detr-resnet-50_finetuned_plant_disease_detection_processed

This model is a fine-tuned version of facebook/detr-resnet-50 on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 0.6402

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: 1e-05
train_batch_size: 8
eval_batch_size: 16
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_ratio: 0.2
num_epochs: 15
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
5.2965	0.19	50	4.4784
4.7649	0.38	100	4.3439
4.4907	0.57	150	4.0077
4.3973	0.76	200	3.2143
3.4084	0.95	250	2.6818
2.7091	1.14	300	2.3603
2.4601	1.33	350	1.9004
2.1096	1.52	400	1.5639
1.6941	1.7	450	1.3240
1.4949	1.89	500	1.1247
1.2246	2.08	550	1.0421
1.4479	2.27	600	1.1546
1.1327	2.46	650	1.1098
1.1184	2.65	700	0.8950
1.0516	2.84	750	0.8601
1.2556	3.03	800	0.8575
1.1216	3.22	850	0.8314
1.1027	3.41	900	1.0676
1.0815	3.6	950	0.9716
1.2254	3.79	1000	1.0091
0.9896	3.98	1050	0.7600
1.0736	4.17	1100	0.8907
1.2462	4.36	1150	0.7506
0.9959	4.55	1200	0.7623
1.0895	4.73	1250	0.7570
1.0736	4.92	1300	0.8248
1.1015	5.11	1350	0.8682
1.1423	5.3	1400	0.8340
1.0906	5.49	1450	0.8372
0.9333	5.68	1500	0.8420
1.1347	5.87	1550	0.8718
0.9407	6.06	1600	0.8270
0.8138	6.25	1650	0.8241
0.8731	6.44	1700	0.8013
1.0146	6.63	1750	0.7704
0.8847	6.82	1800	0.8885
1.0283	7.01	1850	0.8804
1.0359	7.2	1900	0.7907
0.987	7.39	1950	0.7997
1.0279	7.58	2000	0.9095
0.9027	7.77	2050	0.6823
0.927	7.95	2100	0.6728
1.0499	8.14	2150	0.6537
0.9774	8.33	2200	0.6455
0.9171	8.52	2250	0.6456
1.0002	8.71	2300	0.6723
0.9052	8.9	2350	0.6554
0.9029	9.09	2400	0.7272
1.0247	9.28	2450	0.6997
0.8296	9.47	2500	0.6661
1.0659	9.66	2550	0.7914
1.0226	9.85	2600	0.7823
0.9419	10.04	2650	0.7709
0.9008	10.23	2700	0.8114
0.826	10.42	2750	0.7042
0.7957	10.61	2800	0.7764
1.0086	10.8	2850	0.8362
1.0076	10.98	2900	0.8048
0.9613	11.17	2950	0.6945
0.9155	11.36	3000	0.7011
0.9436	11.55	3050	0.6524
0.9134	11.74	3100	0.6582
0.817	11.93	3150	0.6678
0.8545	12.12	3200	0.6520
0.9801	12.31	3250	0.7813
0.8566	12.5	3300	0.7205
0.8966	12.69	3350	0.6326
0.8705	12.88	3400	0.6577
0.8193	13.07	3450	0.6391
0.8099	13.26	3500	0.6658
0.921	13.45	3550	0.6535
0.7915	13.64	3600	0.6576
1.1439	13.83	3650	0.6593
0.8702	14.02	3700	0.6519
0.73	14.2	3750	0.6403
0.8306	14.39	3800	0.6393
0.8678	14.58	3850	0.6405
1.0003	14.77	3900	0.6407
1.023	14.96	3950	0.6402

Framework versions

Transformers 4.36.0.dev0
Pytorch 2.0.1+cu117
Datasets 2.14.5
Tokenizers 0.15.0

Loading…

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

learning_rate: 1e-05

train_batch_size: 8

eval_batch_size: 16

seed: 42

optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08

lr_scheduler_type: linear

lr_scheduler_warmup_ratio: 0.2

num_epochs: 15

mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
5.2965	0.19	50	4.4784
4.7649	0.38	100	4.3439
4.4907	0.57	150	4.0077
4.3973	0.76	200	3.2143
3.4084	0.95	250	2.6818
2.7091	1.14	300	2.3603
2.4601	1.33	350	1.9004
2.1096	1.52	400	1.5639
1.6941	1.7	450	1.3240
1.4949	1.89	500	1.1247
1.2246	2.08	550	1.0421
1.4479	2.27	600	1.1546
1.1327	2.46	650	1.1098
1.1184	2.65	700	0.8950
1.0516	2.84	750	0.8601
1.2556	3.03	800	0.8575
1.1216	3.22	850	0.8314
1.1027	3.41	900	1.0676
1.0815	3.6	950	0.9716
1.2254	3.79	1000	1.0091
0.9896	3.98	1050	0.7600
1.0736	4.17	1100	0.8907
1.2462	4.36	1150	0.7506
0.9959	4.55	1200	0.7623
1.0895	4.73	1250	0.7570
1.0736	4.92	1300	0.8248
1.1015	5.11	1350	0.8682
1.1423	5.3	1400	0.8340
1.0906	5.49	1450	0.8372
0.9333	5.68	1500	0.8420
1.1347	5.87	1550	0.8718
0.9407	6.06	1600	0.8270
0.8138	6.25	1650	0.8241
0.8731	6.44	1700	0.8013
1.0146	6.63	1750	0.7704
0.8847	6.82	1800	0.8885
1.0283	7.01	1850	0.8804
1.0359	7.2	1900	0.7907
0.987	7.39	1950	0.7997
1.0279	7.58	2000	0.9095
0.9027	7.77	2050	0.6823
0.927	7.95	2100	0.6728
1.0499	8.14	2150	0.6537
0.9774	8.33	2200	0.6455
0.9171	8.52	2250	0.6456
1.0002	8.71	2300	0.6723
0.9052	8.9	2350	0.6554
0.9029	9.09	2400	0.7272
1.0247	9.28	2450	0.6997
0.8296	9.47	2500	0.6661
1.0659	9.66	2550	0.7914
1.0226	9.85	2600	0.7823
0.9419	10.04	2650	0.7709
0.9008	10.23	2700	0.8114
0.826	10.42	2750	0.7042
0.7957	10.61	2800	0.7764
1.0086	10.8	2850	0.8362
1.0076	10.98	2900	0.8048
0.9613	11.17	2950	0.6945
0.9155	11.36	3000	0.7011
0.9436	11.55	3050	0.6524
0.9134	11.74	3100	0.6582
0.817	11.93	3150	0.6678
0.8545	12.12	3200	0.6520
0.9801	12.31	3250	0.7813
0.8566	12.5	3300	0.7205
0.8966	12.69	3350	0.6326
0.8705	12.88	3400	0.6577
0.8193	13.07	3450	0.6391
0.8099	13.26	3500	0.6658
0.921	13.45	3550	0.6535
0.7915	13.64	3600	0.6576
1.1439	13.83	3650	0.6593
0.8702	14.02	3700	0.6519
0.73	14.2	3750	0.6403
0.8306	14.39	3800	0.6393
0.8678	14.58	3850	0.6405
1.0003	14.77	3900	0.6407
1.023	14.96	3950	0.6402

Framework versions

Transformers 4.36.0.dev0

Pytorch 2.0.1+cu117

Datasets 2.14.5

Tokenizers 0.15.0

susnato/detr-resnet-50_finetuned_plant_disease_detection_processed

detr-resnet-50_finetuned_plant_disease_detection_processed

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

detr-resnet-50_finetuned_plant_disease_detection_processed

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions