deepghs/wd14_tagger_embedding_denormalize
By deepghs · ↓ 0 · ♥ 2 · Updated 2024-11-30
What's in this repo
This page lists the files in deepghs/wd14_tagger_embedding_denormalize, including model.onnx, model_with_fc.onnx and more.
- 40 ONNX models (.onnx files)
- 122 files, including 40 .json, 40 .onnx, 20 .pt, 4 .1, 4 .0, 4 .2
- Last updated 2024-11-30
ONNX files
| File | Size | Download |
|---|---|---|
| ConvNextV2_mnum1_all/model.onnx | — | Download |
| ConvNextV2_mnum1_all/model_with_fc.onnx | — | Download |
| ConvNextV2_mnum2_all/model.onnx | — | Download |
| ConvNextV2_mnum2_all/model_with_fc.onnx | — | Download |
| ConvNext_mnum1_all/model.onnx | — | Download |
| ConvNext_mnum1_all/model_with_fc.onnx | — | Download |
| ConvNext_mnum2_all/model.onnx | — | Download |
| ConvNext_mnum2_all/model_with_fc.onnx | — | Download |
| ConvNext_v3_mnum1_all/model.onnx | — | Download |
| ConvNext_v3_mnum1_all/model_with_fc.onnx | — | Download |
Showing 10 of 40 .onnx files — see the full file list on the right.
How to use
These ONNX files can be loaded with ONNX Runtime. To inspect a model graph, upload the file to the Netron viewer. Download links point to the original files on Hugging Face.
README
Models of experiment: https://github.com/deepghs/tagger_embedding_aligner
import numpy as np
from imgutils.tagging import get_wd14_tags, convert_wd14_emb_to_prediction, denormalize_wd14_emb
embedding, (r, g, c) = get_wd14_tags(
'/my/image.png',
fmt=('embedding', ('rating', 'general', 'character')),
)
# normal tag results
print('Expected result:')
print(r)
print(g)
print(c)
# normalize embedding
embedding = embedding / np.linalg.norm(embedding)
# bad tag results
br, bg, bc = convert_wd14_emb_to_prediction(embedding)
print('Bad results due to the embedding normalization:')
print(br)
print(bg)
print(bc)
# denormalize this embedding
output = denormalize_wd14_emb(embedding)
print(output.shape)
# should be similar to r, g, c, approx 1e-3 error
rating, general, character = convert_wd14_emb_to_prediction(output)
print('De-normalized result:')
print(rating)
print(general)
print(character)
| Name | Tagger | Embedding Width | Tags Count | FLOPS | Params | EMB Cosine | EMB Norm | Pred Loss | Pred MSE |
|---|---|---|---|---|---|---|---|---|---|
| ViT_v3_mnum2_all | ViT_v3 | 768 | 10861 | 0.000398G | 0.40M | 1 | 0.1712 | 0.004306 | 2.116e-08 |
| ViT_v3_mnum1_all | ViT_v3 | 768 | 10861 | 0.000709G | 0.71M | 1 | 0.2246 | 0.004306 | 3.991e-08 |
| ConvNext_v3_mnum2_all | ConvNext_v3 | 1024 | 10861 | 0.000708G | 0.71M | 1 | 0.1126 | 0.004531 | 2.061e-08 |
| ConvNext_v3_mnum1_all | ConvNext_v3 | 1024 | 10861 | 0.001260G | 1.26M | 1 | 0.1473 | 0.004531 | 3.539e-08 |
| ViT_mnum2_all | ViT | 768 | 9083 | 0.000398G | 0.40M | 1 | 0.08641 | 0.005199 | 3.797e-09 |
| ViT_mnum1_all | ViT | 768 | 9083 | 0.000709G | 0.71M | 1 | 0.1724 | 0.005199 | 1.896e-08 |
| ConvNext_mnum2_all | ConvNext | 1024 | 9083 | 0.000708G | 0.71M | 1 | 0.05776 | 0.005213 | 7.207e-09 |
| ConvNext_mnum1_all | ConvNext | 1024 | 9083 | 0.001260G | 1.26M | 1 | 0.07134 | 0.005214 | 1.292e-08 |
| ViT_Large_mnum2_all | ViT_Large | 1024 | 10861 | 0.000708G | 0.71M | 1 | 1.403 | 0.003966 | 1.617e-07 |
| ViT_Large_mnum1_all | ViT_Large | 1024 | 10861 | 0.001260G | 1.26M | 1 | 1.643 | 0.003966 | 2.24e-07 |
| SwinV2_mnum2_all | SwinV2 | 1024 | 9083 | 0.000708G | 0.71M | 1 | 0.1257 | 0.004726 | 3.797e-08 |
| SwinV2_mnum1_all | SwinV2 | 1024 | 9083 | 0.001260G | 1.26M | 1 | 0.1497 | 0.004727 | 5.487e-08 |
| EVA02_Large_mnum2_all | EVA02_Large | 1024 | 10861 | 0.000708G | 0.71M | 1 | 1.268 | 0.005948 | 5.466e-08 |
| EVA02_Large_mnum1_all | EVA02_Large | 1024 | 10861 | 0.001260G | 1.26M | 1 | 1.713 | 0.005948 | 9.518e-08 |
| ConvNextV2_mnum2_all | ConvNextV2 | 1024 | 9083 | 0.000708G | 0.71M | 1 | 0.09014 | 0.004596 | 1.43e-08 |
| ConvNextV2_mnum1_all | ConvNextV2 | 1024 | 9083 | 0.001260G | 1.26M | 1 | 0.1216 | 0.004596 | 2.76e-08 |
| SwinV2_v3_mnum2_all | SwinV2_v3 | 1024 | 10861 | 0.000708G | 0.71M | 1 | 0.2129 | 0.004128 | 4.035e-08 |
| SwinV2_v3_mnum1_all | SwinV2_v3 | 1024 | 10861 | 0.001260G | 1.26M | 1 | 0.2784 | 0.004129 | 6.893e-08 |
| MOAT_mnum2_all | MOAT | 1024 | 9083 | 0.000708G | 0.71M | 1 | 0.4662 | 0.004998 | 1.855e-08 |
| MOAT_mnum1_all | MOAT | 1024 | 9083 | 0.001260G | 1.26M | 1 | 0.7849 | 0.004998 | 5.549e-08 |
Model info
feature-extraction
dghs-imgutils
dghs-imgutils
tensorboard
onnx
art
anime
cosine_space
feature-extraction
dataset:deepghs/wd14_tagger_inversion
license:openrail
region:us
↓ 0
♥ 2
Created: 2024-11-12
Updated: 2024-11-30
View on Hugging FaceFiles (122)
.gitattributes
ConvNextV2_mnum1_all/events.out.tfevents.1731467912.99115a8e0b9f5c8a69fc2f5fa0161c8670a7b68fb92c149a211a0a72.3434049.4
ConvNextV2_mnum1_all/metrics.json
ConvNextV2_mnum1_all/model.json
ConvNextV2_mnum1_all/model.onnx
ONNX
ConvNextV2_mnum1_all/model.pt
ConvNextV2_mnum1_all/model_with_fc.onnx
ONNX
ConvNextV2_mnum2_all/events.out.tfevents.1731458536.99115a8e0b9f5c8a69fc2f5fa0161c8670a7b68fb92c149a211a0a72.3434049.3
ConvNextV2_mnum2_all/metrics.json
ConvNextV2_mnum2_all/model.json
ConvNextV2_mnum2_all/model.onnx
ONNX
ConvNextV2_mnum2_all/model.pt
ConvNextV2_mnum2_all/model_with_fc.onnx
ONNX
ConvNext_mnum1_all/events.out.tfevents.1731486633.99115a8e0b9f5c8a69fc2f5fa0161c8670a7b68fb92c149a211a0a72.756977.1
ConvNext_mnum1_all/metrics.json
ConvNext_mnum1_all/model.json
ConvNext_mnum1_all/model.onnx
ONNX
ConvNext_mnum1_all/model.pt
ConvNext_mnum1_all/model_with_fc.onnx
ONNX
ConvNext_mnum2_all/events.out.tfevents.1731477314.99115a8e0b9f5c8a69fc2f5fa0161c8670a7b68fb92c149a211a0a72.756977.0
ConvNext_mnum2_all/metrics.json
ConvNext_mnum2_all/model.json
ConvNext_mnum2_all/model.onnx
ONNX
ConvNext_mnum2_all/model.pt
ConvNext_mnum2_all/model_with_fc.onnx
ONNX
ConvNext_v3_mnum1_all/events.out.tfevents.1731524250.99115a8e0b9f5c8a69fc2f5fa0161c8670a7b68fb92c149a211a0a72.756977.5
ConvNext_v3_mnum1_all/metrics.json
ConvNext_v3_mnum1_all/model.json
ConvNext_v3_mnum1_all/model.onnx
ONNX
ConvNext_v3_mnum1_all/model.pt
ConvNext_v3_mnum1_all/model_with_fc.onnx
ONNX
ConvNext_v3_mnum2_all/events.out.tfevents.1731513240.99115a8e0b9f5c8a69fc2f5fa0161c8670a7b68fb92c149a211a0a72.756977.4
ConvNext_v3_mnum2_all/metrics.json
ConvNext_v3_mnum2_all/model.json
ConvNext_v3_mnum2_all/model.onnx
ONNX
ConvNext_v3_mnum2_all/model.pt
ConvNext_v3_mnum2_all/model_with_fc.onnx
ONNX
EVA02_Large_mnum1_all/events.out.tfevents.1731406771.99115a8e0b9f5c8a69fc2f5fa0161c8670a7b68fb92c149a211a0a72.2486943.1
EVA02_Large_mnum1_all/metrics.json
EVA02_Large_mnum1_all/model.json
EVA02_Large_mnum1_all/model.onnx
ONNX
EVA02_Large_mnum1_all/model.pt
EVA02_Large_mnum1_all/model_with_fc.onnx
ONNX
EVA02_Large_mnum2_all/events.out.tfevents.1731396120.99115a8e0b9f5c8a69fc2f5fa0161c8670a7b68fb92c149a211a0a72.2486943.0
EVA02_Large_mnum2_all/metrics.json
EVA02_Large_mnum2_all/model.json
EVA02_Large_mnum2_all/model.onnx
ONNX
EVA02_Large_mnum2_all/model.pt
EVA02_Large_mnum2_all/model_with_fc.onnx
ONNX
MOAT_mnum1_all/events.out.tfevents.1731368146.99115a8e0b9f5c8a69fc2f5fa0161c8670a7b68fb92c149a211a0a72.1224206.2
MOAT_mnum1_all/metrics.json
MOAT_mnum1_all/model.json
MOAT_mnum1_all/model.onnx
ONNX
MOAT_mnum1_all/model.pt
MOAT_mnum1_all/model_with_fc.onnx
ONNX
MOAT_mnum2_all/events.out.tfevents.1731377678.99115a8e0b9f5c8a69fc2f5fa0161c8670a7b68fb92c149a211a0a72.1224206.3
MOAT_mnum2_all/metrics.json
MOAT_mnum2_all/model.json
MOAT_mnum2_all/model.onnx
ONNX
MOAT_mnum2_all/model.pt
MOAT_mnum2_all/model_with_fc.onnx
ONNX
README.md
SwinV2_mnum1_all/events.out.tfevents.1731449244.99115a8e0b9f5c8a69fc2f5fa0161c8670a7b68fb92c149a211a0a72.3434049.2
SwinV2_mnum1_all/metrics.json
SwinV2_mnum1_all/model.json
SwinV2_mnum1_all/model.onnx
ONNX
SwinV2_mnum1_all/model.pt
SwinV2_mnum1_all/model_with_fc.onnx
ONNX
SwinV2_mnum2_all/events.out.tfevents.1731439996.99115a8e0b9f5c8a69fc2f5fa0161c8670a7b68fb92c149a211a0a72.3434049.1
SwinV2_mnum2_all/metrics.json
SwinV2_mnum2_all/model.json
SwinV2_mnum2_all/model.onnx
ONNX
SwinV2_mnum2_all/model.pt
SwinV2_mnum2_all/model_with_fc.onnx
ONNX
SwinV2_v3_mnum1_all/events.out.tfevents.1731346980.99115a8e0b9f5c8a69fc2f5fa0161c8670a7b68fb92c149a211a0a72.1224206.0
SwinV2_v3_mnum1_all/metrics.json
SwinV2_v3_mnum1_all/model.json
SwinV2_v3_mnum1_all/model.onnx
ONNX
SwinV2_v3_mnum1_all/model.pt
SwinV2_v3_mnum1_all/model_with_fc.onnx
ONNX
SwinV2_v3_mnum2_all/events.out.tfevents.1731357825.99115a8e0b9f5c8a69fc2f5fa0161c8670a7b68fb92c149a211a0a72.1224206.1
SwinV2_v3_mnum2_all/metrics.json
SwinV2_v3_mnum2_all/model.json
SwinV2_v3_mnum2_all/model.onnx
ONNX
SwinV2_v3_mnum2_all/model.pt
SwinV2_v3_mnum2_all/model_with_fc.onnx
ONNX
ViT_Large_mnum1_all/events.out.tfevents.1731429112.99115a8e0b9f5c8a69fc2f5fa0161c8670a7b68fb92c149a211a0a72.3434049.0
ViT_Large_mnum1_all/metrics.json
ViT_Large_mnum1_all/model.json
ViT_Large_mnum1_all/model.onnx
ONNX
ViT_Large_mnum1_all/model.pt
ViT_Large_mnum1_all/model_with_fc.onnx
ONNX
ViT_Large_mnum2_all/events.out.tfevents.1731416951.99115a8e0b9f5c8a69fc2f5fa0161c8670a7b68fb92c149a211a0a72.2486943.2
ViT_Large_mnum2_all/metrics.json
ViT_Large_mnum2_all/model.json
ViT_Large_mnum2_all/model.onnx
ONNX
ViT_Large_mnum2_all/model.pt
ViT_Large_mnum2_all/model_with_fc.onnx
ONNX
ViT_mnum1_all/events.out.tfevents.1731504468.99115a8e0b9f5c8a69fc2f5fa0161c8670a7b68fb92c149a211a0a72.756977.3
ViT_mnum1_all/metrics.json
ViT_mnum1_all/model.json
ViT_mnum1_all/model.onnx
ONNX
ViT_mnum1_all/model.pt
ViT_mnum1_all/model_with_fc.onnx
ONNX
ViT_mnum2_all/events.out.tfevents.1731495691.99115a8e0b9f5c8a69fc2f5fa0161c8670a7b68fb92c149a211a0a72.756977.2
ViT_mnum2_all/metrics.json
ViT_mnum2_all/model.json
ViT_mnum2_all/model.onnx
ONNX
ViT_mnum2_all/model.pt
ViT_mnum2_all/model_with_fc.onnx
ONNX
ViT_v3_mnum1_all/events.out.tfevents.1731545594.99115a8e0b9f5c8a69fc2f5fa0161c8670a7b68fb92c149a211a0a72.756977.7
ViT_v3_mnum1_all/metrics.json
ViT_v3_mnum1_all/model.json
ViT_v3_mnum1_all/model.onnx
ONNX
ViT_v3_mnum1_all/model.pt
ViT_v3_mnum1_all/model_with_fc.onnx
ONNX
ViT_v3_mnum2_all/events.out.tfevents.1731534803.99115a8e0b9f5c8a69fc2f5fa0161c8670a7b68fb92c149a211a0a72.756977.6
ViT_v3_mnum2_all/metrics.json
ViT_v3_mnum2_all/model.json
ViT_v3_mnum2_all/model.onnx
ONNX
ViT_v3_mnum2_all/model.pt
ViT_v3_mnum2_all/model_with_fc.onnx
ONNX