vit-base-patch16-224-in21k-finetuned-breast-composition

BTX24/vit-base-patch16-224-in21k-finetuned-breast-composition

vit-base-patch16-224-in21k-finetuned-breast-composition is a machine learning model from BTX24. Use it for the machine learning task on the model card, and read the license before you ship it in a product. It is set up for peft. The card lists the license as apache-2.0.

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

Downloads · 30 days

11% of all-time downloads

All-time downloads

Public

Repo size

16.7 GB

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.safetensors147 MB · 100%

At a glance

Library: peft
License: apache-2.0
Access: Public
Created: Oct 28, 2024
Updated: Oct 29, 2024
SHA: af6519ab

Base models

google/vit-base-patch16-224-in21k

Model card

From the Hugging Face model README

vit-base-patch16-224-in21k-finetuned-breast-composition

This model is a fine-tuned version of google/vit-base-patch16-224-in21k on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 0.4735
Accuracy: 0.8057
F1: 0.8046
Precision: 0.8053
Recall: 0.8057

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: 2e-05
train_batch_size: 16
eval_batch_size: 8
seed: 42
gradient_accumulation_steps: 4
total_train_batch_size: 64
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: cosine_with_restarts
lr_scheduler_warmup_ratio: 0.1
num_epochs: 12
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1	Precision	Recall
1.3733	0.1062	100	1.3716	0.3608	0.2980	0.3783	0.3608
1.3575	0.2125	200	1.3537	0.4026	0.3205	0.3988	0.4026
1.3262	0.3187	300	1.3186	0.4256	0.3125	0.4184	0.4256
1.2714	0.4250	400	1.2666	0.4562	0.3528	0.6333	0.4562
1.212	0.5312	500	1.2172	0.5161	0.4304	0.6528	0.5161
1.1533	0.6375	600	1.1610	0.5655	0.4831	0.6747	0.5655
1.0462	0.7437	700	1.0505	0.6136	0.5278	0.7080	0.6136
0.9731	0.8499	800	0.9478	0.6327	0.5451	0.7219	0.6327
0.9101	0.9562	900	0.8763	0.6438	0.5551	0.7303	0.6438
0.8524	1.0624	1000	0.8397	0.6431	0.5549	0.7130	0.6431
0.8274	1.1687	1100	0.7888	0.6571	0.5795	0.7088	0.6571
0.7976	1.2749	1200	0.7741	0.6578	0.5901	0.7111	0.6578
0.7717	1.3811	1300	0.7198	0.6857	0.6385	0.7203	0.6857
0.7298	1.4874	1400	0.6948	0.7015	0.6731	0.7208	0.7015
0.7192	1.5936	1500	0.6792	0.7124	0.6873	0.7297	0.7124
0.7003	1.6999	1600	0.6539	0.7266	0.7109	0.7354	0.7266
0.69	1.8061	1700	0.6349	0.7422	0.7326	0.7444	0.7422
0.6759	1.9124	1800	0.6228	0.7478	0.7404	0.7505	0.7478
0.6689	2.0186	1900	0.6154	0.7474	0.7429	0.7473	0.7474
0.6505	2.1248	2000	0.6064	0.7557	0.7542	0.7573	0.7557
0.6602	2.2311	2100	0.5888	0.7646	0.7615	0.7657	0.7646
0.6498	2.3373	2200	0.5815	0.7653	0.7622	0.7643	0.7653
0.6484	2.4436	2300	0.5692	0.7716	0.7684	0.7715	0.7716
0.6366	2.5498	2400	0.5698	0.7695	0.7670	0.7703	0.7695
0.6304	2.6560	2500	0.5651	0.7705	0.7668	0.7702	0.7705
0.5906	2.7623	2600	0.5663	0.7691	0.7628	0.7719	0.7691
0.607	2.8685	2700	0.5563	0.7733	0.7699	0.7737	0.7733
0.6211	2.9748	2800	0.5513	0.7745	0.7709	0.7753	0.7745
0.6043	3.0810	2900	0.5478	0.7760	0.7735	0.7761	0.7760
0.5905	3.1873	3000	0.5439	0.7778	0.7765	0.7771	0.7778
0.6309	3.2935	3100	0.5399	0.7774	0.7749	0.7770	0.7774
0.5975	3.3997	3200	0.5431	0.7792	0.7778	0.7786	0.7792
0.6175	3.5060	3300	0.5340	0.7813	0.7779	0.7829	0.7813
0.588	3.6122	3400	0.5293	0.7844	0.7817	0.7850	0.7844
0.5981	3.7185	3500	0.5265	0.7849	0.7832	0.7844	0.7849
0.5926	3.8247	3600	0.5335	0.7784	0.7732	0.7807	0.7784
0.5945	3.9309	3700	0.5249	0.7853	0.7821	0.7866	0.7853
0.5956	4.0372	3800	0.5325	0.7829	0.7788	0.7845	0.7829
0.5943	4.1434	3900	0.5295	0.7860	0.7837	0.7860	0.7860
0.5842	4.2497	4000	0.5227	0.7848	0.7813	0.7857	0.7848
0.5666	4.3559	4100	0.5187	0.7864	0.7833	0.7881	0.7864
0.5762	4.4622	4200	0.5179	0.7876	0.7859	0.7889	0.7876
0.595	4.5684	4300	0.5111	0.7909	0.7898	0.7902	0.7909
0.5641	4.6746	4400	0.5151	0.7888	0.7874	0.7890	0.7888
0.5743	4.7809	4500	0.5113	0.7894	0.7883	0.7907	0.7894
0.564	4.8871	4600	0.5075	0.7919	0.7902	0.7933	0.7919
0.578	4.9934	4700	0.5029	0.7921	0.7905	0.7918	0.7921
0.5643	5.0996	4800	0.5042	0.7931	0.7909	0.7945	0.7931
0.5611	5.2058	4900	0.5012	0.7940	0.7909	0.7959	0.7940
0.5736	5.3121	5000	0.5133	0.7864	0.7812	0.7919	0.7864
0.5635	5.4183	5100	0.5034	0.7947	0.7939	0.7950	0.7947
0.5653	5.5246	5200	0.4981	0.7966	0.7944	0.7982	0.7966
0.5664	5.6308	5300	0.4959	0.7951	0.7932	0.7950	0.7951
0.5689	5.7371	5400	0.4946	0.7972	0.7956	0.7969	0.7972
0.5394	5.8433	5500	0.5022	0.7928	0.7877	0.7979	0.7928
0.5645	5.9495	5600	0.4931	0.7965	0.7944	0.7974	0.7965
0.5588	6.0558	5700	0.4895	0.7990	0.7975	0.7990	0.7990
0.5539	6.1620	5800	0.4874	0.8008	0.7992	0.8010	0.8008
0.5504	6.2683	5900	0.4945	0.7970	0.7941	0.7996	0.7970
0.5683	6.3745	6000	0.4883	0.7985	0.7969	0.7983	0.7985
0.5594	6.4807	6100	0.4883	0.7985	0.7976	0.7980	0.7985
0.5709	6.5870	6200	0.4883	0.7976	0.7959	0.7983	0.7976
0.553	6.6932	6300	0.4907	0.7954	0.7931	0.7964	0.7954
0.5515	6.7995	6400	0.4893	0.7971	0.7945	0.7982	0.7971
0.5501	6.9057	6500	0.4821	0.7982	0.7969	0.7978	0.7982
0.5567	7.0120	6600	0.4851	0.7985	0.7955	0.8005	0.7985
0.5387	7.1182	6700	0.4808	0.8012	0.7995	0.8014	0.8012
0.5257	7.2244	6800	0.4795	0.8020	0.8008	0.8018	0.8020
0.5591	7.3307	6900	0.4809	0.8010	0.7980	0.8030	0.8010
0.5513	7.4369	7000	0.4747	0.8047	0.8033	0.8048	0.8047
0.5636	7.5432	7100	0.4760	0.8031	0.8018	0.8031	0.8031
0.5491	7.6494	7200	0.4787	0.8004	0.7981	0.8014	0.8004
0.5504	7.7556	7300	0.4745	0.8041	0.8026	0.8043	0.8041
0.5393	7.8619	7400	0.4829	0.7983	0.7950	0.8006	0.7983
0.543	7.9681	7500	0.4744	0.8047	0.8036	0.8046	0.8047
0.5308	8.0744	7600	0.4786	0.8018	0.7999	0.8022	0.8018
0.5341	8.1806	7700	0.4740	0.8037	0.8024	0.8036	0.8037
0.5351	8.2869	7800	0.4762	0.8035	0.8015	0.8043	0.8035
0.5309	8.3931	7900	0.4754	0.8053	0.8042	0.8051	0.8053
0.5409	8.4993	8000	0.4761	0.8036	0.8014	0.8049	0.8036
0.5368	8.6056	8100	0.4722	0.8052	0.8038	0.8051	0.8052
0.5276	8.7118	8200	0.4735	0.8057	0.8046	0.8053	0.8057
0.5388	8.8181	8300	0.4766	0.8020	0.8002	0.8022	0.8020
0.5368	8.9243	8400	0.4742	0.8037	0.8021	0.8036	0.8037
0.5574	9.0305	8500	0.4762	0.8034	0.8011	0.8047	0.8034
0.5562	9.1368	8600	0.4735	0.8034	0.8011	0.8046	0.8034
0.5577	9.2430	8700	0.4735	0.8048	0.8042	0.8048	0.8048
0.5398	9.3493	8800	0.4698	0.8056	0.8041	0.8058	0.8056
0.5224	9.4555	8900	0.4734	0.8040	0.8025	0.8041	0.8040
0.5392	9.5618	9000	0.4713	0.8051	0.8038	0.8051	0.8051
0.5346	9.6680	9100	0.4706	0.8048	0.8037	0.8044	0.8048
0.5295	9.7742	9200	0.4713	0.8041	0.8026	0.8040	0.8041
0.5607	9.8805	9300	0.4689	0.8051	0.8039	0.8050	0.8051
0.5354	9.9867	9400	0.4692	0.8057	0.8042	0.8058	0.8057
0.5427	10.0930	9500	0.4678	0.8056	0.8043	0.8053	0.8056
0.5216	10.1992	9600	0.4711	0.8036	0.8020	0.8036	0.8036
0.5348	10.3054	9700	0.4696	0.8054	0.8040	0.8053	0.8054
0.5319	10.4117	9800	0.4710	0.8047	0.8029	0.8049	0.8047
0.5465	10.5179	9900	0.4691	0.8049	0.8035	0.8048	0.8049
0.5387	10.6242	10000	0.4699	0.8042	0.8023	0.8046	0.8042
0.5431	10.7304	10100	0.4697	0.8045	0.8030	0.8044	0.8045
0.5358	10.8367	10200	0.4698	0.8043	0.8025	0.8045	0.8043
0.5506	10.9429	10300	0.4693	0.8047	0.8032	0.8045	0.8047
0.5049	11.0491	10400	0.4686	0.8051	0.8037	0.8050	0.8051
0.5338	11.1554	10500	0.4685	0.8058	0.8044	0.8056	0.8058
0.5411	11.2616	10600	0.4681	0.8056	0.8043	0.8054	0.8056
0.5414	11.3679	10700	0.4684	0.8053	0.8039	0.8050	0.8053
0.5438	11.4741	10800	0.4689	0.8053	0.8038	0.8052	0.8053
0.5271	11.5803	10900	0.4688	0.8050	0.8035	0.8049	0.8050
0.5292	11.6866	11000	0.4688	0.8049	0.8034	0.8049	0.8049
0.5318	11.7928	11100	0.4688	0.8046	0.8031	0.8045	0.8046
0.54	11.8991	11200	0.4688	0.8047	0.8032	0.8046	0.8047

Framework versions

PEFT 0.13.3.dev0
Transformers 4.44.2
Pytorch 2.5.0+cu121
Datasets 3.0.2
Tokenizers 0.19.1

image/png

Library: peft
License: apache-2.0
Created: Oct 28, 2024
Updated: Oct 29, 2024