segformer-b0-finetuned-segments-sidewalk-2

This model is a fine-tuned version of nvidia/mit-b0 on the segments/sidewalk-semantic dataset. It achieves the following results on the evaluation set:

• Loss: 2.6306
• Mean Iou: 0.1027
• Mean Accuracy: 0.1574
• Overall Accuracy: 0.6552
• Per Category Iou: [0.0, 0.40932069741697885, 0.6666047315185674, 0.0015527279135260222, 0.000557997451181134, 0.004734463745284192, 0.0, 0.00024311836753505628, 0.0, 0.0, 0.5448608416905849, 0.005644290758731727, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.4689142754019952, 0.0, 0.00039031599380590526, 0.010175747938072128, 0.0, 0.0, 0.0, 0.0008842445754996234, 0.0, 0.0, 0.6689560919488968, 0.10178439680971307, 0.7089823411348399, 0.0, 0.0, 0.0, 0.0]
• Per Category Accuracy: [nan, 0.6798160901382586, 0.8601972223213155, 0.001563543652833044, 0.0005586801134972854, 0.004789605465686377, nan, 0.00024743825184288725, 0.0, 0.0, 0.8407289173400536, 0.012641370267169317, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.7574833533176979, 0.0, 0.00039110009377117975, 0.013959849889225483, 0.0, nan, 0.0, 0.0009309900323061499, 0.0, 0.0, 0.9337304207449932, 0.12865528611713883, 0.8019892660736478, 0.0, 0.0, 0.0, 0.0]

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: 1

Training results

Training Loss	Epoch	Step	Validation Loss	Mean Iou	Mean Accuracy	Overall Accuracy	Per Category Iou	Per Category Accuracy
2.8872	0.5	20	3.1018	0.0995	0.1523	0.6415	[0.0, 0.3982872425364927, 0.6582689116809847, 0.0, 0.00044314555867048773, 0.019651883205738383, 0.0, 0.0006528617866575068, 0.0, 0.0, 0.4861235900758522, 0.003961411405960721, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.4437814560942763, 0.0, 1.1600860783870164e-06, 0.019965880301918204, 0.0, 0.0, 0.0, 0.0074026601990928, 0.0, 0.0, 0.666238976894996, 0.13012673492067245, 0.6486315429686865, 0.0, 2.0656177918545805e-05, 0.0001944735843164534, 0.0]	[nan, 0.6263716501798601, 0.8841421548179447, 0.0, 0.00044410334445801165, 0.020659891877382746, nan, 0.0006731258604635891, 0.0, 0.0, 0.8403154629142631, 0.017886412063596133, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.6324385775164868, 0.0, 1.160534402881839e-06, 0.06036834410935781, 0.0, nan, 0.0, 0.010232933175604348, 0.0, 0.0, 0.9320173945724101, 0.15828224740687694, 0.6884182010535304, 0.0, 2.3169780427714147e-05, 0.00019505205451704924, 0.0]
2.6167	1.0	40	2.6306	0.1027	0.1574	0.6552	[0.0, 0.40932069741697885, 0.6666047315185674, 0.0015527279135260222, 0.000557997451181134, 0.004734463745284192, 0.0, 0.00024311836753505628, 0.0, 0.0, 0.5448608416905849, 0.005644290758731727, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.4689142754019952, 0.0, 0.00039031599380590526, 0.010175747938072128, 0.0, 0.0, 0.0, 0.0008842445754996234, 0.0, 0.0, 0.6689560919488968, 0.10178439680971307, 0.7089823411348399, 0.0, 0.0, 0.0, 0.0]	[nan, 0.6798160901382586, 0.8601972223213155, 0.001563543652833044, 0.0005586801134972854, 0.004789605465686377, nan, 0.00024743825184288725, 0.0, 0.0, 0.8407289173400536, 0.012641370267169317, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.7574833533176979, 0.0, 0.00039110009377117975, 0.013959849889225483, 0.0, nan, 0.0, 0.0009309900323061499, 0.0, 0.0, 0.9337304207449932, 0.12865528611713883, 0.8019892660736478, 0.0, 0.0, 0.0, 0.0]

Framework versions

• Transformers 4.20.1
• Pytorch 1.11.0
• Datasets 2.1.0
• Tokenizers 0.12.1