swin-tiny-patch4-window7-224-finetuned-ADC-3cls-0922

This model is a fine-tuned version of microsoft/swin-tiny-patch4-window7-224 on the imagefolder dataset. It achieves the following results on the evaluation set:

• Loss: 0.6771
• Accuracy: 0.8286

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: 64
• eval_batch_size: 64
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 256
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.2
• num_epochs: 200

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	1.0	2	0.6875	0.8143
No log	2.0	4	0.6874	0.8143
No log	3.0	6	0.6873	0.8143
No log	4.0	8	0.6871	0.8143
0.7555	5.0	10	0.6869	0.8143
0.7555	6.0	12	0.6866	0.8143
0.7555	7.0	14	0.6862	0.8143
0.7555	8.0	16	0.6858	0.8143
0.7555	9.0	18	0.6853	0.8143
0.7576	10.0	20	0.6848	0.8143
0.7576	11.0	22	0.6842	0.8143
0.7576	12.0	24	0.6836	0.8143
0.7576	13.0	26	0.6830	0.8143
0.7576	14.0	28	0.6823	0.8143
0.769	15.0	30	0.6816	0.8
0.769	16.0	32	0.6808	0.8
0.769	17.0	34	0.6800	0.8143
0.769	18.0	36	0.6791	0.8143
0.769	19.0	38	0.6781	0.8143
0.7564	20.0	40	0.6771	0.8286
0.7564	21.0	42	0.6760	0.8143
0.7564	22.0	44	0.6748	0.8143
0.7564	23.0	46	0.6737	0.8
0.7564	24.0	48	0.6725	0.8
0.7508	25.0	50	0.6713	0.8143
0.7508	26.0	52	0.6701	0.8143
0.7508	27.0	54	0.6689	0.8143
0.7508	28.0	56	0.6674	0.8143
0.7508	29.0	58	0.6660	0.8143
0.747	30.0	60	0.6646	0.8143
0.747	31.0	62	0.6631	0.8143
0.747	32.0	64	0.6616	0.8143
0.747	33.0	66	0.6601	0.8143
0.747	34.0	68	0.6586	0.8143
0.7343	35.0	70	0.6570	0.8143
0.7343	36.0	72	0.6553	0.8143
0.7343	37.0	74	0.6536	0.8143
0.7343	38.0	76	0.6517	0.8143
0.7343	39.0	78	0.6499	0.8143
0.7532	40.0	80	0.6480	0.8143
0.7532	41.0	82	0.6461	0.8143
0.7532	42.0	84	0.6442	0.8143
0.7532	43.0	86	0.6423	0.8143
0.7532	44.0	88	0.6405	0.8143
0.7239	45.0	90	0.6387	0.8143
0.7239	46.0	92	0.6368	0.8143
0.7239	47.0	94	0.6352	0.8143
0.7239	48.0	96	0.6337	0.8143
0.7239	49.0	98	0.6321	0.8286
0.7085	50.0	100	0.6307	0.8286
0.7085	51.0	102	0.6294	0.8286
0.7085	52.0	104	0.6278	0.8286
0.7085	53.0	106	0.6263	0.8286
0.7085	54.0	108	0.6248	0.8143
0.7203	55.0	110	0.6233	0.8143
0.7203	56.0	112	0.6219	0.8143
0.7203	57.0	114	0.6205	0.8143
0.7203	58.0	116	0.6191	0.8143
0.7203	59.0	118	0.6179	0.8143
0.7136	60.0	120	0.6167	0.8143
0.7136	61.0	122	0.6157	0.8143
0.7136	62.0	124	0.6148	0.8
0.7136	63.0	126	0.6138	0.8
0.7136	64.0	128	0.6125	0.8
0.7123	65.0	130	0.6111	0.8
0.7123	66.0	132	0.6096	0.8143
0.7123	67.0	134	0.6083	0.8143
0.7123	68.0	136	0.6070	0.8143
0.7123	69.0	138	0.6057	0.8143
0.7076	70.0	140	0.6046	0.8143
0.7076	71.0	142	0.6035	0.8143
0.7076	72.0	144	0.6023	0.8143
0.7076	73.0	146	0.6011	0.8143
0.7076	74.0	148	0.5999	0.8143
0.6878	75.0	150	0.5988	0.8143
0.6878	76.0	152	0.5975	0.8143
0.6878	77.0	154	0.5964	0.8143
0.6878	78.0	156	0.5953	0.8143
0.6878	79.0	158	0.5942	0.8143
0.6657	80.0	160	0.5932	0.8143
0.6657	81.0	162	0.5923	0.8143
0.6657	82.0	164	0.5914	0.8143
0.6657	83.0	166	0.5906	0.8143
0.6657	84.0	168	0.5897	0.8143
0.6434	85.0	170	0.5888	0.8143
0.6434	86.0	172	0.5878	0.8143
0.6434	87.0	174	0.5868	0.8143
0.6434	88.0	176	0.5859	0.8143
0.6434	89.0	178	0.5851	0.8143
0.6825	90.0	180	0.5843	0.8143
0.6825	91.0	182	0.5836	0.8143
0.6825	92.0	184	0.5828	0.8143
0.6825	93.0	186	0.5823	0.8143
0.6825	94.0	188	0.5817	0.8286
0.6695	95.0	190	0.5809	0.8143
0.6695	96.0	192	0.5801	0.8143
0.6695	97.0	194	0.5793	0.8143
0.6695	98.0	196	0.5787	0.8143
0.6695	99.0	198	0.5780	0.8143
0.6672	100.0	200	0.5772	0.8143
0.6672	101.0	202	0.5762	0.8143
0.6672	102.0	204	0.5754	0.8143
0.6672	103.0	206	0.5746	0.8143
0.6672	104.0	208	0.5738	0.8143
0.6569	105.0	210	0.5731	0.8143
0.6569	106.0	212	0.5724	0.8143
0.6569	107.0	214	0.5716	0.8143
0.6569	108.0	216	0.5708	0.8143
0.6569	109.0	218	0.5701	0.8143
0.6748	110.0	220	0.5694	0.8143
0.6748	111.0	222	0.5687	0.8143
0.6748	112.0	224	0.5680	0.8143
0.6748	113.0	226	0.5674	0.8143
0.6748	114.0	228	0.5668	0.8143
0.6388	115.0	230	0.5662	0.8143
0.6388	116.0	232	0.5657	0.8143
0.6388	117.0	234	0.5652	0.8143
0.6388	118.0	236	0.5648	0.8286
0.6388	119.0	238	0.5645	0.8286
0.6551	120.0	240	0.5641	0.8286
0.6551	121.0	242	0.5636	0.8143
0.6551	122.0	244	0.5631	0.8143
0.6551	123.0	246	0.5627	0.8143
0.6551	124.0	248	0.5624	0.8143
0.6452	125.0	250	0.5622	0.8143
0.6452	126.0	252	0.5620	0.8143
0.6452	127.0	254	0.5618	0.8143
0.6452	128.0	256	0.5615	0.8143
0.6452	129.0	258	0.5613	0.8143
0.645	130.0	260	0.5611	0.8143
0.645	131.0	262	0.5608	0.8143
0.645	132.0	264	0.5606	0.8143
0.645	133.0	266	0.5602	0.8143
0.645	134.0	268	0.5596	0.8143
0.629	135.0	270	0.5590	0.8143
0.629	136.0	272	0.5582	0.8143
0.629	137.0	274	0.5576	0.8143
0.629	138.0	276	0.5571	0.8143
0.629	139.0	278	0.5568	0.8143
0.7126	140.0	280	0.5565	0.8143
0.7126	141.0	282	0.5563	0.8143
0.7126	142.0	284	0.5561	0.8143
0.7126	143.0	286	0.5559	0.8143
0.7126	144.0	288	0.5555	0.8143
0.669	145.0	290	0.5552	0.8143
0.669	146.0	292	0.5547	0.8143
0.669	147.0	294	0.5542	0.8143
0.669	148.0	296	0.5538	0.8143
0.669	149.0	298	0.5534	0.8143
0.6481	150.0	300	0.5530	0.8143
0.6481	151.0	302	0.5526	0.8143
0.6481	152.0	304	0.5522	0.8143
0.6481	153.0	306	0.5519	0.8143
0.6481	154.0	308	0.5515	0.8143
0.6211	155.0	310	0.5510	0.8143
0.6211	156.0	312	0.5506	0.8143
0.6211	157.0	314	0.5502	0.8143
0.6211	158.0	316	0.5499	0.8143
0.6211	159.0	318	0.5496	0.8143
0.6458	160.0	320	0.5492	0.8286
0.6458	161.0	322	0.5490	0.8143
0.6458	162.0	324	0.5488	0.8143
0.6458	163.0	326	0.5486	0.8143
0.6458	164.0	328	0.5484	0.8143
0.6317	165.0	330	0.5481	0.8143
0.6317	166.0	332	0.5479	0.8286
0.6317	167.0	334	0.5476	0.8286
0.6317	168.0	336	0.5473	0.8286
0.6317	169.0	338	0.5471	0.8286
0.6154	170.0	340	0.5470	0.8286
0.6154	171.0	342	0.5468	0.8286
0.6154	172.0	344	0.5466	0.8286
0.6154	173.0	346	0.5464	0.8286
0.6154	174.0	348	0.5462	0.8286
0.6323	175.0	350	0.5460	0.8286
0.6323	176.0	352	0.5459	0.8286
0.6323	177.0	354	0.5457	0.8286
0.6323	178.0	356	0.5456	0.8286
0.6323	179.0	358	0.5455	0.8286
0.6331	180.0	360	0.5453	0.8286
0.6331	181.0	362	0.5452	0.8286
0.6331	182.0	364	0.5451	0.8286
0.6331	183.0	366	0.5449	0.8286
0.6331	184.0	368	0.5448	0.8286
0.6333	185.0	370	0.5447	0.8286
0.6333	186.0	372	0.5447	0.8286
0.6333	187.0	374	0.5446	0.8286
0.6333	188.0	376	0.5445	0.8286
0.6333	189.0	378	0.5445	0.8286
0.608	190.0	380	0.5444	0.8286
0.608	191.0	382	0.5444	0.8286
0.608	192.0	384	0.5443	0.8286
0.608	193.0	386	0.5443	0.8286
0.608	194.0	388	0.5442	0.8286
0.6155	195.0	390	0.5442	0.8286
0.6155	196.0	392	0.5442	0.8286
0.6155	197.0	394	0.5442	0.8286
0.6155	198.0	396	0.5441	0.8286
0.6155	199.0	398	0.5441	0.8286
0.6272	200.0	400	0.5441	0.8286

Framework versions

• Transformers 4.33.2
• Pytorch 2.0.1+cu118
• Datasets 2.14.5
• Tokenizers 0.13.3