File size: 4,455 Bytes
43a2d09 9cceb0b 43a2d09 9cceb0b 43a2d09 9cceb0b 43a2d09 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 |
---
license: apache-2.0
base_model: google/vit-base-patch32-224-in21k
tags:
- image-classification
- vision
- generated_from_trainer
metrics:
- accuracy
- precision
- recall
- f1
model-index:
- name: vit-base-patch32-224-in21k-finetuned-galaxy10-decals
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# vit-base-patch32-224-in21k-finetuned-galaxy10-decals
This model is a fine-tuned version of [google/vit-base-patch32-224-in21k](https://huggingface.co/google/vit-base-patch32-224-in21k) on the matthieulel/galaxy10_decals dataset.
It achieves the following results on the evaluation set:
- Loss: 0.5180
- Accuracy: 0.8382
- Precision: 0.8363
- Recall: 0.8382
- F1: 0.8346
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 0.0001
- train_batch_size: 32
- eval_batch_size: 32
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 128
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 30
### Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1 |
|:-------------:|:-----:|:----:|:---------------:|:--------:|:---------:|:------:|:------:|
| 1.4731 | 0.99 | 124 | 1.3850 | 0.6110 | 0.5791 | 0.6110 | 0.5797 |
| 0.9858 | 2.0 | 249 | 0.8900 | 0.7508 | 0.7578 | 0.7508 | 0.7337 |
| 0.9475 | 3.0 | 374 | 0.7799 | 0.7599 | 0.7667 | 0.7599 | 0.7559 |
| 0.7778 | 4.0 | 499 | 0.6798 | 0.7779 | 0.7825 | 0.7779 | 0.7729 |
| 0.6831 | 4.99 | 623 | 0.6352 | 0.7914 | 0.7916 | 0.7914 | 0.7889 |
| 0.6953 | 6.0 | 748 | 0.5931 | 0.8044 | 0.8076 | 0.8044 | 0.8023 |
| 0.6725 | 7.0 | 873 | 0.7304 | 0.7537 | 0.7671 | 0.7537 | 0.7519 |
| 0.5648 | 8.0 | 998 | 0.6352 | 0.7909 | 0.7961 | 0.7909 | 0.7868 |
| 0.6127 | 8.99 | 1122 | 0.6087 | 0.7858 | 0.7879 | 0.7858 | 0.7820 |
| 0.529 | 10.0 | 1247 | 0.5827 | 0.8072 | 0.8074 | 0.8072 | 0.8041 |
| 0.5212 | 11.0 | 1372 | 0.5787 | 0.8179 | 0.8177 | 0.8179 | 0.8108 |
| 0.4665 | 12.0 | 1497 | 0.5597 | 0.8168 | 0.8213 | 0.8168 | 0.8134 |
| 0.5123 | 12.99 | 1621 | 0.5840 | 0.8044 | 0.8163 | 0.8044 | 0.8044 |
| 0.4918 | 14.0 | 1746 | 0.5592 | 0.8219 | 0.8221 | 0.8219 | 0.8195 |
| 0.4733 | 15.0 | 1871 | 0.5180 | 0.8382 | 0.8363 | 0.8382 | 0.8346 |
| 0.4552 | 16.0 | 1996 | 0.5673 | 0.8174 | 0.8181 | 0.8174 | 0.8153 |
| 0.4004 | 16.99 | 2120 | 0.5711 | 0.8224 | 0.8239 | 0.8224 | 0.8199 |
| 0.3359 | 18.0 | 2245 | 0.5813 | 0.8168 | 0.8153 | 0.8168 | 0.8147 |
| 0.4069 | 19.0 | 2370 | 0.5482 | 0.8343 | 0.8352 | 0.8343 | 0.8307 |
| 0.3783 | 20.0 | 2495 | 0.5658 | 0.8179 | 0.8169 | 0.8179 | 0.8150 |
| 0.3293 | 20.99 | 2619 | 0.5647 | 0.8247 | 0.8234 | 0.8247 | 0.8230 |
| 0.3214 | 22.0 | 2744 | 0.5654 | 0.8309 | 0.8289 | 0.8309 | 0.8293 |
| 0.3285 | 23.0 | 2869 | 0.5943 | 0.8213 | 0.8226 | 0.8213 | 0.8201 |
| 0.2934 | 24.0 | 2994 | 0.5931 | 0.8264 | 0.8287 | 0.8264 | 0.8259 |
| 0.3051 | 24.99 | 3118 | 0.5788 | 0.8309 | 0.8325 | 0.8309 | 0.8303 |
| 0.2911 | 26.0 | 3243 | 0.5700 | 0.8377 | 0.8354 | 0.8377 | 0.8358 |
| 0.2893 | 27.0 | 3368 | 0.5971 | 0.8286 | 0.8320 | 0.8286 | 0.8291 |
| 0.2794 | 28.0 | 3493 | 0.5908 | 0.8315 | 0.8307 | 0.8315 | 0.8303 |
| 0.2506 | 28.99 | 3617 | 0.5914 | 0.8309 | 0.8314 | 0.8309 | 0.8306 |
| 0.2421 | 29.82 | 3720 | 0.5861 | 0.8365 | 0.8366 | 0.8365 | 0.8359 |
### Framework versions
- Transformers 4.37.2
- Pytorch 2.3.0
- Datasets 2.19.1
- Tokenizers 0.15.1
|