Upload unet/config.json with huggingface_hub
Browse files- unet/config.json +61 -0
unet/config.json
ADDED
|
@@ -0,0 +1,61 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_class_name": "UNet2DConditionModel",
|
| 3 |
+
"_diffusers_version": "0.17.0.dev0",
|
| 4 |
+
"act_fn": "silu",
|
| 5 |
+
"addition_embed_type": "image_hint",
|
| 6 |
+
"addition_embed_type_num_heads": 64,
|
| 7 |
+
"attention_head_dim": 64,
|
| 8 |
+
"block_out_channels": [
|
| 9 |
+
384,
|
| 10 |
+
768,
|
| 11 |
+
1152,
|
| 12 |
+
1536
|
| 13 |
+
],
|
| 14 |
+
"center_input_sample": false,
|
| 15 |
+
"class_embed_type": null,
|
| 16 |
+
"class_embeddings_concat": false,
|
| 17 |
+
"conv_in_kernel": 3,
|
| 18 |
+
"conv_out_kernel": 3,
|
| 19 |
+
"cross_attention_dim": 768,
|
| 20 |
+
"cross_attention_norm": null,
|
| 21 |
+
"down_block_types": [
|
| 22 |
+
"ResnetDownsampleBlock2D",
|
| 23 |
+
"SimpleCrossAttnDownBlock2D",
|
| 24 |
+
"SimpleCrossAttnDownBlock2D",
|
| 25 |
+
"SimpleCrossAttnDownBlock2D"
|
| 26 |
+
],
|
| 27 |
+
"downsample_padding": 1,
|
| 28 |
+
"dual_cross_attention": false,
|
| 29 |
+
"encoder_hid_dim": 1280,
|
| 30 |
+
"encoder_hid_dim_type": "image_proj",
|
| 31 |
+
"flip_sin_to_cos": true,
|
| 32 |
+
"freq_shift": 0,
|
| 33 |
+
"in_channels": 8,
|
| 34 |
+
"layers_per_block": 3,
|
| 35 |
+
"mid_block_only_cross_attention": null,
|
| 36 |
+
"mid_block_scale_factor": 1,
|
| 37 |
+
"mid_block_type": "UNetMidBlock2DSimpleCrossAttn",
|
| 38 |
+
"norm_eps": 1e-05,
|
| 39 |
+
"norm_num_groups": 32,
|
| 40 |
+
"num_class_embeds": null,
|
| 41 |
+
"only_cross_attention": false,
|
| 42 |
+
"out_channels": 8,
|
| 43 |
+
"projection_class_embeddings_input_dim": null,
|
| 44 |
+
"resnet_out_scale_factor": 1.0,
|
| 45 |
+
"resnet_skip_time_act": false,
|
| 46 |
+
"resnet_time_scale_shift": "scale_shift",
|
| 47 |
+
"sample_size": 64,
|
| 48 |
+
"time_cond_proj_dim": null,
|
| 49 |
+
"time_embedding_act_fn": null,
|
| 50 |
+
"time_embedding_dim": null,
|
| 51 |
+
"time_embedding_type": "positional",
|
| 52 |
+
"timestep_post_act": null,
|
| 53 |
+
"up_block_types": [
|
| 54 |
+
"SimpleCrossAttnUpBlock2D",
|
| 55 |
+
"SimpleCrossAttnUpBlock2D",
|
| 56 |
+
"SimpleCrossAttnUpBlock2D",
|
| 57 |
+
"ResnetUpsampleBlock2D"
|
| 58 |
+
],
|
| 59 |
+
"upcast_attention": false,
|
| 60 |
+
"use_linear_projection": false
|
| 61 |
+
}
|