Commit
•
76f95c4
1
Parent(s):
440a2ec
Upload folder using huggingface_hub
Browse files- README.md +74 -0
- all_results.json +8 -0
- config.json +40 -0
- model.safetensors +3 -0
- preprocessor_config.json +22 -0
- train_results.json +8 -0
- trainer_state.json +1027 -0
- training_args.bin +3 -0
README.md
ADDED
|
@@ -0,0 +1,74 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
base_model: google/vit-large-patch16-224
|
| 4 |
+
tags:
|
| 5 |
+
- image-classification
|
| 6 |
+
- generated_from_trainer
|
| 7 |
+
metrics:
|
| 8 |
+
- accuracy
|
| 9 |
+
model-index:
|
| 10 |
+
- name: vit-lage-beans-demo-v5
|
| 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-lage-beans-demo-v5
|
| 18 |
+
|
| 19 |
+
This model is a fine-tuned version of [google/vit-large-patch16-224](https://huggingface.co/google/vit-large-patch16-224) on the beans dataset.
|
| 20 |
+
It achieves the following results on the evaluation set:
|
| 21 |
+
- Loss: 0.1027
|
| 22 |
+
- Accuracy: 0.9792
|
| 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 |
+
|
| 49 |
+
### Training results
|
| 50 |
+
|
| 51 |
+
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
| 52 |
+
|:-------------:|:-----:|:----:|:---------------:|:--------:|
|
| 53 |
+
| 0.7861 | 0.28 | 100 | 0.7609 | 0.725 |
|
| 54 |
+
| 0.6276 | 0.56 | 200 | 0.6835 | 0.7764 |
|
| 55 |
+
| 0.6089 | 0.83 | 300 | 0.5568 | 0.7931 |
|
| 56 |
+
| 0.302 | 1.11 | 400 | 0.5887 | 0.8194 |
|
| 57 |
+
| 0.1723 | 1.39 | 500 | 0.2988 | 0.8903 |
|
| 58 |
+
| 0.2522 | 1.67 | 600 | 0.2564 | 0.9167 |
|
| 59 |
+
| 0.1115 | 1.94 | 700 | 0.1680 | 0.9472 |
|
| 60 |
+
| 0.1445 | 2.22 | 800 | 0.2065 | 0.9403 |
|
| 61 |
+
| 0.0648 | 2.5 | 900 | 0.1673 | 0.9569 |
|
| 62 |
+
| 0.0209 | 2.78 | 1000 | 0.1636 | 0.9569 |
|
| 63 |
+
| 0.0003 | 3.06 | 1100 | 0.1293 | 0.9694 |
|
| 64 |
+
| 0.0034 | 3.33 | 1200 | 0.0817 | 0.9792 |
|
| 65 |
+
| 0.0006 | 3.61 | 1300 | 0.0874 | 0.9833 |
|
| 66 |
+
| 0.0023 | 3.89 | 1400 | 0.1076 | 0.9778 |
|
| 67 |
+
|
| 68 |
+
|
| 69 |
+
### Framework versions
|
| 70 |
+
|
| 71 |
+
- Transformers 4.35.2
|
| 72 |
+
- Pytorch 2.1.0+cu118
|
| 73 |
+
- Datasets 2.15.0
|
| 74 |
+
- Tokenizers 0.15.0
|
all_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 4.0,
|
| 3 |
+
"total_flos": 6.311528820941783e+18,
|
| 4 |
+
"train_loss": 0.2221095090903772,
|
| 5 |
+
"train_runtime": 4823.2045,
|
| 6 |
+
"train_samples_per_second": 4.777,
|
| 7 |
+
"train_steps_per_second": 0.299
|
| 8 |
+
}
|
config.json
ADDED
|
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "google/vit-large-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": 1024,
|
| 11 |
+
"id2label": {
|
| 12 |
+
"0": "crack",
|
| 13 |
+
"1": "scratch",
|
| 14 |
+
"2": "tire flat",
|
| 15 |
+
"3": "dent",
|
| 16 |
+
"4": "glass shatter",
|
| 17 |
+
"5": "lamp broken"
|
| 18 |
+
},
|
| 19 |
+
"image_size": 224,
|
| 20 |
+
"initializer_range": 0.02,
|
| 21 |
+
"intermediate_size": 4096,
|
| 22 |
+
"label2id": {
|
| 23 |
+
"crack": "0",
|
| 24 |
+
"dent": "3",
|
| 25 |
+
"glass shatter": "4",
|
| 26 |
+
"lamp broken": "5",
|
| 27 |
+
"scratch": "1",
|
| 28 |
+
"tire flat": "2"
|
| 29 |
+
},
|
| 30 |
+
"layer_norm_eps": 1e-12,
|
| 31 |
+
"model_type": "vit",
|
| 32 |
+
"num_attention_heads": 16,
|
| 33 |
+
"num_channels": 3,
|
| 34 |
+
"num_hidden_layers": 24,
|
| 35 |
+
"patch_size": 16,
|
| 36 |
+
"problem_type": "single_label_classification",
|
| 37 |
+
"qkv_bias": true,
|
| 38 |
+
"torch_dtype": "float32",
|
| 39 |
+
"transformers_version": "4.35.2"
|
| 40 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:766e43cf282cbdc1692727e1501e71085a244441b2157a728228f6f757c2b82a
|
| 3 |
+
size 1213277672
|
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 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 4.0,
|
| 3 |
+
"total_flos": 6.311528820941783e+18,
|
| 4 |
+
"train_loss": 0.2221095090903772,
|
| 5 |
+
"train_runtime": 4823.2045,
|
| 6 |
+
"train_samples_per_second": 4.777,
|
| 7 |
+
"train_steps_per_second": 0.299
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1027 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.08173014223575592,
|
| 3 |
+
"best_model_checkpoint": "./vit-lage-beans-demo-v5/checkpoint-1200",
|
| 4 |
+
"epoch": 4.0,
|
| 5 |
+
"eval_steps": 100,
|
| 6 |
+
"global_step": 1440,
|
| 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": 0.00019861111111111113,
|
| 14 |
+
"loss": 1.4551,
|
| 15 |
+
"step": 10
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 0.06,
|
| 19 |
+
"learning_rate": 0.00019722222222222225,
|
| 20 |
+
"loss": 1.148,
|
| 21 |
+
"step": 20
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
"epoch": 0.08,
|
| 25 |
+
"learning_rate": 0.00019583333333333334,
|
| 26 |
+
"loss": 0.9667,
|
| 27 |
+
"step": 30
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 0.11,
|
| 31 |
+
"learning_rate": 0.00019444444444444446,
|
| 32 |
+
"loss": 0.7212,
|
| 33 |
+
"step": 40
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 0.14,
|
| 37 |
+
"learning_rate": 0.00019305555555555558,
|
| 38 |
+
"loss": 1.0176,
|
| 39 |
+
"step": 50
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.17,
|
| 43 |
+
"learning_rate": 0.00019166666666666667,
|
| 44 |
+
"loss": 0.6942,
|
| 45 |
+
"step": 60
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.19,
|
| 49 |
+
"learning_rate": 0.00019027777777777776,
|
| 50 |
+
"loss": 0.8307,
|
| 51 |
+
"step": 70
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.22,
|
| 55 |
+
"learning_rate": 0.00018888888888888888,
|
| 56 |
+
"loss": 0.5401,
|
| 57 |
+
"step": 80
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 0.25,
|
| 61 |
+
"learning_rate": 0.0001875,
|
| 62 |
+
"loss": 0.7918,
|
| 63 |
+
"step": 90
|
| 64 |
+
},
|
| 65 |
+
{
|
| 66 |
+
"epoch": 0.28,
|
| 67 |
+
"learning_rate": 0.00018611111111111112,
|
| 68 |
+
"loss": 0.7861,
|
| 69 |
+
"step": 100
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 0.28,
|
| 73 |
+
"eval_accuracy": 0.725,
|
| 74 |
+
"eval_loss": 0.7609096765518188,
|
| 75 |
+
"eval_runtime": 45.8495,
|
| 76 |
+
"eval_samples_per_second": 15.704,
|
| 77 |
+
"eval_steps_per_second": 1.963,
|
| 78 |
+
"step": 100
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"epoch": 0.31,
|
| 82 |
+
"learning_rate": 0.00018472222222222224,
|
| 83 |
+
"loss": 0.6622,
|
| 84 |
+
"step": 110
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 0.33,
|
| 88 |
+
"learning_rate": 0.00018333333333333334,
|
| 89 |
+
"loss": 0.8922,
|
| 90 |
+
"step": 120
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 0.36,
|
| 94 |
+
"learning_rate": 0.00018194444444444445,
|
| 95 |
+
"loss": 0.7098,
|
| 96 |
+
"step": 130
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"epoch": 0.39,
|
| 100 |
+
"learning_rate": 0.00018055555555555557,
|
| 101 |
+
"loss": 0.6424,
|
| 102 |
+
"step": 140
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 0.42,
|
| 106 |
+
"learning_rate": 0.0001791666666666667,
|
| 107 |
+
"loss": 0.6945,
|
| 108 |
+
"step": 150
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.44,
|
| 112 |
+
"learning_rate": 0.00017777777777777779,
|
| 113 |
+
"loss": 0.634,
|
| 114 |
+
"step": 160
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.47,
|
| 118 |
+
"learning_rate": 0.0001763888888888889,
|
| 119 |
+
"loss": 0.5476,
|
| 120 |
+
"step": 170
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 0.5,
|
| 124 |
+
"learning_rate": 0.000175,
|
| 125 |
+
"loss": 0.6335,
|
| 126 |
+
"step": 180
|
| 127 |
+
},
|
| 128 |
+
{
|
| 129 |
+
"epoch": 0.53,
|
| 130 |
+
"learning_rate": 0.00017361111111111112,
|
| 131 |
+
"loss": 0.4134,
|
| 132 |
+
"step": 190
|
| 133 |
+
},
|
| 134 |
+
{
|
| 135 |
+
"epoch": 0.56,
|
| 136 |
+
"learning_rate": 0.00017222222222222224,
|
| 137 |
+
"loss": 0.6276,
|
| 138 |
+
"step": 200
|
| 139 |
+
},
|
| 140 |
+
{
|
| 141 |
+
"epoch": 0.56,
|
| 142 |
+
"eval_accuracy": 0.7763888888888889,
|
| 143 |
+
"eval_loss": 0.6835257411003113,
|
| 144 |
+
"eval_runtime": 46.4022,
|
| 145 |
+
"eval_samples_per_second": 15.517,
|
| 146 |
+
"eval_steps_per_second": 1.94,
|
| 147 |
+
"step": 200
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 0.58,
|
| 151 |
+
"learning_rate": 0.00017083333333333333,
|
| 152 |
+
"loss": 0.6095,
|
| 153 |
+
"step": 210
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 0.61,
|
| 157 |
+
"learning_rate": 0.00016944444444444445,
|
| 158 |
+
"loss": 0.5328,
|
| 159 |
+
"step": 220
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 0.64,
|
| 163 |
+
"learning_rate": 0.00016805555555555557,
|
| 164 |
+
"loss": 0.4397,
|
| 165 |
+
"step": 230
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 0.67,
|
| 169 |
+
"learning_rate": 0.0001666666666666667,
|
| 170 |
+
"loss": 0.5338,
|
| 171 |
+
"step": 240
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.69,
|
| 175 |
+
"learning_rate": 0.00016527777777777778,
|
| 176 |
+
"loss": 0.5808,
|
| 177 |
+
"step": 250
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.72,
|
| 181 |
+
"learning_rate": 0.0001638888888888889,
|
| 182 |
+
"loss": 0.601,
|
| 183 |
+
"step": 260
|
| 184 |
+
},
|
| 185 |
+
{
|
| 186 |
+
"epoch": 0.75,
|
| 187 |
+
"learning_rate": 0.00016250000000000002,
|
| 188 |
+
"loss": 0.445,
|
| 189 |
+
"step": 270
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 0.78,
|
| 193 |
+
"learning_rate": 0.0001611111111111111,
|
| 194 |
+
"loss": 0.3964,
|
| 195 |
+
"step": 280
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 0.81,
|
| 199 |
+
"learning_rate": 0.00015972222222222223,
|
| 200 |
+
"loss": 0.5958,
|
| 201 |
+
"step": 290
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 0.83,
|
| 205 |
+
"learning_rate": 0.00015833333333333332,
|
| 206 |
+
"loss": 0.6089,
|
| 207 |
+
"step": 300
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 0.83,
|
| 211 |
+
"eval_accuracy": 0.7930555555555555,
|
| 212 |
+
"eval_loss": 0.5567876100540161,
|
| 213 |
+
"eval_runtime": 46.1606,
|
| 214 |
+
"eval_samples_per_second": 15.598,
|
| 215 |
+
"eval_steps_per_second": 1.95,
|
| 216 |
+
"step": 300
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 0.86,
|
| 220 |
+
"learning_rate": 0.00015694444444444444,
|
| 221 |
+
"loss": 0.518,
|
| 222 |
+
"step": 310
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"epoch": 0.89,
|
| 226 |
+
"learning_rate": 0.00015555555555555556,
|
| 227 |
+
"loss": 0.4721,
|
| 228 |
+
"step": 320
|
| 229 |
+
},
|
| 230 |
+
{
|
| 231 |
+
"epoch": 0.92,
|
| 232 |
+
"learning_rate": 0.00015416666666666668,
|
| 233 |
+
"loss": 0.4247,
|
| 234 |
+
"step": 330
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.94,
|
| 238 |
+
"learning_rate": 0.00015277777777777777,
|
| 239 |
+
"loss": 0.4503,
|
| 240 |
+
"step": 340
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.97,
|
| 244 |
+
"learning_rate": 0.0001513888888888889,
|
| 245 |
+
"loss": 0.344,
|
| 246 |
+
"step": 350
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 1.0,
|
| 250 |
+
"learning_rate": 0.00015000000000000001,
|
| 251 |
+
"loss": 0.3077,
|
| 252 |
+
"step": 360
|
| 253 |
+
},
|
| 254 |
+
{
|
| 255 |
+
"epoch": 1.03,
|
| 256 |
+
"learning_rate": 0.00014861111111111113,
|
| 257 |
+
"loss": 0.2311,
|
| 258 |
+
"step": 370
|
| 259 |
+
},
|
| 260 |
+
{
|
| 261 |
+
"epoch": 1.06,
|
| 262 |
+
"learning_rate": 0.00014722222222222223,
|
| 263 |
+
"loss": 0.2271,
|
| 264 |
+
"step": 380
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 1.08,
|
| 268 |
+
"learning_rate": 0.00014583333333333335,
|
| 269 |
+
"loss": 0.2595,
|
| 270 |
+
"step": 390
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 1.11,
|
| 274 |
+
"learning_rate": 0.00014444444444444444,
|
| 275 |
+
"loss": 0.302,
|
| 276 |
+
"step": 400
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 1.11,
|
| 280 |
+
"eval_accuracy": 0.8194444444444444,
|
| 281 |
+
"eval_loss": 0.5886520147323608,
|
| 282 |
+
"eval_runtime": 46.0306,
|
| 283 |
+
"eval_samples_per_second": 15.642,
|
| 284 |
+
"eval_steps_per_second": 1.955,
|
| 285 |
+
"step": 400
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 1.14,
|
| 289 |
+
"learning_rate": 0.00014305555555555556,
|
| 290 |
+
"loss": 0.2532,
|
| 291 |
+
"step": 410
|
| 292 |
+
},
|
| 293 |
+
{
|
| 294 |
+
"epoch": 1.17,
|
| 295 |
+
"learning_rate": 0.00014166666666666668,
|
| 296 |
+
"loss": 0.3509,
|
| 297 |
+
"step": 420
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 1.19,
|
| 301 |
+
"learning_rate": 0.00014027777777777777,
|
| 302 |
+
"loss": 0.2317,
|
| 303 |
+
"step": 430
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 1.22,
|
| 307 |
+
"learning_rate": 0.0001388888888888889,
|
| 308 |
+
"loss": 0.2371,
|
| 309 |
+
"step": 440
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 1.25,
|
| 313 |
+
"learning_rate": 0.0001375,
|
| 314 |
+
"loss": 0.2115,
|
| 315 |
+
"step": 450
|
| 316 |
+
},
|
| 317 |
+
{
|
| 318 |
+
"epoch": 1.28,
|
| 319 |
+
"learning_rate": 0.00013611111111111113,
|
| 320 |
+
"loss": 0.1403,
|
| 321 |
+
"step": 460
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 1.31,
|
| 325 |
+
"learning_rate": 0.00013472222222222225,
|
| 326 |
+
"loss": 0.163,
|
| 327 |
+
"step": 470
|
| 328 |
+
},
|
| 329 |
+
{
|
| 330 |
+
"epoch": 1.33,
|
| 331 |
+
"learning_rate": 0.00013333333333333334,
|
| 332 |
+
"loss": 0.4951,
|
| 333 |
+
"step": 480
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 1.36,
|
| 337 |
+
"learning_rate": 0.00013194444444444446,
|
| 338 |
+
"loss": 0.274,
|
| 339 |
+
"step": 490
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 1.39,
|
| 343 |
+
"learning_rate": 0.00013055555555555555,
|
| 344 |
+
"loss": 0.1723,
|
| 345 |
+
"step": 500
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 1.39,
|
| 349 |
+
"eval_accuracy": 0.8902777777777777,
|
| 350 |
+
"eval_loss": 0.298811137676239,
|
| 351 |
+
"eval_runtime": 45.7665,
|
| 352 |
+
"eval_samples_per_second": 15.732,
|
| 353 |
+
"eval_steps_per_second": 1.967,
|
| 354 |
+
"step": 500
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 1.42,
|
| 358 |
+
"learning_rate": 0.00012916666666666667,
|
| 359 |
+
"loss": 0.291,
|
| 360 |
+
"step": 510
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 1.44,
|
| 364 |
+
"learning_rate": 0.00012777777777777776,
|
| 365 |
+
"loss": 0.2423,
|
| 366 |
+
"step": 520
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 1.47,
|
| 370 |
+
"learning_rate": 0.00012638888888888888,
|
| 371 |
+
"loss": 0.1181,
|
| 372 |
+
"step": 530
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"epoch": 1.5,
|
| 376 |
+
"learning_rate": 0.000125,
|
| 377 |
+
"loss": 0.2404,
|
| 378 |
+
"step": 540
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"epoch": 1.53,
|
| 382 |
+
"learning_rate": 0.00012361111111111112,
|
| 383 |
+
"loss": 0.1568,
|
| 384 |
+
"step": 550
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 1.56,
|
| 388 |
+
"learning_rate": 0.00012222222222222224,
|
| 389 |
+
"loss": 0.1433,
|
| 390 |
+
"step": 560
|
| 391 |
+
},
|
| 392 |
+
{
|
| 393 |
+
"epoch": 1.58,
|
| 394 |
+
"learning_rate": 0.00012083333333333333,
|
| 395 |
+
"loss": 0.1673,
|
| 396 |
+
"step": 570
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"epoch": 1.61,
|
| 400 |
+
"learning_rate": 0.00011944444444444445,
|
| 401 |
+
"loss": 0.1575,
|
| 402 |
+
"step": 580
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 1.64,
|
| 406 |
+
"learning_rate": 0.00011805555555555556,
|
| 407 |
+
"loss": 0.1429,
|
| 408 |
+
"step": 590
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 1.67,
|
| 412 |
+
"learning_rate": 0.00011666666666666668,
|
| 413 |
+
"loss": 0.2522,
|
| 414 |
+
"step": 600
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 1.67,
|
| 418 |
+
"eval_accuracy": 0.9166666666666666,
|
| 419 |
+
"eval_loss": 0.25636595487594604,
|
| 420 |
+
"eval_runtime": 45.8795,
|
| 421 |
+
"eval_samples_per_second": 15.693,
|
| 422 |
+
"eval_steps_per_second": 1.962,
|
| 423 |
+
"step": 600
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 1.69,
|
| 427 |
+
"learning_rate": 0.00011527777777777777,
|
| 428 |
+
"loss": 0.11,
|
| 429 |
+
"step": 610
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 1.72,
|
| 433 |
+
"learning_rate": 0.00011388888888888889,
|
| 434 |
+
"loss": 0.1725,
|
| 435 |
+
"step": 620
|
| 436 |
+
},
|
| 437 |
+
{
|
| 438 |
+
"epoch": 1.75,
|
| 439 |
+
"learning_rate": 0.00011250000000000001,
|
| 440 |
+
"loss": 0.159,
|
| 441 |
+
"step": 630
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 1.78,
|
| 445 |
+
"learning_rate": 0.00011111111111111112,
|
| 446 |
+
"loss": 0.0854,
|
| 447 |
+
"step": 640
|
| 448 |
+
},
|
| 449 |
+
{
|
| 450 |
+
"epoch": 1.81,
|
| 451 |
+
"learning_rate": 0.00010972222222222224,
|
| 452 |
+
"loss": 0.0999,
|
| 453 |
+
"step": 650
|
| 454 |
+
},
|
| 455 |
+
{
|
| 456 |
+
"epoch": 1.83,
|
| 457 |
+
"learning_rate": 0.00010833333333333333,
|
| 458 |
+
"loss": 0.2217,
|
| 459 |
+
"step": 660
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 1.86,
|
| 463 |
+
"learning_rate": 0.00010694444444444445,
|
| 464 |
+
"loss": 0.1712,
|
| 465 |
+
"step": 670
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 1.89,
|
| 469 |
+
"learning_rate": 0.00010555555555555557,
|
| 470 |
+
"loss": 0.1299,
|
| 471 |
+
"step": 680
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 1.92,
|
| 475 |
+
"learning_rate": 0.00010416666666666667,
|
| 476 |
+
"loss": 0.1376,
|
| 477 |
+
"step": 690
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 1.94,
|
| 481 |
+
"learning_rate": 0.00010277777777777778,
|
| 482 |
+
"loss": 0.1115,
|
| 483 |
+
"step": 700
|
| 484 |
+
},
|
| 485 |
+
{
|
| 486 |
+
"epoch": 1.94,
|
| 487 |
+
"eval_accuracy": 0.9472222222222222,
|
| 488 |
+
"eval_loss": 0.16800908744335175,
|
| 489 |
+
"eval_runtime": 46.034,
|
| 490 |
+
"eval_samples_per_second": 15.641,
|
| 491 |
+
"eval_steps_per_second": 1.955,
|
| 492 |
+
"step": 700
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 1.97,
|
| 496 |
+
"learning_rate": 0.00010138888888888889,
|
| 497 |
+
"loss": 0.1504,
|
| 498 |
+
"step": 710
|
| 499 |
+
},
|
| 500 |
+
{
|
| 501 |
+
"epoch": 2.0,
|
| 502 |
+
"learning_rate": 0.0001,
|
| 503 |
+
"loss": 0.0608,
|
| 504 |
+
"step": 720
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 2.03,
|
| 508 |
+
"learning_rate": 9.861111111111112e-05,
|
| 509 |
+
"loss": 0.0344,
|
| 510 |
+
"step": 730
|
| 511 |
+
},
|
| 512 |
+
{
|
| 513 |
+
"epoch": 2.06,
|
| 514 |
+
"learning_rate": 9.722222222222223e-05,
|
| 515 |
+
"loss": 0.0326,
|
| 516 |
+
"step": 740
|
| 517 |
+
},
|
| 518 |
+
{
|
| 519 |
+
"epoch": 2.08,
|
| 520 |
+
"learning_rate": 9.583333333333334e-05,
|
| 521 |
+
"loss": 0.2028,
|
| 522 |
+
"step": 750
|
| 523 |
+
},
|
| 524 |
+
{
|
| 525 |
+
"epoch": 2.11,
|
| 526 |
+
"learning_rate": 9.444444444444444e-05,
|
| 527 |
+
"loss": 0.0335,
|
| 528 |
+
"step": 760
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 2.14,
|
| 532 |
+
"learning_rate": 9.305555555555556e-05,
|
| 533 |
+
"loss": 0.0554,
|
| 534 |
+
"step": 770
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 2.17,
|
| 538 |
+
"learning_rate": 9.166666666666667e-05,
|
| 539 |
+
"loss": 0.0525,
|
| 540 |
+
"step": 780
|
| 541 |
+
},
|
| 542 |
+
{
|
| 543 |
+
"epoch": 2.19,
|
| 544 |
+
"learning_rate": 9.027777777777779e-05,
|
| 545 |
+
"loss": 0.0358,
|
| 546 |
+
"step": 790
|
| 547 |
+
},
|
| 548 |
+
{
|
| 549 |
+
"epoch": 2.22,
|
| 550 |
+
"learning_rate": 8.888888888888889e-05,
|
| 551 |
+
"loss": 0.1445,
|
| 552 |
+
"step": 800
|
| 553 |
+
},
|
| 554 |
+
{
|
| 555 |
+
"epoch": 2.22,
|
| 556 |
+
"eval_accuracy": 0.9402777777777778,
|
| 557 |
+
"eval_loss": 0.2064695954322815,
|
| 558 |
+
"eval_runtime": 45.6748,
|
| 559 |
+
"eval_samples_per_second": 15.764,
|
| 560 |
+
"eval_steps_per_second": 1.97,
|
| 561 |
+
"step": 800
|
| 562 |
+
},
|
| 563 |
+
{
|
| 564 |
+
"epoch": 2.25,
|
| 565 |
+
"learning_rate": 8.75e-05,
|
| 566 |
+
"loss": 0.0905,
|
| 567 |
+
"step": 810
|
| 568 |
+
},
|
| 569 |
+
{
|
| 570 |
+
"epoch": 2.28,
|
| 571 |
+
"learning_rate": 8.611111111111112e-05,
|
| 572 |
+
"loss": 0.006,
|
| 573 |
+
"step": 820
|
| 574 |
+
},
|
| 575 |
+
{
|
| 576 |
+
"epoch": 2.31,
|
| 577 |
+
"learning_rate": 8.472222222222222e-05,
|
| 578 |
+
"loss": 0.0438,
|
| 579 |
+
"step": 830
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 2.33,
|
| 583 |
+
"learning_rate": 8.333333333333334e-05,
|
| 584 |
+
"loss": 0.0172,
|
| 585 |
+
"step": 840
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 2.36,
|
| 589 |
+
"learning_rate": 8.194444444444445e-05,
|
| 590 |
+
"loss": 0.021,
|
| 591 |
+
"step": 850
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 2.39,
|
| 595 |
+
"learning_rate": 8.055555555555556e-05,
|
| 596 |
+
"loss": 0.0362,
|
| 597 |
+
"step": 860
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 2.42,
|
| 601 |
+
"learning_rate": 7.916666666666666e-05,
|
| 602 |
+
"loss": 0.0007,
|
| 603 |
+
"step": 870
|
| 604 |
+
},
|
| 605 |
+
{
|
| 606 |
+
"epoch": 2.44,
|
| 607 |
+
"learning_rate": 7.777777777777778e-05,
|
| 608 |
+
"loss": 0.0898,
|
| 609 |
+
"step": 880
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 2.47,
|
| 613 |
+
"learning_rate": 7.638888888888889e-05,
|
| 614 |
+
"loss": 0.0117,
|
| 615 |
+
"step": 890
|
| 616 |
+
},
|
| 617 |
+
{
|
| 618 |
+
"epoch": 2.5,
|
| 619 |
+
"learning_rate": 7.500000000000001e-05,
|
| 620 |
+
"loss": 0.0648,
|
| 621 |
+
"step": 900
|
| 622 |
+
},
|
| 623 |
+
{
|
| 624 |
+
"epoch": 2.5,
|
| 625 |
+
"eval_accuracy": 0.9569444444444445,
|
| 626 |
+
"eval_loss": 0.1673145890235901,
|
| 627 |
+
"eval_runtime": 44.888,
|
| 628 |
+
"eval_samples_per_second": 16.04,
|
| 629 |
+
"eval_steps_per_second": 2.005,
|
| 630 |
+
"step": 900
|
| 631 |
+
},
|
| 632 |
+
{
|
| 633 |
+
"epoch": 2.53,
|
| 634 |
+
"learning_rate": 7.361111111111111e-05,
|
| 635 |
+
"loss": 0.0248,
|
| 636 |
+
"step": 910
|
| 637 |
+
},
|
| 638 |
+
{
|
| 639 |
+
"epoch": 2.56,
|
| 640 |
+
"learning_rate": 7.222222222222222e-05,
|
| 641 |
+
"loss": 0.0785,
|
| 642 |
+
"step": 920
|
| 643 |
+
},
|
| 644 |
+
{
|
| 645 |
+
"epoch": 2.58,
|
| 646 |
+
"learning_rate": 7.083333333333334e-05,
|
| 647 |
+
"loss": 0.0009,
|
| 648 |
+
"step": 930
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"epoch": 2.61,
|
| 652 |
+
"learning_rate": 6.944444444444444e-05,
|
| 653 |
+
"loss": 0.0249,
|
| 654 |
+
"step": 940
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 2.64,
|
| 658 |
+
"learning_rate": 6.805555555555556e-05,
|
| 659 |
+
"loss": 0.018,
|
| 660 |
+
"step": 950
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 2.67,
|
| 664 |
+
"learning_rate": 6.666666666666667e-05,
|
| 665 |
+
"loss": 0.0016,
|
| 666 |
+
"step": 960
|
| 667 |
+
},
|
| 668 |
+
{
|
| 669 |
+
"epoch": 2.69,
|
| 670 |
+
"learning_rate": 6.527777777777778e-05,
|
| 671 |
+
"loss": 0.0616,
|
| 672 |
+
"step": 970
|
| 673 |
+
},
|
| 674 |
+
{
|
| 675 |
+
"epoch": 2.72,
|
| 676 |
+
"learning_rate": 6.388888888888888e-05,
|
| 677 |
+
"loss": 0.0318,
|
| 678 |
+
"step": 980
|
| 679 |
+
},
|
| 680 |
+
{
|
| 681 |
+
"epoch": 2.75,
|
| 682 |
+
"learning_rate": 6.25e-05,
|
| 683 |
+
"loss": 0.0253,
|
| 684 |
+
"step": 990
|
| 685 |
+
},
|
| 686 |
+
{
|
| 687 |
+
"epoch": 2.78,
|
| 688 |
+
"learning_rate": 6.111111111111112e-05,
|
| 689 |
+
"loss": 0.0209,
|
| 690 |
+
"step": 1000
|
| 691 |
+
},
|
| 692 |
+
{
|
| 693 |
+
"epoch": 2.78,
|
| 694 |
+
"eval_accuracy": 0.9569444444444445,
|
| 695 |
+
"eval_loss": 0.16363362967967987,
|
| 696 |
+
"eval_runtime": 45.529,
|
| 697 |
+
"eval_samples_per_second": 15.814,
|
| 698 |
+
"eval_steps_per_second": 1.977,
|
| 699 |
+
"step": 1000
|
| 700 |
+
},
|
| 701 |
+
{
|
| 702 |
+
"epoch": 2.81,
|
| 703 |
+
"learning_rate": 5.972222222222223e-05,
|
| 704 |
+
"loss": 0.0209,
|
| 705 |
+
"step": 1010
|
| 706 |
+
},
|
| 707 |
+
{
|
| 708 |
+
"epoch": 2.83,
|
| 709 |
+
"learning_rate": 5.833333333333334e-05,
|
| 710 |
+
"loss": 0.0082,
|
| 711 |
+
"step": 1020
|
| 712 |
+
},
|
| 713 |
+
{
|
| 714 |
+
"epoch": 2.86,
|
| 715 |
+
"learning_rate": 5.6944444444444445e-05,
|
| 716 |
+
"loss": 0.051,
|
| 717 |
+
"step": 1030
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 2.89,
|
| 721 |
+
"learning_rate": 5.555555555555556e-05,
|
| 722 |
+
"loss": 0.0023,
|
| 723 |
+
"step": 1040
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 2.92,
|
| 727 |
+
"learning_rate": 5.4166666666666664e-05,
|
| 728 |
+
"loss": 0.0072,
|
| 729 |
+
"step": 1050
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 2.94,
|
| 733 |
+
"learning_rate": 5.2777777777777784e-05,
|
| 734 |
+
"loss": 0.0777,
|
| 735 |
+
"step": 1060
|
| 736 |
+
},
|
| 737 |
+
{
|
| 738 |
+
"epoch": 2.97,
|
| 739 |
+
"learning_rate": 5.138888888888889e-05,
|
| 740 |
+
"loss": 0.0186,
|
| 741 |
+
"step": 1070
|
| 742 |
+
},
|
| 743 |
+
{
|
| 744 |
+
"epoch": 3.0,
|
| 745 |
+
"learning_rate": 5e-05,
|
| 746 |
+
"loss": 0.0652,
|
| 747 |
+
"step": 1080
|
| 748 |
+
},
|
| 749 |
+
{
|
| 750 |
+
"epoch": 3.03,
|
| 751 |
+
"learning_rate": 4.8611111111111115e-05,
|
| 752 |
+
"loss": 0.0041,
|
| 753 |
+
"step": 1090
|
| 754 |
+
},
|
| 755 |
+
{
|
| 756 |
+
"epoch": 3.06,
|
| 757 |
+
"learning_rate": 4.722222222222222e-05,
|
| 758 |
+
"loss": 0.0003,
|
| 759 |
+
"step": 1100
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 3.06,
|
| 763 |
+
"eval_accuracy": 0.9694444444444444,
|
| 764 |
+
"eval_loss": 0.12931323051452637,
|
| 765 |
+
"eval_runtime": 45.4456,
|
| 766 |
+
"eval_samples_per_second": 15.843,
|
| 767 |
+
"eval_steps_per_second": 1.98,
|
| 768 |
+
"step": 1100
|
| 769 |
+
},
|
| 770 |
+
{
|
| 771 |
+
"epoch": 3.08,
|
| 772 |
+
"learning_rate": 4.5833333333333334e-05,
|
| 773 |
+
"loss": 0.0004,
|
| 774 |
+
"step": 1110
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 3.11,
|
| 778 |
+
"learning_rate": 4.4444444444444447e-05,
|
| 779 |
+
"loss": 0.0121,
|
| 780 |
+
"step": 1120
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 3.14,
|
| 784 |
+
"learning_rate": 4.305555555555556e-05,
|
| 785 |
+
"loss": 0.002,
|
| 786 |
+
"step": 1130
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 3.17,
|
| 790 |
+
"learning_rate": 4.166666666666667e-05,
|
| 791 |
+
"loss": 0.0035,
|
| 792 |
+
"step": 1140
|
| 793 |
+
},
|
| 794 |
+
{
|
| 795 |
+
"epoch": 3.19,
|
| 796 |
+
"learning_rate": 4.027777777777778e-05,
|
| 797 |
+
"loss": 0.0057,
|
| 798 |
+
"step": 1150
|
| 799 |
+
},
|
| 800 |
+
{
|
| 801 |
+
"epoch": 3.22,
|
| 802 |
+
"learning_rate": 3.888888888888889e-05,
|
| 803 |
+
"loss": 0.0002,
|
| 804 |
+
"step": 1160
|
| 805 |
+
},
|
| 806 |
+
{
|
| 807 |
+
"epoch": 3.25,
|
| 808 |
+
"learning_rate": 3.7500000000000003e-05,
|
| 809 |
+
"loss": 0.0005,
|
| 810 |
+
"step": 1170
|
| 811 |
+
},
|
| 812 |
+
{
|
| 813 |
+
"epoch": 3.28,
|
| 814 |
+
"learning_rate": 3.611111111111111e-05,
|
| 815 |
+
"loss": 0.0008,
|
| 816 |
+
"step": 1180
|
| 817 |
+
},
|
| 818 |
+
{
|
| 819 |
+
"epoch": 3.31,
|
| 820 |
+
"learning_rate": 3.472222222222222e-05,
|
| 821 |
+
"loss": 0.017,
|
| 822 |
+
"step": 1190
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 3.33,
|
| 826 |
+
"learning_rate": 3.3333333333333335e-05,
|
| 827 |
+
"loss": 0.0034,
|
| 828 |
+
"step": 1200
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 3.33,
|
| 832 |
+
"eval_accuracy": 0.9791666666666666,
|
| 833 |
+
"eval_loss": 0.08173014223575592,
|
| 834 |
+
"eval_runtime": 44.7405,
|
| 835 |
+
"eval_samples_per_second": 16.093,
|
| 836 |
+
"eval_steps_per_second": 2.012,
|
| 837 |
+
"step": 1200
|
| 838 |
+
},
|
| 839 |
+
{
|
| 840 |
+
"epoch": 3.36,
|
| 841 |
+
"learning_rate": 3.194444444444444e-05,
|
| 842 |
+
"loss": 0.0008,
|
| 843 |
+
"step": 1210
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 3.39,
|
| 847 |
+
"learning_rate": 3.055555555555556e-05,
|
| 848 |
+
"loss": 0.0003,
|
| 849 |
+
"step": 1220
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 3.42,
|
| 853 |
+
"learning_rate": 2.916666666666667e-05,
|
| 854 |
+
"loss": 0.0201,
|
| 855 |
+
"step": 1230
|
| 856 |
+
},
|
| 857 |
+
{
|
| 858 |
+
"epoch": 3.44,
|
| 859 |
+
"learning_rate": 2.777777777777778e-05,
|
| 860 |
+
"loss": 0.0003,
|
| 861 |
+
"step": 1240
|
| 862 |
+
},
|
| 863 |
+
{
|
| 864 |
+
"epoch": 3.47,
|
| 865 |
+
"learning_rate": 2.6388888888888892e-05,
|
| 866 |
+
"loss": 0.0001,
|
| 867 |
+
"step": 1250
|
| 868 |
+
},
|
| 869 |
+
{
|
| 870 |
+
"epoch": 3.5,
|
| 871 |
+
"learning_rate": 2.5e-05,
|
| 872 |
+
"loss": 0.0002,
|
| 873 |
+
"step": 1260
|
| 874 |
+
},
|
| 875 |
+
{
|
| 876 |
+
"epoch": 3.53,
|
| 877 |
+
"learning_rate": 2.361111111111111e-05,
|
| 878 |
+
"loss": 0.0001,
|
| 879 |
+
"step": 1270
|
| 880 |
+
},
|
| 881 |
+
{
|
| 882 |
+
"epoch": 3.56,
|
| 883 |
+
"learning_rate": 2.2222222222222223e-05,
|
| 884 |
+
"loss": 0.0022,
|
| 885 |
+
"step": 1280
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 3.58,
|
| 889 |
+
"learning_rate": 2.0833333333333336e-05,
|
| 890 |
+
"loss": 0.0099,
|
| 891 |
+
"step": 1290
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 3.61,
|
| 895 |
+
"learning_rate": 1.9444444444444445e-05,
|
| 896 |
+
"loss": 0.0006,
|
| 897 |
+
"step": 1300
|
| 898 |
+
},
|
| 899 |
+
{
|
| 900 |
+
"epoch": 3.61,
|
| 901 |
+
"eval_accuracy": 0.9833333333333333,
|
| 902 |
+
"eval_loss": 0.08741980046033859,
|
| 903 |
+
"eval_runtime": 44.9726,
|
| 904 |
+
"eval_samples_per_second": 16.01,
|
| 905 |
+
"eval_steps_per_second": 2.001,
|
| 906 |
+
"step": 1300
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 3.64,
|
| 910 |
+
"learning_rate": 1.8055555555555555e-05,
|
| 911 |
+
"loss": 0.0003,
|
| 912 |
+
"step": 1310
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 3.67,
|
| 916 |
+
"learning_rate": 1.6666666666666667e-05,
|
| 917 |
+
"loss": 0.0003,
|
| 918 |
+
"step": 1320
|
| 919 |
+
},
|
| 920 |
+
{
|
| 921 |
+
"epoch": 3.69,
|
| 922 |
+
"learning_rate": 1.527777777777778e-05,
|
| 923 |
+
"loss": 0.0016,
|
| 924 |
+
"step": 1330
|
| 925 |
+
},
|
| 926 |
+
{
|
| 927 |
+
"epoch": 3.72,
|
| 928 |
+
"learning_rate": 1.388888888888889e-05,
|
| 929 |
+
"loss": 0.001,
|
| 930 |
+
"step": 1340
|
| 931 |
+
},
|
| 932 |
+
{
|
| 933 |
+
"epoch": 3.75,
|
| 934 |
+
"learning_rate": 1.25e-05,
|
| 935 |
+
"loss": 0.0001,
|
| 936 |
+
"step": 1350
|
| 937 |
+
},
|
| 938 |
+
{
|
| 939 |
+
"epoch": 3.78,
|
| 940 |
+
"learning_rate": 1.1111111111111112e-05,
|
| 941 |
+
"loss": 0.0001,
|
| 942 |
+
"step": 1360
|
| 943 |
+
},
|
| 944 |
+
{
|
| 945 |
+
"epoch": 3.81,
|
| 946 |
+
"learning_rate": 9.722222222222223e-06,
|
| 947 |
+
"loss": 0.0005,
|
| 948 |
+
"step": 1370
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 3.83,
|
| 952 |
+
"learning_rate": 8.333333333333334e-06,
|
| 953 |
+
"loss": 0.0001,
|
| 954 |
+
"step": 1380
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 3.86,
|
| 958 |
+
"learning_rate": 6.944444444444445e-06,
|
| 959 |
+
"loss": 0.0002,
|
| 960 |
+
"step": 1390
|
| 961 |
+
},
|
| 962 |
+
{
|
| 963 |
+
"epoch": 3.89,
|
| 964 |
+
"learning_rate": 5.555555555555556e-06,
|
| 965 |
+
"loss": 0.0023,
|
| 966 |
+
"step": 1400
|
| 967 |
+
},
|
| 968 |
+
{
|
| 969 |
+
"epoch": 3.89,
|
| 970 |
+
"eval_accuracy": 0.9777777777777777,
|
| 971 |
+
"eval_loss": 0.10755519568920135,
|
| 972 |
+
"eval_runtime": 51.7521,
|
| 973 |
+
"eval_samples_per_second": 13.912,
|
| 974 |
+
"eval_steps_per_second": 1.739,
|
| 975 |
+
"step": 1400
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 3.92,
|
| 979 |
+
"learning_rate": 4.166666666666667e-06,
|
| 980 |
+
"loss": 0.0266,
|
| 981 |
+
"step": 1410
|
| 982 |
+
},
|
| 983 |
+
{
|
| 984 |
+
"epoch": 3.94,
|
| 985 |
+
"learning_rate": 2.777777777777778e-06,
|
| 986 |
+
"loss": 0.0128,
|
| 987 |
+
"step": 1420
|
| 988 |
+
},
|
| 989 |
+
{
|
| 990 |
+
"epoch": 3.97,
|
| 991 |
+
"learning_rate": 1.388888888888889e-06,
|
| 992 |
+
"loss": 0.0001,
|
| 993 |
+
"step": 1430
|
| 994 |
+
},
|
| 995 |
+
{
|
| 996 |
+
"epoch": 4.0,
|
| 997 |
+
"learning_rate": 0.0,
|
| 998 |
+
"loss": 0.0002,
|
| 999 |
+
"step": 1440
|
| 1000 |
+
},
|
| 1001 |
+
{
|
| 1002 |
+
"epoch": 4.0,
|
| 1003 |
+
"step": 1440,
|
| 1004 |
+
"total_flos": 6.311528820941783e+18,
|
| 1005 |
+
"train_loss": 0.2221095090903772,
|
| 1006 |
+
"train_runtime": 4823.2045,
|
| 1007 |
+
"train_samples_per_second": 4.777,
|
| 1008 |
+
"train_steps_per_second": 0.299
|
| 1009 |
+
},
|
| 1010 |
+
{
|
| 1011 |
+
"epoch": 4.0,
|
| 1012 |
+
"eval_accuracy": 0.9791666666666666,
|
| 1013 |
+
"eval_loss": 0.10265839099884033,
|
| 1014 |
+
"eval_runtime": 48.2725,
|
| 1015 |
+
"eval_samples_per_second": 14.915,
|
| 1016 |
+
"eval_steps_per_second": 1.864,
|
| 1017 |
+
"step": 1440
|
| 1018 |
+
}
|
| 1019 |
+
],
|
| 1020 |
+
"logging_steps": 10,
|
| 1021 |
+
"max_steps": 1440,
|
| 1022 |
+
"num_train_epochs": 4,
|
| 1023 |
+
"save_steps": 100,
|
| 1024 |
+
"total_flos": 6.311528820941783e+18,
|
| 1025 |
+
"trial_name": null,
|
| 1026 |
+
"trial_params": null
|
| 1027 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0c21bcb9e87c2867fade02ad97ba9cd106ffae65bf9bc5e8632f0a7be85ca006
|
| 3 |
+
size 4600
|