File size: 2,984 Bytes
3c0fc42 906f3de 3c0fc42 906f3de 3c0fc42 8522c78 3c0fc42 8522c78 3c0fc42 ffee127 3c0fc42 8522c78 3c0fc42 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 |
This folder contains models trained for the following 4 couplings
1. Suletta and Miorine from Suisei no Majo
2. Maple and Sally from Bofuri
3. Nozomi and Mizore from Sound Euphonium
4. Eila and Sanya from Strike witches
You can prompt them with above names except that for Maple and Sally you should use BoMaple and BoSally.
You can naturally get the trained couplings but it is also possible to pair characters from different series with either the native trained model (above) or lora (below):
### Dataset
Total size 1394
Suremio 285
- Suletta: 81
- Miorine: 85
- Suremio: 119
MapleSally 235
- Maple: 67
- Sally: 22
- MapleSally 33
Augmented with face cropping
Nozomizo 272
- Mizore: 81
- Nozomi: 81
- Nozomizo: 110
Eilanya 326
- Eila 67
- Sanya: 78
- Eilanya: 181
Regularization 276
### Base model
[NMFSAN](https://huggingface.co/Crosstyan/BPModel/blob/main/NMFSAN/README.md) so you can have different styles
### Native training
Trained with [Kohya trainer](https://github.com/Linaqruf/kohya-trainer)
- training of text encoder turned on
- learning rate 1e-6
- batch size 1
- clip skip 2
- number of training steps 64240
*Examples*
### LoRA
Please refer to [LoRA Training Guide](https://rentry.org/lora_train)
- training of text encoder turned on
- network dimension 128
- learning rate 1e-4
- batch size 6
- clip skip 2
- number of training steps 69700/6 (50 epochs)
*Examples*
|