Commit
•
b4a2f9f
1
Parent(s):
fa2503c
Upload model
Browse files- README.md +4 -75
- adapter_config.json +8 -2
- adapter_model.bin +2 -2
README.md
CHANGED
@@ -1,81 +1,10 @@
|
|
1 |
---
|
2 |
-
|
3 |
-
tags:
|
4 |
-
- generated_from_trainer
|
5 |
-
datasets:
|
6 |
-
- food101
|
7 |
-
metrics:
|
8 |
-
- accuracy
|
9 |
-
model-index:
|
10 |
-
- name: vit-base-patch16-224-in21k-finetuned-lora-food101
|
11 |
-
results:
|
12 |
-
- task:
|
13 |
-
name: Image Classification
|
14 |
-
type: image-classification
|
15 |
-
dataset:
|
16 |
-
name: food101
|
17 |
-
type: food101
|
18 |
-
config: default
|
19 |
-
split: train[:5000]
|
20 |
-
args: default
|
21 |
-
metrics:
|
22 |
-
- name: Accuracy
|
23 |
-
type: accuracy
|
24 |
-
value: 0.96
|
25 |
---
|
26 |
-
|
27 |
-
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
28 |
-
should probably proofread and complete it, then remove this comment. -->
|
29 |
-
|
30 |
-
# vit-base-patch16-224-in21k-finetuned-lora-food101
|
31 |
-
|
32 |
-
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 food101 dataset.
|
33 |
-
It achieves the following results on the evaluation set:
|
34 |
-
- Loss: 0.1448
|
35 |
-
- Accuracy: 0.96
|
36 |
-
|
37 |
-
## Model description
|
38 |
-
|
39 |
-
More information needed
|
40 |
-
|
41 |
-
## Intended uses & limitations
|
42 |
-
|
43 |
-
More information needed
|
44 |
-
|
45 |
-
## Training and evaluation data
|
46 |
-
|
47 |
-
More information needed
|
48 |
-
|
49 |
## Training procedure
|
50 |
|
51 |
-
### Training hyperparameters
|
52 |
-
|
53 |
-
The following hyperparameters were used during training:
|
54 |
-
- learning_rate: 0.005
|
55 |
-
- train_batch_size: 128
|
56 |
-
- eval_batch_size: 128
|
57 |
-
- seed: 42
|
58 |
-
- gradient_accumulation_steps: 4
|
59 |
-
- total_train_batch_size: 512
|
60 |
-
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
61 |
-
- lr_scheduler_type: linear
|
62 |
-
- num_epochs: 5
|
63 |
-
- mixed_precision_training: Native AMP
|
64 |
-
|
65 |
-
### Training results
|
66 |
-
|
67 |
-
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
68 |
-
|:-------------:|:-----:|:----:|:---------------:|:--------:|
|
69 |
-
| No log | 1.0 | 9 | 0.5069 | 0.896 |
|
70 |
-
| 2.1627 | 2.0 | 18 | 0.1891 | 0.946 |
|
71 |
-
| 0.3451 | 3.0 | 27 | 0.1448 | 0.96 |
|
72 |
-
| 0.2116 | 4.0 | 36 | 0.1509 | 0.958 |
|
73 |
-
| 0.1711 | 5.0 | 45 | 0.1498 | 0.958 |
|
74 |
-
|
75 |
-
|
76 |
### Framework versions
|
77 |
|
78 |
-
-
|
79 |
-
|
80 |
-
-
|
81 |
-
- Tokenizers 0.13.2
|
|
|
1 |
---
|
2 |
+
library_name: peft
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
4 |
## Training procedure
|
5 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
6 |
### Framework versions
|
7 |
|
8 |
+
- PEFT 0.5.0
|
9 |
+
|
10 |
+
- PEFT 0.5.0
|
|
adapter_config.json
CHANGED
@@ -1,17 +1,23 @@
|
|
1 |
{
|
|
|
|
|
|
|
|
|
2 |
"base_model_name_or_path": "google/vit-base-patch16-224-in21k",
|
3 |
"bias": "none",
|
4 |
-
"enable_lora": null,
|
5 |
"fan_in_fan_out": false,
|
6 |
"inference_mode": true,
|
|
|
|
|
|
|
7 |
"lora_alpha": 16,
|
8 |
"lora_dropout": 0.1,
|
9 |
-
"merge_weights": false,
|
10 |
"modules_to_save": [
|
11 |
"classifier"
|
12 |
],
|
13 |
"peft_type": "LORA",
|
14 |
"r": 16,
|
|
|
15 |
"target_modules": [
|
16 |
"query",
|
17 |
"value"
|
|
|
1 |
{
|
2 |
+
"auto_mapping": {
|
3 |
+
"base_model_class": "ViTForImageClassification",
|
4 |
+
"parent_library": "transformers.models.vit.modeling_vit"
|
5 |
+
},
|
6 |
"base_model_name_or_path": "google/vit-base-patch16-224-in21k",
|
7 |
"bias": "none",
|
|
|
8 |
"fan_in_fan_out": false,
|
9 |
"inference_mode": true,
|
10 |
+
"init_lora_weights": true,
|
11 |
+
"layers_pattern": null,
|
12 |
+
"layers_to_transform": null,
|
13 |
"lora_alpha": 16,
|
14 |
"lora_dropout": 0.1,
|
|
|
15 |
"modules_to_save": [
|
16 |
"classifier"
|
17 |
],
|
18 |
"peft_type": "LORA",
|
19 |
"r": 16,
|
20 |
+
"revision": null,
|
21 |
"target_modules": [
|
22 |
"query",
|
23 |
"value"
|
adapter_model.bin
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
-
size
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:a08052a6661217a6cc6e4f6a39cc27301ba79890406c2f2da3b7a4463e57bda3
|
3 |
+
size 2688708
|