File size: 2,195 Bytes
9a7fe1f
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
168a510
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
openai_unet_sd:
  type: openai_unet
  args:
    image_size: null # no use
    in_channels: 4
    out_channels: 4
    model_channels: 320
    attention_resolutions: [ 4, 2, 1 ]
    num_res_blocks: [ 2, 2, 2, 2 ]
    channel_mult: [ 1, 2, 4, 4 ]
    num_heads: 8
    use_spatial_transformer: True
    transformer_depth: 1
    context_dim: 768
    use_checkpoint: True
    legacy: False

openai_unet_dual_context:
  super_cfg: openai_unet_sd
  type: openai_unet_dual_context

########################
# Code cleaned version #
########################

openai_unet_2d_audio:
  type: openai_unet_2d
  args:
    input_channels: 8
    model_channels: 192
    output_channels: 8
    num_noattn_blocks: [ 2, 2, 2, 2 ]
    channel_mult: [ 1, 2, 4, 4 ]
    with_attn: [true, true, true, false]
    channel_mult_connector: [1, 2, 4]
    num_noattn_blocks_connector: [1, 1, 1]
    with_connector: [True, True, True, False]
    connector_output_channel: 1280
    num_heads: 8
    context_dim: 768
    use_checkpoint: False

openai_unet_2d:
  type: openai_unet_2d
  args:
    input_channels: 4
    model_channels: 320
    output_channels: 4
    num_noattn_blocks: [ 2, 2, 2, 2 ]
    channel_mult: [ 1, 2, 4, 4 ]
    with_attn: [true, true, true, false]
    channel_mult_connector: [1, 2, 4]
    num_noattn_blocks_connector: [1, 1, 1]
    with_connector: [True, True, True, False]
    connector_output_channel: 1280
    num_heads: 8
    context_dim: 768
    use_checkpoint: True
    use_video_architecture: True

openai_unet_0dmd:
  type: openai_unet_0dmd
  args:
    input_channels: 768
    model_channels: 320
    output_channels: 768
    num_noattn_blocks: [ 2, 2, 2, 2 ]
    channel_mult: [ 1, 2, 4, 4 ]
    second_dim: [ 4, 4, 4, 4 ]
    with_attn: [true, true, true, false]
    num_noattn_blocks_connector: [1, 1, 1]
    second_dim_connector: [4, 4, 4]
    with_connector: [True, True, True, False]
    connector_output_channel: 1280
    num_heads: 8
    context_dim: 768
    use_checkpoint: True

prova:
  type: prova
  args:
    unet_frontal_cfg: MODEL(openai_unet_2d)
    unet_lateral_cfg: MODEL(openai_unet_2d)
    unet_text_cfg: MODEL(openai_unet_0dmd)
    model_type: ['frontal', 'lateral', 'text']