segformer-b0-drone-real

This model is a fine-tuned version of nvidia/mit-b0 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 2.5066
• Mean Iou: 0.0277
• Mean Accuracy: 0.0561
• Overall Accuracy: 0.3272
• Per Category Iou: [nan, 0.3359579661339679, 0.0, 0.20517716653585355, 0.003702068923287315, 0.0, 0.0, 0.0, 0.006312134342948017, 0.058787282412541476, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, nan]
• Per Category Accuracy: [nan, 0.6093986043791881, 0.0, 0.4105641817321649, 0.003921086446309355, 0.0, 0.0, 0.0, 0.00706877745310955, 0.20348775859354865, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, nan]

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: 6e-05
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Mean Iou	Mean Accuracy	Overall Accuracy	Per Category Iou	Per Category Accuracy
2.8285	2.0	40	2.8542	0.0229	0.0536	0.2211	[nan, 0.23060686675928652, 3.3649639948852546e-05, 0.16273692278074856, 0.013325250843685254, 4.063223761732559e-05, 0.0, 0.0, 0.035811860573766785, 0.06198642655834834, 0.0, 0.0, 6.128953174797744e-05, 0.0, 0.0, 0.0, 5.1148278860416345e-05, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, nan]	[nan, 0.33385338422824606, 3.373169528473065e-05, 0.299629413956809, 0.014924989967585504, 4.09158863797309e-05, 0.0, 0.0, 0.06156816279101685, 0.46948081490592114, 0.0, 0.0, 7.101768340316739e-05, 0.0, 0.0, 0.0, 5.163155720776539e-05, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, nan]
2.6567	4.0	80	2.6894	0.0262	0.0563	0.2866	[nan, 0.30624516924744655, 0.0, 0.18452743224435408, 0.004815305953306897, 0.0, 0.0, 0.0, 0.013916000286575587, 0.06596235864669749, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, nan]	[nan, 0.49410241551371004, 0.0, 0.3789267923602906, 0.005102681931249924, 0.0, 0.0, 0.0, 0.016502745882809654, 0.3439400151190408, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, nan]
2.3286	6.0	120	2.5923	0.0267	0.0560	0.3328	[nan, 0.35729772739055154, 0.0, 0.1539404924138176, 0.0035934817901112264, 0.0, 0.0, 0.0, 0.005766957703680072, 0.06586525335295575, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, nan]	[nan, 0.7280486819078453, 0.0, 0.23220288304265976, 0.003859608350752127, 0.0, 0.0, 0.0, 0.006166880921555261, 0.2614486947285752, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, nan]
2.3836	8.0	160	2.5890	0.0275	0.0575	0.3244	[nan, 0.349601471481378, 0.0, 0.17496917223449057, 0.0044570694472161245, 0.0, 0.0, 0.0, 0.007437757319012064, 0.0682080200846761, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, nan]	[nan, 0.6614103864255038, 0.0, 0.29170107802854334, 0.004828395043379214, 0.0, 0.0, 0.0, 0.008074588150255963, 0.29956113264795986, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, nan]
2.2849	10.0	200	2.5525	0.0283	0.0569	0.3388	[nan, 0.3351423574874955, 0.0, 0.21293393590619594, 0.002350142484175345, 0.0, 0.0, 0.0, 0.005453996034797477, 0.06749403988523864, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, nan]	[nan, 0.6106816065491506, 0.0, 0.4703984171366385, 0.002485471577527932, 0.0, 0.0, 0.0, 0.0058035081160884615, 0.16273986773434082, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, nan]
2.1539	12.0	240	2.5352	0.0289	0.0579	0.3431	[nan, 0.34790380657207765, 0.0, 0.21813897765519252, 0.0039179488370322925, 0.0, 0.0, 0.0, 0.004069615148127275, 0.06184727957512016, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, nan]	[nan, 0.6213138456628569, 0.0, 0.4695635471053968, 0.004103493982578053, 0.0, 0.0, 0.0, 0.004355898828122703, 0.17384734612397268, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, nan]
2.1894	14.0	280	2.5923	0.0271	0.0567	0.3056	[nan, 0.3019300415296601, 0.0, 0.2108717486241356, 0.003421017543191299, 0.0, 0.0, 0.0, 0.012217783066499338, 0.06721066611554333, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, nan]	[nan, 0.4912292097745947, 0.0, 0.49141307925385636, 0.0036400437237620272, 0.0, 0.0, 0.0, 0.01400423107111977, 0.2475275870599001, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, nan]
1.9499	16.0	320	2.5169	0.0282	0.0569	0.3320	[nan, 0.33589786653498954, 0.0, 0.20789931305652776, 0.004207056605876958, 0.0, 0.0, 0.0, 0.0075080990386678535, 0.06424320004285503, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, nan]	[nan, 0.613616488775929, 0.0, 0.4253884604222434, 0.0044872253922099824, 0.0, 0.0, 0.0, 0.008369991941750555, 0.19915876549744932, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, nan]
2.115	18.0	360	2.5153	0.0277	0.0562	0.3256	[nan, 0.3342799309383031, 0.0, 0.2066227292961174, 0.0034518271833141106, 0.0, 0.0, 0.0, 0.006799912048482982, 0.05915124911379064, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, nan]	[nan, 0.5983876453859164, 0.0, 0.4204451510267332, 0.0036636891451301916, 0.0, 0.0, 0.0, 0.007619718595211191, 0.20718275529815128, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, nan]
2.0119	20.0	400	2.5066	0.0277	0.0561	0.3272	[nan, 0.3359579661339679, 0.0, 0.20517716653585355, 0.003702068923287315, 0.0, 0.0, 0.0, 0.006312134342948017, 0.058787282412541476, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, nan]	[nan, 0.6093986043791881, 0.0, 0.4105641817321649, 0.003921086446309355, 0.0, 0.0, 0.0, 0.00706877745310955, 0.20348775859354865, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, nan]

Framework versions

• Transformers 4.25.1
• Pytorch 1.12.1+cu113
• Datasets 2.7.1
• Tokenizers 0.13.2