Commit
•
303446c
1
Parent(s):
9079a25
josedonoso/ecg-khan-vit
Browse files- README.md +83 -0
- all_results.json +13 -0
- config.json +36 -0
- eval_results.json +8 -0
- model.safetensors +3 -0
- preprocessor_config.json +36 -0
- runs/Jun30_14-34-21_fe288f1a8407/events.out.tfevents.1719758073.fe288f1a8407.618.0 +3 -0
- runs/Jun30_14-47-12_fe288f1a8407/events.out.tfevents.1719758866.fe288f1a8407.618.1 +3 -0
- runs/Jun30_14-54-40_fe288f1a8407/events.out.tfevents.1719759287.fe288f1a8407.618.2 +3 -0
- runs/Jun30_14-54-40_fe288f1a8407/events.out.tfevents.1719759438.fe288f1a8407.618.3 +3 -0
- runs/Jun30_14-54-40_fe288f1a8407/events.out.tfevents.1719759562.fe288f1a8407.618.4 +3 -0
- runs/Jun30_15-00-45_fe288f1a8407/events.out.tfevents.1719759685.fe288f1a8407.618.5 +3 -0
- runs/Jun30_15-00-45_fe288f1a8407/events.out.tfevents.1719760856.fe288f1a8407.618.6 +3 -0
- train_results.json +8 -0
- trainer_state.json +688 -0
- training_args.bin +3 -0
README.md
ADDED
|
@@ -0,0 +1,83 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
base_model: google/vit-base-patch16-224-in21k
|
| 4 |
+
tags:
|
| 5 |
+
- generated_from_trainer
|
| 6 |
+
datasets:
|
| 7 |
+
- imagefolder
|
| 8 |
+
metrics:
|
| 9 |
+
- accuracy
|
| 10 |
+
model-index:
|
| 11 |
+
- name: vit-base-ecg
|
| 12 |
+
results:
|
| 13 |
+
- task:
|
| 14 |
+
name: Image Classification
|
| 15 |
+
type: image-classification
|
| 16 |
+
dataset:
|
| 17 |
+
name: imagefolder
|
| 18 |
+
type: imagefolder
|
| 19 |
+
config: default
|
| 20 |
+
split: validation
|
| 21 |
+
args: default
|
| 22 |
+
metrics:
|
| 23 |
+
- name: Accuracy
|
| 24 |
+
type: accuracy
|
| 25 |
+
value: 0.9642857142857143
|
| 26 |
+
---
|
| 27 |
+
|
| 28 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 29 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 30 |
+
|
| 31 |
+
# vit-base-ecg
|
| 32 |
+
|
| 33 |
+
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 imagefolder dataset.
|
| 34 |
+
It achieves the following results on the evaluation set:
|
| 35 |
+
- Loss: 0.1003
|
| 36 |
+
- Accuracy: 0.9643
|
| 37 |
+
|
| 38 |
+
## Model description
|
| 39 |
+
|
| 40 |
+
More information needed
|
| 41 |
+
|
| 42 |
+
## Intended uses & limitations
|
| 43 |
+
|
| 44 |
+
More information needed
|
| 45 |
+
|
| 46 |
+
## Training and evaluation data
|
| 47 |
+
|
| 48 |
+
More information needed
|
| 49 |
+
|
| 50 |
+
## Training procedure
|
| 51 |
+
|
| 52 |
+
### Training hyperparameters
|
| 53 |
+
|
| 54 |
+
The following hyperparameters were used during training:
|
| 55 |
+
- learning_rate: 0.0002
|
| 56 |
+
- train_batch_size: 16
|
| 57 |
+
- eval_batch_size: 8
|
| 58 |
+
- seed: 42
|
| 59 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 60 |
+
- lr_scheduler_type: linear
|
| 61 |
+
- num_epochs: 20
|
| 62 |
+
- mixed_precision_training: Native AMP
|
| 63 |
+
|
| 64 |
+
### Training results
|
| 65 |
+
|
| 66 |
+
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
| 67 |
+
|:-------------:|:-------:|:----:|:---------------:|:--------:|
|
| 68 |
+
| 0.596 | 2.4390 | 100 | 0.5431 | 0.8214 |
|
| 69 |
+
| 0.0656 | 4.8780 | 200 | 0.1628 | 0.95 |
|
| 70 |
+
| 0.0192 | 7.3171 | 300 | 0.1003 | 0.9643 |
|
| 71 |
+
| 0.0926 | 9.7561 | 400 | 0.1262 | 0.95 |
|
| 72 |
+
| 0.0064 | 12.1951 | 500 | 0.1611 | 0.9643 |
|
| 73 |
+
| 0.0049 | 14.6341 | 600 | 0.1539 | 0.9643 |
|
| 74 |
+
| 0.0044 | 17.0732 | 700 | 0.1509 | 0.9643 |
|
| 75 |
+
| 0.0041 | 19.5122 | 800 | 0.1499 | 0.9643 |
|
| 76 |
+
|
| 77 |
+
|
| 78 |
+
### Framework versions
|
| 79 |
+
|
| 80 |
+
- Transformers 4.41.2
|
| 81 |
+
- Pytorch 2.3.0+cu121
|
| 82 |
+
- Datasets 2.20.0
|
| 83 |
+
- Tokenizers 0.19.1
|
all_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 20.0,
|
| 3 |
+
"eval_accuracy": 0.9642857142857143,
|
| 4 |
+
"eval_loss": 0.10026070475578308,
|
| 5 |
+
"eval_runtime": 13.4022,
|
| 6 |
+
"eval_samples_per_second": 10.446,
|
| 7 |
+
"eval_steps_per_second": 1.343,
|
| 8 |
+
"total_flos": 1.004314187783209e+18,
|
| 9 |
+
"train_loss": 0.1252097422029914,
|
| 10 |
+
"train_runtime": 1111.906,
|
| 11 |
+
"train_samples_per_second": 11.656,
|
| 12 |
+
"train_steps_per_second": 0.737
|
| 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": "Abnormal-Heartbeat",
|
| 13 |
+
"1": "History-of-MI",
|
| 14 |
+
"2": "Myocardial-Infarction",
|
| 15 |
+
"3": "Normal"
|
| 16 |
+
},
|
| 17 |
+
"image_size": 224,
|
| 18 |
+
"initializer_range": 0.02,
|
| 19 |
+
"intermediate_size": 3072,
|
| 20 |
+
"label2id": {
|
| 21 |
+
"Abnormal-Heartbeat": "0",
|
| 22 |
+
"History-of-MI": "1",
|
| 23 |
+
"Myocardial-Infarction": "2",
|
| 24 |
+
"Normal": "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": "4.41.2"
|
| 36 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 20.0,
|
| 3 |
+
"eval_accuracy": 0.9642857142857143,
|
| 4 |
+
"eval_loss": 0.10026070475578308,
|
| 5 |
+
"eval_runtime": 13.4022,
|
| 6 |
+
"eval_samples_per_second": 10.446,
|
| 7 |
+
"eval_steps_per_second": 1.343
|
| 8 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a95a3c8b272113af19d176fe8f3e51fd5a91e037c8385c85987561caf3b3f099
|
| 3 |
+
size 343230128
|
preprocessor_config.json
ADDED
|
@@ -0,0 +1,36 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_valid_processor_keys": [
|
| 3 |
+
"images",
|
| 4 |
+
"do_resize",
|
| 5 |
+
"size",
|
| 6 |
+
"resample",
|
| 7 |
+
"do_rescale",
|
| 8 |
+
"rescale_factor",
|
| 9 |
+
"do_normalize",
|
| 10 |
+
"image_mean",
|
| 11 |
+
"image_std",
|
| 12 |
+
"return_tensors",
|
| 13 |
+
"data_format",
|
| 14 |
+
"input_data_format"
|
| 15 |
+
],
|
| 16 |
+
"do_normalize": true,
|
| 17 |
+
"do_rescale": true,
|
| 18 |
+
"do_resize": true,
|
| 19 |
+
"image_mean": [
|
| 20 |
+
0.5,
|
| 21 |
+
0.5,
|
| 22 |
+
0.5
|
| 23 |
+
],
|
| 24 |
+
"image_processor_type": "ViTImageProcessor",
|
| 25 |
+
"image_std": [
|
| 26 |
+
0.5,
|
| 27 |
+
0.5,
|
| 28 |
+
0.5
|
| 29 |
+
],
|
| 30 |
+
"resample": 2,
|
| 31 |
+
"rescale_factor": 0.00392156862745098,
|
| 32 |
+
"size": {
|
| 33 |
+
"height": 224,
|
| 34 |
+
"width": 224
|
| 35 |
+
}
|
| 36 |
+
}
|
runs/Jun30_14-34-21_fe288f1a8407/events.out.tfevents.1719758073.fe288f1a8407.618.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:116e7173a6fbf02d9723a6b831fcc49b6f30ceaf54af5b62a5150909ba32d28c
|
| 3 |
+
size 6941
|
runs/Jun30_14-47-12_fe288f1a8407/events.out.tfevents.1719758866.fe288f1a8407.618.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8b41873d8b3cf555b4254f255542a5522e4e73bb5be629ea02d8138c376323a2
|
| 3 |
+
size 7035
|
runs/Jun30_14-54-40_fe288f1a8407/events.out.tfevents.1719759287.fe288f1a8407.618.2
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b2510ed64f008c6ac26676b6891b26d13e58b4f799be940175d995d5568966d3
|
| 3 |
+
size 6941
|
runs/Jun30_14-54-40_fe288f1a8407/events.out.tfevents.1719759438.fe288f1a8407.618.3
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a399cb65ef4b3a336b6ae8039f51bfe39251bccd264c9cd440f6de09fcea613b
|
| 3 |
+
size 7035
|
runs/Jun30_14-54-40_fe288f1a8407/events.out.tfevents.1719759562.fe288f1a8407.618.4
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d2efc1c45cece9f37a72e06ee689ff7fbb399978a00474b98df3f6c70ed00989
|
| 3 |
+
size 405
|
runs/Jun30_15-00-45_fe288f1a8407/events.out.tfevents.1719759685.fe288f1a8407.618.5
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:779ad243525b3cc6c2eefe6361b1e015371737aa9313f240e0ca59c0fe731750
|
| 3 |
+
size 25124
|
runs/Jun30_15-00-45_fe288f1a8407/events.out.tfevents.1719760856.fe288f1a8407.618.6
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3b7d1ebd34f2db365d854cc9f6146fce50e6f13cc3f1080c6a0016717bfe518d
|
| 3 |
+
size 411
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 20.0,
|
| 3 |
+
"total_flos": 1.004314187783209e+18,
|
| 4 |
+
"train_loss": 0.1252097422029914,
|
| 5 |
+
"train_runtime": 1111.906,
|
| 6 |
+
"train_samples_per_second": 11.656,
|
| 7 |
+
"train_steps_per_second": 0.737
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,688 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.10026070475578308,
|
| 3 |
+
"best_model_checkpoint": "./vit-base-ecg/checkpoint-300",
|
| 4 |
+
"epoch": 20.0,
|
| 5 |
+
"eval_steps": 100,
|
| 6 |
+
"global_step": 820,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.24390243902439024,
|
| 13 |
+
"grad_norm": 3.298177480697632,
|
| 14 |
+
"learning_rate": 0.0001975609756097561,
|
| 15 |
+
"loss": 1.3469,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.4878048780487805,
|
| 20 |
+
"grad_norm": 1.9206117391586304,
|
| 21 |
+
"learning_rate": 0.0001951219512195122,
|
| 22 |
+
"loss": 1.0603,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.7317073170731707,
|
| 27 |
+
"grad_norm": 3.621140956878662,
|
| 28 |
+
"learning_rate": 0.0001926829268292683,
|
| 29 |
+
"loss": 0.9322,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.975609756097561,
|
| 34 |
+
"grad_norm": 7.4158735275268555,
|
| 35 |
+
"learning_rate": 0.0001902439024390244,
|
| 36 |
+
"loss": 0.7773,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 1.2195121951219512,
|
| 41 |
+
"grad_norm": 4.021040916442871,
|
| 42 |
+
"learning_rate": 0.0001878048780487805,
|
| 43 |
+
"loss": 0.5297,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 1.4634146341463414,
|
| 48 |
+
"grad_norm": 2.9927070140838623,
|
| 49 |
+
"learning_rate": 0.0001853658536585366,
|
| 50 |
+
"loss": 0.4686,
|
| 51 |
+
"step": 60
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 1.7073170731707317,
|
| 55 |
+
"grad_norm": 1.2623863220214844,
|
| 56 |
+
"learning_rate": 0.0001829268292682927,
|
| 57 |
+
"loss": 0.4439,
|
| 58 |
+
"step": 70
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 1.951219512195122,
|
| 62 |
+
"grad_norm": 10.744318962097168,
|
| 63 |
+
"learning_rate": 0.0001804878048780488,
|
| 64 |
+
"loss": 0.3923,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 2.1951219512195124,
|
| 69 |
+
"grad_norm": 8.121638298034668,
|
| 70 |
+
"learning_rate": 0.00017804878048780488,
|
| 71 |
+
"loss": 0.2701,
|
| 72 |
+
"step": 90
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 2.4390243902439024,
|
| 76 |
+
"grad_norm": 6.202785015106201,
|
| 77 |
+
"learning_rate": 0.000175609756097561,
|
| 78 |
+
"loss": 0.596,
|
| 79 |
+
"step": 100
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 2.4390243902439024,
|
| 83 |
+
"eval_accuracy": 0.8214285714285714,
|
| 84 |
+
"eval_loss": 0.5431132912635803,
|
| 85 |
+
"eval_runtime": 8.502,
|
| 86 |
+
"eval_samples_per_second": 16.467,
|
| 87 |
+
"eval_steps_per_second": 2.117,
|
| 88 |
+
"step": 100
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 2.682926829268293,
|
| 92 |
+
"grad_norm": 1.8862333297729492,
|
| 93 |
+
"learning_rate": 0.00017317073170731708,
|
| 94 |
+
"loss": 0.3225,
|
| 95 |
+
"step": 110
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 2.926829268292683,
|
| 99 |
+
"grad_norm": 2.015392780303955,
|
| 100 |
+
"learning_rate": 0.0001707317073170732,
|
| 101 |
+
"loss": 0.3347,
|
| 102 |
+
"step": 120
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 3.1707317073170733,
|
| 106 |
+
"grad_norm": 1.7934260368347168,
|
| 107 |
+
"learning_rate": 0.00016829268292682927,
|
| 108 |
+
"loss": 0.202,
|
| 109 |
+
"step": 130
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 3.4146341463414633,
|
| 113 |
+
"grad_norm": 2.0693511962890625,
|
| 114 |
+
"learning_rate": 0.00016585365853658536,
|
| 115 |
+
"loss": 0.2906,
|
| 116 |
+
"step": 140
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 3.658536585365854,
|
| 120 |
+
"grad_norm": 3.4595236778259277,
|
| 121 |
+
"learning_rate": 0.00016341463414634147,
|
| 122 |
+
"loss": 0.2004,
|
| 123 |
+
"step": 150
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 3.902439024390244,
|
| 127 |
+
"grad_norm": 3.176142692565918,
|
| 128 |
+
"learning_rate": 0.00016097560975609758,
|
| 129 |
+
"loss": 0.1719,
|
| 130 |
+
"step": 160
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 4.146341463414634,
|
| 134 |
+
"grad_norm": 1.0586614608764648,
|
| 135 |
+
"learning_rate": 0.00015853658536585366,
|
| 136 |
+
"loss": 0.0913,
|
| 137 |
+
"step": 170
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 4.390243902439025,
|
| 141 |
+
"grad_norm": 0.48713743686676025,
|
| 142 |
+
"learning_rate": 0.00015609756097560978,
|
| 143 |
+
"loss": 0.0923,
|
| 144 |
+
"step": 180
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 4.634146341463414,
|
| 148 |
+
"grad_norm": 0.5843382477760315,
|
| 149 |
+
"learning_rate": 0.00015365853658536586,
|
| 150 |
+
"loss": 0.1999,
|
| 151 |
+
"step": 190
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 4.878048780487805,
|
| 155 |
+
"grad_norm": 0.11404013633728027,
|
| 156 |
+
"learning_rate": 0.00015121951219512197,
|
| 157 |
+
"loss": 0.0656,
|
| 158 |
+
"step": 200
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 4.878048780487805,
|
| 162 |
+
"eval_accuracy": 0.95,
|
| 163 |
+
"eval_loss": 0.16283166408538818,
|
| 164 |
+
"eval_runtime": 8.0607,
|
| 165 |
+
"eval_samples_per_second": 17.368,
|
| 166 |
+
"eval_steps_per_second": 2.233,
|
| 167 |
+
"step": 200
|
| 168 |
+
},
|
| 169 |
+
{
|
| 170 |
+
"epoch": 5.121951219512195,
|
| 171 |
+
"grad_norm": 0.10728397220373154,
|
| 172 |
+
"learning_rate": 0.00014878048780487806,
|
| 173 |
+
"loss": 0.0267,
|
| 174 |
+
"step": 210
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 5.365853658536586,
|
| 178 |
+
"grad_norm": 3.2944324016571045,
|
| 179 |
+
"learning_rate": 0.00014634146341463414,
|
| 180 |
+
"loss": 0.1612,
|
| 181 |
+
"step": 220
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 5.609756097560975,
|
| 185 |
+
"grad_norm": 0.12557579576969147,
|
| 186 |
+
"learning_rate": 0.00014390243902439025,
|
| 187 |
+
"loss": 0.1216,
|
| 188 |
+
"step": 230
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 5.853658536585366,
|
| 192 |
+
"grad_norm": 6.2914252281188965,
|
| 193 |
+
"learning_rate": 0.00014146341463414634,
|
| 194 |
+
"loss": 0.1067,
|
| 195 |
+
"step": 240
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 6.097560975609756,
|
| 199 |
+
"grad_norm": 3.4353785514831543,
|
| 200 |
+
"learning_rate": 0.00013902439024390245,
|
| 201 |
+
"loss": 0.0795,
|
| 202 |
+
"step": 250
|
| 203 |
+
},
|
| 204 |
+
{
|
| 205 |
+
"epoch": 6.341463414634147,
|
| 206 |
+
"grad_norm": 0.08993230760097504,
|
| 207 |
+
"learning_rate": 0.00013658536585365856,
|
| 208 |
+
"loss": 0.0919,
|
| 209 |
+
"step": 260
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 6.585365853658536,
|
| 213 |
+
"grad_norm": 7.937359809875488,
|
| 214 |
+
"learning_rate": 0.00013414634146341464,
|
| 215 |
+
"loss": 0.0621,
|
| 216 |
+
"step": 270
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 6.829268292682927,
|
| 220 |
+
"grad_norm": 0.10531166195869446,
|
| 221 |
+
"learning_rate": 0.00013170731707317076,
|
| 222 |
+
"loss": 0.0341,
|
| 223 |
+
"step": 280
|
| 224 |
+
},
|
| 225 |
+
{
|
| 226 |
+
"epoch": 7.073170731707317,
|
| 227 |
+
"grad_norm": 0.33791643381118774,
|
| 228 |
+
"learning_rate": 0.00012926829268292684,
|
| 229 |
+
"loss": 0.047,
|
| 230 |
+
"step": 290
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 7.317073170731708,
|
| 234 |
+
"grad_norm": 0.06028667464852333,
|
| 235 |
+
"learning_rate": 0.00012682926829268293,
|
| 236 |
+
"loss": 0.0192,
|
| 237 |
+
"step": 300
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 7.317073170731708,
|
| 241 |
+
"eval_accuracy": 0.9642857142857143,
|
| 242 |
+
"eval_loss": 0.10026070475578308,
|
| 243 |
+
"eval_runtime": 7.8254,
|
| 244 |
+
"eval_samples_per_second": 17.89,
|
| 245 |
+
"eval_steps_per_second": 2.3,
|
| 246 |
+
"step": 300
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 7.560975609756097,
|
| 250 |
+
"grad_norm": 0.11959153413772583,
|
| 251 |
+
"learning_rate": 0.00012439024390243904,
|
| 252 |
+
"loss": 0.0145,
|
| 253 |
+
"step": 310
|
| 254 |
+
},
|
| 255 |
+
{
|
| 256 |
+
"epoch": 7.804878048780488,
|
| 257 |
+
"grad_norm": 0.05471642687916756,
|
| 258 |
+
"learning_rate": 0.00012195121951219512,
|
| 259 |
+
"loss": 0.0987,
|
| 260 |
+
"step": 320
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 8.048780487804878,
|
| 264 |
+
"grad_norm": 0.0556318461894989,
|
| 265 |
+
"learning_rate": 0.00011951219512195122,
|
| 266 |
+
"loss": 0.0447,
|
| 267 |
+
"step": 330
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 8.292682926829269,
|
| 271 |
+
"grad_norm": 0.04877757653594017,
|
| 272 |
+
"learning_rate": 0.00011707317073170732,
|
| 273 |
+
"loss": 0.0226,
|
| 274 |
+
"step": 340
|
| 275 |
+
},
|
| 276 |
+
{
|
| 277 |
+
"epoch": 8.536585365853659,
|
| 278 |
+
"grad_norm": 0.6035917401313782,
|
| 279 |
+
"learning_rate": 0.00011463414634146342,
|
| 280 |
+
"loss": 0.0465,
|
| 281 |
+
"step": 350
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 8.78048780487805,
|
| 285 |
+
"grad_norm": 0.04456908255815506,
|
| 286 |
+
"learning_rate": 0.00011219512195121953,
|
| 287 |
+
"loss": 0.0116,
|
| 288 |
+
"step": 360
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 9.024390243902438,
|
| 292 |
+
"grad_norm": 0.04949244484305382,
|
| 293 |
+
"learning_rate": 0.00010975609756097563,
|
| 294 |
+
"loss": 0.0349,
|
| 295 |
+
"step": 370
|
| 296 |
+
},
|
| 297 |
+
{
|
| 298 |
+
"epoch": 9.268292682926829,
|
| 299 |
+
"grad_norm": 0.047532856464385986,
|
| 300 |
+
"learning_rate": 0.00010731707317073172,
|
| 301 |
+
"loss": 0.0097,
|
| 302 |
+
"step": 380
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 9.512195121951219,
|
| 306 |
+
"grad_norm": 7.109127521514893,
|
| 307 |
+
"learning_rate": 0.00010487804878048781,
|
| 308 |
+
"loss": 0.0565,
|
| 309 |
+
"step": 390
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 9.75609756097561,
|
| 313 |
+
"grad_norm": 0.04661267623305321,
|
| 314 |
+
"learning_rate": 0.0001024390243902439,
|
| 315 |
+
"loss": 0.0926,
|
| 316 |
+
"step": 400
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 9.75609756097561,
|
| 320 |
+
"eval_accuracy": 0.95,
|
| 321 |
+
"eval_loss": 0.1261894255876541,
|
| 322 |
+
"eval_runtime": 7.971,
|
| 323 |
+
"eval_samples_per_second": 17.564,
|
| 324 |
+
"eval_steps_per_second": 2.258,
|
| 325 |
+
"step": 400
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 10.0,
|
| 329 |
+
"grad_norm": 0.12421295046806335,
|
| 330 |
+
"learning_rate": 0.0001,
|
| 331 |
+
"loss": 0.0375,
|
| 332 |
+
"step": 410
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 10.24390243902439,
|
| 336 |
+
"grad_norm": 0.04042995348572731,
|
| 337 |
+
"learning_rate": 9.75609756097561e-05,
|
| 338 |
+
"loss": 0.0199,
|
| 339 |
+
"step": 420
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 10.487804878048781,
|
| 343 |
+
"grad_norm": 0.05262177810072899,
|
| 344 |
+
"learning_rate": 9.51219512195122e-05,
|
| 345 |
+
"loss": 0.0095,
|
| 346 |
+
"step": 430
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 10.731707317073171,
|
| 350 |
+
"grad_norm": 0.03628499433398247,
|
| 351 |
+
"learning_rate": 9.26829268292683e-05,
|
| 352 |
+
"loss": 0.0078,
|
| 353 |
+
"step": 440
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 10.975609756097562,
|
| 357 |
+
"grad_norm": 0.04206101596355438,
|
| 358 |
+
"learning_rate": 9.02439024390244e-05,
|
| 359 |
+
"loss": 0.0369,
|
| 360 |
+
"step": 450
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 11.21951219512195,
|
| 364 |
+
"grad_norm": 0.03796745836734772,
|
| 365 |
+
"learning_rate": 8.78048780487805e-05,
|
| 366 |
+
"loss": 0.0081,
|
| 367 |
+
"step": 460
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 11.463414634146341,
|
| 371 |
+
"grad_norm": 0.03188019618391991,
|
| 372 |
+
"learning_rate": 8.53658536585366e-05,
|
| 373 |
+
"loss": 0.0075,
|
| 374 |
+
"step": 470
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 11.707317073170731,
|
| 378 |
+
"grad_norm": 0.04463886842131615,
|
| 379 |
+
"learning_rate": 8.292682926829268e-05,
|
| 380 |
+
"loss": 0.0071,
|
| 381 |
+
"step": 480
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 11.951219512195122,
|
| 385 |
+
"grad_norm": 0.024716157466173172,
|
| 386 |
+
"learning_rate": 8.048780487804879e-05,
|
| 387 |
+
"loss": 0.0066,
|
| 388 |
+
"step": 490
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 12.195121951219512,
|
| 392 |
+
"grad_norm": 0.030284544453024864,
|
| 393 |
+
"learning_rate": 7.804878048780489e-05,
|
| 394 |
+
"loss": 0.0064,
|
| 395 |
+
"step": 500
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 12.195121951219512,
|
| 399 |
+
"eval_accuracy": 0.9642857142857143,
|
| 400 |
+
"eval_loss": 0.1611199826002121,
|
| 401 |
+
"eval_runtime": 8.7409,
|
| 402 |
+
"eval_samples_per_second": 16.017,
|
| 403 |
+
"eval_steps_per_second": 2.059,
|
| 404 |
+
"step": 500
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 12.439024390243903,
|
| 408 |
+
"grad_norm": 0.027342507615685463,
|
| 409 |
+
"learning_rate": 7.560975609756099e-05,
|
| 410 |
+
"loss": 0.0061,
|
| 411 |
+
"step": 510
|
| 412 |
+
},
|
| 413 |
+
{
|
| 414 |
+
"epoch": 12.682926829268293,
|
| 415 |
+
"grad_norm": 0.03514489158987999,
|
| 416 |
+
"learning_rate": 7.317073170731707e-05,
|
| 417 |
+
"loss": 0.006,
|
| 418 |
+
"step": 520
|
| 419 |
+
},
|
| 420 |
+
{
|
| 421 |
+
"epoch": 12.926829268292684,
|
| 422 |
+
"grad_norm": 0.02839997597038746,
|
| 423 |
+
"learning_rate": 7.073170731707317e-05,
|
| 424 |
+
"loss": 0.0056,
|
| 425 |
+
"step": 530
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 13.170731707317072,
|
| 429 |
+
"grad_norm": 0.025196045637130737,
|
| 430 |
+
"learning_rate": 6.829268292682928e-05,
|
| 431 |
+
"loss": 0.0057,
|
| 432 |
+
"step": 540
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 13.414634146341463,
|
| 436 |
+
"grad_norm": 0.022951927036046982,
|
| 437 |
+
"learning_rate": 6.585365853658538e-05,
|
| 438 |
+
"loss": 0.0053,
|
| 439 |
+
"step": 550
|
| 440 |
+
},
|
| 441 |
+
{
|
| 442 |
+
"epoch": 13.658536585365853,
|
| 443 |
+
"grad_norm": 0.02220899611711502,
|
| 444 |
+
"learning_rate": 6.341463414634146e-05,
|
| 445 |
+
"loss": 0.0056,
|
| 446 |
+
"step": 560
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 13.902439024390244,
|
| 450 |
+
"grad_norm": 0.026446521282196045,
|
| 451 |
+
"learning_rate": 6.097560975609756e-05,
|
| 452 |
+
"loss": 0.0051,
|
| 453 |
+
"step": 570
|
| 454 |
+
},
|
| 455 |
+
{
|
| 456 |
+
"epoch": 14.146341463414634,
|
| 457 |
+
"grad_norm": 0.021074431017041206,
|
| 458 |
+
"learning_rate": 5.853658536585366e-05,
|
| 459 |
+
"loss": 0.0051,
|
| 460 |
+
"step": 580
|
| 461 |
+
},
|
| 462 |
+
{
|
| 463 |
+
"epoch": 14.390243902439025,
|
| 464 |
+
"grad_norm": 0.026110034435987473,
|
| 465 |
+
"learning_rate": 5.6097560975609764e-05,
|
| 466 |
+
"loss": 0.0051,
|
| 467 |
+
"step": 590
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"epoch": 14.634146341463415,
|
| 471 |
+
"grad_norm": 0.024351950734853745,
|
| 472 |
+
"learning_rate": 5.365853658536586e-05,
|
| 473 |
+
"loss": 0.0049,
|
| 474 |
+
"step": 600
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 14.634146341463415,
|
| 478 |
+
"eval_accuracy": 0.9642857142857143,
|
| 479 |
+
"eval_loss": 0.15386557579040527,
|
| 480 |
+
"eval_runtime": 24.339,
|
| 481 |
+
"eval_samples_per_second": 5.752,
|
| 482 |
+
"eval_steps_per_second": 0.74,
|
| 483 |
+
"step": 600
|
| 484 |
+
},
|
| 485 |
+
{
|
| 486 |
+
"epoch": 14.878048780487806,
|
| 487 |
+
"grad_norm": 0.020671434700489044,
|
| 488 |
+
"learning_rate": 5.121951219512195e-05,
|
| 489 |
+
"loss": 0.0048,
|
| 490 |
+
"step": 610
|
| 491 |
+
},
|
| 492 |
+
{
|
| 493 |
+
"epoch": 15.121951219512194,
|
| 494 |
+
"grad_norm": 0.02226124331355095,
|
| 495 |
+
"learning_rate": 4.878048780487805e-05,
|
| 496 |
+
"loss": 0.0046,
|
| 497 |
+
"step": 620
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 15.365853658536585,
|
| 501 |
+
"grad_norm": 0.02063142880797386,
|
| 502 |
+
"learning_rate": 4.634146341463415e-05,
|
| 503 |
+
"loss": 0.0046,
|
| 504 |
+
"step": 630
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 15.609756097560975,
|
| 508 |
+
"grad_norm": 0.02496950887143612,
|
| 509 |
+
"learning_rate": 4.390243902439025e-05,
|
| 510 |
+
"loss": 0.0048,
|
| 511 |
+
"step": 640
|
| 512 |
+
},
|
| 513 |
+
{
|
| 514 |
+
"epoch": 15.853658536585366,
|
| 515 |
+
"grad_norm": 0.023448029533028603,
|
| 516 |
+
"learning_rate": 4.146341463414634e-05,
|
| 517 |
+
"loss": 0.0046,
|
| 518 |
+
"step": 650
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 16.097560975609756,
|
| 522 |
+
"grad_norm": 0.02106618881225586,
|
| 523 |
+
"learning_rate": 3.9024390243902444e-05,
|
| 524 |
+
"loss": 0.0045,
|
| 525 |
+
"step": 660
|
| 526 |
+
},
|
| 527 |
+
{
|
| 528 |
+
"epoch": 16.341463414634145,
|
| 529 |
+
"grad_norm": 0.02371513471007347,
|
| 530 |
+
"learning_rate": 3.6585365853658535e-05,
|
| 531 |
+
"loss": 0.0043,
|
| 532 |
+
"step": 670
|
| 533 |
+
},
|
| 534 |
+
{
|
| 535 |
+
"epoch": 16.585365853658537,
|
| 536 |
+
"grad_norm": 0.02032136172056198,
|
| 537 |
+
"learning_rate": 3.414634146341464e-05,
|
| 538 |
+
"loss": 0.0044,
|
| 539 |
+
"step": 680
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"epoch": 16.829268292682926,
|
| 543 |
+
"grad_norm": 0.020455192774534225,
|
| 544 |
+
"learning_rate": 3.170731707317073e-05,
|
| 545 |
+
"loss": 0.0043,
|
| 546 |
+
"step": 690
|
| 547 |
+
},
|
| 548 |
+
{
|
| 549 |
+
"epoch": 17.073170731707318,
|
| 550 |
+
"grad_norm": 0.023487282916903496,
|
| 551 |
+
"learning_rate": 2.926829268292683e-05,
|
| 552 |
+
"loss": 0.0044,
|
| 553 |
+
"step": 700
|
| 554 |
+
},
|
| 555 |
+
{
|
| 556 |
+
"epoch": 17.073170731707318,
|
| 557 |
+
"eval_accuracy": 0.9642857142857143,
|
| 558 |
+
"eval_loss": 0.15085048973560333,
|
| 559 |
+
"eval_runtime": 11.3536,
|
| 560 |
+
"eval_samples_per_second": 12.331,
|
| 561 |
+
"eval_steps_per_second": 1.585,
|
| 562 |
+
"step": 700
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 17.317073170731707,
|
| 566 |
+
"grad_norm": 0.018334366381168365,
|
| 567 |
+
"learning_rate": 2.682926829268293e-05,
|
| 568 |
+
"loss": 0.0044,
|
| 569 |
+
"step": 710
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 17.5609756097561,
|
| 573 |
+
"grad_norm": 0.01927388273179531,
|
| 574 |
+
"learning_rate": 2.4390243902439026e-05,
|
| 575 |
+
"loss": 0.0042,
|
| 576 |
+
"step": 720
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 17.804878048780488,
|
| 580 |
+
"grad_norm": 0.021918119862675667,
|
| 581 |
+
"learning_rate": 2.1951219512195124e-05,
|
| 582 |
+
"loss": 0.0042,
|
| 583 |
+
"step": 730
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 18.048780487804876,
|
| 587 |
+
"grad_norm": 0.019147571176290512,
|
| 588 |
+
"learning_rate": 1.9512195121951222e-05,
|
| 589 |
+
"loss": 0.0042,
|
| 590 |
+
"step": 740
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 18.29268292682927,
|
| 594 |
+
"grad_norm": 0.020710714161396027,
|
| 595 |
+
"learning_rate": 1.707317073170732e-05,
|
| 596 |
+
"loss": 0.0041,
|
| 597 |
+
"step": 750
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 18.536585365853657,
|
| 601 |
+
"grad_norm": 0.01830894872546196,
|
| 602 |
+
"learning_rate": 1.4634146341463415e-05,
|
| 603 |
+
"loss": 0.004,
|
| 604 |
+
"step": 760
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 18.78048780487805,
|
| 608 |
+
"grad_norm": 0.019078999757766724,
|
| 609 |
+
"learning_rate": 1.2195121951219513e-05,
|
| 610 |
+
"loss": 0.0042,
|
| 611 |
+
"step": 770
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 19.024390243902438,
|
| 615 |
+
"grad_norm": 0.020716849714517593,
|
| 616 |
+
"learning_rate": 9.756097560975611e-06,
|
| 617 |
+
"loss": 0.0042,
|
| 618 |
+
"step": 780
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 19.26829268292683,
|
| 622 |
+
"grad_norm": 0.018130987882614136,
|
| 623 |
+
"learning_rate": 7.317073170731707e-06,
|
| 624 |
+
"loss": 0.0041,
|
| 625 |
+
"step": 790
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 19.51219512195122,
|
| 629 |
+
"grad_norm": 0.018373191356658936,
|
| 630 |
+
"learning_rate": 4.8780487804878055e-06,
|
| 631 |
+
"loss": 0.0041,
|
| 632 |
+
"step": 800
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 19.51219512195122,
|
| 636 |
+
"eval_accuracy": 0.9642857142857143,
|
| 637 |
+
"eval_loss": 0.14993587136268616,
|
| 638 |
+
"eval_runtime": 14.7019,
|
| 639 |
+
"eval_samples_per_second": 9.523,
|
| 640 |
+
"eval_steps_per_second": 1.224,
|
| 641 |
+
"step": 800
|
| 642 |
+
},
|
| 643 |
+
{
|
| 644 |
+
"epoch": 19.75609756097561,
|
| 645 |
+
"grad_norm": 0.01835489086806774,
|
| 646 |
+
"learning_rate": 2.4390243902439027e-06,
|
| 647 |
+
"loss": 0.0039,
|
| 648 |
+
"step": 810
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"epoch": 20.0,
|
| 652 |
+
"grad_norm": 0.02105889283120632,
|
| 653 |
+
"learning_rate": 0.0,
|
| 654 |
+
"loss": 0.0041,
|
| 655 |
+
"step": 820
|
| 656 |
+
},
|
| 657 |
+
{
|
| 658 |
+
"epoch": 20.0,
|
| 659 |
+
"step": 820,
|
| 660 |
+
"total_flos": 1.004314187783209e+18,
|
| 661 |
+
"train_loss": 0.1252097422029914,
|
| 662 |
+
"train_runtime": 1111.906,
|
| 663 |
+
"train_samples_per_second": 11.656,
|
| 664 |
+
"train_steps_per_second": 0.737
|
| 665 |
+
}
|
| 666 |
+
],
|
| 667 |
+
"logging_steps": 10,
|
| 668 |
+
"max_steps": 820,
|
| 669 |
+
"num_input_tokens_seen": 0,
|
| 670 |
+
"num_train_epochs": 20,
|
| 671 |
+
"save_steps": 100,
|
| 672 |
+
"stateful_callbacks": {
|
| 673 |
+
"TrainerControl": {
|
| 674 |
+
"args": {
|
| 675 |
+
"should_epoch_stop": false,
|
| 676 |
+
"should_evaluate": false,
|
| 677 |
+
"should_log": false,
|
| 678 |
+
"should_save": true,
|
| 679 |
+
"should_training_stop": false
|
| 680 |
+
},
|
| 681 |
+
"attributes": {}
|
| 682 |
+
}
|
| 683 |
+
},
|
| 684 |
+
"total_flos": 1.004314187783209e+18,
|
| 685 |
+
"train_batch_size": 16,
|
| 686 |
+
"trial_name": null,
|
| 687 |
+
"trial_params": null
|
| 688 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2b87d34e39562259d7b9726043582bf66d43d61651b4037f9783e4b6b8c5689f
|
| 3 |
+
size 5112
|