initial commit
Browse files- README.md +85 -0
- all_results.json +13 -0
- config.json +36 -0
- eval_results.json +8 -0
- model.safetensors +3 -0
- preprocessor_config.json +22 -0
- runs/Aug23_10-45-51_c663354eca06/events.out.tfevents.1724409962.c663354eca06.443.0 +3 -0
- runs/Aug23_10-45-51_c663354eca06/events.out.tfevents.1724410509.c663354eca06.443.1 +3 -0
- train_results.json +8 -0
- trainer_state.json +781 -0
- training_args.bin +3 -0
README.md
ADDED
|
@@ -0,0 +1,85 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
base_model: google/vit-base-patch16-224-in21k
|
| 4 |
+
tags:
|
| 5 |
+
- image-classification
|
| 6 |
+
- generated_from_trainer
|
| 7 |
+
datasets:
|
| 8 |
+
- imagefolder
|
| 9 |
+
metrics:
|
| 10 |
+
- accuracy
|
| 11 |
+
model-index:
|
| 12 |
+
- name: vit-brain-tumour
|
| 13 |
+
results:
|
| 14 |
+
- task:
|
| 15 |
+
name: Image Classification
|
| 16 |
+
type: image-classification
|
| 17 |
+
dataset:
|
| 18 |
+
name: Simezu/brain-tumour-MRI-scan
|
| 19 |
+
type: imagefolder
|
| 20 |
+
config: default
|
| 21 |
+
split: validation
|
| 22 |
+
args: default
|
| 23 |
+
metrics:
|
| 24 |
+
- name: Accuracy
|
| 25 |
+
type: accuracy
|
| 26 |
+
value: 0.9925442684063374
|
| 27 |
+
---
|
| 28 |
+
|
| 29 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 30 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 31 |
+
|
| 32 |
+
# vit-brain-tumour
|
| 33 |
+
|
| 34 |
+
This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on the Simezu/brain-tumour-MRI-scan dataset.
|
| 35 |
+
It achieves the following results on the evaluation set:
|
| 36 |
+
- Loss: 0.0309
|
| 37 |
+
- Accuracy: 0.9925
|
| 38 |
+
|
| 39 |
+
## Model description
|
| 40 |
+
|
| 41 |
+
More information needed
|
| 42 |
+
|
| 43 |
+
## Intended uses & limitations
|
| 44 |
+
|
| 45 |
+
More information needed
|
| 46 |
+
|
| 47 |
+
## Training and evaluation data
|
| 48 |
+
|
| 49 |
+
More information needed
|
| 50 |
+
|
| 51 |
+
## Training procedure
|
| 52 |
+
|
| 53 |
+
### Training hyperparameters
|
| 54 |
+
|
| 55 |
+
The following hyperparameters were used during training:
|
| 56 |
+
- learning_rate: 0.0002
|
| 57 |
+
- train_batch_size: 16
|
| 58 |
+
- eval_batch_size: 8
|
| 59 |
+
- seed: 42
|
| 60 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 61 |
+
- lr_scheduler_type: linear
|
| 62 |
+
- num_epochs: 4
|
| 63 |
+
- mixed_precision_training: Native AMP
|
| 64 |
+
|
| 65 |
+
### Training results
|
| 66 |
+
|
| 67 |
+
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
| 68 |
+
|:-------------:|:------:|:----:|:---------------:|:--------:|
|
| 69 |
+
| 0.227 | 0.4255 | 100 | 0.3067 | 0.8910 |
|
| 70 |
+
| 0.0659 | 0.8511 | 200 | 0.1109 | 0.9627 |
|
| 71 |
+
| 0.0404 | 1.2766 | 300 | 0.0900 | 0.9776 |
|
| 72 |
+
| 0.05 | 1.7021 | 400 | 0.1082 | 0.9748 |
|
| 73 |
+
| 0.006 | 2.1277 | 500 | 0.0374 | 0.9888 |
|
| 74 |
+
| 0.0147 | 2.5532 | 600 | 0.0541 | 0.9888 |
|
| 75 |
+
| 0.0105 | 2.9787 | 700 | 0.0359 | 0.9907 |
|
| 76 |
+
| 0.0032 | 3.4043 | 800 | 0.0392 | 0.9907 |
|
| 77 |
+
| 0.0055 | 3.8298 | 900 | 0.0309 | 0.9925 |
|
| 78 |
+
|
| 79 |
+
|
| 80 |
+
### Framework versions
|
| 81 |
+
|
| 82 |
+
- Transformers 4.42.4
|
| 83 |
+
- Pytorch 2.3.1+cu121
|
| 84 |
+
- Datasets 2.21.0
|
| 85 |
+
- Tokenizers 0.19.1
|
all_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 4.0,
|
| 3 |
+
"eval_accuracy": 0.9925442684063374,
|
| 4 |
+
"eval_loss": 0.030865401029586792,
|
| 5 |
+
"eval_runtime": 5.4127,
|
| 6 |
+
"eval_samples_per_second": 198.237,
|
| 7 |
+
"eval_steps_per_second": 24.941,
|
| 8 |
+
"total_flos": 1.1651904419373711e+18,
|
| 9 |
+
"train_loss": 0.07046630619926021,
|
| 10 |
+
"train_runtime": 463.0785,
|
| 11 |
+
"train_samples_per_second": 32.47,
|
| 12 |
+
"train_steps_per_second": 2.03
|
| 13 |
+
}
|
config.json
ADDED
|
@@ -0,0 +1,36 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "google/vit-base-patch16-224-in21k",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"ViTForImageClassification"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.0,
|
| 7 |
+
"encoder_stride": 16,
|
| 8 |
+
"hidden_act": "gelu",
|
| 9 |
+
"hidden_dropout_prob": 0.0,
|
| 10 |
+
"hidden_size": 768,
|
| 11 |
+
"id2label": {
|
| 12 |
+
"0": "glioma",
|
| 13 |
+
"1": "meningioma",
|
| 14 |
+
"2": "no-tumor",
|
| 15 |
+
"3": "pituitary"
|
| 16 |
+
},
|
| 17 |
+
"image_size": 224,
|
| 18 |
+
"initializer_range": 0.02,
|
| 19 |
+
"intermediate_size": 3072,
|
| 20 |
+
"label2id": {
|
| 21 |
+
"glioma": "0",
|
| 22 |
+
"meningioma": "1",
|
| 23 |
+
"no-tumor": "2",
|
| 24 |
+
"pituitary": "3"
|
| 25 |
+
},
|
| 26 |
+
"layer_norm_eps": 1e-12,
|
| 27 |
+
"model_type": "vit",
|
| 28 |
+
"num_attention_heads": 12,
|
| 29 |
+
"num_channels": 3,
|
| 30 |
+
"num_hidden_layers": 12,
|
| 31 |
+
"patch_size": 16,
|
| 32 |
+
"problem_type": "single_label_classification",
|
| 33 |
+
"qkv_bias": true,
|
| 34 |
+
"torch_dtype": "float32",
|
| 35 |
+
"transformers_version": "4.42.4"
|
| 36 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 4.0,
|
| 3 |
+
"eval_accuracy": 0.9925442684063374,
|
| 4 |
+
"eval_loss": 0.030865401029586792,
|
| 5 |
+
"eval_runtime": 5.4127,
|
| 6 |
+
"eval_samples_per_second": 198.237,
|
| 7 |
+
"eval_steps_per_second": 24.941
|
| 8 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ab0b5ee834f3babe7532669890b91e8fec88082c80826217c200bc916b838791
|
| 3 |
+
size 343230128
|
preprocessor_config.json
ADDED
|
@@ -0,0 +1,22 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"do_normalize": true,
|
| 3 |
+
"do_rescale": true,
|
| 4 |
+
"do_resize": true,
|
| 5 |
+
"image_mean": [
|
| 6 |
+
0.5,
|
| 7 |
+
0.5,
|
| 8 |
+
0.5
|
| 9 |
+
],
|
| 10 |
+
"image_processor_type": "ViTImageProcessor",
|
| 11 |
+
"image_std": [
|
| 12 |
+
0.5,
|
| 13 |
+
0.5,
|
| 14 |
+
0.5
|
| 15 |
+
],
|
| 16 |
+
"resample": 2,
|
| 17 |
+
"rescale_factor": 0.00392156862745098,
|
| 18 |
+
"size": {
|
| 19 |
+
"height": 224,
|
| 20 |
+
"width": 224
|
| 21 |
+
}
|
| 22 |
+
}
|
runs/Aug23_10-45-51_c663354eca06/events.out.tfevents.1724409962.c663354eca06.443.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:86f5452dd2d9c15694339e9b5970903a8606772d15c1f9a68ebeb8f9d04c80fb
|
| 3 |
+
size 27966
|
runs/Aug23_10-45-51_c663354eca06/events.out.tfevents.1724410509.c663354eca06.443.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:755183e22cfde8dea884bb99c5b282a2c92b4e8907e5c33becc64d22a507bc3e
|
| 3 |
+
size 1703
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 4.0,
|
| 3 |
+
"total_flos": 1.1651904419373711e+18,
|
| 4 |
+
"train_loss": 0.07046630619926021,
|
| 5 |
+
"train_runtime": 463.0785,
|
| 6 |
+
"train_samples_per_second": 32.47,
|
| 7 |
+
"train_steps_per_second": 2.03
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,781 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.030865401029586792,
|
| 3 |
+
"best_model_checkpoint": "./vit-brain-tumour/checkpoint-900",
|
| 4 |
+
"epoch": 4.0,
|
| 5 |
+
"eval_steps": 100,
|
| 6 |
+
"global_step": 940,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.0425531914893617,
|
| 13 |
+
"grad_norm": 2.4241483211517334,
|
| 14 |
+
"learning_rate": 0.00019787234042553193,
|
| 15 |
+
"loss": 0.9058,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.0851063829787234,
|
| 20 |
+
"grad_norm": 1.0025970935821533,
|
| 21 |
+
"learning_rate": 0.00019574468085106384,
|
| 22 |
+
"loss": 0.4081,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.1276595744680851,
|
| 27 |
+
"grad_norm": 1.7178714275360107,
|
| 28 |
+
"learning_rate": 0.00019361702127659576,
|
| 29 |
+
"loss": 0.221,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.1702127659574468,
|
| 34 |
+
"grad_norm": 1.0210686922073364,
|
| 35 |
+
"learning_rate": 0.00019148936170212768,
|
| 36 |
+
"loss": 0.4859,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.2127659574468085,
|
| 41 |
+
"grad_norm": 1.8148099184036255,
|
| 42 |
+
"learning_rate": 0.00018936170212765957,
|
| 43 |
+
"loss": 0.1614,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.2553191489361702,
|
| 48 |
+
"grad_norm": 2.7210752964019775,
|
| 49 |
+
"learning_rate": 0.0001872340425531915,
|
| 50 |
+
"loss": 0.1979,
|
| 51 |
+
"step": 60
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.2978723404255319,
|
| 55 |
+
"grad_norm": 1.6892952919006348,
|
| 56 |
+
"learning_rate": 0.0001851063829787234,
|
| 57 |
+
"loss": 0.1603,
|
| 58 |
+
"step": 70
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.3404255319148936,
|
| 62 |
+
"grad_norm": 0.6372403502464294,
|
| 63 |
+
"learning_rate": 0.00018297872340425532,
|
| 64 |
+
"loss": 0.2073,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.3829787234042553,
|
| 69 |
+
"grad_norm": 0.22926707565784454,
|
| 70 |
+
"learning_rate": 0.00018085106382978726,
|
| 71 |
+
"loss": 0.1834,
|
| 72 |
+
"step": 90
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.425531914893617,
|
| 76 |
+
"grad_norm": 3.5747313499450684,
|
| 77 |
+
"learning_rate": 0.00017872340425531915,
|
| 78 |
+
"loss": 0.227,
|
| 79 |
+
"step": 100
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.425531914893617,
|
| 83 |
+
"eval_accuracy": 0.8909599254426841,
|
| 84 |
+
"eval_loss": 0.3066631555557251,
|
| 85 |
+
"eval_runtime": 4.984,
|
| 86 |
+
"eval_samples_per_second": 215.29,
|
| 87 |
+
"eval_steps_per_second": 27.087,
|
| 88 |
+
"step": 100
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 0.46808510638297873,
|
| 92 |
+
"grad_norm": 2.685378074645996,
|
| 93 |
+
"learning_rate": 0.00017659574468085107,
|
| 94 |
+
"loss": 0.1576,
|
| 95 |
+
"step": 110
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 0.5106382978723404,
|
| 99 |
+
"grad_norm": 2.99109148979187,
|
| 100 |
+
"learning_rate": 0.00017446808510638298,
|
| 101 |
+
"loss": 0.1967,
|
| 102 |
+
"step": 120
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 0.5531914893617021,
|
| 106 |
+
"grad_norm": 0.5496222376823425,
|
| 107 |
+
"learning_rate": 0.0001723404255319149,
|
| 108 |
+
"loss": 0.2398,
|
| 109 |
+
"step": 130
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 0.5957446808510638,
|
| 113 |
+
"grad_norm": 2.8761274814605713,
|
| 114 |
+
"learning_rate": 0.00017021276595744682,
|
| 115 |
+
"loss": 0.1012,
|
| 116 |
+
"step": 140
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 0.6382978723404256,
|
| 120 |
+
"grad_norm": 1.6604396104812622,
|
| 121 |
+
"learning_rate": 0.00016808510638297873,
|
| 122 |
+
"loss": 0.0942,
|
| 123 |
+
"step": 150
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 0.6808510638297872,
|
| 127 |
+
"grad_norm": 0.680061399936676,
|
| 128 |
+
"learning_rate": 0.00016595744680851065,
|
| 129 |
+
"loss": 0.1741,
|
| 130 |
+
"step": 160
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 0.723404255319149,
|
| 134 |
+
"grad_norm": 0.24274976551532745,
|
| 135 |
+
"learning_rate": 0.00016382978723404257,
|
| 136 |
+
"loss": 0.1793,
|
| 137 |
+
"step": 170
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 0.7659574468085106,
|
| 141 |
+
"grad_norm": 2.806928873062134,
|
| 142 |
+
"learning_rate": 0.00016170212765957446,
|
| 143 |
+
"loss": 0.1423,
|
| 144 |
+
"step": 180
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 0.8085106382978723,
|
| 148 |
+
"grad_norm": 1.0465861558914185,
|
| 149 |
+
"learning_rate": 0.00015957446808510637,
|
| 150 |
+
"loss": 0.1028,
|
| 151 |
+
"step": 190
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 0.851063829787234,
|
| 155 |
+
"grad_norm": 0.5258079171180725,
|
| 156 |
+
"learning_rate": 0.00015744680851063832,
|
| 157 |
+
"loss": 0.0659,
|
| 158 |
+
"step": 200
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 0.851063829787234,
|
| 162 |
+
"eval_accuracy": 0.9627213420316869,
|
| 163 |
+
"eval_loss": 0.11090704798698425,
|
| 164 |
+
"eval_runtime": 5.4644,
|
| 165 |
+
"eval_samples_per_second": 196.362,
|
| 166 |
+
"eval_steps_per_second": 24.705,
|
| 167 |
+
"step": 200
|
| 168 |
+
},
|
| 169 |
+
{
|
| 170 |
+
"epoch": 0.8936170212765957,
|
| 171 |
+
"grad_norm": 6.45848274230957,
|
| 172 |
+
"learning_rate": 0.0001553191489361702,
|
| 173 |
+
"loss": 0.0879,
|
| 174 |
+
"step": 210
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 0.9361702127659575,
|
| 178 |
+
"grad_norm": 1.5740026235580444,
|
| 179 |
+
"learning_rate": 0.00015319148936170213,
|
| 180 |
+
"loss": 0.081,
|
| 181 |
+
"step": 220
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 0.9787234042553191,
|
| 185 |
+
"grad_norm": 1.8585551977157593,
|
| 186 |
+
"learning_rate": 0.00015106382978723407,
|
| 187 |
+
"loss": 0.095,
|
| 188 |
+
"step": 230
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 1.0212765957446808,
|
| 192 |
+
"grad_norm": 0.10410909354686737,
|
| 193 |
+
"learning_rate": 0.00014893617021276596,
|
| 194 |
+
"loss": 0.1254,
|
| 195 |
+
"step": 240
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 1.0638297872340425,
|
| 199 |
+
"grad_norm": 0.05694637447595596,
|
| 200 |
+
"learning_rate": 0.00014680851063829788,
|
| 201 |
+
"loss": 0.025,
|
| 202 |
+
"step": 250
|
| 203 |
+
},
|
| 204 |
+
{
|
| 205 |
+
"epoch": 1.1063829787234043,
|
| 206 |
+
"grad_norm": 0.053941983729600906,
|
| 207 |
+
"learning_rate": 0.0001446808510638298,
|
| 208 |
+
"loss": 0.0336,
|
| 209 |
+
"step": 260
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 1.148936170212766,
|
| 213 |
+
"grad_norm": 5.261789798736572,
|
| 214 |
+
"learning_rate": 0.0001425531914893617,
|
| 215 |
+
"loss": 0.0965,
|
| 216 |
+
"step": 270
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 1.1914893617021276,
|
| 220 |
+
"grad_norm": 1.4031267166137695,
|
| 221 |
+
"learning_rate": 0.00014042553191489363,
|
| 222 |
+
"loss": 0.0207,
|
| 223 |
+
"step": 280
|
| 224 |
+
},
|
| 225 |
+
{
|
| 226 |
+
"epoch": 1.2340425531914894,
|
| 227 |
+
"grad_norm": 0.06696026772260666,
|
| 228 |
+
"learning_rate": 0.00013829787234042554,
|
| 229 |
+
"loss": 0.0992,
|
| 230 |
+
"step": 290
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 1.2765957446808511,
|
| 234 |
+
"grad_norm": 0.055372320115566254,
|
| 235 |
+
"learning_rate": 0.00013617021276595746,
|
| 236 |
+
"loss": 0.0404,
|
| 237 |
+
"step": 300
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 1.2765957446808511,
|
| 241 |
+
"eval_accuracy": 0.9776328052190121,
|
| 242 |
+
"eval_loss": 0.09002945572137833,
|
| 243 |
+
"eval_runtime": 5.5899,
|
| 244 |
+
"eval_samples_per_second": 191.953,
|
| 245 |
+
"eval_steps_per_second": 24.151,
|
| 246 |
+
"step": 300
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 1.3191489361702127,
|
| 250 |
+
"grad_norm": 0.5205884575843811,
|
| 251 |
+
"learning_rate": 0.00013404255319148938,
|
| 252 |
+
"loss": 0.1412,
|
| 253 |
+
"step": 310
|
| 254 |
+
},
|
| 255 |
+
{
|
| 256 |
+
"epoch": 1.3617021276595744,
|
| 257 |
+
"grad_norm": 0.23773987591266632,
|
| 258 |
+
"learning_rate": 0.00013191489361702127,
|
| 259 |
+
"loss": 0.0442,
|
| 260 |
+
"step": 320
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 1.4042553191489362,
|
| 264 |
+
"grad_norm": 4.623290061950684,
|
| 265 |
+
"learning_rate": 0.00012978723404255318,
|
| 266 |
+
"loss": 0.0839,
|
| 267 |
+
"step": 330
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 1.4468085106382977,
|
| 271 |
+
"grad_norm": 0.059595488011837006,
|
| 272 |
+
"learning_rate": 0.00012765957446808513,
|
| 273 |
+
"loss": 0.0131,
|
| 274 |
+
"step": 340
|
| 275 |
+
},
|
| 276 |
+
{
|
| 277 |
+
"epoch": 1.4893617021276595,
|
| 278 |
+
"grad_norm": 5.406118392944336,
|
| 279 |
+
"learning_rate": 0.00012553191489361702,
|
| 280 |
+
"loss": 0.0347,
|
| 281 |
+
"step": 350
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 1.5319148936170213,
|
| 285 |
+
"grad_norm": 0.03830511495471001,
|
| 286 |
+
"learning_rate": 0.00012340425531914893,
|
| 287 |
+
"loss": 0.0145,
|
| 288 |
+
"step": 360
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 1.574468085106383,
|
| 292 |
+
"grad_norm": 0.043490536510944366,
|
| 293 |
+
"learning_rate": 0.00012127659574468086,
|
| 294 |
+
"loss": 0.0299,
|
| 295 |
+
"step": 370
|
| 296 |
+
},
|
| 297 |
+
{
|
| 298 |
+
"epoch": 1.6170212765957448,
|
| 299 |
+
"grad_norm": 0.046317484229803085,
|
| 300 |
+
"learning_rate": 0.00011914893617021277,
|
| 301 |
+
"loss": 0.0225,
|
| 302 |
+
"step": 380
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 1.6595744680851063,
|
| 306 |
+
"grad_norm": 0.2598080337047577,
|
| 307 |
+
"learning_rate": 0.00011702127659574468,
|
| 308 |
+
"loss": 0.0475,
|
| 309 |
+
"step": 390
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 1.702127659574468,
|
| 313 |
+
"grad_norm": 0.43866652250289917,
|
| 314 |
+
"learning_rate": 0.00011489361702127661,
|
| 315 |
+
"loss": 0.05,
|
| 316 |
+
"step": 400
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 1.702127659574468,
|
| 320 |
+
"eval_accuracy": 0.9748369058713886,
|
| 321 |
+
"eval_loss": 0.10819696635007858,
|
| 322 |
+
"eval_runtime": 5.1592,
|
| 323 |
+
"eval_samples_per_second": 207.977,
|
| 324 |
+
"eval_steps_per_second": 26.167,
|
| 325 |
+
"step": 400
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 1.7446808510638299,
|
| 329 |
+
"grad_norm": 4.7892985343933105,
|
| 330 |
+
"learning_rate": 0.00011276595744680852,
|
| 331 |
+
"loss": 0.059,
|
| 332 |
+
"step": 410
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 1.7872340425531914,
|
| 336 |
+
"grad_norm": 0.0325588583946228,
|
| 337 |
+
"learning_rate": 0.00011063829787234043,
|
| 338 |
+
"loss": 0.0592,
|
| 339 |
+
"step": 420
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 1.8297872340425532,
|
| 343 |
+
"grad_norm": 14.177051544189453,
|
| 344 |
+
"learning_rate": 0.00010851063829787234,
|
| 345 |
+
"loss": 0.1221,
|
| 346 |
+
"step": 430
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 1.872340425531915,
|
| 350 |
+
"grad_norm": 0.03597332164645195,
|
| 351 |
+
"learning_rate": 0.00010638297872340425,
|
| 352 |
+
"loss": 0.071,
|
| 353 |
+
"step": 440
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 1.9148936170212765,
|
| 357 |
+
"grad_norm": 0.058201372623443604,
|
| 358 |
+
"learning_rate": 0.00010425531914893618,
|
| 359 |
+
"loss": 0.0227,
|
| 360 |
+
"step": 450
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 1.9574468085106385,
|
| 364 |
+
"grad_norm": 1.1982567310333252,
|
| 365 |
+
"learning_rate": 0.00010212765957446809,
|
| 366 |
+
"loss": 0.0354,
|
| 367 |
+
"step": 460
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 2.0,
|
| 371 |
+
"grad_norm": 0.06208239495754242,
|
| 372 |
+
"learning_rate": 0.0001,
|
| 373 |
+
"loss": 0.0258,
|
| 374 |
+
"step": 470
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 2.0425531914893615,
|
| 378 |
+
"grad_norm": 0.04629204794764519,
|
| 379 |
+
"learning_rate": 9.787234042553192e-05,
|
| 380 |
+
"loss": 0.0322,
|
| 381 |
+
"step": 480
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 2.0851063829787235,
|
| 385 |
+
"grad_norm": 0.029703835025429726,
|
| 386 |
+
"learning_rate": 9.574468085106384e-05,
|
| 387 |
+
"loss": 0.0064,
|
| 388 |
+
"step": 490
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 2.127659574468085,
|
| 392 |
+
"grad_norm": 0.036140162497758865,
|
| 393 |
+
"learning_rate": 9.361702127659576e-05,
|
| 394 |
+
"loss": 0.006,
|
| 395 |
+
"step": 500
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 2.127659574468085,
|
| 399 |
+
"eval_accuracy": 0.9888164026095061,
|
| 400 |
+
"eval_loss": 0.037363313138484955,
|
| 401 |
+
"eval_runtime": 5.6218,
|
| 402 |
+
"eval_samples_per_second": 190.863,
|
| 403 |
+
"eval_steps_per_second": 24.013,
|
| 404 |
+
"step": 500
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 2.1702127659574466,
|
| 408 |
+
"grad_norm": 0.023606618866324425,
|
| 409 |
+
"learning_rate": 9.148936170212766e-05,
|
| 410 |
+
"loss": 0.02,
|
| 411 |
+
"step": 510
|
| 412 |
+
},
|
| 413 |
+
{
|
| 414 |
+
"epoch": 2.2127659574468086,
|
| 415 |
+
"grad_norm": 0.04733359441161156,
|
| 416 |
+
"learning_rate": 8.936170212765958e-05,
|
| 417 |
+
"loss": 0.0371,
|
| 418 |
+
"step": 520
|
| 419 |
+
},
|
| 420 |
+
{
|
| 421 |
+
"epoch": 2.25531914893617,
|
| 422 |
+
"grad_norm": 0.03030354529619217,
|
| 423 |
+
"learning_rate": 8.723404255319149e-05,
|
| 424 |
+
"loss": 0.0056,
|
| 425 |
+
"step": 530
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 2.297872340425532,
|
| 429 |
+
"grad_norm": 0.21325312554836273,
|
| 430 |
+
"learning_rate": 8.510638297872341e-05,
|
| 431 |
+
"loss": 0.0146,
|
| 432 |
+
"step": 540
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 2.3404255319148937,
|
| 436 |
+
"grad_norm": 0.018344180658459663,
|
| 437 |
+
"learning_rate": 8.297872340425533e-05,
|
| 438 |
+
"loss": 0.0103,
|
| 439 |
+
"step": 550
|
| 440 |
+
},
|
| 441 |
+
{
|
| 442 |
+
"epoch": 2.382978723404255,
|
| 443 |
+
"grad_norm": 0.022901207208633423,
|
| 444 |
+
"learning_rate": 8.085106382978723e-05,
|
| 445 |
+
"loss": 0.0052,
|
| 446 |
+
"step": 560
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 2.425531914893617,
|
| 450 |
+
"grad_norm": 0.023875948041677475,
|
| 451 |
+
"learning_rate": 7.872340425531916e-05,
|
| 452 |
+
"loss": 0.0048,
|
| 453 |
+
"step": 570
|
| 454 |
+
},
|
| 455 |
+
{
|
| 456 |
+
"epoch": 2.4680851063829787,
|
| 457 |
+
"grad_norm": 0.01968464069068432,
|
| 458 |
+
"learning_rate": 7.659574468085106e-05,
|
| 459 |
+
"loss": 0.0047,
|
| 460 |
+
"step": 580
|
| 461 |
+
},
|
| 462 |
+
{
|
| 463 |
+
"epoch": 2.5106382978723403,
|
| 464 |
+
"grad_norm": 0.11908008903265,
|
| 465 |
+
"learning_rate": 7.446808510638298e-05,
|
| 466 |
+
"loss": 0.0047,
|
| 467 |
+
"step": 590
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"epoch": 2.5531914893617023,
|
| 471 |
+
"grad_norm": 0.020715612918138504,
|
| 472 |
+
"learning_rate": 7.23404255319149e-05,
|
| 473 |
+
"loss": 0.0147,
|
| 474 |
+
"step": 600
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 2.5531914893617023,
|
| 478 |
+
"eval_accuracy": 0.9888164026095061,
|
| 479 |
+
"eval_loss": 0.0540616437792778,
|
| 480 |
+
"eval_runtime": 5.1006,
|
| 481 |
+
"eval_samples_per_second": 210.367,
|
| 482 |
+
"eval_steps_per_second": 26.467,
|
| 483 |
+
"step": 600
|
| 484 |
+
},
|
| 485 |
+
{
|
| 486 |
+
"epoch": 2.595744680851064,
|
| 487 |
+
"grad_norm": 0.019655458629131317,
|
| 488 |
+
"learning_rate": 7.021276595744681e-05,
|
| 489 |
+
"loss": 0.0041,
|
| 490 |
+
"step": 610
|
| 491 |
+
},
|
| 492 |
+
{
|
| 493 |
+
"epoch": 2.6382978723404253,
|
| 494 |
+
"grad_norm": 0.04054981470108032,
|
| 495 |
+
"learning_rate": 6.808510638297873e-05,
|
| 496 |
+
"loss": 0.0625,
|
| 497 |
+
"step": 620
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 2.6808510638297873,
|
| 501 |
+
"grad_norm": 0.18255342543125153,
|
| 502 |
+
"learning_rate": 6.595744680851063e-05,
|
| 503 |
+
"loss": 0.0171,
|
| 504 |
+
"step": 630
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 2.723404255319149,
|
| 508 |
+
"grad_norm": 0.02170875295996666,
|
| 509 |
+
"learning_rate": 6.382978723404256e-05,
|
| 510 |
+
"loss": 0.0044,
|
| 511 |
+
"step": 640
|
| 512 |
+
},
|
| 513 |
+
{
|
| 514 |
+
"epoch": 2.7659574468085104,
|
| 515 |
+
"grad_norm": 0.01993757300078869,
|
| 516 |
+
"learning_rate": 6.170212765957447e-05,
|
| 517 |
+
"loss": 0.0042,
|
| 518 |
+
"step": 650
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 2.8085106382978724,
|
| 522 |
+
"grad_norm": 0.017473889514803886,
|
| 523 |
+
"learning_rate": 5.9574468085106384e-05,
|
| 524 |
+
"loss": 0.0302,
|
| 525 |
+
"step": 660
|
| 526 |
+
},
|
| 527 |
+
{
|
| 528 |
+
"epoch": 2.851063829787234,
|
| 529 |
+
"grad_norm": 0.018037663772702217,
|
| 530 |
+
"learning_rate": 5.744680851063831e-05,
|
| 531 |
+
"loss": 0.0044,
|
| 532 |
+
"step": 670
|
| 533 |
+
},
|
| 534 |
+
{
|
| 535 |
+
"epoch": 2.8936170212765955,
|
| 536 |
+
"grad_norm": 0.016940144822001457,
|
| 537 |
+
"learning_rate": 5.531914893617022e-05,
|
| 538 |
+
"loss": 0.0314,
|
| 539 |
+
"step": 680
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"epoch": 2.9361702127659575,
|
| 543 |
+
"grad_norm": 0.018456028774380684,
|
| 544 |
+
"learning_rate": 5.319148936170213e-05,
|
| 545 |
+
"loss": 0.0039,
|
| 546 |
+
"step": 690
|
| 547 |
+
},
|
| 548 |
+
{
|
| 549 |
+
"epoch": 2.978723404255319,
|
| 550 |
+
"grad_norm": 0.02064073272049427,
|
| 551 |
+
"learning_rate": 5.1063829787234044e-05,
|
| 552 |
+
"loss": 0.0105,
|
| 553 |
+
"step": 700
|
| 554 |
+
},
|
| 555 |
+
{
|
| 556 |
+
"epoch": 2.978723404255319,
|
| 557 |
+
"eval_accuracy": 0.9906803355079217,
|
| 558 |
+
"eval_loss": 0.03594699874520302,
|
| 559 |
+
"eval_runtime": 5.0939,
|
| 560 |
+
"eval_samples_per_second": 210.644,
|
| 561 |
+
"eval_steps_per_second": 26.502,
|
| 562 |
+
"step": 700
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 3.021276595744681,
|
| 566 |
+
"grad_norm": 0.018245166167616844,
|
| 567 |
+
"learning_rate": 4.893617021276596e-05,
|
| 568 |
+
"loss": 0.0041,
|
| 569 |
+
"step": 710
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 3.0638297872340425,
|
| 573 |
+
"grad_norm": 0.01676538586616516,
|
| 574 |
+
"learning_rate": 4.680851063829788e-05,
|
| 575 |
+
"loss": 0.0047,
|
| 576 |
+
"step": 720
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 3.106382978723404,
|
| 580 |
+
"grad_norm": 0.018977968022227287,
|
| 581 |
+
"learning_rate": 4.468085106382979e-05,
|
| 582 |
+
"loss": 0.0035,
|
| 583 |
+
"step": 730
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 3.148936170212766,
|
| 587 |
+
"grad_norm": 0.01823911815881729,
|
| 588 |
+
"learning_rate": 4.2553191489361704e-05,
|
| 589 |
+
"loss": 0.0048,
|
| 590 |
+
"step": 740
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 3.1914893617021276,
|
| 594 |
+
"grad_norm": 0.017579803243279457,
|
| 595 |
+
"learning_rate": 4.0425531914893614e-05,
|
| 596 |
+
"loss": 0.0035,
|
| 597 |
+
"step": 750
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 3.2340425531914896,
|
| 601 |
+
"grad_norm": 0.013467966578900814,
|
| 602 |
+
"learning_rate": 3.829787234042553e-05,
|
| 603 |
+
"loss": 0.004,
|
| 604 |
+
"step": 760
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 3.276595744680851,
|
| 608 |
+
"grad_norm": 0.03895196318626404,
|
| 609 |
+
"learning_rate": 3.617021276595745e-05,
|
| 610 |
+
"loss": 0.0036,
|
| 611 |
+
"step": 770
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 3.3191489361702127,
|
| 615 |
+
"grad_norm": 0.019369514659047127,
|
| 616 |
+
"learning_rate": 3.4042553191489365e-05,
|
| 617 |
+
"loss": 0.0034,
|
| 618 |
+
"step": 780
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 3.3617021276595747,
|
| 622 |
+
"grad_norm": 0.014404061250388622,
|
| 623 |
+
"learning_rate": 3.191489361702128e-05,
|
| 624 |
+
"loss": 0.0042,
|
| 625 |
+
"step": 790
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 3.404255319148936,
|
| 629 |
+
"grad_norm": 0.016735393553972244,
|
| 630 |
+
"learning_rate": 2.9787234042553192e-05,
|
| 631 |
+
"loss": 0.0032,
|
| 632 |
+
"step": 800
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 3.404255319148936,
|
| 636 |
+
"eval_accuracy": 0.9906803355079217,
|
| 637 |
+
"eval_loss": 0.039232004433870316,
|
| 638 |
+
"eval_runtime": 5.1085,
|
| 639 |
+
"eval_samples_per_second": 210.041,
|
| 640 |
+
"eval_steps_per_second": 26.426,
|
| 641 |
+
"step": 800
|
| 642 |
+
},
|
| 643 |
+
{
|
| 644 |
+
"epoch": 3.4468085106382977,
|
| 645 |
+
"grad_norm": 0.01882547326385975,
|
| 646 |
+
"learning_rate": 2.765957446808511e-05,
|
| 647 |
+
"loss": 0.0033,
|
| 648 |
+
"step": 810
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"epoch": 3.4893617021276597,
|
| 652 |
+
"grad_norm": 0.018505167216062546,
|
| 653 |
+
"learning_rate": 2.5531914893617022e-05,
|
| 654 |
+
"loss": 0.0039,
|
| 655 |
+
"step": 820
|
| 656 |
+
},
|
| 657 |
+
{
|
| 658 |
+
"epoch": 3.5319148936170213,
|
| 659 |
+
"grad_norm": 0.013403504155576229,
|
| 660 |
+
"learning_rate": 2.340425531914894e-05,
|
| 661 |
+
"loss": 0.008,
|
| 662 |
+
"step": 830
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 3.574468085106383,
|
| 666 |
+
"grad_norm": 0.013858008198440075,
|
| 667 |
+
"learning_rate": 2.1276595744680852e-05,
|
| 668 |
+
"loss": 0.0033,
|
| 669 |
+
"step": 840
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 3.617021276595745,
|
| 673 |
+
"grad_norm": 0.01901717483997345,
|
| 674 |
+
"learning_rate": 1.9148936170212766e-05,
|
| 675 |
+
"loss": 0.0031,
|
| 676 |
+
"step": 850
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"epoch": 3.6595744680851063,
|
| 680 |
+
"grad_norm": 0.015291731804609299,
|
| 681 |
+
"learning_rate": 1.7021276595744682e-05,
|
| 682 |
+
"loss": 0.0031,
|
| 683 |
+
"step": 860
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 3.702127659574468,
|
| 687 |
+
"grad_norm": 0.015405308455228806,
|
| 688 |
+
"learning_rate": 1.4893617021276596e-05,
|
| 689 |
+
"loss": 0.0037,
|
| 690 |
+
"step": 870
|
| 691 |
+
},
|
| 692 |
+
{
|
| 693 |
+
"epoch": 3.74468085106383,
|
| 694 |
+
"grad_norm": 0.01782970316708088,
|
| 695 |
+
"learning_rate": 1.2765957446808511e-05,
|
| 696 |
+
"loss": 0.0032,
|
| 697 |
+
"step": 880
|
| 698 |
+
},
|
| 699 |
+
{
|
| 700 |
+
"epoch": 3.7872340425531914,
|
| 701 |
+
"grad_norm": 0.013772828504443169,
|
| 702 |
+
"learning_rate": 1.0638297872340426e-05,
|
| 703 |
+
"loss": 0.0032,
|
| 704 |
+
"step": 890
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 3.829787234042553,
|
| 708 |
+
"grad_norm": 3.2476401329040527,
|
| 709 |
+
"learning_rate": 8.510638297872341e-06,
|
| 710 |
+
"loss": 0.0055,
|
| 711 |
+
"step": 900
|
| 712 |
+
},
|
| 713 |
+
{
|
| 714 |
+
"epoch": 3.829787234042553,
|
| 715 |
+
"eval_accuracy": 0.9925442684063374,
|
| 716 |
+
"eval_loss": 0.030865401029586792,
|
| 717 |
+
"eval_runtime": 5.6777,
|
| 718 |
+
"eval_samples_per_second": 188.986,
|
| 719 |
+
"eval_steps_per_second": 23.777,
|
| 720 |
+
"step": 900
|
| 721 |
+
},
|
| 722 |
+
{
|
| 723 |
+
"epoch": 3.872340425531915,
|
| 724 |
+
"grad_norm": 0.01216566190123558,
|
| 725 |
+
"learning_rate": 6.3829787234042555e-06,
|
| 726 |
+
"loss": 0.0029,
|
| 727 |
+
"step": 910
|
| 728 |
+
},
|
| 729 |
+
{
|
| 730 |
+
"epoch": 3.9148936170212765,
|
| 731 |
+
"grad_norm": 0.02088315039873123,
|
| 732 |
+
"learning_rate": 4.255319148936171e-06,
|
| 733 |
+
"loss": 0.0032,
|
| 734 |
+
"step": 920
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 3.9574468085106385,
|
| 738 |
+
"grad_norm": 0.014219781383872032,
|
| 739 |
+
"learning_rate": 2.1276595744680853e-06,
|
| 740 |
+
"loss": 0.0031,
|
| 741 |
+
"step": 930
|
| 742 |
+
},
|
| 743 |
+
{
|
| 744 |
+
"epoch": 4.0,
|
| 745 |
+
"grad_norm": 0.048915717750787735,
|
| 746 |
+
"learning_rate": 0.0,
|
| 747 |
+
"loss": 0.0031,
|
| 748 |
+
"step": 940
|
| 749 |
+
},
|
| 750 |
+
{
|
| 751 |
+
"epoch": 4.0,
|
| 752 |
+
"step": 940,
|
| 753 |
+
"total_flos": 1.1651904419373711e+18,
|
| 754 |
+
"train_loss": 0.07046630619926021,
|
| 755 |
+
"train_runtime": 463.0785,
|
| 756 |
+
"train_samples_per_second": 32.47,
|
| 757 |
+
"train_steps_per_second": 2.03
|
| 758 |
+
}
|
| 759 |
+
],
|
| 760 |
+
"logging_steps": 10,
|
| 761 |
+
"max_steps": 940,
|
| 762 |
+
"num_input_tokens_seen": 0,
|
| 763 |
+
"num_train_epochs": 4,
|
| 764 |
+
"save_steps": 100,
|
| 765 |
+
"stateful_callbacks": {
|
| 766 |
+
"TrainerControl": {
|
| 767 |
+
"args": {
|
| 768 |
+
"should_epoch_stop": false,
|
| 769 |
+
"should_evaluate": false,
|
| 770 |
+
"should_log": false,
|
| 771 |
+
"should_save": true,
|
| 772 |
+
"should_training_stop": true
|
| 773 |
+
},
|
| 774 |
+
"attributes": {}
|
| 775 |
+
}
|
| 776 |
+
},
|
| 777 |
+
"total_flos": 1.1651904419373711e+18,
|
| 778 |
+
"train_batch_size": 16,
|
| 779 |
+
"trial_name": null,
|
| 780 |
+
"trial_params": null
|
| 781 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d213b459eca3287dd1c86ee1a26f3c8f1025f6ae761af1a6ab873ffcb0d269f5
|
| 3 |
+
size 5112
|