vit-cxr4-384

Coelhomatias/vit-cxr4-384

vit-cxr4-384 is a image classification model from Coelhomatias. Use it when you need a label for an image. It is set up for transformers. The card lists the license as apache-2.0.

should probably proofread and complete it, then remove this comment. --

Downloads · 30 days

2% of all-time downloads

All-time downloads

354

Public

Parameters

86.1M

20.3 GB on disk

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.safetensors344 MB · 100%

At a glance

Task: Image Classification
Library: transformers
License: apache-2.0
Model type: vit
Access: Public
Created: Jan 11, 2024
Updated: Jan 12, 2024
SHA: a98fd09f

Base models

google/vit-base-patch16-384

Model card

From the Hugging Face model README

vit-cxr4-384

This model is a fine-tuned version of google/vit-base-patch16-384 on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 0.2413
Precision: 0.8525
Recall: 0.9419
F1: 0.8950
Accuracy: 0.8926

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: 5e-05
train_batch_size: 24
eval_batch_size: 24
seed: 17
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 2
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
0.3694	0.08	100	0.3851	0.8076	0.8435	0.8251	0.8234
0.3084	0.16	200	0.4478	0.7556	0.9889	0.8566	0.8364
0.3177	0.24	300	0.3073	0.8062	0.9572	0.8752	0.8652
0.3413	0.31	400	0.2936	0.8172	0.9434	0.8758	0.8678
0.2612	0.39	500	0.2936	0.8402	0.9122	0.8747	0.8709
0.3607	0.47	600	0.2717	0.8210	0.9603	0.8852	0.8769
0.274	0.55	700	0.2875	0.8373	0.9196	0.8765	0.8720
0.3127	0.63	800	0.2664	0.8156	0.9683	0.8854	0.8761
0.2875	0.71	900	0.2643	0.8369	0.9334	0.8825	0.8772
0.2652	0.78	1000	0.2659	0.8134	0.9683	0.8841	0.8746
0.2661	0.86	1100	0.2591	0.8334	0.9445	0.8855	0.8793
0.3019	0.94	1200	0.2729	0.8851	0.8599	0.8723	0.8756
0.229	1.02	1300	0.2548	0.8357	0.9603	0.8937	0.8871
0.1841	1.1	1400	0.2438	0.8586	0.9217	0.8891	0.8863
0.2257	1.18	1500	0.2365	0.8629	0.9254	0.8931	0.8905
0.2217	1.25	1600	0.2509	0.8888	0.8662	0.8773	0.8803
0.2619	1.33	1700	0.2588	0.8373	0.9582	0.8937	0.8874
0.2222	1.41	1800	0.2521	0.8644	0.9238	0.8931	0.8908
0.2044	1.49	1900	0.2598	0.8409	0.9588	0.8960	0.8900
0.2238	1.57	2000	0.2641	0.9117	0.8302	0.8691	0.8764
0.249	1.65	2100	0.2368	0.8464	0.9561	0.8979	0.8926
0.1773	1.72	2200	0.2233	0.8682	0.9265	0.8964	0.8942
0.1447	1.8	2300	0.2269	0.8760	0.9191	0.8970	0.8957
0.245	1.88	2400	0.2355	0.8578	0.9445	0.8991	0.8952
0.1685	1.96	2500	0.2312	0.8615	0.9344	0.8965	0.8934

Framework versions

Transformers 4.35.2
Pytorch 2.1.0+cu121
Datasets 2.16.1
Tokenizers 0.15.0

Task: Image Classification
Library: transformers
Type: vit
License: apache-2.0
Created: Jan 11, 2024
Updated: Jan 12, 2024