End of training
Browse files- .gitattributes +1 -0
- README.md +47 -55
- all_results.json +16 -16
- classification_report.png +0 -0
- config.json +1 -1
- eval_results.json +11 -11
- integrated_gradients_grid.jpg +3 -0
- model.safetensors +1 -1
- train_and_eval.jpg +0 -0
- train_results.json +6 -6
- trainer_state.json +588 -238
- training_args.bin +2 -2
.gitattributes
CHANGED
|
@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
|
| 33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
| 33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
| 36 |
+
integrated_gradients_grid.jpg filter=lfs diff=lfs merge=lfs -text
|
README.md
CHANGED
|
@@ -4,8 +4,6 @@ license: apache-2.0
|
|
| 4 |
base_model: facebook/convnext-tiny-224
|
| 5 |
tags:
|
| 6 |
- generated_from_trainer
|
| 7 |
-
datasets:
|
| 8 |
-
- imagefolder
|
| 9 |
metrics:
|
| 10 |
- precision
|
| 11 |
- recall
|
|
@@ -13,29 +11,7 @@ metrics:
|
|
| 13 |
- accuracy
|
| 14 |
model-index:
|
| 15 |
- name: convnext-tiny-224-finetuned-barkley
|
| 16 |
-
results:
|
| 17 |
-
- task:
|
| 18 |
-
name: Image Classification
|
| 19 |
-
type: image-classification
|
| 20 |
-
dataset:
|
| 21 |
-
name: imagefolder
|
| 22 |
-
type: imagefolder
|
| 23 |
-
config: default
|
| 24 |
-
split: train
|
| 25 |
-
args: default
|
| 26 |
-
metrics:
|
| 27 |
-
- name: Precision
|
| 28 |
-
type: precision
|
| 29 |
-
value: 0.9936145510835913
|
| 30 |
-
- name: Recall
|
| 31 |
-
type: recall
|
| 32 |
-
value: 0.993421052631579
|
| 33 |
-
- name: F1
|
| 34 |
-
type: f1
|
| 35 |
-
value: 0.993419541966282
|
| 36 |
-
- name: Accuracy
|
| 37 |
-
type: accuracy
|
| 38 |
-
value: 0.9939393939393939
|
| 39 |
---
|
| 40 |
|
| 41 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
@@ -43,15 +19,15 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 43 |
|
| 44 |
# convnext-tiny-224-finetuned-barkley
|
| 45 |
|
| 46 |
-
This model is a fine-tuned version of [facebook/convnext-tiny-224](https://huggingface.co/facebook/convnext-tiny-224) on
|
| 47 |
It achieves the following results on the evaluation set:
|
| 48 |
-
- Loss: 0.
|
| 49 |
-
- Precision: 0
|
| 50 |
-
- Recall: 0
|
| 51 |
-
- F1: 0
|
| 52 |
-
- Accuracy: 0
|
| 53 |
-
- Top1 Accuracy: 0
|
| 54 |
-
- Error Rate: 0.
|
| 55 |
|
| 56 |
## Model description
|
| 57 |
|
|
@@ -71,9 +47,11 @@ More information needed
|
|
| 71 |
|
| 72 |
The following hyperparameters were used during training:
|
| 73 |
- learning_rate: 0.0002
|
| 74 |
-
- train_batch_size:
|
| 75 |
-
- eval_batch_size:
|
| 76 |
- seed: 42
|
|
|
|
|
|
|
| 77 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 78 |
- lr_scheduler_type: linear
|
| 79 |
- lr_scheduler_warmup_ratio: 0.1
|
|
@@ -82,29 +60,43 @@ The following hyperparameters were used during training:
|
|
| 82 |
|
| 83 |
### Training results
|
| 84 |
|
| 85 |
-
| Training Loss | Epoch | Step |
|
| 86 |
-
|
| 87 |
-
| 1.
|
| 88 |
-
| 1.
|
| 89 |
-
| 1.
|
| 90 |
-
| 1.
|
| 91 |
-
| 1.
|
| 92 |
-
| 1.
|
| 93 |
-
|
|
| 94 |
-
|
|
| 95 |
-
|
|
| 96 |
-
|
|
| 97 |
-
|
|
| 98 |
-
| 0.
|
| 99 |
-
| 0.
|
| 100 |
-
| 0.
|
| 101 |
-
| 0.
|
| 102 |
-
| 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 103 |
|
| 104 |
|
| 105 |
### Framework versions
|
| 106 |
|
| 107 |
-
- Transformers 4.
|
| 108 |
- Pytorch 2.3.1+cu121
|
| 109 |
- Datasets 3.0.1
|
| 110 |
-
- Tokenizers 0.
|
|
|
|
| 4 |
base_model: facebook/convnext-tiny-224
|
| 5 |
tags:
|
| 6 |
- generated_from_trainer
|
|
|
|
|
|
|
| 7 |
metrics:
|
| 8 |
- precision
|
| 9 |
- recall
|
|
|
|
| 11 |
- accuracy
|
| 12 |
model-index:
|
| 13 |
- name: convnext-tiny-224-finetuned-barkley
|
| 14 |
+
results: []
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 15 |
---
|
| 16 |
|
| 17 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
|
| 19 |
|
| 20 |
# convnext-tiny-224-finetuned-barkley
|
| 21 |
|
| 22 |
+
This model is a fine-tuned version of [facebook/convnext-tiny-224](https://huggingface.co/facebook/convnext-tiny-224) on an unknown dataset.
|
| 23 |
It achieves the following results on the evaluation set:
|
| 24 |
+
- Loss: 0.0128
|
| 25 |
+
- Precision: 1.0
|
| 26 |
+
- Recall: 1.0
|
| 27 |
+
- F1: 1.0
|
| 28 |
+
- Accuracy: 1.0
|
| 29 |
+
- Top1 Accuracy: 1.0
|
| 30 |
+
- Error Rate: 0.0
|
| 31 |
|
| 32 |
## Model description
|
| 33 |
|
|
|
|
| 47 |
|
| 48 |
The following hyperparameters were used during training:
|
| 49 |
- learning_rate: 0.0002
|
| 50 |
+
- train_batch_size: 8
|
| 51 |
+
- eval_batch_size: 8
|
| 52 |
- seed: 42
|
| 53 |
+
- gradient_accumulation_steps: 4
|
| 54 |
+
- total_train_batch_size: 32
|
| 55 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 56 |
- lr_scheduler_type: linear
|
| 57 |
- lr_scheduler_warmup_ratio: 0.1
|
|
|
|
| 60 |
|
| 61 |
### Training results
|
| 62 |
|
| 63 |
+
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy | Top1 Accuracy | Error Rate |
|
| 64 |
+
|:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|:-------------:|:----------:|
|
| 65 |
+
| 1.6288 | 1.0 | 38 | 1.6005 | 0.2133 | 0.2697 | 0.2043 | 0.2371 | 0.2697 | 0.7629 |
|
| 66 |
+
| 1.6059 | 2.0 | 76 | 1.5802 | 0.2384 | 0.2763 | 0.2243 | 0.2473 | 0.2763 | 0.7527 |
|
| 67 |
+
| 1.5808 | 3.0 | 114 | 1.5570 | 0.2778 | 0.3026 | 0.2595 | 0.2744 | 0.3026 | 0.7256 |
|
| 68 |
+
| 1.5555 | 4.0 | 152 | 1.5291 | 0.3831 | 0.375 | 0.3491 | 0.3511 | 0.375 | 0.6489 |
|
| 69 |
+
| 1.5232 | 5.0 | 190 | 1.4933 | 0.4252 | 0.4408 | 0.4154 | 0.4147 | 0.4408 | 0.5853 |
|
| 70 |
+
| 1.4784 | 6.0 | 228 | 1.4484 | 0.5076 | 0.5197 | 0.4926 | 0.4972 | 0.5197 | 0.5028 |
|
| 71 |
+
| 1.4242 | 7.0 | 266 | 1.3902 | 0.6857 | 0.6382 | 0.6307 | 0.6249 | 0.6382 | 0.3751 |
|
| 72 |
+
| 1.3586 | 8.0 | 304 | 1.3186 | 0.7728 | 0.7171 | 0.7166 | 0.7134 | 0.7171 | 0.2866 |
|
| 73 |
+
| 1.276 | 9.0 | 342 | 1.2236 | 0.8547 | 0.8026 | 0.8109 | 0.8060 | 0.8026 | 0.1940 |
|
| 74 |
+
| 1.1778 | 10.0 | 380 | 1.1122 | 0.8899 | 0.8553 | 0.8609 | 0.8601 | 0.8553 | 0.1399 |
|
| 75 |
+
| 1.0543 | 11.0 | 418 | 0.9839 | 0.9064 | 0.8947 | 0.8958 | 0.9005 | 0.8947 | 0.0995 |
|
| 76 |
+
| 0.921 | 12.0 | 456 | 0.8418 | 0.9541 | 0.9539 | 0.9537 | 0.9575 | 0.9539 | 0.0425 |
|
| 77 |
+
| 0.773 | 13.0 | 494 | 0.6935 | 0.9624 | 0.9605 | 0.9605 | 0.9652 | 0.9605 | 0.0348 |
|
| 78 |
+
| 0.6204 | 14.0 | 532 | 0.5515 | 0.9688 | 0.9671 | 0.9672 | 0.9708 | 0.9671 | 0.0292 |
|
| 79 |
+
| 0.4835 | 15.0 | 570 | 0.4146 | 0.9704 | 0.9671 | 0.9676 | 0.9697 | 0.9671 | 0.0303 |
|
| 80 |
+
| 0.3641 | 16.0 | 608 | 0.3043 | 0.9805 | 0.9803 | 0.9802 | 0.9830 | 0.9803 | 0.0170 |
|
| 81 |
+
| 0.2706 | 17.0 | 646 | 0.2247 | 0.9805 | 0.9803 | 0.9802 | 0.9830 | 0.9803 | 0.0170 |
|
| 82 |
+
| 0.1998 | 18.0 | 684 | 0.1705 | 0.9873 | 0.9868 | 0.9868 | 0.9889 | 0.9868 | 0.0111 |
|
| 83 |
+
| 0.1446 | 19.0 | 722 | 0.1271 | 0.9937 | 0.9934 | 0.9934 | 0.9944 | 0.9934 | 0.0056 |
|
| 84 |
+
| 0.1106 | 20.0 | 760 | 0.1047 | 0.9873 | 0.9868 | 0.9868 | 0.9889 | 0.9868 | 0.0111 |
|
| 85 |
+
| 0.0872 | 21.0 | 798 | 0.0780 | 0.9937 | 0.9934 | 0.9934 | 0.9944 | 0.9934 | 0.0056 |
|
| 86 |
+
| 0.0614 | 22.0 | 836 | 0.0739 | 0.9873 | 0.9868 | 0.9868 | 0.9889 | 0.9868 | 0.0111 |
|
| 87 |
+
| 0.0491 | 23.0 | 874 | 0.0517 | 0.9937 | 0.9934 | 0.9934 | 0.9944 | 0.9934 | 0.0056 |
|
| 88 |
+
| 0.0365 | 24.0 | 912 | 0.0401 | 0.9871 | 0.9868 | 0.9868 | 0.9878 | 0.9868 | 0.0122 |
|
| 89 |
+
| 0.0255 | 25.0 | 950 | 0.0336 | 0.9937 | 0.9934 | 0.9934 | 0.9944 | 0.9934 | 0.0056 |
|
| 90 |
+
| 0.0212 | 26.0 | 988 | 0.0377 | 0.9873 | 0.9868 | 0.9868 | 0.9889 | 0.9868 | 0.0111 |
|
| 91 |
+
| 0.0175 | 27.0 | 1026 | 0.0195 | 0.9937 | 0.9934 | 0.9934 | 0.9944 | 0.9934 | 0.0056 |
|
| 92 |
+
| 0.0125 | 28.0 | 1064 | 0.0214 | 0.9936 | 0.9934 | 0.9934 | 0.9933 | 0.9934 | 0.0067 |
|
| 93 |
+
| 0.0155 | 29.0 | 1102 | 0.0128 | 1.0 | 1.0 | 1.0 | 1.0 | 1.0 | 0.0 |
|
| 94 |
+
| 0.0104 | 30.0 | 1140 | 0.0159 | 0.9937 | 0.9934 | 0.9934 | 0.9944 | 0.9934 | 0.0056 |
|
| 95 |
|
| 96 |
|
| 97 |
### Framework versions
|
| 98 |
|
| 99 |
+
- Transformers 4.44.2
|
| 100 |
- Pytorch 2.3.1+cu121
|
| 101 |
- Datasets 3.0.1
|
| 102 |
+
- Tokenizers 0.19.1
|
all_results.json
CHANGED
|
@@ -1,18 +1,18 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_accuracy": 0
|
| 4 |
-
"eval_error_rate": 0.
|
| 5 |
-
"eval_f1": 0
|
| 6 |
-
"eval_loss": 0.
|
| 7 |
-
"eval_precision": 0
|
| 8 |
-
"eval_recall": 0
|
| 9 |
-
"eval_runtime":
|
| 10 |
-
"eval_samples_per_second":
|
| 11 |
-
"eval_steps_per_second": 0.
|
| 12 |
-
"eval_top1_accuracy": 0
|
| 13 |
-
"total_flos":
|
| 14 |
-
"train_loss": 0.
|
| 15 |
-
"train_runtime":
|
| 16 |
-
"train_samples_per_second":
|
| 17 |
-
"train_steps_per_second": 0.
|
| 18 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 30.0,
|
| 3 |
+
"eval_accuracy": 1.0,
|
| 4 |
+
"eval_error_rate": 0.0,
|
| 5 |
+
"eval_f1": 1.0,
|
| 6 |
+
"eval_loss": 0.012786287814378738,
|
| 7 |
+
"eval_precision": 1.0,
|
| 8 |
+
"eval_recall": 1.0,
|
| 9 |
+
"eval_runtime": 55.4352,
|
| 10 |
+
"eval_samples_per_second": 2.742,
|
| 11 |
+
"eval_steps_per_second": 0.343,
|
| 12 |
+
"eval_top1_accuracy": 1.0,
|
| 13 |
+
"total_flos": 9.167322602550067e+17,
|
| 14 |
+
"train_loss": 0.6629255272840199,
|
| 15 |
+
"train_runtime": 13519.4546,
|
| 16 |
+
"train_samples_per_second": 2.698,
|
| 17 |
+
"train_steps_per_second": 0.084
|
| 18 |
}
|
classification_report.png
ADDED
|
config.json
CHANGED
|
@@ -54,5 +54,5 @@
|
|
| 54 |
"stage4"
|
| 55 |
],
|
| 56 |
"torch_dtype": "float32",
|
| 57 |
-
"transformers_version": "4.
|
| 58 |
}
|
|
|
|
| 54 |
"stage4"
|
| 55 |
],
|
| 56 |
"torch_dtype": "float32",
|
| 57 |
+
"transformers_version": "4.44.2"
|
| 58 |
}
|
eval_results.json
CHANGED
|
@@ -1,13 +1,13 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_accuracy": 0
|
| 4 |
-
"eval_error_rate": 0.
|
| 5 |
-
"eval_f1": 0
|
| 6 |
-
"eval_loss": 0.
|
| 7 |
-
"eval_precision": 0
|
| 8 |
-
"eval_recall": 0
|
| 9 |
-
"eval_runtime":
|
| 10 |
-
"eval_samples_per_second":
|
| 11 |
-
"eval_steps_per_second": 0.
|
| 12 |
-
"eval_top1_accuracy": 0
|
| 13 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 30.0,
|
| 3 |
+
"eval_accuracy": 1.0,
|
| 4 |
+
"eval_error_rate": 0.0,
|
| 5 |
+
"eval_f1": 1.0,
|
| 6 |
+
"eval_loss": 0.012786287814378738,
|
| 7 |
+
"eval_precision": 1.0,
|
| 8 |
+
"eval_recall": 1.0,
|
| 9 |
+
"eval_runtime": 55.4352,
|
| 10 |
+
"eval_samples_per_second": 2.742,
|
| 11 |
+
"eval_steps_per_second": 0.343,
|
| 12 |
+
"eval_top1_accuracy": 1.0
|
| 13 |
}
|
integrated_gradients_grid.jpg
ADDED
|
Git LFS Details
|
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 111317164
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7e5faa27c00aa42aa187e40f60ad05ba472a4242928fe76fc51a447cebef11be
|
| 3 |
size 111317164
|
train_and_eval.jpg
ADDED
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"total_flos":
|
| 4 |
-
"train_loss": 0.
|
| 5 |
-
"train_runtime":
|
| 6 |
-
"train_samples_per_second":
|
| 7 |
-
"train_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 30.0,
|
| 3 |
+
"total_flos": 9.167322602550067e+17,
|
| 4 |
+
"train_loss": 0.6629255272840199,
|
| 5 |
+
"train_runtime": 13519.4546,
|
| 6 |
+
"train_samples_per_second": 2.698,
|
| 7 |
+
"train_steps_per_second": 0.084
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,424 +1,774 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 0.
|
| 3 |
-
"best_model_checkpoint": "convnext-tiny-224-finetuned-barkley\\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": 1.0,
|
| 13 |
-
"train_accuracy": 0.
|
| 14 |
},
|
| 15 |
{
|
| 16 |
"epoch": 1.0,
|
| 17 |
-
"grad_norm": 2.
|
| 18 |
-
"learning_rate":
|
| 19 |
-
"loss": 1.
|
| 20 |
"step": 38
|
| 21 |
},
|
| 22 |
{
|
| 23 |
"epoch": 1.0,
|
| 24 |
-
"eval_accuracy": 0.
|
| 25 |
-
"eval_error_rate": 0.
|
| 26 |
-
"eval_f1": 0.
|
| 27 |
-
"eval_loss": 1.
|
| 28 |
-
"eval_precision": 0.
|
| 29 |
-
"eval_recall": 0.
|
| 30 |
-
"eval_runtime":
|
| 31 |
-
"eval_samples_per_second":
|
| 32 |
-
"eval_steps_per_second": 0.
|
| 33 |
-
"eval_top1_accuracy": 0.
|
| 34 |
"step": 38
|
| 35 |
},
|
| 36 |
{
|
| 37 |
"epoch": 2.0,
|
| 38 |
-
"train_accuracy": 0.
|
| 39 |
},
|
| 40 |
{
|
| 41 |
"epoch": 2.0,
|
| 42 |
-
"grad_norm": 2.
|
| 43 |
-
"learning_rate":
|
| 44 |
-
"loss": 1.
|
| 45 |
"step": 76
|
| 46 |
},
|
| 47 |
{
|
| 48 |
"epoch": 2.0,
|
| 49 |
-
"eval_accuracy": 0.
|
| 50 |
-
"eval_error_rate": 0.
|
| 51 |
-
"eval_f1": 0.
|
| 52 |
-
"eval_loss": 1.
|
| 53 |
-
"eval_precision": 0.
|
| 54 |
-
"eval_recall": 0.
|
| 55 |
-
"eval_runtime":
|
| 56 |
-
"eval_samples_per_second":
|
| 57 |
-
"eval_steps_per_second": 0.
|
| 58 |
-
"eval_top1_accuracy": 0.
|
| 59 |
"step": 76
|
| 60 |
},
|
| 61 |
{
|
| 62 |
"epoch": 3.0,
|
| 63 |
-
"train_accuracy": 0.
|
| 64 |
},
|
| 65 |
{
|
| 66 |
"epoch": 3.0,
|
| 67 |
-
"grad_norm": 2.
|
| 68 |
-
"learning_rate":
|
| 69 |
-
"loss": 1.
|
| 70 |
"step": 114
|
| 71 |
},
|
| 72 |
{
|
| 73 |
"epoch": 3.0,
|
| 74 |
-
"eval_accuracy": 0.
|
| 75 |
-
"eval_error_rate": 0.
|
| 76 |
-
"eval_f1": 0.
|
| 77 |
-
"eval_loss": 1.
|
| 78 |
-
"eval_precision": 0.
|
| 79 |
-
"eval_recall": 0.
|
| 80 |
-
"eval_runtime":
|
| 81 |
-
"eval_samples_per_second":
|
| 82 |
-
"eval_steps_per_second": 0.
|
| 83 |
-
"eval_top1_accuracy": 0.
|
| 84 |
"step": 114
|
| 85 |
},
|
| 86 |
{
|
| 87 |
"epoch": 4.0,
|
| 88 |
-
"train_accuracy": 0.
|
| 89 |
},
|
| 90 |
{
|
| 91 |
"epoch": 4.0,
|
| 92 |
-
"grad_norm": 2.
|
| 93 |
-
"learning_rate":
|
| 94 |
-
"loss": 1.
|
| 95 |
"step": 152
|
| 96 |
},
|
| 97 |
{
|
| 98 |
"epoch": 4.0,
|
| 99 |
-
"eval_accuracy": 0.
|
| 100 |
-
"eval_error_rate": 0.
|
| 101 |
-
"eval_f1": 0.
|
| 102 |
-
"eval_loss": 1.
|
| 103 |
-
"eval_precision": 0.
|
| 104 |
-
"eval_recall": 0.
|
| 105 |
-
"eval_runtime":
|
| 106 |
-
"eval_samples_per_second":
|
| 107 |
-
"eval_steps_per_second": 0.
|
| 108 |
-
"eval_top1_accuracy": 0.
|
| 109 |
"step": 152
|
| 110 |
},
|
| 111 |
{
|
| 112 |
"epoch": 5.0,
|
| 113 |
-
"train_accuracy": 0.
|
| 114 |
},
|
| 115 |
{
|
| 116 |
"epoch": 5.0,
|
| 117 |
-
"grad_norm":
|
| 118 |
-
"learning_rate":
|
| 119 |
-
"loss": 1.
|
| 120 |
"step": 190
|
| 121 |
},
|
| 122 |
{
|
| 123 |
"epoch": 5.0,
|
| 124 |
-
"eval_accuracy": 0.
|
| 125 |
-
"eval_error_rate": 0.
|
| 126 |
-
"eval_f1": 0.
|
| 127 |
-
"eval_loss": 1.
|
| 128 |
-
"eval_precision": 0.
|
| 129 |
-
"eval_recall": 0.
|
| 130 |
-
"eval_runtime":
|
| 131 |
-
"eval_samples_per_second":
|
| 132 |
-
"eval_steps_per_second": 0.
|
| 133 |
-
"eval_top1_accuracy": 0.
|
| 134 |
"step": 190
|
| 135 |
},
|
| 136 |
{
|
| 137 |
"epoch": 6.0,
|
| 138 |
-
"train_accuracy": 0.
|
| 139 |
},
|
| 140 |
{
|
| 141 |
"epoch": 6.0,
|
| 142 |
-
"grad_norm": 2.
|
| 143 |
-
"learning_rate":
|
| 144 |
-
"loss": 1.
|
| 145 |
"step": 228
|
| 146 |
},
|
| 147 |
{
|
| 148 |
"epoch": 6.0,
|
| 149 |
-
"eval_accuracy": 0.
|
| 150 |
-
"eval_error_rate": 0.
|
| 151 |
-
"eval_f1": 0.
|
| 152 |
-
"eval_loss":
|
| 153 |
-
"eval_precision": 0.
|
| 154 |
-
"eval_recall": 0.
|
| 155 |
-
"eval_runtime":
|
| 156 |
-
"eval_samples_per_second": 2.
|
| 157 |
-
"eval_steps_per_second": 0.
|
| 158 |
-
"eval_top1_accuracy": 0.
|
| 159 |
"step": 228
|
| 160 |
},
|
| 161 |
{
|
| 162 |
"epoch": 7.0,
|
| 163 |
-
"train_accuracy": 0.
|
| 164 |
},
|
| 165 |
{
|
| 166 |
"epoch": 7.0,
|
| 167 |
-
"grad_norm": 3.
|
| 168 |
-
"learning_rate":
|
| 169 |
-
"loss":
|
| 170 |
"step": 266
|
| 171 |
},
|
| 172 |
{
|
| 173 |
"epoch": 7.0,
|
| 174 |
-
"eval_accuracy": 0.
|
| 175 |
-
"eval_error_rate": 0.
|
| 176 |
-
"eval_f1": 0.
|
| 177 |
-
"eval_loss":
|
| 178 |
-
"eval_precision": 0.
|
| 179 |
-
"eval_recall": 0.
|
| 180 |
-
"eval_runtime":
|
| 181 |
-
"eval_samples_per_second": 2.
|
| 182 |
-
"eval_steps_per_second": 0.
|
| 183 |
-
"eval_top1_accuracy": 0.
|
| 184 |
"step": 266
|
| 185 |
},
|
| 186 |
{
|
| 187 |
"epoch": 8.0,
|
| 188 |
-
"train_accuracy": 0.
|
| 189 |
},
|
| 190 |
{
|
| 191 |
"epoch": 8.0,
|
| 192 |
-
"grad_norm":
|
| 193 |
-
"learning_rate":
|
| 194 |
-
"loss":
|
| 195 |
"step": 304
|
| 196 |
},
|
| 197 |
{
|
| 198 |
"epoch": 8.0,
|
| 199 |
-
"eval_accuracy": 0.
|
| 200 |
-
"eval_error_rate": 0.
|
| 201 |
-
"eval_f1": 0.
|
| 202 |
-
"eval_loss":
|
| 203 |
-
"eval_precision": 0.
|
| 204 |
-
"eval_recall": 0.
|
| 205 |
-
"eval_runtime":
|
| 206 |
-
"eval_samples_per_second":
|
| 207 |
-
"eval_steps_per_second": 0.
|
| 208 |
-
"eval_top1_accuracy": 0.
|
| 209 |
"step": 304
|
| 210 |
},
|
| 211 |
{
|
| 212 |
"epoch": 9.0,
|
| 213 |
-
"train_accuracy": 0.
|
| 214 |
},
|
| 215 |
{
|
| 216 |
"epoch": 9.0,
|
| 217 |
-
"grad_norm": 2.
|
| 218 |
-
"learning_rate":
|
| 219 |
-
"loss":
|
| 220 |
"step": 342
|
| 221 |
},
|
| 222 |
{
|
| 223 |
"epoch": 9.0,
|
| 224 |
-
"eval_accuracy": 0.
|
| 225 |
-
"eval_error_rate": 0.
|
| 226 |
-
"eval_f1": 0.
|
| 227 |
-
"eval_loss":
|
| 228 |
-
"eval_precision": 0.
|
| 229 |
-
"eval_recall": 0.
|
| 230 |
-
"eval_runtime":
|
| 231 |
-
"eval_samples_per_second":
|
| 232 |
-
"eval_steps_per_second": 0.
|
| 233 |
-
"eval_top1_accuracy": 0.
|
| 234 |
"step": 342
|
| 235 |
},
|
| 236 |
{
|
| 237 |
"epoch": 10.0,
|
| 238 |
-
"train_accuracy": 0.
|
| 239 |
},
|
| 240 |
{
|
| 241 |
"epoch": 10.0,
|
| 242 |
-
"grad_norm": 3.
|
| 243 |
-
"learning_rate":
|
| 244 |
-
"loss":
|
| 245 |
"step": 380
|
| 246 |
},
|
| 247 |
{
|
| 248 |
"epoch": 10.0,
|
| 249 |
-
"eval_accuracy": 0.
|
| 250 |
-
"eval_error_rate": 0.
|
| 251 |
-
"eval_f1": 0.
|
| 252 |
-
"eval_loss":
|
| 253 |
-
"eval_precision": 0.
|
| 254 |
-
"eval_recall": 0.
|
| 255 |
-
"eval_runtime":
|
| 256 |
-
"eval_samples_per_second":
|
| 257 |
-
"eval_steps_per_second": 0.
|
| 258 |
-
"eval_top1_accuracy": 0.
|
| 259 |
"step": 380
|
| 260 |
},
|
| 261 |
{
|
| 262 |
"epoch": 11.0,
|
| 263 |
-
"train_accuracy": 0.
|
| 264 |
},
|
| 265 |
{
|
| 266 |
"epoch": 11.0,
|
| 267 |
-
"grad_norm":
|
| 268 |
-
"learning_rate":
|
| 269 |
-
"loss":
|
| 270 |
"step": 418
|
| 271 |
},
|
| 272 |
{
|
| 273 |
"epoch": 11.0,
|
| 274 |
-
"eval_accuracy": 0.
|
| 275 |
-
"eval_error_rate": 0.
|
| 276 |
-
"eval_f1": 0.
|
| 277 |
-
"eval_loss": 0.
|
| 278 |
-
"eval_precision": 0.
|
| 279 |
-
"eval_recall": 0.
|
| 280 |
-
"eval_runtime":
|
| 281 |
-
"eval_samples_per_second": 2.
|
| 282 |
-
"eval_steps_per_second": 0.
|
| 283 |
-
"eval_top1_accuracy": 0.
|
| 284 |
"step": 418
|
| 285 |
},
|
| 286 |
{
|
| 287 |
"epoch": 12.0,
|
| 288 |
-
"train_accuracy": 0.
|
| 289 |
},
|
| 290 |
{
|
| 291 |
"epoch": 12.0,
|
| 292 |
-
"grad_norm":
|
| 293 |
-
"learning_rate":
|
| 294 |
-
"loss": 0.
|
| 295 |
"step": 456
|
| 296 |
},
|
| 297 |
{
|
| 298 |
"epoch": 12.0,
|
| 299 |
-
"eval_accuracy": 0.
|
| 300 |
-
"eval_error_rate": 0.
|
| 301 |
-
"eval_f1": 0.
|
| 302 |
-
"eval_loss": 0.
|
| 303 |
-
"eval_precision": 0.
|
| 304 |
-
"eval_recall": 0.
|
| 305 |
-
"eval_runtime":
|
| 306 |
-
"eval_samples_per_second": 2.
|
| 307 |
-
"eval_steps_per_second": 0.
|
| 308 |
-
"eval_top1_accuracy": 0.
|
| 309 |
"step": 456
|
| 310 |
},
|
| 311 |
{
|
| 312 |
"epoch": 13.0,
|
| 313 |
-
"train_accuracy": 0.
|
| 314 |
},
|
| 315 |
{
|
| 316 |
"epoch": 13.0,
|
| 317 |
-
"grad_norm":
|
| 318 |
-
"learning_rate":
|
| 319 |
-
"loss": 0.
|
| 320 |
"step": 494
|
| 321 |
},
|
| 322 |
{
|
| 323 |
"epoch": 13.0,
|
| 324 |
-
"eval_accuracy": 0.
|
| 325 |
-
"eval_error_rate": 0.
|
| 326 |
-
"eval_f1": 0.
|
| 327 |
-
"eval_loss": 0.
|
| 328 |
-
"eval_precision": 0.
|
| 329 |
-
"eval_recall": 0.
|
| 330 |
-
"eval_runtime":
|
| 331 |
-
"eval_samples_per_second": 2.
|
| 332 |
-
"eval_steps_per_second": 0.
|
| 333 |
-
"eval_top1_accuracy": 0.
|
| 334 |
"step": 494
|
| 335 |
},
|
| 336 |
{
|
| 337 |
"epoch": 14.0,
|
| 338 |
-
"train_accuracy": 0.
|
| 339 |
},
|
| 340 |
{
|
| 341 |
"epoch": 14.0,
|
| 342 |
-
"grad_norm":
|
| 343 |
-
"learning_rate":
|
| 344 |
-
"loss": 0.
|
| 345 |
"step": 532
|
| 346 |
},
|
| 347 |
{
|
| 348 |
"epoch": 14.0,
|
| 349 |
-
"eval_accuracy": 0.
|
| 350 |
-
"eval_error_rate": 0.
|
| 351 |
-
"eval_f1": 0.
|
| 352 |
-
"eval_loss": 0.
|
| 353 |
-
"eval_precision": 0.
|
| 354 |
-
"eval_recall": 0.
|
| 355 |
-
"eval_runtime":
|
| 356 |
-
"eval_samples_per_second": 2.
|
| 357 |
-
"eval_steps_per_second": 0.
|
| 358 |
-
"eval_top1_accuracy": 0.
|
| 359 |
"step": 532
|
| 360 |
},
|
| 361 |
{
|
| 362 |
"epoch": 15.0,
|
| 363 |
-
"train_accuracy": 0.
|
| 364 |
},
|
| 365 |
{
|
| 366 |
"epoch": 15.0,
|
| 367 |
-
"grad_norm":
|
| 368 |
-
"learning_rate":
|
| 369 |
-
"loss": 0.
|
| 370 |
"step": 570
|
| 371 |
},
|
| 372 |
{
|
| 373 |
"epoch": 15.0,
|
| 374 |
-
"eval_accuracy": 0.
|
| 375 |
-
"eval_error_rate": 0.
|
| 376 |
-
"eval_f1": 0.
|
| 377 |
-
"eval_loss": 0.
|
| 378 |
-
"eval_precision": 0.
|
| 379 |
-
"eval_recall": 0.
|
| 380 |
-
"eval_runtime":
|
| 381 |
-
"eval_samples_per_second":
|
| 382 |
-
"eval_steps_per_second": 0.
|
| 383 |
-
"eval_top1_accuracy": 0.
|
| 384 |
"step": 570
|
| 385 |
},
|
| 386 |
{
|
| 387 |
"epoch": 16.0,
|
| 388 |
-
"train_accuracy": 0.
|
| 389 |
},
|
| 390 |
{
|
| 391 |
"epoch": 16.0,
|
| 392 |
-
"grad_norm":
|
| 393 |
-
"learning_rate":
|
| 394 |
-
"loss": 0.
|
| 395 |
"step": 608
|
| 396 |
},
|
| 397 |
{
|
| 398 |
"epoch": 16.0,
|
| 399 |
-
"eval_accuracy": 0.
|
| 400 |
-
"eval_error_rate": 0.
|
| 401 |
-
"eval_f1": 0.
|
| 402 |
-
"eval_loss": 0.
|
| 403 |
-
"eval_precision": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 404 |
"eval_recall": 0.9868421052631579,
|
| 405 |
-
"eval_runtime":
|
| 406 |
-
"eval_samples_per_second":
|
| 407 |
-
"eval_steps_per_second": 0.
|
| 408 |
"eval_top1_accuracy": 0.9868421052631579,
|
| 409 |
-
"step":
|
| 410 |
},
|
| 411 |
{
|
| 412 |
-
"epoch":
|
| 413 |
-
"
|
| 414 |
-
|
| 415 |
-
|
| 416 |
-
"
|
| 417 |
-
"
|
| 418 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 419 |
}
|
| 420 |
],
|
| 421 |
-
"logging_steps":
|
| 422 |
"max_steps": 1140,
|
| 423 |
"num_input_tokens_seen": 0,
|
| 424 |
"num_train_epochs": 30,
|
|
@@ -430,7 +780,7 @@
|
|
| 430 |
"early_stopping_threshold": 0.0
|
| 431 |
},
|
| 432 |
"attributes": {
|
| 433 |
-
"early_stopping_patience_counter":
|
| 434 |
}
|
| 435 |
},
|
| 436 |
"TrainerControl": {
|
|
@@ -444,8 +794,8 @@
|
|
| 444 |
"attributes": {}
|
| 445 |
}
|
| 446 |
},
|
| 447 |
-
"total_flos":
|
| 448 |
-
"train_batch_size":
|
| 449 |
"trial_name": null,
|
| 450 |
"trial_params": null
|
| 451 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 0.012786287814378738,
|
| 3 |
+
"best_model_checkpoint": "convnext-tiny-224-finetuned-barkley\\checkpoint-1102",
|
| 4 |
+
"epoch": 30.0,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 1140,
|
| 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 |
+
"train_accuracy": 0.2138157894736842
|
| 14 |
},
|
| 15 |
{
|
| 16 |
"epoch": 1.0,
|
| 17 |
+
"grad_norm": 2.7230629920959473,
|
| 18 |
+
"learning_rate": 8.365843306988109e-07,
|
| 19 |
+
"loss": 1.6288,
|
| 20 |
"step": 38
|
| 21 |
},
|
| 22 |
{
|
| 23 |
"epoch": 1.0,
|
| 24 |
+
"eval_accuracy": 0.2371292106586224,
|
| 25 |
+
"eval_error_rate": 0.7628707893413775,
|
| 26 |
+
"eval_f1": 0.20430842832369756,
|
| 27 |
+
"eval_loss": 1.6004831790924072,
|
| 28 |
+
"eval_precision": 0.2132792246940375,
|
| 29 |
+
"eval_recall": 0.26973684210526316,
|
| 30 |
+
"eval_runtime": 41.666,
|
| 31 |
+
"eval_samples_per_second": 3.648,
|
| 32 |
+
"eval_steps_per_second": 0.456,
|
| 33 |
+
"eval_top1_accuracy": 0.26973684210526316,
|
| 34 |
"step": 38
|
| 35 |
},
|
| 36 |
{
|
| 37 |
"epoch": 2.0,
|
| 38 |
+
"train_accuracy": 0.25
|
| 39 |
},
|
| 40 |
{
|
| 41 |
"epoch": 2.0,
|
| 42 |
+
"grad_norm": 2.8333628177642822,
|
| 43 |
+
"learning_rate": 9.460584867009364e-07,
|
| 44 |
+
"loss": 1.6059,
|
| 45 |
"step": 76
|
| 46 |
},
|
| 47 |
{
|
| 48 |
"epoch": 2.0,
|
| 49 |
+
"eval_accuracy": 0.24728506787330323,
|
| 50 |
+
"eval_error_rate": 0.7527149321266968,
|
| 51 |
+
"eval_f1": 0.2243479907737179,
|
| 52 |
+
"eval_loss": 1.5801552534103394,
|
| 53 |
+
"eval_precision": 0.2383676582761251,
|
| 54 |
+
"eval_recall": 0.27631578947368424,
|
| 55 |
+
"eval_runtime": 40.3741,
|
| 56 |
+
"eval_samples_per_second": 3.765,
|
| 57 |
+
"eval_steps_per_second": 0.471,
|
| 58 |
+
"eval_top1_accuracy": 0.27631578947368424,
|
| 59 |
"step": 76
|
| 60 |
},
|
| 61 |
{
|
| 62 |
"epoch": 3.0,
|
| 63 |
+
"train_accuracy": 0.2894736842105263
|
| 64 |
},
|
| 65 |
{
|
| 66 |
"epoch": 3.0,
|
| 67 |
+
"grad_norm": 2.8583333492279053,
|
| 68 |
+
"learning_rate": 1.1275880849384341e-06,
|
| 69 |
+
"loss": 1.5808,
|
| 70 |
"step": 114
|
| 71 |
},
|
| 72 |
{
|
| 73 |
"epoch": 3.0,
|
| 74 |
+
"eval_accuracy": 0.2744343891402715,
|
| 75 |
+
"eval_error_rate": 0.7255656108597285,
|
| 76 |
+
"eval_f1": 0.25945030547058157,
|
| 77 |
+
"eval_loss": 1.5570068359375,
|
| 78 |
+
"eval_precision": 0.2778035204289794,
|
| 79 |
+
"eval_recall": 0.3026315789473684,
|
| 80 |
+
"eval_runtime": 25.2172,
|
| 81 |
+
"eval_samples_per_second": 6.028,
|
| 82 |
+
"eval_steps_per_second": 0.753,
|
| 83 |
+
"eval_top1_accuracy": 0.3026315789473684,
|
| 84 |
"step": 114
|
| 85 |
},
|
| 86 |
{
|
| 87 |
"epoch": 4.0,
|
| 88 |
+
"train_accuracy": 0.32383040935672514
|
| 89 |
},
|
| 90 |
{
|
| 91 |
"epoch": 4.0,
|
| 92 |
+
"grad_norm": 2.166118860244751,
|
| 93 |
+
"learning_rate": 1.3797895548168056e-06,
|
| 94 |
+
"loss": 1.5555,
|
| 95 |
"step": 152
|
| 96 |
},
|
| 97 |
{
|
| 98 |
"epoch": 4.0,
|
| 99 |
+
"eval_accuracy": 0.3510859728506787,
|
| 100 |
+
"eval_error_rate": 0.6489140271493212,
|
| 101 |
+
"eval_f1": 0.3490548373858623,
|
| 102 |
+
"eval_loss": 1.5290976762771606,
|
| 103 |
+
"eval_precision": 0.3831372574793627,
|
| 104 |
+
"eval_recall": 0.375,
|
| 105 |
+
"eval_runtime": 28.1947,
|
| 106 |
+
"eval_samples_per_second": 5.391,
|
| 107 |
+
"eval_steps_per_second": 0.674,
|
| 108 |
+
"eval_top1_accuracy": 0.375,
|
| 109 |
"step": 152
|
| 110 |
},
|
| 111 |
{
|
| 112 |
"epoch": 5.0,
|
| 113 |
+
"train_accuracy": 0.38742690058479534
|
| 114 |
},
|
| 115 |
{
|
| 116 |
"epoch": 5.0,
|
| 117 |
+
"grad_norm": 2.677858591079712,
|
| 118 |
+
"learning_rate": 1.7007406834242322e-06,
|
| 119 |
+
"loss": 1.5232,
|
| 120 |
"step": 190
|
| 121 |
},
|
| 122 |
{
|
| 123 |
"epoch": 5.0,
|
| 124 |
+
"eval_accuracy": 0.41472599296128704,
|
| 125 |
+
"eval_error_rate": 0.585274007038713,
|
| 126 |
+
"eval_f1": 0.41544045031957555,
|
| 127 |
+
"eval_loss": 1.4932990074157715,
|
| 128 |
+
"eval_precision": 0.4251577933805101,
|
| 129 |
+
"eval_recall": 0.4407894736842105,
|
| 130 |
+
"eval_runtime": 25.5314,
|
| 131 |
+
"eval_samples_per_second": 5.953,
|
| 132 |
+
"eval_steps_per_second": 0.744,
|
| 133 |
+
"eval_top1_accuracy": 0.4407894736842105,
|
| 134 |
"step": 190
|
| 135 |
},
|
| 136 |
{
|
| 137 |
"epoch": 6.0,
|
| 138 |
+
"train_accuracy": 0.47076023391812866
|
| 139 |
},
|
| 140 |
{
|
| 141 |
"epoch": 6.0,
|
| 142 |
+
"grad_norm": 2.4222543239593506,
|
| 143 |
+
"learning_rate": 2.087995266130065e-06,
|
| 144 |
+
"loss": 1.4784,
|
| 145 |
"step": 228
|
| 146 |
},
|
| 147 |
{
|
| 148 |
"epoch": 6.0,
|
| 149 |
+
"eval_accuracy": 0.49721468074409253,
|
| 150 |
+
"eval_error_rate": 0.5027853192559075,
|
| 151 |
+
"eval_f1": 0.4925881060058483,
|
| 152 |
+
"eval_loss": 1.4484349489212036,
|
| 153 |
+
"eval_precision": 0.5076020884114232,
|
| 154 |
+
"eval_recall": 0.5197368421052632,
|
| 155 |
+
"eval_runtime": 56.6448,
|
| 156 |
+
"eval_samples_per_second": 2.683,
|
| 157 |
+
"eval_steps_per_second": 0.335,
|
| 158 |
+
"eval_top1_accuracy": 0.5197368421052632,
|
| 159 |
"step": 228
|
| 160 |
},
|
| 161 |
{
|
| 162 |
"epoch": 7.0,
|
| 163 |
+
"train_accuracy": 0.5701754385964912
|
| 164 |
},
|
| 165 |
{
|
| 166 |
"epoch": 7.0,
|
| 167 |
+
"grad_norm": 3.0168983936309814,
|
| 168 |
+
"learning_rate": 2.5386017509096417e-06,
|
| 169 |
+
"loss": 1.4242,
|
| 170 |
"step": 266
|
| 171 |
},
|
| 172 |
{
|
| 173 |
"epoch": 7.0,
|
| 174 |
+
"eval_accuracy": 0.6248768225238813,
|
| 175 |
+
"eval_error_rate": 0.3751231774761187,
|
| 176 |
+
"eval_f1": 0.6307132735050356,
|
| 177 |
+
"eval_loss": 1.390175223350525,
|
| 178 |
+
"eval_precision": 0.6856800923548139,
|
| 179 |
+
"eval_recall": 0.6381578947368421,
|
| 180 |
+
"eval_runtime": 59.0359,
|
| 181 |
+
"eval_samples_per_second": 2.575,
|
| 182 |
+
"eval_steps_per_second": 0.322,
|
| 183 |
+
"eval_top1_accuracy": 0.6381578947368421,
|
| 184 |
"step": 266
|
| 185 |
},
|
| 186 |
{
|
| 187 |
"epoch": 8.0,
|
| 188 |
+
"train_accuracy": 0.6498538011695907
|
| 189 |
},
|
| 190 |
{
|
| 191 |
"epoch": 8.0,
|
| 192 |
+
"grad_norm": 2.808101177215576,
|
| 193 |
+
"learning_rate": 3.049125734293339e-06,
|
| 194 |
+
"loss": 1.3586,
|
| 195 |
"step": 304
|
| 196 |
},
|
| 197 |
{
|
| 198 |
"epoch": 8.0,
|
| 199 |
+
"eval_accuracy": 0.7134087481146305,
|
| 200 |
+
"eval_error_rate": 0.2865912518853695,
|
| 201 |
+
"eval_f1": 0.7165677244674166,
|
| 202 |
+
"eval_loss": 1.318568229675293,
|
| 203 |
+
"eval_precision": 0.7728468899521532,
|
| 204 |
+
"eval_recall": 0.7171052631578947,
|
| 205 |
+
"eval_runtime": 55.4154,
|
| 206 |
+
"eval_samples_per_second": 2.743,
|
| 207 |
+
"eval_steps_per_second": 0.343,
|
| 208 |
+
"eval_top1_accuracy": 0.7171052631578947,
|
| 209 |
"step": 304
|
| 210 |
},
|
| 211 |
{
|
| 212 |
"epoch": 9.0,
|
| 213 |
+
"train_accuracy": 0.7149122807017544
|
| 214 |
},
|
| 215 |
{
|
| 216 |
"epoch": 9.0,
|
| 217 |
+
"grad_norm": 2.8706772327423096,
|
| 218 |
+
"learning_rate": 3.6156761374816205e-06,
|
| 219 |
+
"loss": 1.276,
|
| 220 |
"step": 342
|
| 221 |
},
|
| 222 |
{
|
| 223 |
"epoch": 9.0,
|
| 224 |
+
"eval_accuracy": 0.805972850678733,
|
| 225 |
+
"eval_error_rate": 0.19402714932126697,
|
| 226 |
+
"eval_f1": 0.8108964691671459,
|
| 227 |
+
"eval_loss": 1.223597526550293,
|
| 228 |
+
"eval_precision": 0.8547017577806745,
|
| 229 |
+
"eval_recall": 0.8026315789473685,
|
| 230 |
+
"eval_runtime": 55.5305,
|
| 231 |
+
"eval_samples_per_second": 2.737,
|
| 232 |
+
"eval_steps_per_second": 0.342,
|
| 233 |
+
"eval_top1_accuracy": 0.8026315789473685,
|
| 234 |
"step": 342
|
| 235 |
},
|
| 236 |
{
|
| 237 |
"epoch": 10.0,
|
| 238 |
+
"train_accuracy": 0.8033625730994152
|
| 239 |
},
|
| 240 |
{
|
| 241 |
"epoch": 10.0,
|
| 242 |
+
"grad_norm": 3.2001399993896484,
|
| 243 |
+
"learning_rate": 4.233934863118696e-06,
|
| 244 |
+
"loss": 1.1778,
|
| 245 |
"step": 380
|
| 246 |
},
|
| 247 |
{
|
| 248 |
"epoch": 10.0,
|
| 249 |
+
"eval_accuracy": 0.8600754147812971,
|
| 250 |
+
"eval_error_rate": 0.13992458521870288,
|
| 251 |
+
"eval_f1": 0.8609113468479812,
|
| 252 |
+
"eval_loss": 1.1122019290924072,
|
| 253 |
+
"eval_precision": 0.8898553355968974,
|
| 254 |
+
"eval_recall": 0.8552631578947368,
|
| 255 |
+
"eval_runtime": 53.869,
|
| 256 |
+
"eval_samples_per_second": 2.822,
|
| 257 |
+
"eval_steps_per_second": 0.353,
|
| 258 |
+
"eval_top1_accuracy": 0.8552631578947368,
|
| 259 |
"step": 380
|
| 260 |
},
|
| 261 |
{
|
| 262 |
"epoch": 11.0,
|
| 263 |
+
"train_accuracy": 0.868421052631579
|
| 264 |
},
|
| 265 |
{
|
| 266 |
"epoch": 11.0,
|
| 267 |
+
"grad_norm": 3.1884777545928955,
|
| 268 |
+
"learning_rate": 4.899189706688707e-06,
|
| 269 |
+
"loss": 1.0543,
|
| 270 |
"step": 418
|
| 271 |
},
|
| 272 |
{
|
| 273 |
"epoch": 11.0,
|
| 274 |
+
"eval_accuracy": 0.9004725992961287,
|
| 275 |
+
"eval_error_rate": 0.09952740070387134,
|
| 276 |
+
"eval_f1": 0.8958336111023342,
|
| 277 |
+
"eval_loss": 0.9839252829551697,
|
| 278 |
+
"eval_precision": 0.9063624095203042,
|
| 279 |
+
"eval_recall": 0.8947368421052632,
|
| 280 |
+
"eval_runtime": 54.1079,
|
| 281 |
+
"eval_samples_per_second": 2.809,
|
| 282 |
+
"eval_steps_per_second": 0.351,
|
| 283 |
+
"eval_top1_accuracy": 0.8947368421052632,
|
| 284 |
"step": 418
|
| 285 |
},
|
| 286 |
{
|
| 287 |
"epoch": 12.0,
|
| 288 |
+
"train_accuracy": 0.902046783625731
|
| 289 |
},
|
| 290 |
{
|
| 291 |
"epoch": 12.0,
|
| 292 |
+
"grad_norm": 3.692692279815674,
|
| 293 |
+
"learning_rate": 5.6063702716924525e-06,
|
| 294 |
+
"loss": 0.921,
|
| 295 |
"step": 456
|
| 296 |
},
|
| 297 |
{
|
| 298 |
"epoch": 12.0,
|
| 299 |
+
"eval_accuracy": 0.9575364504776269,
|
| 300 |
+
"eval_error_rate": 0.04246354952237308,
|
| 301 |
+
"eval_f1": 0.9537314326645565,
|
| 302 |
+
"eval_loss": 0.8418065309524536,
|
| 303 |
+
"eval_precision": 0.9540851522650184,
|
| 304 |
+
"eval_recall": 0.9539473684210527,
|
| 305 |
+
"eval_runtime": 54.166,
|
| 306 |
+
"eval_samples_per_second": 2.806,
|
| 307 |
+
"eval_steps_per_second": 0.351,
|
| 308 |
+
"eval_top1_accuracy": 0.9539473684210527,
|
| 309 |
"step": 456
|
| 310 |
},
|
| 311 |
{
|
| 312 |
"epoch": 13.0,
|
| 313 |
+
"train_accuracy": 0.9371345029239766
|
| 314 |
},
|
| 315 |
{
|
| 316 |
"epoch": 13.0,
|
| 317 |
+
"grad_norm": 3.067211389541626,
|
| 318 |
+
"learning_rate": 6.350086614868708e-06,
|
| 319 |
+
"loss": 0.773,
|
| 320 |
"step": 494
|
| 321 |
},
|
| 322 |
{
|
| 323 |
"epoch": 13.0,
|
| 324 |
+
"eval_accuracy": 0.9652287581699346,
|
| 325 |
+
"eval_error_rate": 0.034771241830065414,
|
| 326 |
+
"eval_f1": 0.9604621710987601,
|
| 327 |
+
"eval_loss": 0.6935167908668518,
|
| 328 |
+
"eval_precision": 0.9623944201691405,
|
| 329 |
+
"eval_recall": 0.9605263157894737,
|
| 330 |
+
"eval_runtime": 67.8683,
|
| 331 |
+
"eval_samples_per_second": 2.24,
|
| 332 |
+
"eval_steps_per_second": 0.28,
|
| 333 |
+
"eval_top1_accuracy": 0.9605263157894737,
|
| 334 |
"step": 494
|
| 335 |
},
|
| 336 |
{
|
| 337 |
"epoch": 14.0,
|
| 338 |
+
"train_accuracy": 0.9576023391812866
|
| 339 |
},
|
| 340 |
{
|
| 341 |
"epoch": 14.0,
|
| 342 |
+
"grad_norm": 2.8400356769561768,
|
| 343 |
+
"learning_rate": 7.124670326916465e-06,
|
| 344 |
+
"loss": 0.6204,
|
| 345 |
"step": 532
|
| 346 |
},
|
| 347 |
{
|
| 348 |
"epoch": 14.0,
|
| 349 |
+
"eval_accuracy": 0.9707843137254901,
|
| 350 |
+
"eval_error_rate": 0.029215686274509878,
|
| 351 |
+
"eval_f1": 0.96721850419808,
|
| 352 |
+
"eval_loss": 0.5515281558036804,
|
| 353 |
+
"eval_precision": 0.9687600085406214,
|
| 354 |
+
"eval_recall": 0.9671052631578947,
|
| 355 |
+
"eval_runtime": 75.408,
|
| 356 |
+
"eval_samples_per_second": 2.016,
|
| 357 |
+
"eval_steps_per_second": 0.252,
|
| 358 |
+
"eval_top1_accuracy": 0.9671052631578947,
|
| 359 |
"step": 532
|
| 360 |
},
|
| 361 |
{
|
| 362 |
"epoch": 15.0,
|
| 363 |
+
"train_accuracy": 0.9707602339181286
|
| 364 |
},
|
| 365 |
{
|
| 366 |
"epoch": 15.0,
|
| 367 |
+
"grad_norm": 2.3800432682037354,
|
| 368 |
+
"learning_rate": 7.924217735611663e-06,
|
| 369 |
+
"loss": 0.4835,
|
| 370 |
"step": 570
|
| 371 |
},
|
| 372 |
{
|
| 373 |
"epoch": 15.0,
|
| 374 |
+
"eval_accuracy": 0.9696732026143791,
|
| 375 |
+
"eval_error_rate": 0.03032679738562094,
|
| 376 |
+
"eval_f1": 0.9676301640599072,
|
| 377 |
+
"eval_loss": 0.4145788848400116,
|
| 378 |
+
"eval_precision": 0.9704260651629072,
|
| 379 |
+
"eval_recall": 0.9671052631578947,
|
| 380 |
+
"eval_runtime": 43.8505,
|
| 381 |
+
"eval_samples_per_second": 3.466,
|
| 382 |
+
"eval_steps_per_second": 0.433,
|
| 383 |
+
"eval_top1_accuracy": 0.9671052631578947,
|
| 384 |
"step": 570
|
| 385 |
},
|
| 386 |
{
|
| 387 |
"epoch": 16.0,
|
| 388 |
+
"train_accuracy": 0.9736842105263158
|
| 389 |
},
|
| 390 |
{
|
| 391 |
"epoch": 16.0,
|
| 392 |
+
"grad_norm": 3.108427047729492,
|
| 393 |
+
"learning_rate": 8.742634902035743e-06,
|
| 394 |
+
"loss": 0.3641,
|
| 395 |
"step": 608
|
| 396 |
},
|
| 397 |
{
|
| 398 |
"epoch": 16.0,
|
| 399 |
+
"eval_accuracy": 0.9830065359477125,
|
| 400 |
+
"eval_error_rate": 0.01699346405228752,
|
| 401 |
+
"eval_f1": 0.980191060766086,
|
| 402 |
+
"eval_loss": 0.3042638599872589,
|
| 403 |
+
"eval_precision": 0.9805310767959324,
|
| 404 |
+
"eval_recall": 0.9802631578947368,
|
| 405 |
+
"eval_runtime": 32.0995,
|
| 406 |
+
"eval_samples_per_second": 4.735,
|
| 407 |
+
"eval_steps_per_second": 0.592,
|
| 408 |
+
"eval_top1_accuracy": 0.9802631578947368,
|
| 409 |
+
"step": 608
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 17.0,
|
| 413 |
+
"train_accuracy": 0.9788011695906432
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 17.0,
|
| 417 |
+
"grad_norm": 2.5407485961914062,
|
| 418 |
+
"learning_rate": 9.573684066966612e-06,
|
| 419 |
+
"loss": 0.2706,
|
| 420 |
+
"step": 646
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"epoch": 17.0,
|
| 424 |
+
"eval_accuracy": 0.9830065359477125,
|
| 425 |
+
"eval_error_rate": 0.01699346405228752,
|
| 426 |
+
"eval_f1": 0.980191060766086,
|
| 427 |
+
"eval_loss": 0.22474148869514465,
|
| 428 |
+
"eval_precision": 0.9805310767959324,
|
| 429 |
+
"eval_recall": 0.9802631578947368,
|
| 430 |
+
"eval_runtime": 44.9328,
|
| 431 |
+
"eval_samples_per_second": 3.383,
|
| 432 |
+
"eval_steps_per_second": 0.423,
|
| 433 |
+
"eval_top1_accuracy": 0.9802631578947368,
|
| 434 |
+
"step": 646
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 18.0,
|
| 438 |
+
"train_accuracy": 0.9817251461988304
|
| 439 |
+
},
|
| 440 |
+
{
|
| 441 |
+
"epoch": 18.0,
|
| 442 |
+
"grad_norm": 3.097386598587036,
|
| 443 |
+
"learning_rate": 1.0411031193429937e-05,
|
| 444 |
+
"loss": 0.1998,
|
| 445 |
+
"step": 684
|
| 446 |
+
},
|
| 447 |
+
{
|
| 448 |
+
"epoch": 18.0,
|
| 449 |
+
"eval_accuracy": 0.9888888888888889,
|
| 450 |
+
"eval_error_rate": 0.011111111111111072,
|
| 451 |
+
"eval_f1": 0.9867701266776593,
|
| 452 |
+
"eval_loss": 0.17049287259578705,
|
| 453 |
+
"eval_precision": 0.9872979940891655,
|
| 454 |
"eval_recall": 0.9868421052631579,
|
| 455 |
+
"eval_runtime": 39.0056,
|
| 456 |
+
"eval_samples_per_second": 3.897,
|
| 457 |
+
"eval_steps_per_second": 0.487,
|
| 458 |
"eval_top1_accuracy": 0.9868421052631579,
|
| 459 |
+
"step": 684
|
| 460 |
},
|
| 461 |
{
|
| 462 |
+
"epoch": 19.0,
|
| 463 |
+
"train_accuracy": 0.9890350877192983
|
| 464 |
+
},
|
| 465 |
+
{
|
| 466 |
+
"epoch": 19.0,
|
| 467 |
+
"grad_norm": 2.5862479209899902,
|
| 468 |
+
"learning_rate": 1.1248294243054004e-05,
|
| 469 |
+
"loss": 0.1446,
|
| 470 |
+
"step": 722
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 19.0,
|
| 474 |
+
"eval_accuracy": 0.9944444444444445,
|
| 475 |
+
"eval_error_rate": 0.005555555555555536,
|
| 476 |
+
"eval_f1": 0.99343678755752,
|
| 477 |
+
"eval_loss": 0.12706294655799866,
|
| 478 |
+
"eval_precision": 0.9936647173489279,
|
| 479 |
+
"eval_recall": 0.993421052631579,
|
| 480 |
+
"eval_runtime": 35.4207,
|
| 481 |
+
"eval_samples_per_second": 4.291,
|
| 482 |
+
"eval_steps_per_second": 0.536,
|
| 483 |
+
"eval_top1_accuracy": 0.993421052631579,
|
| 484 |
+
"step": 722
|
| 485 |
+
},
|
| 486 |
+
{
|
| 487 |
+
"epoch": 20.0,
|
| 488 |
+
"train_accuracy": 0.9897660818713451
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 20.0,
|
| 492 |
+
"grad_norm": 4.520482540130615,
|
| 493 |
+
"learning_rate": 1.2079091818278531e-05,
|
| 494 |
+
"loss": 0.1106,
|
| 495 |
+
"step": 760
|
| 496 |
+
},
|
| 497 |
+
{
|
| 498 |
+
"epoch": 20.0,
|
| 499 |
+
"eval_accuracy": 0.9888888888888889,
|
| 500 |
+
"eval_error_rate": 0.011111111111111072,
|
| 501 |
+
"eval_f1": 0.9867701266776593,
|
| 502 |
+
"eval_loss": 0.1046518012881279,
|
| 503 |
+
"eval_precision": 0.9872979940891655,
|
| 504 |
+
"eval_recall": 0.9868421052631579,
|
| 505 |
+
"eval_runtime": 41.2892,
|
| 506 |
+
"eval_samples_per_second": 3.681,
|
| 507 |
+
"eval_steps_per_second": 0.46,
|
| 508 |
+
"eval_top1_accuracy": 0.9868421052631579,
|
| 509 |
+
"step": 760
|
| 510 |
+
},
|
| 511 |
+
{
|
| 512 |
+
"epoch": 21.0,
|
| 513 |
+
"train_accuracy": 0.9883040935672515
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 21.0,
|
| 517 |
+
"grad_norm": 9.188246726989746,
|
| 518 |
+
"learning_rate": 1.2897091799679402e-05,
|
| 519 |
+
"loss": 0.0872,
|
| 520 |
+
"step": 798
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 21.0,
|
| 524 |
+
"eval_accuracy": 0.9944444444444445,
|
| 525 |
+
"eval_error_rate": 0.005555555555555536,
|
| 526 |
+
"eval_f1": 0.99343678755752,
|
| 527 |
+
"eval_loss": 0.0779847577214241,
|
| 528 |
+
"eval_precision": 0.9936647173489279,
|
| 529 |
+
"eval_recall": 0.993421052631579,
|
| 530 |
+
"eval_runtime": 68.0671,
|
| 531 |
+
"eval_samples_per_second": 2.233,
|
| 532 |
+
"eval_steps_per_second": 0.279,
|
| 533 |
+
"eval_top1_accuracy": 0.993421052631579,
|
| 534 |
+
"step": 798
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 22.0,
|
| 538 |
+
"train_accuracy": 0.993421052631579
|
| 539 |
+
},
|
| 540 |
+
{
|
| 541 |
+
"epoch": 22.0,
|
| 542 |
+
"grad_norm": 1.4848785400390625,
|
| 543 |
+
"learning_rate": 1.3696059607708444e-05,
|
| 544 |
+
"loss": 0.0614,
|
| 545 |
+
"step": 836
|
| 546 |
+
},
|
| 547 |
+
{
|
| 548 |
+
"epoch": 22.0,
|
| 549 |
+
"eval_accuracy": 0.9888888888888889,
|
| 550 |
+
"eval_error_rate": 0.011111111111111072,
|
| 551 |
+
"eval_f1": 0.9867701266776593,
|
| 552 |
+
"eval_loss": 0.07387673109769821,
|
| 553 |
+
"eval_precision": 0.9872979940891655,
|
| 554 |
+
"eval_recall": 0.9868421052631579,
|
| 555 |
+
"eval_runtime": 57.9913,
|
| 556 |
+
"eval_samples_per_second": 2.621,
|
| 557 |
+
"eval_steps_per_second": 0.328,
|
| 558 |
+
"eval_top1_accuracy": 0.9868421052631579,
|
| 559 |
+
"step": 836
|
| 560 |
+
},
|
| 561 |
+
{
|
| 562 |
+
"epoch": 23.0,
|
| 563 |
+
"train_accuracy": 0.9948830409356725
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 23.0,
|
| 567 |
+
"grad_norm": 0.7044534683227539,
|
| 568 |
+
"learning_rate": 1.4469905721010107e-05,
|
| 569 |
+
"loss": 0.0491,
|
| 570 |
+
"step": 874
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 23.0,
|
| 574 |
+
"eval_accuracy": 0.9944444444444445,
|
| 575 |
+
"eval_error_rate": 0.005555555555555536,
|
| 576 |
+
"eval_f1": 0.99343678755752,
|
| 577 |
+
"eval_loss": 0.05167735368013382,
|
| 578 |
+
"eval_precision": 0.9936647173489279,
|
| 579 |
+
"eval_recall": 0.993421052631579,
|
| 580 |
+
"eval_runtime": 43.1178,
|
| 581 |
+
"eval_samples_per_second": 3.525,
|
| 582 |
+
"eval_steps_per_second": 0.441,
|
| 583 |
+
"eval_top1_accuracy": 0.993421052631579,
|
| 584 |
+
"step": 874
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 24.0,
|
| 588 |
+
"train_accuracy": 0.9956140350877193
|
| 589 |
+
},
|
| 590 |
+
{
|
| 591 |
+
"epoch": 24.0,
|
| 592 |
+
"grad_norm": 4.517460346221924,
|
| 593 |
+
"learning_rate": 1.5212732089142938e-05,
|
| 594 |
+
"loss": 0.0365,
|
| 595 |
+
"step": 912
|
| 596 |
+
},
|
| 597 |
+
{
|
| 598 |
+
"epoch": 24.0,
|
| 599 |
+
"eval_accuracy": 0.9877777777777779,
|
| 600 |
+
"eval_error_rate": 0.012222222222222134,
|
| 601 |
+
"eval_f1": 0.9868484170131115,
|
| 602 |
+
"eval_loss": 0.04010358452796936,
|
| 603 |
+
"eval_precision": 0.9870857699805068,
|
| 604 |
+
"eval_recall": 0.9868421052631579,
|
| 605 |
+
"eval_runtime": 37.7666,
|
| 606 |
+
"eval_samples_per_second": 4.025,
|
| 607 |
+
"eval_steps_per_second": 0.503,
|
| 608 |
+
"eval_top1_accuracy": 0.9868421052631579,
|
| 609 |
+
"step": 912
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 25.0,
|
| 613 |
+
"train_accuracy": 0.9956140350877193
|
| 614 |
+
},
|
| 615 |
+
{
|
| 616 |
+
"epoch": 25.0,
|
| 617 |
+
"grad_norm": 0.25067076086997986,
|
| 618 |
+
"learning_rate": 1.590081029331129e-05,
|
| 619 |
+
"loss": 0.0255,
|
| 620 |
+
"step": 950
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"epoch": 25.0,
|
| 624 |
+
"eval_accuracy": 0.9944444444444445,
|
| 625 |
+
"eval_error_rate": 0.005555555555555536,
|
| 626 |
+
"eval_f1": 0.99343678755752,
|
| 627 |
+
"eval_loss": 0.033598948270082474,
|
| 628 |
+
"eval_precision": 0.9936647173489279,
|
| 629 |
+
"eval_recall": 0.993421052631579,
|
| 630 |
+
"eval_runtime": 41.2598,
|
| 631 |
+
"eval_samples_per_second": 3.684,
|
| 632 |
+
"eval_steps_per_second": 0.46,
|
| 633 |
+
"eval_top1_accuracy": 0.993421052631579,
|
| 634 |
+
"step": 950
|
| 635 |
+
},
|
| 636 |
+
{
|
| 637 |
+
"epoch": 26.0,
|
| 638 |
+
"train_accuracy": 0.9963450292397661
|
| 639 |
+
},
|
| 640 |
+
{
|
| 641 |
+
"epoch": 26.0,
|
| 642 |
+
"grad_norm": 3.2572007179260254,
|
| 643 |
+
"learning_rate": 1.65660651368813e-05,
|
| 644 |
+
"loss": 0.0212,
|
| 645 |
+
"step": 988
|
| 646 |
+
},
|
| 647 |
+
{
|
| 648 |
+
"epoch": 26.0,
|
| 649 |
+
"eval_accuracy": 0.9888888888888889,
|
| 650 |
+
"eval_error_rate": 0.011111111111111072,
|
| 651 |
+
"eval_f1": 0.9867701266776593,
|
| 652 |
+
"eval_loss": 0.037725575268268585,
|
| 653 |
+
"eval_precision": 0.9872979940891655,
|
| 654 |
+
"eval_recall": 0.9868421052631579,
|
| 655 |
+
"eval_runtime": 39.6757,
|
| 656 |
+
"eval_samples_per_second": 3.831,
|
| 657 |
+
"eval_steps_per_second": 0.479,
|
| 658 |
+
"eval_top1_accuracy": 0.9868421052631579,
|
| 659 |
+
"step": 988
|
| 660 |
+
},
|
| 661 |
+
{
|
| 662 |
+
"epoch": 27.0,
|
| 663 |
+
"train_accuracy": 0.9956140350877193
|
| 664 |
+
},
|
| 665 |
+
{
|
| 666 |
+
"epoch": 27.0,
|
| 667 |
+
"grad_norm": 0.9294455051422119,
|
| 668 |
+
"learning_rate": 1.7184323862518377e-05,
|
| 669 |
+
"loss": 0.0175,
|
| 670 |
+
"step": 1026
|
| 671 |
+
},
|
| 672 |
+
{
|
| 673 |
+
"epoch": 27.0,
|
| 674 |
+
"eval_accuracy": 0.9944444444444445,
|
| 675 |
+
"eval_error_rate": 0.005555555555555536,
|
| 676 |
+
"eval_f1": 0.99343678755752,
|
| 677 |
+
"eval_loss": 0.019463175907731056,
|
| 678 |
+
"eval_precision": 0.9936647173489279,
|
| 679 |
+
"eval_recall": 0.993421052631579,
|
| 680 |
+
"eval_runtime": 41.079,
|
| 681 |
+
"eval_samples_per_second": 3.7,
|
| 682 |
+
"eval_steps_per_second": 0.463,
|
| 683 |
+
"eval_top1_accuracy": 0.993421052631579,
|
| 684 |
+
"step": 1026
|
| 685 |
+
},
|
| 686 |
+
{
|
| 687 |
+
"epoch": 28.0,
|
| 688 |
+
"train_accuracy": 0.9978070175438597
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 28.0,
|
| 692 |
+
"grad_norm": 3.1146209239959717,
|
| 693 |
+
"learning_rate": 1.775087426570666e-05,
|
| 694 |
+
"loss": 0.0125,
|
| 695 |
+
"step": 1064
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 28.0,
|
| 699 |
+
"eval_accuracy": 0.9933333333333334,
|
| 700 |
+
"eval_error_rate": 0.006666666666666599,
|
| 701 |
+
"eval_f1": 0.9934103601236665,
|
| 702 |
+
"eval_loss": 0.02142958901822567,
|
| 703 |
+
"eval_precision": 0.9935988620199147,
|
| 704 |
+
"eval_recall": 0.993421052631579,
|
| 705 |
+
"eval_runtime": 71.7993,
|
| 706 |
+
"eval_samples_per_second": 2.117,
|
| 707 |
+
"eval_steps_per_second": 0.265,
|
| 708 |
+
"eval_top1_accuracy": 0.993421052631579,
|
| 709 |
+
"step": 1064
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 29.0,
|
| 713 |
+
"train_accuracy": 0.9948830409356725
|
| 714 |
+
},
|
| 715 |
+
{
|
| 716 |
+
"epoch": 29.0,
|
| 717 |
+
"grad_norm": 0.6761703491210938,
|
| 718 |
+
"learning_rate": 1.826139824909036e-05,
|
| 719 |
+
"loss": 0.0155,
|
| 720 |
+
"step": 1102
|
| 721 |
+
},
|
| 722 |
+
{
|
| 723 |
+
"epoch": 29.0,
|
| 724 |
+
"eval_accuracy": 1.0,
|
| 725 |
+
"eval_error_rate": 0.0,
|
| 726 |
+
"eval_f1": 1.0,
|
| 727 |
+
"eval_loss": 0.012786287814378738,
|
| 728 |
+
"eval_precision": 1.0,
|
| 729 |
+
"eval_recall": 1.0,
|
| 730 |
+
"eval_runtime": 70.5861,
|
| 731 |
+
"eval_samples_per_second": 2.153,
|
| 732 |
+
"eval_steps_per_second": 0.269,
|
| 733 |
+
"eval_top1_accuracy": 1.0,
|
| 734 |
+
"step": 1102
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 30.0,
|
| 738 |
+
"train_accuracy": 0.9985380116959064
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 30.0,
|
| 742 |
+
"grad_norm": 0.24630075693130493,
|
| 743 |
+
"learning_rate": 1.8712004733869936e-05,
|
| 744 |
+
"loss": 0.0104,
|
| 745 |
+
"step": 1140
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 30.0,
|
| 749 |
+
"eval_accuracy": 0.9944444444444445,
|
| 750 |
+
"eval_error_rate": 0.005555555555555536,
|
| 751 |
+
"eval_f1": 0.99343678755752,
|
| 752 |
+
"eval_loss": 0.0159451961517334,
|
| 753 |
+
"eval_precision": 0.9936647173489279,
|
| 754 |
+
"eval_recall": 0.993421052631579,
|
| 755 |
+
"eval_runtime": 68.8635,
|
| 756 |
+
"eval_samples_per_second": 2.207,
|
| 757 |
+
"eval_steps_per_second": 0.276,
|
| 758 |
+
"eval_top1_accuracy": 0.993421052631579,
|
| 759 |
+
"step": 1140
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 30.0,
|
| 763 |
+
"step": 1140,
|
| 764 |
+
"total_flos": 9.167322602550067e+17,
|
| 765 |
+
"train_loss": 0.6629255272840199,
|
| 766 |
+
"train_runtime": 13519.4546,
|
| 767 |
+
"train_samples_per_second": 2.698,
|
| 768 |
+
"train_steps_per_second": 0.084
|
| 769 |
}
|
| 770 |
],
|
| 771 |
+
"logging_steps": 10,
|
| 772 |
"max_steps": 1140,
|
| 773 |
"num_input_tokens_seen": 0,
|
| 774 |
"num_train_epochs": 30,
|
|
|
|
| 780 |
"early_stopping_threshold": 0.0
|
| 781 |
},
|
| 782 |
"attributes": {
|
| 783 |
+
"early_stopping_patience_counter": 0
|
| 784 |
}
|
| 785 |
},
|
| 786 |
"TrainerControl": {
|
|
|
|
| 794 |
"attributes": {}
|
| 795 |
}
|
| 796 |
},
|
| 797 |
+
"total_flos": 9.167322602550067e+17,
|
| 798 |
+
"train_batch_size": 8,
|
| 799 |
"trial_name": null,
|
| 800 |
"trial_params": null
|
| 801 |
}
|
training_args.bin
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 sha256:591d261e4698fb2aced103c9b4eff40a248d2fd6e9b86b4707a073f3e92348b1
|
| 3 |
+
size 5112
|