Instructions to use VictorSanh/vqvae_ckpts with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use VictorSanh/vqvae_ckpts with Transformers:
# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("VictorSanh/vqvae_ckpts", dtype="auto") - Notebooks
- Google Colab
- Kaggle
| { | |
| "_class_name": "VQGANModel", | |
| "_version": "0.0.1", | |
| "attn_resolutions": [], | |
| "channel_mult": [ | |
| 1, | |
| 2, | |
| 2, | |
| 4, | |
| 6 | |
| ], | |
| "commitment_cost": 0.25, | |
| "dropout": 0.0, | |
| "hidden_channels": 128, | |
| "no_attn_mid_block": true, | |
| "num_channels": 3, | |
| "num_embeddings": 8192, | |
| "num_res_blocks": 2, | |
| "quantized_embed_dim": 64, | |
| "resample_with_conv": true, | |
| "resolution": 256, | |
| "z_channels": 64 | |
| } | |