{ | |
"aln": 1, | |
"aln_gamma_init": 0.001, | |
"attn_drop_rate": 0.0, | |
"cond_drop_rate": 0.1, | |
"cos_attn": false, | |
"depth": 16, | |
"drop_path_rate": 0.0, | |
"drop_rate": 0.0, | |
"embed_dim": 1024, | |
"flash_if_available": true, | |
"fused_if_available": true, | |
"layer_scale": -1, | |
"mlp_ratio": 4.0, | |
"norm_eps": 1e-06, | |
"num_classes": 1000, | |
"num_heads": 16, | |
"patch_nums": [ | |
1, | |
2, | |
3, | |
4, | |
5, | |
6, | |
8, | |
10, | |
13, | |
16 | |
], | |
"shared_aln": false, | |
"tau": 4, | |
"vae_kwargs": { | |
"ch": 160, | |
"share_quant_resi": 4, | |
"test_mode": true, | |
"v_patch_nums": [ | |
1, | |
2, | |
3, | |
4, | |
5, | |
6, | |
8, | |
10, | |
13, | |
16 | |
], | |
"vocab_size": 4096, | |
"z_channels": 32 | |
} | |
} |