🍻 cheers
Browse files- README.md +73 -0
- all_results.json +13 -0
- config.json +42 -0
- eval_results.json +8 -0
- model.safetensors +3 -0
- preprocessor_config.json +22 -0
- runs/Feb22_17-18-56_259230989c1f/events.out.tfevents.1708622346.259230989c1f.3969.1 +3 -0
- runs/Feb22_17-18-56_259230989c1f/events.out.tfevents.1708623094.259230989c1f.3969.2 +3 -0
- train_results.json +8 -0
- trainer_state.json +906 -0
- training_args.bin +3 -0
README.md
ADDED
|
@@ -0,0 +1,73 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
base_model: google/vit-base-patch16-224
|
| 4 |
+
tags:
|
| 5 |
+
- image-classification
|
| 6 |
+
- generated_from_trainer
|
| 7 |
+
metrics:
|
| 8 |
+
- accuracy
|
| 9 |
+
model-index:
|
| 10 |
+
- name: vit-skin-demo-v1
|
| 11 |
+
results: []
|
| 12 |
+
---
|
| 13 |
+
|
| 14 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 15 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 16 |
+
|
| 17 |
+
# vit-skin-demo-v1
|
| 18 |
+
|
| 19 |
+
This model is a fine-tuned version of [google/vit-base-patch16-224](https://huggingface.co/google/vit-base-patch16-224) on the skin-cancer dataset.
|
| 20 |
+
It achieves the following results on the evaluation set:
|
| 21 |
+
- Loss: 0.4302
|
| 22 |
+
- Accuracy: 0.8558
|
| 23 |
+
|
| 24 |
+
## Model description
|
| 25 |
+
|
| 26 |
+
More information needed
|
| 27 |
+
|
| 28 |
+
## Intended uses & limitations
|
| 29 |
+
|
| 30 |
+
More information needed
|
| 31 |
+
|
| 32 |
+
## Training and evaluation data
|
| 33 |
+
|
| 34 |
+
More information needed
|
| 35 |
+
|
| 36 |
+
## Training procedure
|
| 37 |
+
|
| 38 |
+
### Training hyperparameters
|
| 39 |
+
|
| 40 |
+
The following hyperparameters were used during training:
|
| 41 |
+
- learning_rate: 0.0002
|
| 42 |
+
- train_batch_size: 16
|
| 43 |
+
- eval_batch_size: 8
|
| 44 |
+
- seed: 42
|
| 45 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 46 |
+
- lr_scheduler_type: linear
|
| 47 |
+
- num_epochs: 4
|
| 48 |
+
- mixed_precision_training: Native AMP
|
| 49 |
+
|
| 50 |
+
### Training results
|
| 51 |
+
|
| 52 |
+
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
| 53 |
+
|:-------------:|:-----:|:----:|:---------------:|:--------:|
|
| 54 |
+
| 0.7377 | 0.31 | 100 | 0.7305 | 0.7553 |
|
| 55 |
+
| 0.8988 | 0.62 | 200 | 0.6799 | 0.7541 |
|
| 56 |
+
| 0.7157 | 0.93 | 300 | 0.6039 | 0.7772 |
|
| 57 |
+
| 0.5569 | 1.25 | 400 | 0.6506 | 0.7578 |
|
| 58 |
+
| 0.5342 | 1.56 | 500 | 0.5929 | 0.7846 |
|
| 59 |
+
| 0.6498 | 1.87 | 600 | 0.5553 | 0.7953 |
|
| 60 |
+
| 0.4956 | 2.18 | 700 | 0.5429 | 0.7921 |
|
| 61 |
+
| 0.5216 | 2.49 | 800 | 0.4704 | 0.8302 |
|
| 62 |
+
| 0.3468 | 2.8 | 900 | 0.4669 | 0.8327 |
|
| 63 |
+
| 0.4862 | 3.12 | 1000 | 0.4615 | 0.8421 |
|
| 64 |
+
| 0.4018 | 3.43 | 1100 | 0.4526 | 0.8458 |
|
| 65 |
+
| 0.302 | 3.74 | 1200 | 0.4302 | 0.8558 |
|
| 66 |
+
|
| 67 |
+
|
| 68 |
+
### Framework versions
|
| 69 |
+
|
| 70 |
+
- Transformers 4.37.2
|
| 71 |
+
- Pytorch 2.1.0+cu121
|
| 72 |
+
- Datasets 2.17.1
|
| 73 |
+
- Tokenizers 0.15.2
|
all_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 4.0,
|
| 3 |
+
"eval_accuracy": 0.8558052434456929,
|
| 4 |
+
"eval_loss": 0.4301721155643463,
|
| 5 |
+
"eval_runtime": 19.4217,
|
| 6 |
+
"eval_samples_per_second": 82.485,
|
| 7 |
+
"eval_steps_per_second": 10.349,
|
| 8 |
+
"total_flos": 1.5895869226063626e+18,
|
| 9 |
+
"train_loss": 0.5172773925036285,
|
| 10 |
+
"train_runtime": 727.4962,
|
| 11 |
+
"train_samples_per_second": 28.195,
|
| 12 |
+
"train_steps_per_second": 1.765
|
| 13 |
+
}
|
config.json
ADDED
|
@@ -0,0 +1,42 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "google/vit-base-patch16-224",
|
| 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": "akiec",
|
| 13 |
+
"1": "bcc",
|
| 14 |
+
"2": "bkl",
|
| 15 |
+
"3": "df",
|
| 16 |
+
"4": "mel",
|
| 17 |
+
"5": "nv",
|
| 18 |
+
"6": "vasc"
|
| 19 |
+
},
|
| 20 |
+
"image_size": 224,
|
| 21 |
+
"initializer_range": 0.02,
|
| 22 |
+
"intermediate_size": 3072,
|
| 23 |
+
"label2id": {
|
| 24 |
+
"akiec": 0,
|
| 25 |
+
"bcc": 1,
|
| 26 |
+
"bkl": 2,
|
| 27 |
+
"df": 3,
|
| 28 |
+
"mel": 4,
|
| 29 |
+
"nv": 5,
|
| 30 |
+
"vasc": 6
|
| 31 |
+
},
|
| 32 |
+
"layer_norm_eps": 1e-12,
|
| 33 |
+
"model_type": "vit",
|
| 34 |
+
"num_attention_heads": 12,
|
| 35 |
+
"num_channels": 3,
|
| 36 |
+
"num_hidden_layers": 12,
|
| 37 |
+
"patch_size": 16,
|
| 38 |
+
"problem_type": "single_label_classification",
|
| 39 |
+
"qkv_bias": true,
|
| 40 |
+
"torch_dtype": "float32",
|
| 41 |
+
"transformers_version": "4.37.2"
|
| 42 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 4.0,
|
| 3 |
+
"eval_accuracy": 0.8558052434456929,
|
| 4 |
+
"eval_loss": 0.4301721155643463,
|
| 5 |
+
"eval_runtime": 19.4217,
|
| 6 |
+
"eval_samples_per_second": 82.485,
|
| 7 |
+
"eval_steps_per_second": 10.349
|
| 8 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e8e06e4cda7cd6a18264034d13c0f80f6475d4bad45bba50ac54c95430f6a0aa
|
| 3 |
+
size 343239356
|
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": "ViTFeatureExtractor",
|
| 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/Feb22_17-18-56_259230989c1f/events.out.tfevents.1708622346.259230989c1f.3969.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d71bf5fc137ec5110c399a636392465a65f1c1510c2bc8b15ee7eabb4cd04be8
|
| 3 |
+
size 28779
|
runs/Feb22_17-18-56_259230989c1f/events.out.tfevents.1708623094.259230989c1f.3969.2
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:625d60ef69bf7ce852a5e00950f811482a697825b9612e6b28455c00da2d47f2
|
| 3 |
+
size 411
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 4.0,
|
| 3 |
+
"total_flos": 1.5895869226063626e+18,
|
| 4 |
+
"train_loss": 0.5172773925036285,
|
| 5 |
+
"train_runtime": 727.4962,
|
| 6 |
+
"train_samples_per_second": 28.195,
|
| 7 |
+
"train_steps_per_second": 1.765
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,906 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.4301721155643463,
|
| 3 |
+
"best_model_checkpoint": "./vit-skin-demo-v1/checkpoint-1200",
|
| 4 |
+
"epoch": 4.0,
|
| 5 |
+
"eval_steps": 100,
|
| 6 |
+
"global_step": 1284,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.03,
|
| 13 |
+
"learning_rate": 0.00019859813084112152,
|
| 14 |
+
"loss": 1.3761,
|
| 15 |
+
"step": 10
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 0.06,
|
| 19 |
+
"learning_rate": 0.00019704049844236762,
|
| 20 |
+
"loss": 0.9425,
|
| 21 |
+
"step": 20
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
"epoch": 0.09,
|
| 25 |
+
"learning_rate": 0.00019548286604361372,
|
| 26 |
+
"loss": 0.9102,
|
| 27 |
+
"step": 30
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 0.12,
|
| 31 |
+
"learning_rate": 0.00019392523364485982,
|
| 32 |
+
"loss": 0.8726,
|
| 33 |
+
"step": 40
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 0.16,
|
| 37 |
+
"learning_rate": 0.00019236760124610595,
|
| 38 |
+
"loss": 0.8659,
|
| 39 |
+
"step": 50
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.19,
|
| 43 |
+
"learning_rate": 0.00019080996884735202,
|
| 44 |
+
"loss": 0.8507,
|
| 45 |
+
"step": 60
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.22,
|
| 49 |
+
"learning_rate": 0.00018925233644859814,
|
| 50 |
+
"loss": 0.8446,
|
| 51 |
+
"step": 70
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.25,
|
| 55 |
+
"learning_rate": 0.00018769470404984424,
|
| 56 |
+
"loss": 0.941,
|
| 57 |
+
"step": 80
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 0.28,
|
| 61 |
+
"learning_rate": 0.00018613707165109034,
|
| 62 |
+
"loss": 0.7384,
|
| 63 |
+
"step": 90
|
| 64 |
+
},
|
| 65 |
+
{
|
| 66 |
+
"epoch": 0.31,
|
| 67 |
+
"learning_rate": 0.00018457943925233644,
|
| 68 |
+
"loss": 0.7377,
|
| 69 |
+
"step": 100
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 0.31,
|
| 73 |
+
"eval_accuracy": 0.7553058676654182,
|
| 74 |
+
"eval_loss": 0.7304826974868774,
|
| 75 |
+
"eval_runtime": 19.4854,
|
| 76 |
+
"eval_samples_per_second": 82.215,
|
| 77 |
+
"eval_steps_per_second": 10.315,
|
| 78 |
+
"step": 100
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"epoch": 0.34,
|
| 82 |
+
"learning_rate": 0.00018302180685358257,
|
| 83 |
+
"loss": 0.7471,
|
| 84 |
+
"step": 110
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 0.37,
|
| 88 |
+
"learning_rate": 0.00018146417445482867,
|
| 89 |
+
"loss": 0.6848,
|
| 90 |
+
"step": 120
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 0.4,
|
| 94 |
+
"learning_rate": 0.00017990654205607477,
|
| 95 |
+
"loss": 0.7872,
|
| 96 |
+
"step": 130
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"epoch": 0.44,
|
| 100 |
+
"learning_rate": 0.00017834890965732087,
|
| 101 |
+
"loss": 0.7769,
|
| 102 |
+
"step": 140
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 0.47,
|
| 106 |
+
"learning_rate": 0.000176791277258567,
|
| 107 |
+
"loss": 0.7049,
|
| 108 |
+
"step": 150
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.5,
|
| 112 |
+
"learning_rate": 0.00017523364485981307,
|
| 113 |
+
"loss": 0.4937,
|
| 114 |
+
"step": 160
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.53,
|
| 118 |
+
"learning_rate": 0.0001736760124610592,
|
| 119 |
+
"loss": 0.7708,
|
| 120 |
+
"step": 170
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 0.56,
|
| 124 |
+
"learning_rate": 0.0001721183800623053,
|
| 125 |
+
"loss": 0.891,
|
| 126 |
+
"step": 180
|
| 127 |
+
},
|
| 128 |
+
{
|
| 129 |
+
"epoch": 0.59,
|
| 130 |
+
"learning_rate": 0.00017056074766355142,
|
| 131 |
+
"loss": 0.7497,
|
| 132 |
+
"step": 190
|
| 133 |
+
},
|
| 134 |
+
{
|
| 135 |
+
"epoch": 0.62,
|
| 136 |
+
"learning_rate": 0.0001690031152647975,
|
| 137 |
+
"loss": 0.8988,
|
| 138 |
+
"step": 200
|
| 139 |
+
},
|
| 140 |
+
{
|
| 141 |
+
"epoch": 0.62,
|
| 142 |
+
"eval_accuracy": 0.7540574282147315,
|
| 143 |
+
"eval_loss": 0.6799274682998657,
|
| 144 |
+
"eval_runtime": 19.407,
|
| 145 |
+
"eval_samples_per_second": 82.548,
|
| 146 |
+
"eval_steps_per_second": 10.357,
|
| 147 |
+
"step": 200
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 0.65,
|
| 151 |
+
"learning_rate": 0.00016744548286604362,
|
| 152 |
+
"loss": 0.6315,
|
| 153 |
+
"step": 210
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 0.69,
|
| 157 |
+
"learning_rate": 0.00016588785046728972,
|
| 158 |
+
"loss": 0.8089,
|
| 159 |
+
"step": 220
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 0.72,
|
| 163 |
+
"learning_rate": 0.00016433021806853582,
|
| 164 |
+
"loss": 0.6893,
|
| 165 |
+
"step": 230
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 0.75,
|
| 169 |
+
"learning_rate": 0.00016277258566978192,
|
| 170 |
+
"loss": 0.7498,
|
| 171 |
+
"step": 240
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.78,
|
| 175 |
+
"learning_rate": 0.00016121495327102805,
|
| 176 |
+
"loss": 0.8541,
|
| 177 |
+
"step": 250
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.81,
|
| 181 |
+
"learning_rate": 0.00015965732087227415,
|
| 182 |
+
"loss": 0.6608,
|
| 183 |
+
"step": 260
|
| 184 |
+
},
|
| 185 |
+
{
|
| 186 |
+
"epoch": 0.84,
|
| 187 |
+
"learning_rate": 0.00015809968847352025,
|
| 188 |
+
"loss": 0.6824,
|
| 189 |
+
"step": 270
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 0.87,
|
| 193 |
+
"learning_rate": 0.00015654205607476635,
|
| 194 |
+
"loss": 0.7043,
|
| 195 |
+
"step": 280
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 0.9,
|
| 199 |
+
"learning_rate": 0.00015498442367601248,
|
| 200 |
+
"loss": 0.6091,
|
| 201 |
+
"step": 290
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 0.93,
|
| 205 |
+
"learning_rate": 0.00015342679127725855,
|
| 206 |
+
"loss": 0.7157,
|
| 207 |
+
"step": 300
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 0.93,
|
| 211 |
+
"eval_accuracy": 0.7771535580524345,
|
| 212 |
+
"eval_loss": 0.6038566827774048,
|
| 213 |
+
"eval_runtime": 19.6645,
|
| 214 |
+
"eval_samples_per_second": 81.467,
|
| 215 |
+
"eval_steps_per_second": 10.221,
|
| 216 |
+
"step": 300
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 0.97,
|
| 220 |
+
"learning_rate": 0.00015186915887850468,
|
| 221 |
+
"loss": 0.517,
|
| 222 |
+
"step": 310
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"epoch": 1.0,
|
| 226 |
+
"learning_rate": 0.00015031152647975078,
|
| 227 |
+
"loss": 0.6309,
|
| 228 |
+
"step": 320
|
| 229 |
+
},
|
| 230 |
+
{
|
| 231 |
+
"epoch": 1.03,
|
| 232 |
+
"learning_rate": 0.0001487538940809969,
|
| 233 |
+
"loss": 0.7726,
|
| 234 |
+
"step": 330
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 1.06,
|
| 238 |
+
"learning_rate": 0.000147196261682243,
|
| 239 |
+
"loss": 0.6196,
|
| 240 |
+
"step": 340
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 1.09,
|
| 244 |
+
"learning_rate": 0.0001456386292834891,
|
| 245 |
+
"loss": 0.4591,
|
| 246 |
+
"step": 350
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 1.12,
|
| 250 |
+
"learning_rate": 0.00014408099688473523,
|
| 251 |
+
"loss": 0.5935,
|
| 252 |
+
"step": 360
|
| 253 |
+
},
|
| 254 |
+
{
|
| 255 |
+
"epoch": 1.15,
|
| 256 |
+
"learning_rate": 0.0001425233644859813,
|
| 257 |
+
"loss": 0.5606,
|
| 258 |
+
"step": 370
|
| 259 |
+
},
|
| 260 |
+
{
|
| 261 |
+
"epoch": 1.18,
|
| 262 |
+
"learning_rate": 0.00014096573208722743,
|
| 263 |
+
"loss": 0.5984,
|
| 264 |
+
"step": 380
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 1.21,
|
| 268 |
+
"learning_rate": 0.00013940809968847353,
|
| 269 |
+
"loss": 0.5357,
|
| 270 |
+
"step": 390
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 1.25,
|
| 274 |
+
"learning_rate": 0.00013785046728971963,
|
| 275 |
+
"loss": 0.5569,
|
| 276 |
+
"step": 400
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 1.25,
|
| 280 |
+
"eval_accuracy": 0.7578027465667915,
|
| 281 |
+
"eval_loss": 0.6506312489509583,
|
| 282 |
+
"eval_runtime": 18.5005,
|
| 283 |
+
"eval_samples_per_second": 86.592,
|
| 284 |
+
"eval_steps_per_second": 10.865,
|
| 285 |
+
"step": 400
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 1.28,
|
| 289 |
+
"learning_rate": 0.00013629283489096573,
|
| 290 |
+
"loss": 0.5707,
|
| 291 |
+
"step": 410
|
| 292 |
+
},
|
| 293 |
+
{
|
| 294 |
+
"epoch": 1.31,
|
| 295 |
+
"learning_rate": 0.00013473520249221186,
|
| 296 |
+
"loss": 0.5719,
|
| 297 |
+
"step": 420
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 1.34,
|
| 301 |
+
"learning_rate": 0.00013317757009345796,
|
| 302 |
+
"loss": 0.5915,
|
| 303 |
+
"step": 430
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 1.37,
|
| 307 |
+
"learning_rate": 0.00013161993769470406,
|
| 308 |
+
"loss": 0.5368,
|
| 309 |
+
"step": 440
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 1.4,
|
| 313 |
+
"learning_rate": 0.00013006230529595016,
|
| 314 |
+
"loss": 0.6008,
|
| 315 |
+
"step": 450
|
| 316 |
+
},
|
| 317 |
+
{
|
| 318 |
+
"epoch": 1.43,
|
| 319 |
+
"learning_rate": 0.00012850467289719628,
|
| 320 |
+
"loss": 0.5129,
|
| 321 |
+
"step": 460
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 1.46,
|
| 325 |
+
"learning_rate": 0.00012694704049844238,
|
| 326 |
+
"loss": 0.4853,
|
| 327 |
+
"step": 470
|
| 328 |
+
},
|
| 329 |
+
{
|
| 330 |
+
"epoch": 1.5,
|
| 331 |
+
"learning_rate": 0.00012538940809968848,
|
| 332 |
+
"loss": 0.6041,
|
| 333 |
+
"step": 480
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 1.53,
|
| 337 |
+
"learning_rate": 0.00012383177570093458,
|
| 338 |
+
"loss": 0.701,
|
| 339 |
+
"step": 490
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 1.56,
|
| 343 |
+
"learning_rate": 0.0001222741433021807,
|
| 344 |
+
"loss": 0.5342,
|
| 345 |
+
"step": 500
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 1.56,
|
| 349 |
+
"eval_accuracy": 0.7846441947565543,
|
| 350 |
+
"eval_loss": 0.5928722620010376,
|
| 351 |
+
"eval_runtime": 19.6731,
|
| 352 |
+
"eval_samples_per_second": 81.431,
|
| 353 |
+
"eval_steps_per_second": 10.217,
|
| 354 |
+
"step": 500
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 1.59,
|
| 358 |
+
"learning_rate": 0.0001207165109034268,
|
| 359 |
+
"loss": 0.6391,
|
| 360 |
+
"step": 510
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 1.62,
|
| 364 |
+
"learning_rate": 0.00011915887850467291,
|
| 365 |
+
"loss": 0.5824,
|
| 366 |
+
"step": 520
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 1.65,
|
| 370 |
+
"learning_rate": 0.00011760124610591901,
|
| 371 |
+
"loss": 0.5496,
|
| 372 |
+
"step": 530
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"epoch": 1.68,
|
| 376 |
+
"learning_rate": 0.00011604361370716512,
|
| 377 |
+
"loss": 0.709,
|
| 378 |
+
"step": 540
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"epoch": 1.71,
|
| 382 |
+
"learning_rate": 0.00011448598130841121,
|
| 383 |
+
"loss": 0.6867,
|
| 384 |
+
"step": 550
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 1.74,
|
| 388 |
+
"learning_rate": 0.00011292834890965734,
|
| 389 |
+
"loss": 0.4701,
|
| 390 |
+
"step": 560
|
| 391 |
+
},
|
| 392 |
+
{
|
| 393 |
+
"epoch": 1.78,
|
| 394 |
+
"learning_rate": 0.00011137071651090342,
|
| 395 |
+
"loss": 0.6941,
|
| 396 |
+
"step": 570
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"epoch": 1.81,
|
| 400 |
+
"learning_rate": 0.00010981308411214955,
|
| 401 |
+
"loss": 0.5546,
|
| 402 |
+
"step": 580
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 1.84,
|
| 406 |
+
"learning_rate": 0.00010825545171339564,
|
| 407 |
+
"loss": 0.3854,
|
| 408 |
+
"step": 590
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 1.87,
|
| 412 |
+
"learning_rate": 0.00010669781931464175,
|
| 413 |
+
"loss": 0.6498,
|
| 414 |
+
"step": 600
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 1.87,
|
| 418 |
+
"eval_accuracy": 0.7952559300873908,
|
| 419 |
+
"eval_loss": 0.5553368926048279,
|
| 420 |
+
"eval_runtime": 18.8966,
|
| 421 |
+
"eval_samples_per_second": 84.777,
|
| 422 |
+
"eval_steps_per_second": 10.637,
|
| 423 |
+
"step": 600
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 1.9,
|
| 427 |
+
"learning_rate": 0.00010514018691588785,
|
| 428 |
+
"loss": 0.5885,
|
| 429 |
+
"step": 610
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 1.93,
|
| 433 |
+
"learning_rate": 0.00010358255451713396,
|
| 434 |
+
"loss": 0.5098,
|
| 435 |
+
"step": 620
|
| 436 |
+
},
|
| 437 |
+
{
|
| 438 |
+
"epoch": 1.96,
|
| 439 |
+
"learning_rate": 0.00010202492211838006,
|
| 440 |
+
"loss": 0.5097,
|
| 441 |
+
"step": 630
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 1.99,
|
| 445 |
+
"learning_rate": 0.00010046728971962618,
|
| 446 |
+
"loss": 0.5651,
|
| 447 |
+
"step": 640
|
| 448 |
+
},
|
| 449 |
+
{
|
| 450 |
+
"epoch": 2.02,
|
| 451 |
+
"learning_rate": 9.890965732087229e-05,
|
| 452 |
+
"loss": 0.3913,
|
| 453 |
+
"step": 650
|
| 454 |
+
},
|
| 455 |
+
{
|
| 456 |
+
"epoch": 2.06,
|
| 457 |
+
"learning_rate": 9.735202492211839e-05,
|
| 458 |
+
"loss": 0.5351,
|
| 459 |
+
"step": 660
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 2.09,
|
| 463 |
+
"learning_rate": 9.579439252336449e-05,
|
| 464 |
+
"loss": 0.3227,
|
| 465 |
+
"step": 670
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 2.12,
|
| 469 |
+
"learning_rate": 9.42367601246106e-05,
|
| 470 |
+
"loss": 0.4168,
|
| 471 |
+
"step": 680
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 2.15,
|
| 475 |
+
"learning_rate": 9.26791277258567e-05,
|
| 476 |
+
"loss": 0.4675,
|
| 477 |
+
"step": 690
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 2.18,
|
| 481 |
+
"learning_rate": 9.112149532710282e-05,
|
| 482 |
+
"loss": 0.4956,
|
| 483 |
+
"step": 700
|
| 484 |
+
},
|
| 485 |
+
{
|
| 486 |
+
"epoch": 2.18,
|
| 487 |
+
"eval_accuracy": 0.7921348314606742,
|
| 488 |
+
"eval_loss": 0.5429254174232483,
|
| 489 |
+
"eval_runtime": 18.9872,
|
| 490 |
+
"eval_samples_per_second": 84.373,
|
| 491 |
+
"eval_steps_per_second": 10.586,
|
| 492 |
+
"step": 700
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 2.21,
|
| 496 |
+
"learning_rate": 8.956386292834892e-05,
|
| 497 |
+
"loss": 0.4092,
|
| 498 |
+
"step": 710
|
| 499 |
+
},
|
| 500 |
+
{
|
| 501 |
+
"epoch": 2.24,
|
| 502 |
+
"learning_rate": 8.800623052959502e-05,
|
| 503 |
+
"loss": 0.4268,
|
| 504 |
+
"step": 720
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 2.27,
|
| 508 |
+
"learning_rate": 8.644859813084113e-05,
|
| 509 |
+
"loss": 0.4388,
|
| 510 |
+
"step": 730
|
| 511 |
+
},
|
| 512 |
+
{
|
| 513 |
+
"epoch": 2.31,
|
| 514 |
+
"learning_rate": 8.489096573208723e-05,
|
| 515 |
+
"loss": 0.3978,
|
| 516 |
+
"step": 740
|
| 517 |
+
},
|
| 518 |
+
{
|
| 519 |
+
"epoch": 2.34,
|
| 520 |
+
"learning_rate": 8.333333333333334e-05,
|
| 521 |
+
"loss": 0.5106,
|
| 522 |
+
"step": 750
|
| 523 |
+
},
|
| 524 |
+
{
|
| 525 |
+
"epoch": 2.37,
|
| 526 |
+
"learning_rate": 8.177570093457944e-05,
|
| 527 |
+
"loss": 0.4768,
|
| 528 |
+
"step": 760
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 2.4,
|
| 532 |
+
"learning_rate": 8.021806853582556e-05,
|
| 533 |
+
"loss": 0.5389,
|
| 534 |
+
"step": 770
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 2.43,
|
| 538 |
+
"learning_rate": 7.866043613707166e-05,
|
| 539 |
+
"loss": 0.4351,
|
| 540 |
+
"step": 780
|
| 541 |
+
},
|
| 542 |
+
{
|
| 543 |
+
"epoch": 2.46,
|
| 544 |
+
"learning_rate": 7.710280373831776e-05,
|
| 545 |
+
"loss": 0.4425,
|
| 546 |
+
"step": 790
|
| 547 |
+
},
|
| 548 |
+
{
|
| 549 |
+
"epoch": 2.49,
|
| 550 |
+
"learning_rate": 7.554517133956387e-05,
|
| 551 |
+
"loss": 0.5216,
|
| 552 |
+
"step": 800
|
| 553 |
+
},
|
| 554 |
+
{
|
| 555 |
+
"epoch": 2.49,
|
| 556 |
+
"eval_accuracy": 0.8302122347066168,
|
| 557 |
+
"eval_loss": 0.4704440236091614,
|
| 558 |
+
"eval_runtime": 18.9227,
|
| 559 |
+
"eval_samples_per_second": 84.66,
|
| 560 |
+
"eval_steps_per_second": 10.622,
|
| 561 |
+
"step": 800
|
| 562 |
+
},
|
| 563 |
+
{
|
| 564 |
+
"epoch": 2.52,
|
| 565 |
+
"learning_rate": 7.398753894080997e-05,
|
| 566 |
+
"loss": 0.4263,
|
| 567 |
+
"step": 810
|
| 568 |
+
},
|
| 569 |
+
{
|
| 570 |
+
"epoch": 2.55,
|
| 571 |
+
"learning_rate": 7.242990654205608e-05,
|
| 572 |
+
"loss": 0.346,
|
| 573 |
+
"step": 820
|
| 574 |
+
},
|
| 575 |
+
{
|
| 576 |
+
"epoch": 2.59,
|
| 577 |
+
"learning_rate": 7.087227414330218e-05,
|
| 578 |
+
"loss": 0.4664,
|
| 579 |
+
"step": 830
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 2.62,
|
| 583 |
+
"learning_rate": 6.93146417445483e-05,
|
| 584 |
+
"loss": 0.3724,
|
| 585 |
+
"step": 840
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 2.65,
|
| 589 |
+
"learning_rate": 6.77570093457944e-05,
|
| 590 |
+
"loss": 0.4354,
|
| 591 |
+
"step": 850
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 2.68,
|
| 595 |
+
"learning_rate": 6.61993769470405e-05,
|
| 596 |
+
"loss": 0.534,
|
| 597 |
+
"step": 860
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 2.71,
|
| 601 |
+
"learning_rate": 6.464174454828661e-05,
|
| 602 |
+
"loss": 0.5173,
|
| 603 |
+
"step": 870
|
| 604 |
+
},
|
| 605 |
+
{
|
| 606 |
+
"epoch": 2.74,
|
| 607 |
+
"learning_rate": 6.308411214953271e-05,
|
| 608 |
+
"loss": 0.4208,
|
| 609 |
+
"step": 880
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 2.77,
|
| 613 |
+
"learning_rate": 6.152647975077882e-05,
|
| 614 |
+
"loss": 0.4199,
|
| 615 |
+
"step": 890
|
| 616 |
+
},
|
| 617 |
+
{
|
| 618 |
+
"epoch": 2.8,
|
| 619 |
+
"learning_rate": 5.9968847352024924e-05,
|
| 620 |
+
"loss": 0.3468,
|
| 621 |
+
"step": 900
|
| 622 |
+
},
|
| 623 |
+
{
|
| 624 |
+
"epoch": 2.8,
|
| 625 |
+
"eval_accuracy": 0.83270911360799,
|
| 626 |
+
"eval_loss": 0.46694162487983704,
|
| 627 |
+
"eval_runtime": 19.215,
|
| 628 |
+
"eval_samples_per_second": 83.372,
|
| 629 |
+
"eval_steps_per_second": 10.461,
|
| 630 |
+
"step": 900
|
| 631 |
+
},
|
| 632 |
+
{
|
| 633 |
+
"epoch": 2.83,
|
| 634 |
+
"learning_rate": 5.841121495327103e-05,
|
| 635 |
+
"loss": 0.3722,
|
| 636 |
+
"step": 910
|
| 637 |
+
},
|
| 638 |
+
{
|
| 639 |
+
"epoch": 2.87,
|
| 640 |
+
"learning_rate": 5.685358255451714e-05,
|
| 641 |
+
"loss": 0.3283,
|
| 642 |
+
"step": 920
|
| 643 |
+
},
|
| 644 |
+
{
|
| 645 |
+
"epoch": 2.9,
|
| 646 |
+
"learning_rate": 5.5295950155763244e-05,
|
| 647 |
+
"loss": 0.4791,
|
| 648 |
+
"step": 930
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"epoch": 2.93,
|
| 652 |
+
"learning_rate": 5.373831775700935e-05,
|
| 653 |
+
"loss": 0.3859,
|
| 654 |
+
"step": 940
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 2.96,
|
| 658 |
+
"learning_rate": 5.218068535825546e-05,
|
| 659 |
+
"loss": 0.2583,
|
| 660 |
+
"step": 950
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 2.99,
|
| 664 |
+
"learning_rate": 5.062305295950156e-05,
|
| 665 |
+
"loss": 0.318,
|
| 666 |
+
"step": 960
|
| 667 |
+
},
|
| 668 |
+
{
|
| 669 |
+
"epoch": 3.02,
|
| 670 |
+
"learning_rate": 4.9065420560747664e-05,
|
| 671 |
+
"loss": 0.3943,
|
| 672 |
+
"step": 970
|
| 673 |
+
},
|
| 674 |
+
{
|
| 675 |
+
"epoch": 3.05,
|
| 676 |
+
"learning_rate": 4.750778816199377e-05,
|
| 677 |
+
"loss": 0.2449,
|
| 678 |
+
"step": 980
|
| 679 |
+
},
|
| 680 |
+
{
|
| 681 |
+
"epoch": 3.08,
|
| 682 |
+
"learning_rate": 4.595015576323988e-05,
|
| 683 |
+
"loss": 0.204,
|
| 684 |
+
"step": 990
|
| 685 |
+
},
|
| 686 |
+
{
|
| 687 |
+
"epoch": 3.12,
|
| 688 |
+
"learning_rate": 4.4392523364485984e-05,
|
| 689 |
+
"loss": 0.4862,
|
| 690 |
+
"step": 1000
|
| 691 |
+
},
|
| 692 |
+
{
|
| 693 |
+
"epoch": 3.12,
|
| 694 |
+
"eval_accuracy": 0.8420724094881398,
|
| 695 |
+
"eval_loss": 0.4615306556224823,
|
| 696 |
+
"eval_runtime": 19.2799,
|
| 697 |
+
"eval_samples_per_second": 83.092,
|
| 698 |
+
"eval_steps_per_second": 10.425,
|
| 699 |
+
"step": 1000
|
| 700 |
+
},
|
| 701 |
+
{
|
| 702 |
+
"epoch": 3.15,
|
| 703 |
+
"learning_rate": 4.283489096573209e-05,
|
| 704 |
+
"loss": 0.2649,
|
| 705 |
+
"step": 1010
|
| 706 |
+
},
|
| 707 |
+
{
|
| 708 |
+
"epoch": 3.18,
|
| 709 |
+
"learning_rate": 4.12772585669782e-05,
|
| 710 |
+
"loss": 0.3136,
|
| 711 |
+
"step": 1020
|
| 712 |
+
},
|
| 713 |
+
{
|
| 714 |
+
"epoch": 3.21,
|
| 715 |
+
"learning_rate": 3.97196261682243e-05,
|
| 716 |
+
"loss": 0.2549,
|
| 717 |
+
"step": 1030
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 3.24,
|
| 721 |
+
"learning_rate": 3.8161993769470404e-05,
|
| 722 |
+
"loss": 0.2944,
|
| 723 |
+
"step": 1040
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 3.27,
|
| 727 |
+
"learning_rate": 3.660436137071651e-05,
|
| 728 |
+
"loss": 0.3809,
|
| 729 |
+
"step": 1050
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 3.3,
|
| 733 |
+
"learning_rate": 3.504672897196262e-05,
|
| 734 |
+
"loss": 0.2504,
|
| 735 |
+
"step": 1060
|
| 736 |
+
},
|
| 737 |
+
{
|
| 738 |
+
"epoch": 3.33,
|
| 739 |
+
"learning_rate": 3.3489096573208724e-05,
|
| 740 |
+
"loss": 0.4471,
|
| 741 |
+
"step": 1070
|
| 742 |
+
},
|
| 743 |
+
{
|
| 744 |
+
"epoch": 3.36,
|
| 745 |
+
"learning_rate": 3.193146417445483e-05,
|
| 746 |
+
"loss": 0.2706,
|
| 747 |
+
"step": 1080
|
| 748 |
+
},
|
| 749 |
+
{
|
| 750 |
+
"epoch": 3.4,
|
| 751 |
+
"learning_rate": 3.0373831775700934e-05,
|
| 752 |
+
"loss": 0.2495,
|
| 753 |
+
"step": 1090
|
| 754 |
+
},
|
| 755 |
+
{
|
| 756 |
+
"epoch": 3.43,
|
| 757 |
+
"learning_rate": 2.881619937694704e-05,
|
| 758 |
+
"loss": 0.4018,
|
| 759 |
+
"step": 1100
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 3.43,
|
| 763 |
+
"eval_accuracy": 0.8458177278401997,
|
| 764 |
+
"eval_loss": 0.4525544345378876,
|
| 765 |
+
"eval_runtime": 18.7511,
|
| 766 |
+
"eval_samples_per_second": 85.435,
|
| 767 |
+
"eval_steps_per_second": 10.719,
|
| 768 |
+
"step": 1100
|
| 769 |
+
},
|
| 770 |
+
{
|
| 771 |
+
"epoch": 3.46,
|
| 772 |
+
"learning_rate": 2.7258566978193147e-05,
|
| 773 |
+
"loss": 0.2398,
|
| 774 |
+
"step": 1110
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 3.49,
|
| 778 |
+
"learning_rate": 2.570093457943925e-05,
|
| 779 |
+
"loss": 0.3493,
|
| 780 |
+
"step": 1120
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 3.52,
|
| 784 |
+
"learning_rate": 2.414330218068536e-05,
|
| 785 |
+
"loss": 0.1811,
|
| 786 |
+
"step": 1130
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 3.55,
|
| 790 |
+
"learning_rate": 2.2585669781931463e-05,
|
| 791 |
+
"loss": 0.2008,
|
| 792 |
+
"step": 1140
|
| 793 |
+
},
|
| 794 |
+
{
|
| 795 |
+
"epoch": 3.58,
|
| 796 |
+
"learning_rate": 2.102803738317757e-05,
|
| 797 |
+
"loss": 0.2805,
|
| 798 |
+
"step": 1150
|
| 799 |
+
},
|
| 800 |
+
{
|
| 801 |
+
"epoch": 3.61,
|
| 802 |
+
"learning_rate": 1.9470404984423677e-05,
|
| 803 |
+
"loss": 0.2142,
|
| 804 |
+
"step": 1160
|
| 805 |
+
},
|
| 806 |
+
{
|
| 807 |
+
"epoch": 3.64,
|
| 808 |
+
"learning_rate": 1.7912772585669783e-05,
|
| 809 |
+
"loss": 0.2079,
|
| 810 |
+
"step": 1170
|
| 811 |
+
},
|
| 812 |
+
{
|
| 813 |
+
"epoch": 3.68,
|
| 814 |
+
"learning_rate": 1.6355140186915887e-05,
|
| 815 |
+
"loss": 0.3097,
|
| 816 |
+
"step": 1180
|
| 817 |
+
},
|
| 818 |
+
{
|
| 819 |
+
"epoch": 3.71,
|
| 820 |
+
"learning_rate": 1.4797507788161993e-05,
|
| 821 |
+
"loss": 0.2846,
|
| 822 |
+
"step": 1190
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 3.74,
|
| 826 |
+
"learning_rate": 1.3239875389408098e-05,
|
| 827 |
+
"loss": 0.302,
|
| 828 |
+
"step": 1200
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 3.74,
|
| 832 |
+
"eval_accuracy": 0.8558052434456929,
|
| 833 |
+
"eval_loss": 0.4301721155643463,
|
| 834 |
+
"eval_runtime": 19.5305,
|
| 835 |
+
"eval_samples_per_second": 82.026,
|
| 836 |
+
"eval_steps_per_second": 10.292,
|
| 837 |
+
"step": 1200
|
| 838 |
+
},
|
| 839 |
+
{
|
| 840 |
+
"epoch": 3.77,
|
| 841 |
+
"learning_rate": 1.1682242990654207e-05,
|
| 842 |
+
"loss": 0.3489,
|
| 843 |
+
"step": 1210
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 3.8,
|
| 847 |
+
"learning_rate": 1.0124610591900312e-05,
|
| 848 |
+
"loss": 0.3015,
|
| 849 |
+
"step": 1220
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 3.83,
|
| 853 |
+
"learning_rate": 8.566978193146417e-06,
|
| 854 |
+
"loss": 0.231,
|
| 855 |
+
"step": 1230
|
| 856 |
+
},
|
| 857 |
+
{
|
| 858 |
+
"epoch": 3.86,
|
| 859 |
+
"learning_rate": 7.009345794392523e-06,
|
| 860 |
+
"loss": 0.192,
|
| 861 |
+
"step": 1240
|
| 862 |
+
},
|
| 863 |
+
{
|
| 864 |
+
"epoch": 3.89,
|
| 865 |
+
"learning_rate": 5.451713395638629e-06,
|
| 866 |
+
"loss": 0.2904,
|
| 867 |
+
"step": 1250
|
| 868 |
+
},
|
| 869 |
+
{
|
| 870 |
+
"epoch": 3.93,
|
| 871 |
+
"learning_rate": 3.894080996884735e-06,
|
| 872 |
+
"loss": 0.412,
|
| 873 |
+
"step": 1260
|
| 874 |
+
},
|
| 875 |
+
{
|
| 876 |
+
"epoch": 3.96,
|
| 877 |
+
"learning_rate": 2.3364485981308413e-06,
|
| 878 |
+
"loss": 0.2304,
|
| 879 |
+
"step": 1270
|
| 880 |
+
},
|
| 881 |
+
{
|
| 882 |
+
"epoch": 3.99,
|
| 883 |
+
"learning_rate": 7.788161993769471e-07,
|
| 884 |
+
"loss": 0.293,
|
| 885 |
+
"step": 1280
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 4.0,
|
| 889 |
+
"step": 1284,
|
| 890 |
+
"total_flos": 1.5895869226063626e+18,
|
| 891 |
+
"train_loss": 0.5172773925036285,
|
| 892 |
+
"train_runtime": 727.4962,
|
| 893 |
+
"train_samples_per_second": 28.195,
|
| 894 |
+
"train_steps_per_second": 1.765
|
| 895 |
+
}
|
| 896 |
+
],
|
| 897 |
+
"logging_steps": 10,
|
| 898 |
+
"max_steps": 1284,
|
| 899 |
+
"num_input_tokens_seen": 0,
|
| 900 |
+
"num_train_epochs": 4,
|
| 901 |
+
"save_steps": 100,
|
| 902 |
+
"total_flos": 1.5895869226063626e+18,
|
| 903 |
+
"train_batch_size": 16,
|
| 904 |
+
"trial_name": null,
|
| 905 |
+
"trial_params": null
|
| 906 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:704b66cd088744eb21498752bc94663801f430d0d6b42ab9a28dbbcbb7337107
|
| 3 |
+
size 4664
|