Model save
Browse files
README.md
ADDED
|
@@ -0,0 +1,116 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: other
|
| 3 |
+
base_model: facebook/mask2former-swin-tiny-coco-instance
|
| 4 |
+
tags:
|
| 5 |
+
- generated_from_trainer
|
| 6 |
+
model-index:
|
| 7 |
+
- name: finetune-instance-segmentation-ade20k-mini-mask2former-v1
|
| 8 |
+
results: []
|
| 9 |
+
---
|
| 10 |
+
|
| 11 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 12 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 13 |
+
|
| 14 |
+
[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="200" height="32"/>](https://wandb.ai/qubvel-hf-co/huggingface/runs/tpf8jz7h)
|
| 15 |
+
# finetune-instance-segmentation-ade20k-mini-mask2former-v1
|
| 16 |
+
|
| 17 |
+
This model is a fine-tuned version of [facebook/mask2former-swin-tiny-coco-instance](https://huggingface.co/facebook/mask2former-swin-tiny-coco-instance) on an unknown dataset.
|
| 18 |
+
It achieves the following results on the evaluation set:
|
| 19 |
+
- Loss: 27.5494
|
| 20 |
+
- Map: 0.2315
|
| 21 |
+
- Map 50: 0.4495
|
| 22 |
+
- Map 75: 0.2185
|
| 23 |
+
- Map Small: 0.1535
|
| 24 |
+
- Map Medium: 0.6606
|
| 25 |
+
- Map Large: 0.8161
|
| 26 |
+
- Mar 1: 0.0981
|
| 27 |
+
- Mar 10: 0.2576
|
| 28 |
+
- Mar 100: 0.3
|
| 29 |
+
- Mar Small: 0.2272
|
| 30 |
+
- Mar Medium: 0.7189
|
| 31 |
+
- Mar Large: 0.8618
|
| 32 |
+
- Map Person: 0.1626
|
| 33 |
+
- Mar 100 Person: 0.2224
|
| 34 |
+
- Map Car: 0.3003
|
| 35 |
+
- Mar 100 Car: 0.3776
|
| 36 |
+
|
| 37 |
+
## Model description
|
| 38 |
+
|
| 39 |
+
More information needed
|
| 40 |
+
|
| 41 |
+
## Intended uses & limitations
|
| 42 |
+
|
| 43 |
+
More information needed
|
| 44 |
+
|
| 45 |
+
## Training and evaluation data
|
| 46 |
+
|
| 47 |
+
More information needed
|
| 48 |
+
|
| 49 |
+
## Training procedure
|
| 50 |
+
|
| 51 |
+
### Training hyperparameters
|
| 52 |
+
|
| 53 |
+
The following hyperparameters were used during training:
|
| 54 |
+
- learning_rate: 1e-05
|
| 55 |
+
- train_batch_size: 8
|
| 56 |
+
- eval_batch_size: 8
|
| 57 |
+
- seed: 42
|
| 58 |
+
- gradient_accumulation_steps: 2
|
| 59 |
+
- total_train_batch_size: 16
|
| 60 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 61 |
+
- lr_scheduler_type: constant
|
| 62 |
+
- num_epochs: 40.0
|
| 63 |
+
- mixed_precision_training: Native AMP
|
| 64 |
+
|
| 65 |
+
### Training results
|
| 66 |
+
|
| 67 |
+
| 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 Person | Mar 100 Person | Map Car | Mar 100 Car |
|
| 68 |
+
|:-------------:|:-----:|:----:|:---------------:|:------:|:------:|:------:|:---------:|:----------:|:---------:|:------:|:------:|:-------:|:---------:|:----------:|:---------:|:----------:|:--------------:|:-------:|:-----------:|
|
| 69 |
+
| 36.7831 | 1.0 | 100 | 33.2768 | 0.1838 | 0.3677 | 0.174 | 0.1175 | 0.6012 | 0.7974 | 0.0884 | 0.2431 | 0.284 | 0.2104 | 0.7053 | 0.8712 | 0.1175 | 0.2014 | 0.25 | 0.3665 |
|
| 70 |
+
| 30.2324 | 2.0 | 200 | 30.8268 | 0.198 | 0.4007 | 0.1831 | 0.1321 | 0.6183 | 0.8028 | 0.0916 | 0.25 | 0.2885 | 0.2151 | 0.7125 | 0.8354 | 0.1331 | 0.2079 | 0.263 | 0.3691 |
|
| 71 |
+
| 28.4136 | 3.0 | 300 | 29.8261 | 0.2036 | 0.416 | 0.1849 | 0.1337 | 0.6332 | 0.7969 | 0.0934 | 0.2472 | 0.2905 | 0.2169 | 0.7162 | 0.8323 | 0.1381 | 0.2112 | 0.269 | 0.3697 |
|
| 72 |
+
| 27.5659 | 4.0 | 400 | 29.2926 | 0.2101 | 0.4176 | 0.1918 | 0.1371 | 0.6352 | 0.8051 | 0.094 | 0.25 | 0.2884 | 0.2143 | 0.7174 | 0.8354 | 0.1456 | 0.2107 | 0.2745 | 0.3661 |
|
| 73 |
+
| 26.9971 | 5.0 | 500 | 28.8044 | 0.213 | 0.4209 | 0.2016 | 0.1379 | 0.6419 | 0.8094 | 0.093 | 0.2499 | 0.2894 | 0.2148 | 0.7207 | 0.8441 | 0.1475 | 0.2096 | 0.2785 | 0.3692 |
|
| 74 |
+
| 26.42 | 6.0 | 600 | 28.4848 | 0.2196 | 0.4224 | 0.2062 | 0.1426 | 0.647 | 0.8046 | 0.0944 | 0.2523 | 0.2925 | 0.2188 | 0.7196 | 0.8354 | 0.15 | 0.2106 | 0.2892 | 0.3745 |
|
| 75 |
+
| 25.9065 | 7.0 | 700 | 28.2601 | 0.2212 | 0.4261 | 0.207 | 0.1444 | 0.6442 | 0.8049 | 0.0943 | 0.2527 | 0.2902 | 0.2176 | 0.7103 | 0.8323 | 0.153 | 0.2102 | 0.2893 | 0.3703 |
|
| 76 |
+
| 25.6766 | 8.0 | 800 | 28.2581 | 0.2209 | 0.4276 | 0.2076 | 0.1434 | 0.6485 | 0.8201 | 0.0943 | 0.2532 | 0.294 | 0.2197 | 0.7212 | 0.8681 | 0.1532 | 0.2122 | 0.2885 | 0.3758 |
|
| 77 |
+
| 25.3111 | 9.0 | 900 | 27.8623 | 0.2234 | 0.4318 | 0.2163 | 0.1451 | 0.649 | 0.8252 | 0.0951 | 0.2519 | 0.2953 | 0.2212 | 0.721 | 0.8649 | 0.1561 | 0.2148 | 0.2907 | 0.3757 |
|
| 78 |
+
| 24.9424 | 10.0 | 1000 | 27.8925 | 0.2256 | 0.4367 | 0.2129 | 0.1479 | 0.6476 | 0.8314 | 0.0953 | 0.2556 | 0.2973 | 0.2244 | 0.7159 | 0.8712 | 0.1588 | 0.2153 | 0.2923 | 0.3793 |
|
| 79 |
+
| 24.6502 | 11.0 | 1100 | 27.7524 | 0.2254 | 0.441 | 0.2163 | 0.1486 | 0.6468 | 0.8186 | 0.0952 | 0.2556 | 0.2963 | 0.2231 | 0.7167 | 0.8681 | 0.1578 | 0.2153 | 0.2929 | 0.3772 |
|
| 80 |
+
| 24.5278 | 12.0 | 1200 | 27.7122 | 0.2252 | 0.4349 | 0.2167 | 0.1473 | 0.6462 | 0.8237 | 0.0927 | 0.2549 | 0.2979 | 0.2251 | 0.7162 | 0.8649 | 0.1583 | 0.2165 | 0.2921 | 0.3793 |
|
| 81 |
+
| 24.3514 | 13.0 | 1300 | 27.5382 | 0.224 | 0.4345 | 0.2156 | 0.1459 | 0.6554 | 0.8324 | 0.0958 | 0.2554 | 0.2988 | 0.2251 | 0.722 | 0.8806 | 0.1583 | 0.2191 | 0.2897 | 0.3785 |
|
| 82 |
+
| 24.3422 | 14.0 | 1400 | 27.5665 | 0.226 | 0.4374 | 0.2172 | 0.1488 | 0.6505 | 0.8059 | 0.0974 | 0.2551 | 0.2964 | 0.2241 | 0.7141 | 0.8434 | 0.1592 | 0.2158 | 0.2928 | 0.377 |
|
| 83 |
+
| 23.9768 | 15.0 | 1500 | 27.7770 | 0.2281 | 0.4379 | 0.2215 | 0.1499 | 0.6553 | 0.8188 | 0.096 | 0.2553 | 0.2978 | 0.2244 | 0.72 | 0.8632 | 0.1599 | 0.2163 | 0.2963 | 0.3793 |
|
| 84 |
+
| 23.7005 | 16.0 | 1600 | 27.5535 | 0.227 | 0.4392 | 0.2167 | 0.1485 | 0.6509 | 0.8165 | 0.0965 | 0.255 | 0.2972 | 0.2241 | 0.7175 | 0.8656 | 0.1608 | 0.2164 | 0.2932 | 0.3779 |
|
| 85 |
+
| 23.579 | 17.0 | 1700 | 27.4894 | 0.2286 | 0.44 | 0.2209 | 0.1511 | 0.6488 | 0.8152 | 0.097 | 0.2583 | 0.2965 | 0.2243 | 0.7113 | 0.8601 | 0.162 | 0.2144 | 0.2952 | 0.3785 |
|
| 86 |
+
| 23.5004 | 18.0 | 1800 | 27.2188 | 0.2274 | 0.4374 | 0.216 | 0.1498 | 0.6512 | 0.7954 | 0.0962 | 0.2562 | 0.2969 | 0.2251 | 0.712 | 0.8323 | 0.1614 | 0.215 | 0.2933 | 0.3788 |
|
| 87 |
+
| 23.1744 | 19.0 | 1900 | 27.3523 | 0.2286 | 0.4391 | 0.2166 | 0.1494 | 0.6559 | 0.8203 | 0.0962 | 0.2565 | 0.2998 | 0.2274 | 0.7156 | 0.8656 | 0.1602 | 0.2174 | 0.297 | 0.3821 |
|
| 88 |
+
| 23.1884 | 20.0 | 2000 | 27.1185 | 0.2304 | 0.4395 | 0.2204 | 0.1521 | 0.6533 | 0.8004 | 0.0968 | 0.2558 | 0.299 | 0.2273 | 0.7131 | 0.8347 | 0.1611 | 0.217 | 0.2998 | 0.3809 |
|
| 89 |
+
| 22.9136 | 21.0 | 2100 | 27.4296 | 0.2301 | 0.4386 | 0.2197 | 0.1518 | 0.6545 | 0.8185 | 0.0968 | 0.2552 | 0.2979 | 0.2256 | 0.7123 | 0.8712 | 0.1609 | 0.2179 | 0.2992 | 0.3778 |
|
| 90 |
+
| 22.6863 | 22.0 | 2200 | 26.9978 | 0.2309 | 0.444 | 0.2196 | 0.1519 | 0.657 | 0.7955 | 0.0976 | 0.2543 | 0.2982 | 0.2264 | 0.714 | 0.8316 | 0.1624 | 0.2181 | 0.2994 | 0.3784 |
|
| 91 |
+
| 22.7741 | 23.0 | 2300 | 27.0703 | 0.23 | 0.4436 | 0.2183 | 0.1519 | 0.6508 | 0.8029 | 0.0966 | 0.2562 | 0.3001 | 0.229 | 0.7106 | 0.8434 | 0.162 | 0.218 | 0.2979 | 0.3823 |
|
| 92 |
+
| 22.4779 | 24.0 | 2400 | 27.0394 | 0.2335 | 0.4521 | 0.2252 | 0.1552 | 0.656 | 0.8318 | 0.0962 | 0.2598 | 0.3026 | 0.231 | 0.7143 | 0.8601 | 0.1624 | 0.2187 | 0.3045 | 0.3865 |
|
| 93 |
+
| 22.357 | 25.0 | 2500 | 27.1483 | 0.2304 | 0.4456 | 0.2189 | 0.1517 | 0.6586 | 0.8065 | 0.0967 | 0.2554 | 0.2996 | 0.2278 | 0.7143 | 0.8378 | 0.162 | 0.2187 | 0.2989 | 0.3805 |
|
| 94 |
+
| 22.3167 | 26.0 | 2600 | 27.3299 | 0.232 | 0.4438 | 0.2193 | 0.1534 | 0.6572 | 0.8221 | 0.0977 | 0.2564 | 0.2989 | 0.2267 | 0.7134 | 0.8681 | 0.1624 | 0.2176 | 0.3016 | 0.3802 |
|
| 95 |
+
| 22.0958 | 27.0 | 2700 | 27.2571 | 0.232 | 0.4438 | 0.2171 | 0.1535 | 0.6539 | 0.8268 | 0.0974 | 0.2591 | 0.2986 | 0.226 | 0.7153 | 0.8774 | 0.1622 | 0.2185 | 0.3018 | 0.3788 |
|
| 96 |
+
| 22.0902 | 28.0 | 2800 | 27.5156 | 0.2315 | 0.4482 | 0.2177 | 0.1539 | 0.6566 | 0.8265 | 0.0978 | 0.2583 | 0.3021 | 0.23 | 0.716 | 0.8719 | 0.1626 | 0.22 | 0.3004 | 0.3842 |
|
| 97 |
+
| 21.9943 | 29.0 | 2900 | 27.0142 | 0.2288 | 0.4449 | 0.2155 | 0.1511 | 0.6536 | 0.8176 | 0.097 | 0.2557 | 0.2984 | 0.2257 | 0.7169 | 0.8569 | 0.1616 | 0.2202 | 0.2961 | 0.3766 |
|
| 98 |
+
| 21.8843 | 30.0 | 3000 | 27.1738 | 0.2314 | 0.4456 | 0.2192 | 0.1534 | 0.6557 | 0.8263 | 0.0973 | 0.2587 | 0.3026 | 0.23 | 0.7204 | 0.8625 | 0.1629 | 0.2203 | 0.2999 | 0.3848 |
|
| 99 |
+
| 21.8635 | 31.0 | 3100 | 27.0658 | 0.2316 | 0.4461 | 0.22 | 0.1534 | 0.6582 | 0.8166 | 0.0987 | 0.2581 | 0.3013 | 0.2292 | 0.7156 | 0.8625 | 0.163 | 0.2188 | 0.3003 | 0.3838 |
|
| 100 |
+
| 21.473 | 32.0 | 3200 | 27.1354 | 0.2323 | 0.4493 | 0.219 | 0.1545 | 0.6569 | 0.8077 | 0.0966 | 0.259 | 0.3024 | 0.2305 | 0.7172 | 0.8507 | 0.1619 | 0.2182 | 0.3026 | 0.3866 |
|
| 101 |
+
| 21.6879 | 33.0 | 3300 | 26.9810 | 0.2306 | 0.4461 | 0.2178 | 0.1533 | 0.6572 | 0.8095 | 0.0983 | 0.2581 | 0.3004 | 0.2285 | 0.7146 | 0.8476 | 0.1624 | 0.2194 | 0.2989 | 0.3814 |
|
| 102 |
+
| 21.3771 | 34.0 | 3400 | 27.5323 | 0.23 | 0.4476 | 0.2149 | 0.1536 | 0.6593 | 0.8185 | 0.0968 | 0.2577 | 0.2996 | 0.2265 | 0.7204 | 0.8618 | 0.162 | 0.2212 | 0.298 | 0.3781 |
|
| 103 |
+
| 21.2772 | 35.0 | 3500 | 27.1451 | 0.2327 | 0.4465 | 0.2172 | 0.1544 | 0.6641 | 0.8195 | 0.0988 | 0.2597 | 0.3028 | 0.2294 | 0.7262 | 0.8594 | 0.1616 | 0.221 | 0.3038 | 0.3847 |
|
| 104 |
+
| 21.3682 | 36.0 | 3600 | 27.4698 | 0.2334 | 0.4503 | 0.2184 | 0.155 | 0.6608 | 0.8088 | 0.0985 | 0.2574 | 0.3013 | 0.2292 | 0.7164 | 0.8594 | 0.1657 | 0.223 | 0.3011 | 0.3797 |
|
| 105 |
+
| 21.0417 | 37.0 | 3700 | 27.2499 | 0.2354 | 0.4523 | 0.2211 | 0.1569 | 0.6643 | 0.8224 | 0.0998 | 0.2604 | 0.3037 | 0.2307 | 0.7243 | 0.8562 | 0.1654 | 0.2209 | 0.3054 | 0.3865 |
|
| 106 |
+
| 21.0664 | 38.0 | 3800 | 27.3426 | 0.2304 | 0.4437 | 0.2159 | 0.1516 | 0.6568 | 0.8071 | 0.0986 | 0.2566 | 0.2993 | 0.227 | 0.7164 | 0.8451 | 0.1641 | 0.2198 | 0.2967 | 0.3788 |
|
| 107 |
+
| 21.0042 | 39.0 | 3900 | 27.7720 | 0.2315 | 0.4449 | 0.2182 | 0.1528 | 0.6611 | 0.8214 | 0.0994 | 0.2594 | 0.2994 | 0.2265 | 0.7191 | 0.8594 | 0.1604 | 0.2161 | 0.3026 | 0.3827 |
|
| 108 |
+
| 20.8548 | 40.0 | 4000 | 27.5494 | 0.2315 | 0.4495 | 0.2185 | 0.1535 | 0.6606 | 0.8161 | 0.0981 | 0.2576 | 0.3 | 0.2272 | 0.7189 | 0.8618 | 0.1626 | 0.2224 | 0.3003 | 0.3776 |
|
| 109 |
+
|
| 110 |
+
|
| 111 |
+
### Framework versions
|
| 112 |
+
|
| 113 |
+
- Transformers 4.42.0.dev0
|
| 114 |
+
- Pytorch 1.13.0+cu117
|
| 115 |
+
- Datasets 2.18.0
|
| 116 |
+
- Tokenizers 0.19.1
|