End of training
Browse files
README.md
CHANGED
|
@@ -15,38 +15,6 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
# detr-resnet-50-dc5-fashionpedia-finetuned
|
| 16 |
|
| 17 |
This model is a fine-tuned version of [facebook/detr-resnet-50-dc5](https://huggingface.co/facebook/detr-resnet-50-dc5) on the None dataset.
|
| 18 |
-
It achieves the following results on the evaluation set:
|
| 19 |
-
- Loss: 2.5675
|
| 20 |
-
- Map: 0.0275
|
| 21 |
-
- Map 50: 0.0634
|
| 22 |
-
- Map 75: 0.0182
|
| 23 |
-
- Map Small: 0.0
|
| 24 |
-
- Map Medium: 0.032
|
| 25 |
-
- Map Large: 0.0752
|
| 26 |
-
- Mar 1: 0.0195
|
| 27 |
-
- Mar 10: 0.0979
|
| 28 |
-
- Mar 100: 0.1196
|
| 29 |
-
- Mar Small: 0.0
|
| 30 |
-
- Mar Medium: 0.1307
|
| 31 |
-
- Mar Large: 0.3012
|
| 32 |
-
- Map Aortic enlargement: 0.1155
|
| 33 |
-
- Mar 100 Aortic enlargement: 0.4321
|
| 34 |
-
- Map Cardiomegaly: 0.132
|
| 35 |
-
- Mar 100 Cardiomegaly: 0.5943
|
| 36 |
-
- Map Ild: 0.0
|
| 37 |
-
- Mar 100 Ild: 0.0
|
| 38 |
-
- Map Lung opacity: 0.0001
|
| 39 |
-
- Mar 100 Lung opacity: 0.05
|
| 40 |
-
- Map Nodule/mass: 0.0
|
| 41 |
-
- Mar 100 Nodule/mass: 0.0
|
| 42 |
-
- Map Other lesion: 0.0
|
| 43 |
-
- Mar 100 Other lesion: 0.0
|
| 44 |
-
- Map Pleural effusion: 0.0
|
| 45 |
-
- Mar 100 Pleural effusion: 0.0
|
| 46 |
-
- Map Pleural thickening: 0.0
|
| 47 |
-
- Mar 100 Pleural thickening: 0.0
|
| 48 |
-
- Map Pulmonary fibrosis: 0.0
|
| 49 |
-
- Mar 100 Pulmonary fibrosis: 0.0
|
| 50 |
|
| 51 |
## Model description
|
| 52 |
|
|
@@ -66,38 +34,16 @@ More information needed
|
|
| 66 |
|
| 67 |
The following hyperparameters were used during training:
|
| 68 |
- learning_rate: 1e-05
|
| 69 |
-
- train_batch_size:
|
| 70 |
-
- eval_batch_size:
|
| 71 |
- seed: 42
|
| 72 |
- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
|
| 73 |
- lr_scheduler_type: linear
|
| 74 |
-
- training_steps:
|
| 75 |
- mixed_precision_training: Native AMP
|
| 76 |
|
| 77 |
### Training results
|
| 78 |
|
| 79 |
-
| Training Loss | Epoch | Step | Validation Loss | Map | Map 50 | Map 75 | Map Small | Map Medium | Map Large | Mar 1 | Mar 10 | Mar 100 | Mar Small | Mar Medium | Mar Large | Map Aortic enlargement | Mar 100 Aortic enlargement | Map Atelectasis | Mar 100 Atelectasis | Map Cardiomegaly | Mar 100 Cardiomegaly | Map Ild | Mar 100 Ild | Map Infiltration | Mar 100 Infiltration | Map Lung opacity | Mar 100 Lung opacity | Map Nodule/mass | Mar 100 Nodule/mass | Map Other lesion | Mar 100 Other lesion | Map Pleural effusion | Mar 100 Pleural effusion | Map Pleural thickening | Mar 100 Pleural thickening | Map Pulmonary fibrosis | Mar 100 Pulmonary fibrosis |
|
| 80 |
-
|:-------------:|:-------:|:----:|:---------------:|:------:|:------:|:------:|:---------:|:----------:|:---------:|:------:|:------:|:-------:|:---------:|:----------:|:---------:|:----------------------:|:--------------------------:|:---------------:|:-------------------:|:----------------:|:--------------------:|:-------:|:-----------:|:----------------:|:--------------------:|:----------------:|:--------------------:|:---------------:|:-------------------:|:----------------:|:--------------------:|:--------------------:|:------------------------:|:----------------------:|:--------------------------:|:----------------------:|:--------------------------:|
|
| 81 |
-
| No log | 2.7778 | 50 | 3.9076 | 0.0006 | 0.0028 | 0.0002 | 0.0 | 0.0003 | 0.005 | 0.0 | 0.0057 | 0.021 | 0.0 | 0.0143 | 0.0845 | 0.0006 | 0.0415 | -1.0 | -1.0 | 0.0047 | 0.1472 | 0.0 | 0.0 | -1.0 | -1.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 |
|
| 82 |
-
| 4.3075 | 5.5556 | 100 | 3.5001 | 0.0047 | 0.0127 | 0.0028 | 0.0 | 0.0033 | 0.0271 | 0.0 | 0.0317 | 0.0849 | 0.0 | 0.0774 | 0.2357 | 0.0223 | 0.3377 | -1.0 | -1.0 | 0.0199 | 0.4264 | 0.0 | 0.0 | -1.0 | -1.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 |
|
| 83 |
-
| 4.3075 | 8.3333 | 150 | 3.2511 | 0.0079 | 0.0197 | 0.0044 | 0.0 | 0.005 | 0.0459 | 0.0029 | 0.0465 | 0.099 | 0.0 | 0.0962 | 0.3083 | 0.0296 | 0.3057 | 0.0411 | 0.5849 | 0.0 | 0.0 | -1.0 | -1.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 |
|
| 84 |
-
| 3.7002 | 11.1111 | 200 | 3.0549 | 0.0084 | 0.0247 | 0.0047 | 0.0 | 0.0073 | 0.0297 | 0.0055 | 0.053 | 0.096 | 0.0 | 0.1022 | 0.2488 | 0.0317 | 0.3811 | 0.0439 | 0.483 | 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 | 0.0 |
|
| 85 |
-
| 3.7002 | 13.8889 | 250 | 2.8268 | 0.0121 | 0.0276 | 0.0078 | 0.0 | 0.0073 | 0.052 | 0.0101 | 0.0591 | 0.1115 | 0.0 | 0.1193 | 0.3083 | 0.0356 | 0.4 | 0.0733 | 0.6038 | 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 | 0.0 |
|
| 86 |
-
| 3.3402 | 16.6667 | 300 | 2.8859 | 0.016 | 0.0383 | 0.0123 | 0.0 | 0.0303 | 0.0413 | 0.0073 | 0.0857 | 0.1061 | 0.0 | 0.1212 | 0.2476 | 0.0591 | 0.4604 | 0.0846 | 0.4943 | 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 | 0.0 |
|
| 87 |
-
| 3.3402 | 19.4444 | 350 | 2.8537 | 0.0166 | 0.0363 | 0.0124 | 0.0 | 0.0342 | 0.0553 | 0.0084 | 0.0748 | 0.1 | 0.0 | 0.1113 | 0.2393 | 0.0381 | 0.4264 | 0.1117 | 0.4736 | 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 | 0.0 |
|
| 88 |
-
| 3.2237 | 22.2222 | 400 | 2.7335 | 0.0177 | 0.0419 | 0.0131 | 0.0 | 0.0277 | 0.0508 | 0.0096 | 0.0925 | 0.1107 | 0.0 | 0.1143 | 0.2833 | 0.0629 | 0.4509 | 0.0963 | 0.5453 | 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 | 0.0 |
|
| 89 |
-
| 3.2237 | 25.0 | 450 | 2.7128 | 0.0209 | 0.0499 | 0.0155 | 0.0 | 0.0236 | 0.0642 | 0.0157 | 0.0922 | 0.1145 | 0.0 | 0.1176 | 0.2964 | 0.0741 | 0.4679 | 0.114 | 0.5623 | 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 | 0.0 |
|
| 90 |
-
| 3.0321 | 27.7778 | 500 | 2.6574 | 0.0197 | 0.0433 | 0.0137 | 0.0 | 0.035 | 0.0498 | 0.0143 | 0.0973 | 0.1143 | 0.0 | 0.1192 | 0.2881 | 0.082 | 0.4717 | 0.095 | 0.5566 | 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 | 0.0 |
|
| 91 |
-
| 3.0321 | 30.5556 | 550 | 2.6041 | 0.0206 | 0.0504 | 0.0158 | 0.0 | 0.0375 | 0.0616 | 0.0184 | 0.0939 | 0.1202 | 0.0 | 0.1204 | 0.3036 | 0.0693 | 0.4585 | 0.1157 | 0.5736 | 0.0 | 0.0 | 0.0001 | 0.05 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 |
|
| 92 |
-
| 2.9455 | 33.3333 | 600 | 2.6164 | 0.0252 | 0.0548 | 0.0214 | 0.0 | 0.0413 | 0.0575 | 0.0212 | 0.0979 | 0.1174 | 0.0 | 0.1286 | 0.3083 | 0.1141 | 0.4472 | 0.1131 | 0.6094 | 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 | 0.0 |
|
| 93 |
-
| 2.9455 | 36.1111 | 650 | 2.5827 | 0.023 | 0.0516 | 0.0172 | 0.0 | 0.0321 | 0.0486 | 0.0191 | 0.0948 | 0.1212 | 0.0 | 0.1322 | 0.3119 | 0.1166 | 0.4755 | 0.0908 | 0.6151 | 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 | 0.0 |
|
| 94 |
-
| 2.8715 | 38.8889 | 700 | 2.5590 | 0.028 | 0.0587 | 0.0232 | 0.0 | 0.0333 | 0.0607 | 0.0247 | 0.0992 | 0.1413 | 0.0 | 0.1453 | 0.3143 | 0.1387 | 0.4717 | 0.1123 | 0.6 | 0.0 | 0.0 | 0.0006 | 0.2 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 |
|
| 95 |
-
| 2.8715 | 41.6667 | 750 | 2.5453 | 0.0264 | 0.0579 | 0.0192 | 0.0 | 0.0273 | 0.0678 | 0.0243 | 0.0975 | 0.1193 | 0.0 | 0.1275 | 0.3155 | 0.1166 | 0.4566 | 0.121 | 0.617 | 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 | 0.0 |
|
| 96 |
-
| 2.7836 | 44.4444 | 800 | 2.5490 | 0.0249 | 0.0559 | 0.0173 | 0.0 | 0.0291 | 0.0634 | 0.0226 | 0.0992 | 0.1166 | 0.0 | 0.1223 | 0.3083 | 0.1078 | 0.4509 | 0.1163 | 0.5981 | 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 | 0.0 |
|
| 97 |
-
| 2.7836 | 47.2222 | 850 | 2.5600 | 0.0256 | 0.0593 | 0.0187 | 0.0 | 0.034 | 0.0628 | 0.0189 | 0.0962 | 0.1195 | 0.0 | 0.1292 | 0.3083 | 0.1153 | 0.4698 | 0.1152 | 0.6057 | 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 | 0.0 |
|
| 98 |
-
| 2.7591 | 50.0 | 900 | 2.5640 | 0.0258 | 0.0574 | 0.0185 | 0.0 | 0.0336 | 0.0636 | 0.0205 | 0.1039 | 0.1219 | 0.0 | 0.1319 | 0.3036 | 0.1159 | 0.4509 | 0.1162 | 0.5962 | 0.0 | 0.0 | 0.0001 | 0.05 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 |
|
| 99 |
-
| 2.7591 | 52.7778 | 950 | 2.5807 | 0.0272 | 0.0621 | 0.0175 | 0.0 | 0.0308 | 0.0777 | 0.0203 | 0.0981 | 0.1149 | 0.0 | 0.1252 | 0.3036 | 0.109 | 0.4377 | 0.1361 | 0.5962 | 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 | 0.0 |
|
| 100 |
-
| 2.7396 | 55.5556 | 1000 | 2.5675 | 0.0275 | 0.0634 | 0.0182 | 0.0 | 0.032 | 0.0752 | 0.0195 | 0.0979 | 0.1196 | 0.0 | 0.1307 | 0.3012 | 0.1155 | 0.4321 | 0.132 | 0.5943 | 0.0 | 0.0 | 0.0001 | 0.05 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 |
|
| 101 |
|
| 102 |
|
| 103 |
### Framework versions
|
|
|
|
| 15 |
# detr-resnet-50-dc5-fashionpedia-finetuned
|
| 16 |
|
| 17 |
This model is a fine-tuned version of [facebook/detr-resnet-50-dc5](https://huggingface.co/facebook/detr-resnet-50-dc5) on the None dataset.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 18 |
|
| 19 |
## Model description
|
| 20 |
|
|
|
|
| 34 |
|
| 35 |
The following hyperparameters were used during training:
|
| 36 |
- learning_rate: 1e-05
|
| 37 |
+
- train_batch_size: 2
|
| 38 |
+
- eval_batch_size: 2
|
| 39 |
- seed: 42
|
| 40 |
- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
|
| 41 |
- lr_scheduler_type: linear
|
| 42 |
+
- training_steps: 3
|
| 43 |
- mixed_precision_training: Native AMP
|
| 44 |
|
| 45 |
### Training results
|
| 46 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 47 |
|
| 48 |
|
| 49 |
### Framework versions
|