-
Notifications
You must be signed in to change notification settings - Fork 2.8k
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
Adds streamlit demo, fixes simple sample script
- Loading branch information
Vikram Voleti
committed
Mar 18, 2024
1 parent
73a044f
commit b41860f
Showing
13 changed files
with
1,049 additions
and
65 deletions.
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,119 @@ | ||
| model: | ||
| target: sgm.models.diffusion.DiffusionEngine | ||
| params: | ||
| scale_factor: 0.18215 | ||
| disable_first_stage_autocast: True | ||
|
|
||
| denoiser_config: | ||
| target: sgm.modules.diffusionmodules.denoiser.Denoiser | ||
| params: | ||
| scaling_config: | ||
| target: sgm.modules.diffusionmodules.denoiser_scaling.VScalingWithEDMcNoise | ||
|
|
||
| network_config: | ||
| target: sgm.modules.diffusionmodules.video_model.VideoUNet | ||
| params: | ||
| adm_in_channels: 1280 | ||
| num_classes: sequential | ||
| use_checkpoint: True | ||
| in_channels: 8 | ||
| out_channels: 4 | ||
| model_channels: 320 | ||
| attention_resolutions: [4, 2, 1] | ||
| num_res_blocks: 2 | ||
| channel_mult: [1, 2, 4, 4] | ||
| num_head_channels: 64 | ||
| use_linear_in_transformer: True | ||
| transformer_depth: 1 | ||
| context_dim: 1024 | ||
| spatial_transformer_attn_type: softmax-xformers | ||
| extra_ff_mix_layer: True | ||
| use_spatial_context: True | ||
| merge_strategy: learned_with_images | ||
| video_kernel_size: [3, 1, 1] | ||
|
|
||
| conditioner_config: | ||
| target: sgm.modules.GeneralConditioner | ||
| params: | ||
| emb_models: | ||
| - input_key: cond_frames_without_noise | ||
| is_trainable: False | ||
| target: sgm.modules.encoders.modules.FrozenOpenCLIPImagePredictionEmbedder | ||
| params: | ||
| n_cond_frames: 1 | ||
| n_copies: 1 | ||
| open_clip_embedding_config: | ||
| target: sgm.modules.encoders.modules.FrozenOpenCLIPImageEmbedder | ||
| params: | ||
| freeze: True | ||
|
|
||
| - input_key: cond_frames | ||
| is_trainable: False | ||
| target: sgm.modules.encoders.modules.VideoPredictionEmbedderWithEncoder | ||
| params: | ||
| disable_encoder_autocast: True | ||
| n_cond_frames: 1 | ||
| n_copies: 1 | ||
| is_ae: True | ||
| encoder_config: | ||
| target: sgm.models.autoencoder.AutoencoderKLModeOnly | ||
| params: | ||
| embed_dim: 4 | ||
| monitor: val/rec_loss | ||
| ddconfig: | ||
| attn_type: vanilla-xformers | ||
| double_z: True | ||
| z_channels: 4 | ||
| resolution: 256 | ||
| in_channels: 3 | ||
| out_ch: 3 | ||
| ch: 128 | ||
| ch_mult: [1, 2, 4, 4] | ||
| num_res_blocks: 2 | ||
| attn_resolutions: [] | ||
| dropout: 0.0 | ||
| lossconfig: | ||
| target: torch.nn.Identity | ||
|
|
||
| - input_key: cond_aug | ||
| is_trainable: False | ||
| target: sgm.modules.encoders.modules.ConcatTimestepEmbedderND | ||
| params: | ||
| outdim: 256 | ||
|
|
||
| - input_key: polars_rad | ||
| is_trainable: False | ||
| target: sgm.modules.encoders.modules.ConcatTimestepEmbedderND | ||
| params: | ||
| outdim: 512 | ||
|
|
||
| - input_key: azimuths_rad | ||
| is_trainable: False | ||
| target: sgm.modules.encoders.modules.ConcatTimestepEmbedderND | ||
| params: | ||
| outdim: 512 | ||
|
|
||
| first_stage_config: | ||
| target: sgm.models.autoencoder.AutoencodingEngine | ||
| params: | ||
| loss_config: | ||
| target: torch.nn.Identity | ||
| regularizer_config: | ||
| target: sgm.modules.autoencoding.regularizers.DiagonalGaussianRegularizer | ||
| encoder_config: | ||
| target: torch.nn.Identity | ||
| decoder_config: | ||
| target: sgm.modules.autoencoding.temporal_ae.VideoDecoder | ||
| params: | ||
| attn_type: vanilla | ||
| double_z: True | ||
| z_channels: 4 | ||
| resolution: 256 | ||
| in_channels: 3 | ||
| out_ch: 3 | ||
| ch: 128 | ||
| ch_mult: [1, 2, 4, 4] | ||
| num_res_blocks: 2 | ||
| attn_resolutions: [] | ||
| dropout: 0.0 | ||
| video_kernel_size: [3, 1, 1] |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,118 @@ | ||
| model: | ||
| target: sgm.models.diffusion.DiffusionEngine | ||
| params: | ||
| scale_factor: 0.18215 | ||
| disable_first_stage_autocast: True | ||
|
|
||
| denoiser_config: | ||
| target: sgm.modules.diffusionmodules.denoiser.Denoiser | ||
| params: | ||
| scaling_config: | ||
| target: sgm.modules.diffusionmodules.denoiser_scaling.VScalingWithEDMcNoise | ||
|
|
||
| network_config: | ||
| target: sgm.modules.diffusionmodules.video_model.VideoUNet | ||
| params: | ||
| adm_in_channels: 1280 | ||
| num_classes: sequential | ||
| use_checkpoint: True | ||
| in_channels: 8 | ||
| out_channels: 4 | ||
| model_channels: 320 | ||
| attention_resolutions: [4, 2, 1] | ||
| num_res_blocks: 2 | ||
| channel_mult: [1, 2, 4, 4] | ||
| num_head_channels: 64 | ||
| use_linear_in_transformer: True | ||
| transformer_depth: 1 | ||
| context_dim: 1024 | ||
| spatial_transformer_attn_type: softmax-xformers | ||
| extra_ff_mix_layer: True | ||
| use_spatial_context: True | ||
| merge_strategy: learned_with_images | ||
| video_kernel_size: [3, 1, 1] | ||
|
|
||
| conditioner_config: | ||
| target: sgm.modules.GeneralConditioner | ||
| params: | ||
| emb_models: | ||
| - input_key: cond_frames_without_noise | ||
| is_trainable: False | ||
| target: sgm.modules.encoders.modules.FrozenOpenCLIPImagePredictionEmbedder | ||
| params: | ||
| n_cond_frames: 1 | ||
| n_copies: 1 | ||
| open_clip_embedding_config: | ||
| target: sgm.modules.encoders.modules.FrozenOpenCLIPImageEmbedder | ||
| params: | ||
| freeze: True | ||
|
|
||
| - input_key: cond_frames | ||
| is_trainable: False | ||
| target: sgm.modules.encoders.modules.VideoPredictionEmbedderWithEncoder | ||
| params: | ||
| disable_encoder_autocast: True | ||
| n_cond_frames: 1 | ||
| n_copies: 1 | ||
| is_ae: True | ||
| encoder_config: | ||
| target: sgm.models.autoencoder.AutoencoderKLModeOnly | ||
| params: | ||
| embed_dim: 4 | ||
| monitor: val/rec_loss | ||
| ddconfig: | ||
| attn_type: vanilla-xformers | ||
| double_z: True | ||
| z_channels: 4 | ||
| resolution: 256 | ||
| in_channels: 3 | ||
| out_ch: 3 | ||
| ch: 128 | ||
| ch_mult: [1, 2, 4, 4] | ||
| num_res_blocks: 2 | ||
| attn_resolutions: [] | ||
| dropout: 0.0 | ||
| lossconfig: | ||
| target: torch.nn.Identity | ||
|
|
||
| - input_key: cond_aug | ||
| is_trainable: False | ||
| target: sgm.modules.encoders.modules.ConcatTimestepEmbedderND | ||
| params: | ||
| outdim: 256 | ||
|
|
||
| - input_key: polars_rad | ||
| is_trainable: False | ||
| target: sgm.modules.encoders.modules.ConcatTimestepEmbedderND | ||
| params: | ||
| outdim: 512 | ||
|
|
||
| - input_key: azimuths_rad | ||
| is_trainable: False | ||
| target: sgm.modules.encoders.modules.ConcatTimestepEmbedderND | ||
| params: | ||
| outdim: 512 | ||
|
|
||
| first_stage_config: | ||
| target: sgm.models.autoencoder.AutoencodingEngine | ||
| params: | ||
| loss_config: | ||
| target: torch.nn.Identity | ||
| regularizer_config: | ||
| target: sgm.modules.autoencoding.regularizers.DiagonalGaussianRegularizer | ||
| encoder_config: | ||
| target: torch.nn.Identity | ||
| decoder_config: | ||
| target: sgm.modules.diffusionmodules.model.Decoder | ||
| params: | ||
| attn_type: vanilla-xformers | ||
| double_z: True | ||
| z_channels: 4 | ||
| resolution: 256 | ||
| in_channels: 3 | ||
| out_ch: 3 | ||
| ch: 128 | ||
| ch_mult: [ 1, 2, 4, 4 ] | ||
| num_res_blocks: 2 | ||
| attn_resolutions: [ ] | ||
| dropout: 0.0 |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,107 @@ | ||
| model: | ||
| target: sgm.models.diffusion.DiffusionEngine | ||
| params: | ||
| scale_factor: 0.18215 | ||
| disable_first_stage_autocast: True | ||
|
|
||
| denoiser_config: | ||
| target: sgm.modules.diffusionmodules.denoiser.Denoiser | ||
| params: | ||
| scaling_config: | ||
| target: sgm.modules.diffusionmodules.denoiser_scaling.VScalingWithEDMcNoise | ||
|
|
||
| network_config: | ||
| target: sgm.modules.diffusionmodules.video_model.VideoUNet | ||
| params: | ||
| adm_in_channels: 256 | ||
| num_classes: sequential | ||
| use_checkpoint: True | ||
| in_channels: 8 | ||
| out_channels: 4 | ||
| model_channels: 320 | ||
| attention_resolutions: [4, 2, 1] | ||
| num_res_blocks: 2 | ||
| channel_mult: [1, 2, 4, 4] | ||
| num_head_channels: 64 | ||
| use_linear_in_transformer: True | ||
| transformer_depth: 1 | ||
| context_dim: 1024 | ||
| spatial_transformer_attn_type: softmax-xformers | ||
| extra_ff_mix_layer: True | ||
| use_spatial_context: True | ||
| merge_strategy: learned_with_images | ||
| video_kernel_size: [3, 1, 1] | ||
|
|
||
| conditioner_config: | ||
| target: sgm.modules.GeneralConditioner | ||
| params: | ||
| emb_models: | ||
| - input_key: cond_frames_without_noise | ||
| is_trainable: False | ||
| target: sgm.modules.encoders.modules.FrozenOpenCLIPImagePredictionEmbedder | ||
| params: | ||
| n_cond_frames: 1 | ||
| n_copies: 1 | ||
| open_clip_embedding_config: | ||
| target: sgm.modules.encoders.modules.FrozenOpenCLIPImageEmbedder | ||
| params: | ||
| freeze: True | ||
|
|
||
| - input_key: cond_frames | ||
| is_trainable: False | ||
| target: sgm.modules.encoders.modules.VideoPredictionEmbedderWithEncoder | ||
| params: | ||
| disable_encoder_autocast: True | ||
| n_cond_frames: 1 | ||
| n_copies: 1 | ||
| is_ae: True | ||
| encoder_config: | ||
| target: sgm.models.autoencoder.AutoencoderKLModeOnly | ||
| params: | ||
| embed_dim: 4 | ||
| monitor: val/rec_loss | ||
| ddconfig: | ||
| attn_type: vanilla-xformers | ||
| double_z: True | ||
| z_channels: 4 | ||
| resolution: 256 | ||
| in_channels: 3 | ||
| out_ch: 3 | ||
| ch: 128 | ||
| ch_mult: [1, 2, 4, 4] | ||
| num_res_blocks: 2 | ||
| attn_resolutions: [] | ||
| dropout: 0.0 | ||
| lossconfig: | ||
| target: torch.nn.Identity | ||
|
|
||
| - input_key: cond_aug | ||
| is_trainable: False | ||
| target: sgm.modules.encoders.modules.ConcatTimestepEmbedderND | ||
| params: | ||
| outdim: 256 | ||
|
|
||
| first_stage_config: | ||
| target: sgm.models.autoencoder.AutoencodingEngine | ||
| params: | ||
| loss_config: | ||
| target: torch.nn.Identity | ||
| regularizer_config: | ||
| target: sgm.modules.autoencoding.regularizers.DiagonalGaussianRegularizer | ||
| encoder_config: | ||
| target: torch.nn.Identity | ||
| decoder_config: | ||
| target: sgm.modules.autoencoding.temporal_ae.VideoDecoder | ||
| params: | ||
| attn_type: vanilla | ||
| double_z: True | ||
| z_channels: 4 | ||
| resolution: 256 | ||
| in_channels: 3 | ||
| out_ch: 3 | ||
| ch: 128 | ||
| ch_mult: [1, 2, 4, 4] | ||
| num_res_blocks: 2 | ||
| attn_resolutions: [] | ||
| dropout: 0.0 | ||
| video_kernel_size: [3, 1, 1] |
Oops, something went wrong.