upload trashify box detection model trained on purely synthetic data
Browse files- README.md +104 -69
- model.safetensors +1 -1
- training_args.bin +1 -1
README.md
CHANGED
|
@@ -15,7 +15,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
This model is a fine-tuned version of [microsoft/conditional-detr-resnet-50](https://huggingface.co/microsoft/conditional-detr-resnet-50) on the None dataset.
|
| 17 |
It achieves the following results on the evaluation set:
|
| 18 |
-
- Loss: 0.
|
| 19 |
|
| 20 |
## Model description
|
| 21 |
|
|
@@ -41,78 +41,113 @@ The following hyperparameters were used during training:
|
|
| 41 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 42 |
- lr_scheduler_type: linear
|
| 43 |
- lr_scheduler_warmup_ratio: 0.05
|
| 44 |
-
- num_epochs:
|
| 45 |
- mixed_precision_training: Native AMP
|
| 46 |
|
| 47 |
### Training results
|
| 48 |
|
| 49 |
-
| Training Loss | Epoch | Step
|
| 50 |
-
|
| 51 |
-
|
|
| 52 |
-
| 2.
|
| 53 |
-
| 2.
|
| 54 |
-
|
|
| 55 |
-
|
|
| 56 |
-
| 1.
|
| 57 |
-
| 1.
|
| 58 |
-
| 1.
|
| 59 |
-
| 1.
|
| 60 |
-
| 1.
|
| 61 |
-
| 1.
|
| 62 |
-
| 1.
|
| 63 |
-
| 1.
|
| 64 |
-
| 1.
|
| 65 |
-
| 1.
|
| 66 |
-
| 1.
|
| 67 |
-
| 1.
|
| 68 |
-
| 1.
|
| 69 |
-
| 1.
|
| 70 |
-
| 1.
|
| 71 |
-
| 1.
|
| 72 |
-
| 1.
|
| 73 |
-
| 1.
|
| 74 |
-
| 1.
|
| 75 |
-
| 1.
|
| 76 |
-
| 1.
|
| 77 |
-
| 1.
|
| 78 |
-
| 1.
|
| 79 |
-
| 1.
|
| 80 |
-
| 1.
|
| 81 |
-
| 1.
|
| 82 |
-
| 1.
|
| 83 |
-
| 1.
|
| 84 |
-
| 1.
|
| 85 |
-
| 1.
|
| 86 |
-
| 1.
|
| 87 |
-
| 1.
|
| 88 |
-
| 1.
|
| 89 |
-
| 1.
|
| 90 |
-
| 1.
|
| 91 |
-
| 1.
|
| 92 |
-
| 1.
|
| 93 |
-
| 1.
|
| 94 |
-
| 1.
|
| 95 |
-
| 1.
|
| 96 |
-
| 1.
|
| 97 |
-
| 1.
|
| 98 |
-
| 1.
|
| 99 |
-
| 1.
|
| 100 |
-
| 1.
|
| 101 |
-
| 1.
|
| 102 |
-
| 1.
|
| 103 |
-
| 1.
|
| 104 |
-
| 1.
|
| 105 |
-
|
|
| 106 |
-
|
|
| 107 |
-
|
|
| 108 |
-
|
|
| 109 |
-
|
|
| 110 |
-
|
|
| 111 |
-
|
|
| 112 |
-
|
|
| 113 |
-
|
|
| 114 |
-
|
|
| 115 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 116 |
|
| 117 |
|
| 118 |
### Framework versions
|
|
|
|
| 15 |
|
| 16 |
This model is a fine-tuned version of [microsoft/conditional-detr-resnet-50](https://huggingface.co/microsoft/conditional-detr-resnet-50) on the None dataset.
|
| 17 |
It achieves the following results on the evaluation set:
|
| 18 |
+
- Loss: 0.8292
|
| 19 |
|
| 20 |
## Model description
|
| 21 |
|
|
|
|
| 41 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 42 |
- lr_scheduler_type: linear
|
| 43 |
- lr_scheduler_warmup_ratio: 0.05
|
| 44 |
+
- num_epochs: 100
|
| 45 |
- mixed_precision_training: Native AMP
|
| 46 |
|
| 47 |
### Training results
|
| 48 |
|
| 49 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 50 |
+
|:-------------:|:-----:|:-----:|:---------------:|
|
| 51 |
+
| 47.9307 | 1.0 | 151 | 3.7270 |
|
| 52 |
+
| 2.9183 | 2.0 | 302 | 2.1431 |
|
| 53 |
+
| 2.0809 | 3.0 | 453 | 1.5949 |
|
| 54 |
+
| 1.6817 | 4.0 | 604 | 1.2977 |
|
| 55 |
+
| 1.5111 | 5.0 | 755 | 1.1754 |
|
| 56 |
+
| 1.4827 | 6.0 | 906 | 1.2093 |
|
| 57 |
+
| 1.4104 | 7.0 | 1057 | 1.1245 |
|
| 58 |
+
| 1.3974 | 8.0 | 1208 | 1.1345 |
|
| 59 |
+
| 1.3833 | 9.0 | 1359 | 1.0921 |
|
| 60 |
+
| 1.3964 | 10.0 | 1510 | 1.1397 |
|
| 61 |
+
| 1.3626 | 11.0 | 1661 | 1.0392 |
|
| 62 |
+
| 1.3321 | 12.0 | 1812 | 1.0339 |
|
| 63 |
+
| 1.3478 | 13.0 | 1963 | 1.0831 |
|
| 64 |
+
| 1.3223 | 14.0 | 2114 | 1.0881 |
|
| 65 |
+
| 1.3128 | 15.0 | 2265 | 1.0432 |
|
| 66 |
+
| 1.3079 | 16.0 | 2416 | 1.0585 |
|
| 67 |
+
| 1.2738 | 17.0 | 2567 | 1.0196 |
|
| 68 |
+
| 1.27 | 18.0 | 2718 | 0.9945 |
|
| 69 |
+
| 1.2396 | 19.0 | 2869 | 0.9707 |
|
| 70 |
+
| 1.2669 | 20.0 | 3020 | 1.0081 |
|
| 71 |
+
| 1.24 | 21.0 | 3171 | 0.9685 |
|
| 72 |
+
| 1.2028 | 22.0 | 3322 | 0.9739 |
|
| 73 |
+
| 1.2123 | 23.0 | 3473 | 0.9891 |
|
| 74 |
+
| 1.212 | 24.0 | 3624 | 0.9930 |
|
| 75 |
+
| 1.1998 | 25.0 | 3775 | 0.9591 |
|
| 76 |
+
| 1.1869 | 26.0 | 3926 | 0.9751 |
|
| 77 |
+
| 1.1942 | 27.0 | 4077 | 0.9734 |
|
| 78 |
+
| 1.1895 | 28.0 | 4228 | 0.9486 |
|
| 79 |
+
| 1.191 | 29.0 | 4379 | 0.9523 |
|
| 80 |
+
| 1.1749 | 30.0 | 4530 | 0.9293 |
|
| 81 |
+
| 1.1729 | 31.0 | 4681 | 0.9333 |
|
| 82 |
+
| 1.1641 | 32.0 | 4832 | 0.9146 |
|
| 83 |
+
| 1.1402 | 33.0 | 4983 | 0.9348 |
|
| 84 |
+
| 1.156 | 34.0 | 5134 | 0.9283 |
|
| 85 |
+
| 1.1562 | 35.0 | 5285 | 0.9428 |
|
| 86 |
+
| 1.1273 | 36.0 | 5436 | 0.9155 |
|
| 87 |
+
| 1.1258 | 37.0 | 5587 | 0.9424 |
|
| 88 |
+
| 1.1467 | 38.0 | 5738 | 0.9435 |
|
| 89 |
+
| 1.1284 | 39.0 | 5889 | 0.9089 |
|
| 90 |
+
| 1.1113 | 40.0 | 6040 | 0.9038 |
|
| 91 |
+
| 1.1025 | 41.0 | 6191 | 0.8817 |
|
| 92 |
+
| 1.0797 | 42.0 | 6342 | 0.8926 |
|
| 93 |
+
| 1.0803 | 43.0 | 6493 | 0.8805 |
|
| 94 |
+
| 1.0867 | 44.0 | 6644 | 0.8845 |
|
| 95 |
+
| 1.0646 | 45.0 | 6795 | 0.8696 |
|
| 96 |
+
| 1.0766 | 46.0 | 6946 | 0.8768 |
|
| 97 |
+
| 1.0653 | 47.0 | 7097 | 0.8643 |
|
| 98 |
+
| 1.0615 | 48.0 | 7248 | 0.8796 |
|
| 99 |
+
| 1.0514 | 49.0 | 7399 | 0.8712 |
|
| 100 |
+
| 1.0423 | 50.0 | 7550 | 0.8654 |
|
| 101 |
+
| 1.0291 | 51.0 | 7701 | 0.8345 |
|
| 102 |
+
| 1.046 | 52.0 | 7852 | 0.8521 |
|
| 103 |
+
| 1.0095 | 53.0 | 8003 | 0.8407 |
|
| 104 |
+
| 1.028 | 54.0 | 8154 | 0.8673 |
|
| 105 |
+
| 0.9932 | 55.0 | 8305 | 0.8398 |
|
| 106 |
+
| 0.9966 | 56.0 | 8456 | 0.8441 |
|
| 107 |
+
| 0.9961 | 57.0 | 8607 | 0.8362 |
|
| 108 |
+
| 0.9902 | 58.0 | 8758 | 0.8379 |
|
| 109 |
+
| 0.9802 | 59.0 | 8909 | 0.8337 |
|
| 110 |
+
| 0.9806 | 60.0 | 9060 | 0.8284 |
|
| 111 |
+
| 0.9671 | 61.0 | 9211 | 0.8259 |
|
| 112 |
+
| 0.9627 | 62.0 | 9362 | 0.8297 |
|
| 113 |
+
| 0.9535 | 63.0 | 9513 | 0.8434 |
|
| 114 |
+
| 0.953 | 64.0 | 9664 | 0.8236 |
|
| 115 |
+
| 0.9444 | 65.0 | 9815 | 0.8190 |
|
| 116 |
+
| 0.9329 | 66.0 | 9966 | 0.8189 |
|
| 117 |
+
| 0.9461 | 67.0 | 10117 | 0.8255 |
|
| 118 |
+
| 0.9235 | 68.0 | 10268 | 0.8205 |
|
| 119 |
+
| 0.9328 | 69.0 | 10419 | 0.8311 |
|
| 120 |
+
| 0.9128 | 70.0 | 10570 | 0.8234 |
|
| 121 |
+
| 0.9007 | 71.0 | 10721 | 0.8207 |
|
| 122 |
+
| 0.9064 | 72.0 | 10872 | 0.8331 |
|
| 123 |
+
| 0.8921 | 73.0 | 11023 | 0.8198 |
|
| 124 |
+
| 0.8883 | 74.0 | 11174 | 0.8236 |
|
| 125 |
+
| 0.8919 | 75.0 | 11325 | 0.8233 |
|
| 126 |
+
| 0.89 | 76.0 | 11476 | 0.8205 |
|
| 127 |
+
| 0.8815 | 77.0 | 11627 | 0.8196 |
|
| 128 |
+
| 0.8949 | 78.0 | 11778 | 0.8284 |
|
| 129 |
+
| 0.8816 | 79.0 | 11929 | 0.8186 |
|
| 130 |
+
| 0.8641 | 80.0 | 12080 | 0.8247 |
|
| 131 |
+
| 0.8697 | 81.0 | 12231 | 0.8133 |
|
| 132 |
+
| 0.8641 | 82.0 | 12382 | 0.8247 |
|
| 133 |
+
| 0.8437 | 83.0 | 12533 | 0.8293 |
|
| 134 |
+
| 0.8503 | 84.0 | 12684 | 0.8229 |
|
| 135 |
+
| 0.8446 | 85.0 | 12835 | 0.8158 |
|
| 136 |
+
| 0.835 | 86.0 | 12986 | 0.8099 |
|
| 137 |
+
| 0.8328 | 87.0 | 13137 | 0.8143 |
|
| 138 |
+
| 0.8332 | 88.0 | 13288 | 0.8273 |
|
| 139 |
+
| 0.8316 | 89.0 | 13439 | 0.8336 |
|
| 140 |
+
| 0.8223 | 90.0 | 13590 | 0.8273 |
|
| 141 |
+
| 0.8097 | 91.0 | 13741 | 0.8208 |
|
| 142 |
+
| 0.8063 | 92.0 | 13892 | 0.8248 |
|
| 143 |
+
| 0.7967 | 93.0 | 14043 | 0.8263 |
|
| 144 |
+
| 0.7916 | 94.0 | 14194 | 0.8302 |
|
| 145 |
+
| 0.7918 | 95.0 | 14345 | 0.8303 |
|
| 146 |
+
| 0.7935 | 96.0 | 14496 | 0.8321 |
|
| 147 |
+
| 0.8028 | 97.0 | 14647 | 0.8300 |
|
| 148 |
+
| 0.7961 | 98.0 | 14798 | 0.8301 |
|
| 149 |
+
| 0.7773 | 99.0 | 14949 | 0.8285 |
|
| 150 |
+
| 0.7741 | 100.0 | 15100 | 0.8292 |
|
| 151 |
|
| 152 |
|
| 153 |
### Framework versions
|
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 174080824
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5254c6fef814500a0b1dbb10c956981185477e72856db40aaa84cfe4f53bb83a
|
| 3 |
size 174080824
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 5240
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4b072b1e2341c502ff15e905bd41cceb74d6eb2926326e6507c19c73bfd9c3d4
|
| 3 |
size 5240
|