End of training
Browse files
README.md
CHANGED
|
@@ -1,6 +1,6 @@
|
|
| 1 |
---
|
| 2 |
license: apache-2.0
|
| 3 |
-
base_model: google/vit-
|
| 4 |
tags:
|
| 5 |
- generated_from_trainer
|
| 6 |
metrics:
|
|
@@ -15,10 +15,10 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
# FASHION-vision
|
| 17 |
|
| 18 |
-
This model is a fine-tuned version of [google/vit-
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
-
- Loss: 0.
|
| 21 |
-
- Accuracy: 0.
|
| 22 |
|
| 23 |
## Model description
|
| 24 |
|
|
@@ -38,120 +38,30 @@ More information needed
|
|
| 38 |
|
| 39 |
The following hyperparameters were used during training:
|
| 40 |
- learning_rate: 5e-05
|
| 41 |
-
- train_batch_size:
|
| 42 |
-
- eval_batch_size:
|
| 43 |
- seed: 42
|
| 44 |
- gradient_accumulation_steps: 4
|
| 45 |
-
- total_train_batch_size:
|
| 46 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 47 |
- lr_scheduler_type: linear
|
| 48 |
- lr_scheduler_warmup_ratio: 0.1
|
| 49 |
-
- num_epochs:
|
| 50 |
|
| 51 |
### Training results
|
| 52 |
|
| 53 |
-
| Training Loss | Epoch | Step
|
| 54 |
-
|
| 55 |
-
|
|
| 56 |
-
| 0.
|
| 57 |
-
| 0.
|
| 58 |
-
| 0.
|
| 59 |
-
| 0.
|
| 60 |
-
| 0.
|
| 61 |
-
| 0.
|
| 62 |
-
| 0.
|
| 63 |
-
| 0.
|
| 64 |
-
| 0.
|
| 65 |
-
| 0.2885 | 11.0 | 4125 | 0.3352 | 0.8823 |
|
| 66 |
-
| 0.3217 | 12.0 | 4500 | 0.3236 | 0.8833 |
|
| 67 |
-
| 0.2861 | 13.0 | 4875 | 0.3292 | 0.8811 |
|
| 68 |
-
| 0.263 | 14.0 | 5250 | 0.3083 | 0.8944 |
|
| 69 |
-
| 0.2265 | 15.0 | 5625 | 0.3035 | 0.8938 |
|
| 70 |
-
| 0.2407 | 16.0 | 6000 | 0.3094 | 0.8897 |
|
| 71 |
-
| 0.251 | 17.0 | 6375 | 0.3113 | 0.8894 |
|
| 72 |
-
| 0.2251 | 18.0 | 6750 | 0.2934 | 0.8951 |
|
| 73 |
-
| 0.2124 | 19.0 | 7125 | 0.3084 | 0.895 |
|
| 74 |
-
| 0.1974 | 20.0 | 7500 | 0.3144 | 0.8936 |
|
| 75 |
-
| 0.1907 | 21.0 | 7875 | 0.3048 | 0.8972 |
|
| 76 |
-
| 0.1857 | 22.0 | 8250 | 0.3046 | 0.896 |
|
| 77 |
-
| 0.1696 | 23.0 | 8625 | 0.3014 | 0.8982 |
|
| 78 |
-
| 0.2066 | 24.0 | 9000 | 0.2943 | 0.8985 |
|
| 79 |
-
| 0.2106 | 25.0 | 9375 | 0.3057 | 0.8984 |
|
| 80 |
-
| 0.2036 | 26.0 | 9750 | 0.3103 | 0.8968 |
|
| 81 |
-
| 0.1629 | 27.0 | 10125 | 0.3100 | 0.9003 |
|
| 82 |
-
| 0.1711 | 28.0 | 10500 | 0.3112 | 0.8978 |
|
| 83 |
-
| 0.144 | 29.0 | 10875 | 0.3285 | 0.897 |
|
| 84 |
-
| 0.1738 | 30.0 | 11250 | 0.3250 | 0.8968 |
|
| 85 |
-
| 0.1616 | 31.0 | 11625 | 0.3205 | 0.8979 |
|
| 86 |
-
| 0.1504 | 32.0 | 12000 | 0.3321 | 0.8947 |
|
| 87 |
-
| 0.1894 | 33.0 | 12375 | 0.3121 | 0.8963 |
|
| 88 |
-
| 0.1346 | 34.0 | 12750 | 0.3079 | 0.9017 |
|
| 89 |
-
| 0.1538 | 35.0 | 13125 | 0.3131 | 0.9045 |
|
| 90 |
-
| 0.1453 | 36.0 | 13500 | 0.3180 | 0.9042 |
|
| 91 |
-
| 0.1467 | 37.0 | 13875 | 0.3125 | 0.9042 |
|
| 92 |
-
| 0.1667 | 38.0 | 14250 | 0.3107 | 0.9035 |
|
| 93 |
-
| 0.1149 | 39.0 | 14625 | 0.3427 | 0.899 |
|
| 94 |
-
| 0.1248 | 40.0 | 15000 | 0.3152 | 0.9033 |
|
| 95 |
-
| 0.155 | 41.0 | 15375 | 0.3235 | 0.9015 |
|
| 96 |
-
| 0.1321 | 42.0 | 15750 | 0.3220 | 0.9065 |
|
| 97 |
-
| 0.156 | 43.0 | 16125 | 0.3326 | 0.9024 |
|
| 98 |
-
| 0.1511 | 44.0 | 16500 | 0.3351 | 0.8988 |
|
| 99 |
-
| 0.1039 | 45.0 | 16875 | 0.3309 | 0.9052 |
|
| 100 |
-
| 0.1277 | 46.0 | 17250 | 0.3552 | 0.9001 |
|
| 101 |
-
| 0.1147 | 47.0 | 17625 | 0.3462 | 0.9032 |
|
| 102 |
-
| 0.13 | 48.0 | 18000 | 0.3374 | 0.9009 |
|
| 103 |
-
| 0.1348 | 49.0 | 18375 | 0.3475 | 0.9006 |
|
| 104 |
-
| 0.1188 | 50.0 | 18750 | 0.3419 | 0.9067 |
|
| 105 |
-
| 0.1532 | 51.0 | 19125 | 0.3444 | 0.9025 |
|
| 106 |
-
| 0.1173 | 52.0 | 19500 | 0.3387 | 0.9034 |
|
| 107 |
-
| 0.1189 | 53.0 | 19875 | 0.3407 | 0.9033 |
|
| 108 |
-
| 0.13 | 54.0 | 20250 | 0.3614 | 0.9016 |
|
| 109 |
-
| 0.1206 | 55.0 | 20625 | 0.3404 | 0.9047 |
|
| 110 |
-
| 0.0989 | 56.0 | 21000 | 0.3560 | 0.903 |
|
| 111 |
-
| 0.1036 | 57.0 | 21375 | 0.3462 | 0.9056 |
|
| 112 |
-
| 0.1095 | 58.0 | 21750 | 0.3497 | 0.9031 |
|
| 113 |
-
| 0.143 | 59.0 | 22125 | 0.3364 | 0.9064 |
|
| 114 |
-
| 0.0889 | 60.0 | 22500 | 0.3544 | 0.9047 |
|
| 115 |
-
| 0.1008 | 61.0 | 22875 | 0.3510 | 0.904 |
|
| 116 |
-
| 0.1343 | 62.0 | 23250 | 0.3461 | 0.9069 |
|
| 117 |
-
| 0.1019 | 63.0 | 23625 | 0.3365 | 0.9058 |
|
| 118 |
-
| 0.1125 | 64.0 | 24000 | 0.3372 | 0.9086 |
|
| 119 |
-
| 0.143 | 65.0 | 24375 | 0.3433 | 0.9072 |
|
| 120 |
-
| 0.0971 | 66.0 | 24750 | 0.3390 | 0.9102 |
|
| 121 |
-
| 0.1147 | 67.0 | 25125 | 0.3493 | 0.9091 |
|
| 122 |
-
| 0.0931 | 68.0 | 25500 | 0.3469 | 0.9093 |
|
| 123 |
-
| 0.1127 | 69.0 | 25875 | 0.3421 | 0.9069 |
|
| 124 |
-
| 0.0935 | 70.0 | 26250 | 0.3535 | 0.9058 |
|
| 125 |
-
| 0.1152 | 71.0 | 26625 | 0.3313 | 0.9093 |
|
| 126 |
-
| 0.1288 | 72.0 | 27000 | 0.3661 | 0.9069 |
|
| 127 |
-
| 0.1244 | 73.0 | 27375 | 0.3405 | 0.9103 |
|
| 128 |
-
| 0.1158 | 74.0 | 27750 | 0.3345 | 0.9104 |
|
| 129 |
-
| 0.1255 | 75.0 | 28125 | 0.3367 | 0.9091 |
|
| 130 |
-
| 0.0886 | 76.0 | 28500 | 0.3657 | 0.9096 |
|
| 131 |
-
| 0.1008 | 77.0 | 28875 | 0.3468 | 0.9086 |
|
| 132 |
-
| 0.1209 | 78.0 | 29250 | 0.3489 | 0.9096 |
|
| 133 |
-
| 0.0944 | 79.0 | 29625 | 0.3511 | 0.9058 |
|
| 134 |
-
| 0.0928 | 80.0 | 30000 | 0.3509 | 0.9097 |
|
| 135 |
-
| 0.0932 | 81.0 | 30375 | 0.3485 | 0.9097 |
|
| 136 |
-
| 0.0973 | 82.0 | 30750 | 0.3584 | 0.9075 |
|
| 137 |
-
| 0.0964 | 83.0 | 31125 | 0.3459 | 0.9107 |
|
| 138 |
-
| 0.1262 | 84.0 | 31500 | 0.3648 | 0.9107 |
|
| 139 |
-
| 0.113 | 85.0 | 31875 | 0.3483 | 0.9083 |
|
| 140 |
-
| 0.0828 | 86.0 | 32250 | 0.3396 | 0.9116 |
|
| 141 |
-
| 0.1104 | 87.0 | 32625 | 0.3370 | 0.9119 |
|
| 142 |
-
| 0.0804 | 88.0 | 33000 | 0.3596 | 0.9117 |
|
| 143 |
-
| 0.0905 | 89.0 | 33375 | 0.3538 | 0.912 |
|
| 144 |
-
| 0.1064 | 90.0 | 33750 | 0.3497 | 0.9112 |
|
| 145 |
-
| 0.0917 | 91.0 | 34125 | 0.3392 | 0.9139 |
|
| 146 |
-
| 0.0813 | 92.0 | 34500 | 0.3561 | 0.9109 |
|
| 147 |
-
| 0.074 | 93.0 | 34875 | 0.3475 | 0.9098 |
|
| 148 |
-
| 0.0922 | 94.0 | 35250 | 0.3482 | 0.9114 |
|
| 149 |
-
| 0.0752 | 95.0 | 35625 | 0.3751 | 0.9097 |
|
| 150 |
-
| 0.0751 | 96.0 | 36000 | 0.3530 | 0.9103 |
|
| 151 |
-
| 0.0818 | 97.0 | 36375 | 0.3477 | 0.9137 |
|
| 152 |
-
| 0.0677 | 98.0 | 36750 | 0.3495 | 0.9115 |
|
| 153 |
-
| 0.0838 | 99.0 | 37125 | 0.3533 | 0.9114 |
|
| 154 |
-
| 0.0772 | 100.0 | 37500 | 0.3565 | 0.9122 |
|
| 155 |
|
| 156 |
|
| 157 |
### Framework versions
|
|
|
|
| 1 |
---
|
| 2 |
license: apache-2.0
|
| 3 |
+
base_model: google/vit-huge-patch14-224-in21k
|
| 4 |
tags:
|
| 5 |
- generated_from_trainer
|
| 6 |
metrics:
|
|
|
|
| 15 |
|
| 16 |
# FASHION-vision
|
| 17 |
|
| 18 |
+
This model is a fine-tuned version of [google/vit-huge-patch14-224-in21k](https://huggingface.co/google/vit-huge-patch14-224-in21k) on an unknown dataset.
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
+
- Loss: 0.2821
|
| 21 |
+
- Accuracy: 0.9034
|
| 22 |
|
| 23 |
## Model description
|
| 24 |
|
|
|
|
| 38 |
|
| 39 |
The following hyperparameters were used during training:
|
| 40 |
- learning_rate: 5e-05
|
| 41 |
+
- train_batch_size: 16
|
| 42 |
+
- eval_batch_size: 16
|
| 43 |
- seed: 42
|
| 44 |
- gradient_accumulation_steps: 4
|
| 45 |
+
- total_train_batch_size: 64
|
| 46 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 47 |
- lr_scheduler_type: linear
|
| 48 |
- lr_scheduler_warmup_ratio: 0.1
|
| 49 |
+
- num_epochs: 10
|
| 50 |
|
| 51 |
### Training results
|
| 52 |
|
| 53 |
+
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
| 54 |
+
|:-------------:|:-----:|:----:|:---------------:|:--------:|
|
| 55 |
+
| 0.5334 | 1.0 | 750 | 0.5595 | 0.8302 |
|
| 56 |
+
| 0.3389 | 2.0 | 1500 | 0.4090 | 0.8602 |
|
| 57 |
+
| 0.358 | 3.0 | 2250 | 0.3631 | 0.8717 |
|
| 58 |
+
| 0.3672 | 4.0 | 3000 | 0.3368 | 0.8815 |
|
| 59 |
+
| 0.3458 | 5.0 | 3750 | 0.3231 | 0.8842 |
|
| 60 |
+
| 0.2721 | 6.0 | 4500 | 0.3075 | 0.8885 |
|
| 61 |
+
| 0.2397 | 7.0 | 5250 | 0.3035 | 0.8899 |
|
| 62 |
+
| 0.2779 | 8.0 | 6000 | 0.2893 | 0.8963 |
|
| 63 |
+
| 0.2046 | 9.0 | 6750 | 0.2868 | 0.8991 |
|
| 64 |
+
| 0.2599 | 10.0 | 7500 | 0.2821 | 0.9034 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 65 |
|
| 66 |
|
| 67 |
### Framework versions
|