| { | |
| "experiment": { | |
| "tokenizer_checkpoint": "tatitok_bl32_vq.bin", | |
| "output_dir": "tatitok_bl32_vq" | |
| }, | |
| "model": { | |
| "vq_model": { | |
| "quantize_mode": "vq", | |
| "codebook_size": 8192, | |
| "token_size": 64, | |
| "use_l2_norm": false, | |
| "commitment_cost": 0.25, | |
| "clustering_vq": true, | |
| "vit_enc_model_size": "base", | |
| "vit_dec_model_size": "large", | |
| "vit_enc_patch_size": 16, | |
| "vit_dec_patch_size": 16, | |
| "num_latent_tokens": 32, | |
| "finetune_decoder": false, | |
| "is_legacy": false | |
| } | |
| }, | |
| "dataset": { | |
| "preprocessing": { | |
| "crop_size": 256 | |
| } | |
| } | |
| } |