Push model using huggingface_hub.
Browse files- README.md +13 -0
- config.json +24 -0
- model.safetensors +3 -0
README.md
ADDED
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: apache-2.0
|
3 |
+
tags:
|
4 |
+
- arxiv:2406.07550
|
5 |
+
- arxiv:2501.07730
|
6 |
+
- image-tokenization
|
7 |
+
- model_hub_mixin
|
8 |
+
- pytorch_model_hub_mixin
|
9 |
+
---
|
10 |
+
|
11 |
+
This model has been pushed to the Hub using the [PytorchModelHubMixin](https://huggingface.co/docs/huggingface_hub/package_reference/mixins#huggingface_hub.PyTorchModelHubMixin) integration:
|
12 |
+
- Library: https://github.com/bytedance/1d-tokenizer
|
13 |
+
- Docs: [More Information Needed]
|
config.json
ADDED
@@ -0,0 +1,24 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"experiment": {
|
3 |
+
"tokenizer_checkpoint": "tatitok_bl32_vae.bin",
|
4 |
+
"output_dir": "tatitok_bl32_vae"
|
5 |
+
},
|
6 |
+
"model": {
|
7 |
+
"vq_model": {
|
8 |
+
"quantize_mode": "vae",
|
9 |
+
"token_size": 16,
|
10 |
+
"vit_enc_model_size": "base",
|
11 |
+
"vit_dec_model_size": "large",
|
12 |
+
"vit_enc_patch_size": 16,
|
13 |
+
"vit_dec_patch_size": 16,
|
14 |
+
"num_latent_tokens": 32,
|
15 |
+
"finetune_decoder": false,
|
16 |
+
"is_legacy": false
|
17 |
+
}
|
18 |
+
},
|
19 |
+
"dataset": {
|
20 |
+
"preprocessing": {
|
21 |
+
"crop_size": 256
|
22 |
+
}
|
23 |
+
}
|
24 |
+
}
|
model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:9a2d7df0587fc265646e3d4f21c88d376e20852b7a2b5d81b075c42785501c4c
|
3 |
+
size 1560860240
|