test-timm

This model is a fine-tuned version of timm/mobilenetv3_large_100.miil_in21k on the davanstrien/zenodo-presentations-open-labels dataset. It achieves the following results on the evaluation set:

  • Loss: 0.4897
  • Accuracy: 0.7913

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: 128
  • eval_batch_size: 128
  • seed: 1337
  • optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
  • lr_scheduler_type: linear
  • num_epochs: 200.0

Training results

Training Loss Epoch Step Accuracy Validation Loss
0.6794 1.0 23 0.6063 0.6560
0.6215 2.0 46 0.7362 0.5833
0.5784 3.0 69 0.7598 0.5490
0.5347 4.0 92 0.7638 0.5306
0.5307 5.0 115 0.7638 0.5235
0.5391 6.0 138 0.7677 0.5090
0.48 7.0 161 0.7717 0.5108
0.473 8.0 184 0.7756 0.5028
0.5014 9.0 207 0.7717 0.5054
0.496 10.0 230 0.7717 0.5040
0.4688 11.0 253 0.7677 0.4972
0.4943 12.0 276 0.7638 0.4977
0.5012 13.0 299 0.7717 0.5057
0.4639 14.0 322 0.7717 0.5010
0.4709 15.0 345 0.7795 0.4949
0.4888 16.0 368 0.7835 0.4955
0.4594 17.0 391 0.7717 0.4986
0.4745 18.0 414 0.7677 0.5011
0.4667 19.0 437 0.7756 0.4928
0.4551 20.0 460 0.7795 0.5055
0.4657 21.0 483 0.7756 0.4928
0.4818 22.0 506 0.7756 0.5002
0.4633 23.0 529 0.7835 0.4946
0.4779 24.0 552 0.7795 0.4942
0.4718 25.0 575 0.7835 0.4963
0.4511 26.0 598 0.7717 0.5011
0.4798 27.0 621 0.7874 0.4904
0.4868 28.0 644 0.7835 0.4982
0.4653 29.0 667 0.7874 0.4988
0.4613 30.0 690 0.7795 0.4985
0.4675 31.0 713 0.7717 0.5060
0.4587 32.0 736 0.7717 0.5059
0.464 33.0 759 0.7795 0.5042
0.4374 34.0 782 0.7677 0.5063
0.4864 35.0 805 0.7677 0.5040
0.4354 36.0 828 0.7717 0.5109
0.4655 37.0 851 0.7717 0.5107
0.4691 38.0 874 0.7677 0.5093
0.4826 39.0 897 0.7717 0.5044
0.4577 40.0 920 0.7795 0.5000
0.4636 41.0 943 0.7717 0.4963
0.4361 42.0 966 0.7717 0.4958
0.4534 43.0 989 0.7795 0.5008
0.4559 44.0 1012 0.7795 0.5025
0.4189 45.0 1035 0.7756 0.5014
0.4861 46.0 1058 0.7677 0.5004
0.4709 47.0 1081 0.7795 0.5005
0.4726 48.0 1104 0.7717 0.5008
0.4441 49.0 1127 0.7756 0.4988
0.4579 50.0 1150 0.7756 0.5000
0.4366 51.0 1173 0.4980 0.7756
0.4467 52.0 1196 0.4947 0.7795
0.4797 53.0 1219 0.4950 0.7756
0.4544 54.0 1242 0.4998 0.7717
0.4466 55.0 1265 0.4980 0.7795
0.4599 56.0 1288 0.4963 0.7835
0.4458 57.0 1311 0.4956 0.7874
0.4296 58.0 1334 0.4994 0.7874
0.4415 59.0 1357 0.4998 0.7835
0.4036 60.0 1380 0.4996 0.7795
0.4406 61.0 1403 0.5022 0.7913
0.4235 62.0 1426 0.5018 0.7913
0.4492 63.0 1449 0.4964 0.8031
0.4065 64.0 1472 0.4953 0.7874
0.4474 65.0 1495 0.4897 0.7913
0.4605 66.0 1518 0.5039 0.7795
0.436 67.0 1541 0.5024 0.7756
0.4746 68.0 1564 0.5007 0.7874
0.4555 69.0 1587 0.5054 0.7874
0.433 70.0 1610 0.4974 0.7874
0.4503 71.0 1633 0.5096 0.7795
0.4424 72.0 1656 0.5040 0.7756
0.4331 73.0 1679 0.5056 0.7913
0.4263 74.0 1702 0.5026 0.7874
0.4305 75.0 1725 0.5033 0.7835
0.4271 76.0 1748 0.5015 0.7874
0.4635 77.0 1771 0.4988 0.7913
0.4212 78.0 1794 0.4994 0.7913
0.4154 79.0 1817 0.5044 0.7874
0.4288 80.0 1840 0.5033 0.7913
0.4211 81.0 1863 0.5050 0.7835
0.4022 82.0 1886 0.5021 0.7835
0.4477 83.0 1909 0.5096 0.7756
0.4091 84.0 1932 0.5017 0.7913
0.4284 85.0 1955 0.5094 0.7795
0.4317 86.0 1978 0.5056 0.7874
0.4011 87.0 2001 0.4992 0.7953
0.4043 88.0 2024 0.5106 0.7874
0.4233 89.0 2047 0.5083 0.7835
0.4383 90.0 2070 0.5016 0.7913
0.4328 91.0 2093 0.5062 0.7874
0.3978 92.0 2116 0.5026 0.7874
0.4052 93.0 2139 0.4964 0.7913
0.3938 94.0 2162 0.5036 0.7874
0.393 95.0 2185 0.5102 0.7835
0.4294 96.0 2208 0.5003 0.7874
0.4122 97.0 2231 0.5013 0.7913
0.4207 98.0 2254 0.5076 0.7874
0.4127 99.0 2277 0.5040 0.7835
0.441 100.0 2300 0.5022 0.7835
0.3938 101.0 2323 0.4975 0.7992
0.4109 102.0 2346 0.5019 0.7913
0.4299 103.0 2369 0.5060 0.7874
0.4148 104.0 2392 0.5038 0.7874
0.4179 105.0 2415 0.5064 0.7835
0.4352 106.0 2438 0.5059 0.7874
0.4027 107.0 2461 0.5025 0.7953
0.4002 108.0 2484 0.5020 0.7874
0.3988 109.0 2507 0.5063 0.7874
0.4095 110.0 2530 0.5034 0.7913
0.4001 111.0 2553 0.5054 0.7874
0.4201 112.0 2576 0.5076 0.7992
0.4134 113.0 2599 0.5070 0.7953
0.3614 114.0 2622 0.5033 0.7835
0.3928 115.0 2645 0.5043 0.7874
0.435 116.0 2668 0.4999 0.7874
0.4162 117.0 2691 0.5132 0.7874
0.4078 118.0 2714 0.5088 0.7795
0.4025 119.0 2737 0.5075 0.7835
0.4096 120.0 2760 0.5023 0.7835
0.3879 121.0 2783 0.5063 0.7835
0.4033 122.0 2806 0.5001 0.7874
0.3927 123.0 2829 0.5087 0.7795
0.3803 124.0 2852 0.5150 0.7913
0.4248 125.0 2875 0.5150 0.7835
0.3874 126.0 2898 0.5158 0.7874
0.3646 127.0 2921 0.4980 0.8031
0.4115 128.0 2944 0.5077 0.7913
0.385 129.0 2967 0.5153 0.7913
0.4064 130.0 2990 0.5114 0.7953
0.4168 131.0 3013 0.5057 0.7992
0.4319 132.0 3036 0.5041 0.7953
0.4234 133.0 3059 0.5119 0.7992
0.3721 134.0 3082 0.5118 0.7874
0.3709 135.0 3105 0.5078 0.7913
0.4149 136.0 3128 0.5164 0.7795
0.416 137.0 3151 0.5123 0.7835
0.406 138.0 3174 0.5116 0.7913
0.3613 139.0 3197 0.5170 0.7913
0.3786 140.0 3220 0.5099 0.8031
0.3976 141.0 3243 0.5111 0.7913
0.371 142.0 3266 0.5081 0.7953
0.4056 143.0 3289 0.5098 0.7913
0.4214 144.0 3312 0.5085 0.7953
0.3832 145.0 3335 0.5084 0.7953
0.3762 146.0 3358 0.5061 0.7913
0.4118 147.0 3381 0.5111 0.7992
0.3866 148.0 3404 0.5092 0.8071
0.3869 149.0 3427 0.5122 0.7953
0.3734 150.0 3450 0.5117 0.7953
0.4061 151.0 3473 0.5095 0.7913
0.3705 152.0 3496 0.5171 0.7953
0.3873 153.0 3519 0.5179 0.7953
0.3927 154.0 3542 0.5117 0.7992
0.3807 155.0 3565 0.5133 0.7953
0.3761 156.0 3588 0.5140 0.7913
0.3964 157.0 3611 0.5118 0.7953
0.39 158.0 3634 0.5122 0.8031
0.3943 159.0 3657 0.5126 0.8031
0.3417 160.0 3680 0.5097 0.7992
0.3996 161.0 3703 0.5048 0.7913
0.4 162.0 3726 0.5148 0.7953
0.4051 163.0 3749 0.5150 0.7874
0.3973 164.0 3772 0.5037 0.8031
0.3963 165.0 3795 0.5048 0.7953
0.3568 166.0 3818 0.5168 0.7913
0.3995 167.0 3841 0.5096 0.7913
0.3628 168.0 3864 0.5102 0.7953
0.3836 169.0 3887 0.5133 0.7953
0.3646 170.0 3910 0.5099 0.8031
0.3789 171.0 3933 0.5151 0.7874
0.3832 172.0 3956 0.5149 0.8031
0.3476 173.0 3979 0.5178 0.7835
0.3806 174.0 4002 0.5081 0.7992
0.4053 175.0 4025 0.5100 0.7874
0.3986 176.0 4048 0.5189 0.7992
0.3827 177.0 4071 0.5129 0.7992
0.3892 178.0 4094 0.5099 0.7874
0.3955 179.0 4117 0.5212 0.7992
0.4077 180.0 4140 0.5102 0.7953
0.3579 181.0 4163 0.5100 0.7953
0.3666 182.0 4186 0.5248 0.7835
0.3746 183.0 4209 0.5220 0.7874
0.3867 184.0 4232 0.5173 0.7913
0.4024 185.0 4255 0.5248 0.7874
0.4014 186.0 4278 0.5085 0.7913
0.3445 187.0 4301 0.5137 0.8031
0.382 188.0 4324 0.5213 0.7913
0.3673 189.0 4347 0.5242 0.7913
0.3631 190.0 4370 0.5146 0.7913
0.393 191.0 4393 0.5098 0.7835
0.3806 192.0 4416 0.5134 0.7992
0.3789 193.0 4439 0.5127 0.7992
0.3717 194.0 4462 0.5184 0.7913
0.361 195.0 4485 0.5186 0.7835
0.3722 196.0 4508 0.5107 0.7953
0.3551 197.0 4531 0.5175 0.7953
0.3649 198.0 4554 0.5136 0.7992
0.3749 199.0 4577 0.5193 0.7913
0.3782 200.0 4600 0.5182 0.7992

Framework versions

  • Transformers 4.46.0.dev0
  • Pytorch 2.4.1+cu121
  • Datasets 3.0.1
  • Tokenizers 0.20.1
Downloads last month
113
Safetensors
Model size
4.23M params
Tensor type
F32
·
Inference Examples
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.

Model tree for davanstrien/test-timm

Finetuned
(1)
this model