 314 |
+
"loss": 0.4122,
|
| 315 |
+
"step": 120
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 316 |
},
|
| 317 |
{
|
| 318 |
+
"epoch": 26.89,
|
| 319 |
+
"eval_accuracy": 0.8064516129032258,
|
| 320 |
+
"eval_loss": 0.4585064947605133,
|
| 321 |
+
"eval_runtime": 2.2065,
|
| 322 |
+
"eval_samples_per_second": 28.098,
|
| 323 |
+
"eval_steps_per_second": 1.813,
|
| 324 |
+
"step": 121
|
| 325 |
},
|
| 326 |
{
|
| 327 |
"epoch": 28.0,
|
| 328 |
+
"eval_accuracy": 0.7096774193548387,
|
| 329 |
+
"eval_loss": 0.5910329222679138,
|
| 330 |
+
"eval_runtime": 2.23,
|
| 331 |
+
"eval_samples_per_second": 27.802,
|
| 332 |
+
"eval_steps_per_second": 1.794,
|
| 333 |
+
"step": 126
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 334 |
},
|
| 335 |
{
|
| 336 |
+
"epoch": 28.89,
|
| 337 |
+
"learning_rate": 1.0416666666666668e-05,
|
| 338 |
+
"loss": 0.3739,
|
| 339 |
+
"step": 130
|
|
|
|
|
|
|
|
|
|
| 340 |
},
|
| 341 |
{
|
| 342 |
+
"epoch": 28.89,
|
| 343 |
+
"eval_accuracy": 0.7580645161290323,
|
| 344 |
+
"eval_loss": 0.5821260213851929,
|
| 345 |
+
"eval_runtime": 2.2145,
|
| 346 |
+
"eval_samples_per_second": 27.997,
|
| 347 |
+
"eval_steps_per_second": 1.806,
|
| 348 |
+
"step": 130
|
| 349 |
},
|
| 350 |
{
|
| 351 |
"epoch": 30.0,
|
| 352 |
+
"eval_accuracy": 0.7741935483870968,
|
| 353 |
+
"eval_loss": 0.5329306125640869,
|
| 354 |
+
"eval_runtime": 2.3526,
|
| 355 |
+
"eval_samples_per_second": 26.354,
|
| 356 |
+
"eval_steps_per_second": 1.7,
|
| 357 |
+
"step": 135
|
| 358 |
},
|
| 359 |
{
|
| 360 |
+
"epoch": 30.89,
|
| 361 |
+
"eval_accuracy": 0.8225806451612904,
|
| 362 |
+
"eval_loss": 0.4422537386417389,
|
| 363 |
+
"eval_runtime": 2.1625,
|
| 364 |
+
"eval_samples_per_second": 28.67,
|
| 365 |
+
"eval_steps_per_second": 1.85,
|
| 366 |
+
"step": 139
|
| 367 |
},
|
| 368 |
{
|
| 369 |
+
"epoch": 31.11,
|
| 370 |
+
"learning_rate": 6.944444444444445e-06,
|
| 371 |
+
"loss": 0.3896,
|
| 372 |
+
"step": 140
|
| 373 |
},
|
| 374 |
{
|
| 375 |
"epoch": 32.0,
|
| 376 |
+
"eval_accuracy": 0.7580645161290323,
|
| 377 |
+
"eval_loss": 0.47155243158340454,
|
| 378 |
+
"eval_runtime": 2.1865,
|
| 379 |
+
"eval_samples_per_second": 28.356,
|
| 380 |
+
"eval_steps_per_second": 1.829,
|
| 381 |
+
"step": 144
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 382 |
},
|
| 383 |
{
|
| 384 |
+
"epoch": 32.89,
|
| 385 |
+
"eval_accuracy": 0.7903225806451613,
|
| 386 |
+
"eval_loss": 0.4785827100276947,
|
| 387 |
+
"eval_runtime": 2.232,
|
| 388 |
+
"eval_samples_per_second": 27.777,
|
| 389 |
+
"eval_steps_per_second": 1.792,
|
| 390 |
+
"step": 148
|
| 391 |
},
|
| 392 |
{
|
| 393 |
+
"epoch": 33.33,
|
| 394 |
+
"learning_rate": 3.4722222222222224e-06,
|
| 395 |
+
"loss": 0.3472,
|
| 396 |
+
"step": 150
|
| 397 |
},
|
| 398 |
{
|
| 399 |
"epoch": 34.0,
|
| 400 |
+
"eval_accuracy": 0.7903225806451613,
|
| 401 |
+
"eval_loss": 0.45382845401763916,
|
| 402 |
+
"eval_runtime": 2.19,
|
| 403 |
+
"eval_samples_per_second": 28.31,
|
| 404 |
+
"eval_steps_per_second": 1.826,
|
| 405 |
+
"step": 153
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 406 |
},
|
| 407 |
{
|
| 408 |
+
"epoch": 34.89,
|
| 409 |
+
"eval_accuracy": 0.7903225806451613,
|
| 410 |
+
"eval_loss": 0.455282062292099,
|
| 411 |
+
"eval_runtime": 2.2165,
|
| 412 |
+
"eval_samples_per_second": 27.972,
|
| 413 |
+
"eval_steps_per_second": 1.805,
|
| 414 |
+
"step": 157
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 415 |
},
|
| 416 |
{
|
| 417 |
+
"epoch": 35.56,
|
| 418 |
+
"learning_rate": 0.0,
|
| 419 |
+
"loss": 0.3349,
|
| 420 |
+
"step": 160
|
| 421 |
},
|
| 422 |
{
|
| 423 |
+
"epoch": 35.56,
|
| 424 |
+
"eval_accuracy": 0.7903225806451613,
|
| 425 |
+
"eval_loss": 0.45279815793037415,
|
| 426 |
+
"eval_runtime": 2.4931,
|
| 427 |
+
"eval_samples_per_second": 24.869,
|
| 428 |
+
"eval_steps_per_second": 1.604,
|
| 429 |
+
"step": 160
|
| 430 |
},
|
| 431 |
{
|
| 432 |
+
"epoch": 35.56,
|
| 433 |
+
"step": 160,
|
| 434 |
+
"total_flos": 7.931930389512192e+17,
|
| 435 |
+
"train_loss": 0.7482577681541442,
|
| 436 |
+
"train_runtime": 589.7762,
|
| 437 |
+
"train_samples_per_second": 19.533,
|
| 438 |
+
"train_steps_per_second": 0.271
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 439 |
}
|
| 440 |
],
|
| 441 |
"logging_steps": 10,
|
| 442 |
+
"max_steps": 160,
|
| 443 |
"num_input_tokens_seen": 0,
|
| 444 |
"num_train_epochs": 40,
|
| 445 |
"save_steps": 500,
|
| 446 |
+
"total_flos": 7.931930389512192e+17,
|
| 447 |
"train_batch_size": 16,
|
| 448 |
"trial_name": null,
|
| 449 |
"trial_params": null
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4728
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:70281fb12f068732959d20071bec1711ddb5ad7a6484dbe72999c1aab81491a0
|
| 3 |
size 4728
|