Lincoln Stein
9df0980c46
parallel processing working on single-GPU, not tested on multi
2024-04-01 00:07:47 -04:00
Lincoln Stein
83356ec74c
fix merge conflicts
2024-03-31 17:04:57 -04:00
Lincoln Stein
eaa2c68693
remove vram_cache and don't move VRAM models back into CPU
2024-03-31 16:37:13 -04:00
psychedelicious
5b016bf376
fix(nodes): esrgan model name typo
2024-03-22 02:22:19 -07:00
blessedcoolant
eafc85cfe3
feat: Add Mask from ID Node
2024-03-22 06:23:51 +05:30
psychedelicious
c36d12a50f
feat: adaptation of Lineart Anime processor
...
Adapted from https://github.com/huggingface/controlnet_aux
2024-03-21 07:02:57 -07:00
psychedelicious
c7f8fe4d5e
feat: adaptation of Lineart processor
...
Adapted from https://github.com/huggingface/controlnet_aux
2024-03-21 07:02:57 -07:00
psychedelicious
ffb41c3616
feat: adaptation of HED processor
...
Adapted from controlnet repo
2024-03-21 07:02:57 -07:00
psychedelicious
611006b692
feat: adaptation of Canny processor
...
Adapted from controlnet processors package
fix: do final resize in canny processor
canny
2024-03-21 07:02:57 -07:00
psychedelicious
01d8ab04a5
feat(nodes): add missing detect_resolution to processors
...
Some processors, like Canny, didn't use `detect_resolution`. The resultant control images were then resized by the processors from 512x512 to the desired dimensions. The result is that the control images are the right size, but very low quality.
Using detect_resolution fixes this.
2024-03-21 07:02:57 -07:00
psychedelicious
fabef8b45b
feat(mm): download upscaling & lama models as they are requested
2024-03-20 15:05:25 +11:00
psychedelicious
29b04b7e83
chore: bump nodes versions
...
Bump all nodes in prep for v4.0.0.
2024-03-20 10:28:07 +11:00
maryhipp
820614e4d8
ruff
2024-03-19 21:59:51 +11:00
maryhipp
4e9207a10b
fix(worker): remove resolution from zoe as it seems to break it
2024-03-19 21:59:51 +11:00
maryhipp
ed0f9f7d66
feat(worker): add image_resolution as option for all cnet procesors
2024-03-19 21:59:51 +11:00
psychedelicious
dedce2d896
fix(config): remove unnecessary resolve on config path
2024-03-19 09:24:28 +11:00
psychedelicious
897fe497dc
fix(config): use new get_config across the app, use correct settings
2024-03-19 09:24:28 +11:00
Lincoln Stein
7387b0bdc9
install missing clip_vision encoders if required by an ip adapter ( #5982 )
...
Co-authored-by: Lincoln Stein <lstein@gmail.com>
2024-03-18 02:19:53 +00:00
psychedelicious
297408d67e
feat(nodes): add control adapter processed images to metadata
...
In the client, a controlnet or t2i adapter has two images:
- The source control image: the image the user selected (required)
- The processed control image: the user's image after we've processed it (optional)
The processed image is optional because a user may provide a pre-processed image.
We only actually use one of these images when building the graph, and until this change, we only stored one of the in image metadata. This created a situation where only a processed image was stored in metadata - say, a canny edge map - and the user-selected image wasn't provided.
By adding the processed image to metadata, we can recall both the control image and optional processed image.
This commit is followed by a UI-facing changes to support the change.
2024-03-14 12:34:03 -07:00
psychedelicious
4492bedd19
tidy(nodes): use ModelIdentifierField for model metadata
...
Until recently, this had a different shape than the ModelMetadataField. They are now the same, so we can re-use the ModelIdentifierField.
2024-03-14 10:53:57 +11:00
blessedcoolant
af660163ca
chore: cleanup DepthAnything code
2024-03-13 20:35:52 +05:30
Brandon Rising
8d2a4db902
Found another instance of expecting a mid_block on the decoder in a vae
2024-03-12 12:11:38 -04:00
Brandon Rising
7b393656de
Update l2i invoke and seamless to support AutoencoderTiny, remove attention processors if no mid_block is detected
2024-03-12 12:00:24 -04:00
psychedelicious
8c2ff794d5
fix(nodes): ip adapter uses valid ModelIdentifierField for image encoder model
...
- Add class method to `ModelIdentifierField` to construct the field from a model config
- Use this to construct a valid IP adapter model field
2024-03-10 17:28:58 -05:00
Ryan Dick
145bb45858
Remove dead code related to an old symmetry feature.
2024-03-10 00:13:18 -06:00
dunkeroni
631e789195
fix(canvas): create masked latents when None
2024-03-10 11:58:41 +11:00
psychedelicious
92b0d13d0e
feat(nodes): "ModelField" -> "ModelIdentifierField", add hash/name/base/type
2024-03-10 11:03:38 +11:00
psychedelicious
effbd8a1ba
chore: ruff
2024-03-08 11:10:44 -05:00
psychedelicious
ddde355b09
fix(mm): add ui_type to model fields
...
Recently the schema for models was changed to a generic `ModelField`, and the UI was unable to derive the type of those fields. This didn't affect functionality, but it did break the styling of handles.
Add `ui_type` to the affected fields and update the UI to use the correct capitalizations.
2024-03-08 11:10:44 -05:00
psychedelicious
132790eebe
tidy(nodes): use canonical capitalizations
2024-03-07 10:56:59 +11:00
psychedelicious
528ac5dd25
refactor(nodes): model identifiers
...
- All models are identified by a key and optionally a submodel type via new model `ModelField`. Previously, a few model types had their own class, but not all of them. This inconsistency just added complexity without any benefit.
- Update all invocation to use the new format.
- In the node API, models are loaded by key or an instance of `ModelField` as a convenience.
- Add an enriched model schema for metadata. It includes key, hash, name, base and type.
2024-03-07 10:56:59 +11:00
psychedelicious
7c9128b253
tidy(mm): use canonical capitalization for all model-related enums, classes
...
For example, "Lora" -> "LoRA", "Vae" -> "VAE".
2024-03-05 23:50:19 +11:00
dunkeroni
735857479d
fix(canvas): use corrected mask for pasteback
2024-03-03 12:58:47 -05:00
Ryan Dick
cc45007dc4
Remove unused code for attention map saving.
2024-03-02 08:25:41 -05:00
Ryan Dick
ad96857e0f
Fix avoid storing extra conditioning info in two places.
2024-03-01 15:12:03 -05:00
psychedelicious
0b0128647b
feat(nodes): revise model load API args
2024-03-01 10:42:33 +11:00
blessedcoolant
ae34bcfbc0
fix: Assertion issue with SDXL Compel
2024-03-01 10:42:33 +11:00
Brandon Rising
f475b78734
Ruff check
2024-03-01 10:42:33 +11:00
Brandon Rising
ca9b815c89
Extract TI loading logic into util, disallow it from ever failing a generation
2024-03-01 10:42:33 +11:00
Brandon Rising
8efd4284e9
Fix one last reference to the uncasted model
2024-03-01 10:42:33 +11:00
Brandon Rising
5922cee541
Allow TIs to be either a key or a name in the prompt during our transition to using keys
2024-03-01 10:42:33 +11:00
psychedelicious
80697a71de
feat(nodes): update LoRAMetadataItem model
...
LoRA model now at under `model` not `lora.
2024-03-01 10:42:33 +11:00
psychedelicious
82249cc634
tidy(nodes): rename canvas paste back
2024-03-01 10:42:33 +11:00
blessedcoolant
cc82ce820a
fix: outpaint result not getting pasted back correctly
2024-03-01 10:42:33 +11:00
blessedcoolant
8e1fbd6ed1
fix: lint errors
2024-03-01 10:42:33 +11:00
blessedcoolant
68d79c002d
canvas: improve paste back (or try to)
2024-03-01 10:42:33 +11:00
dunkeroni
30a374a70f
chore: typing
2024-03-01 10:42:33 +11:00
dunkeroni
07dde92664
chore: typing fix
2024-03-01 10:42:33 +11:00
dunkeroni
06cc57d82a
feat(nodes): added gradient mask node
2024-03-01 10:42:33 +11:00
psychedelicious
34f3a39cc9
fix(nodes): fix TI loading
2024-03-01 10:42:33 +11:00
