Mirrored_Repos/InvokeAI
1
0
Fork 0
mirror of https://github.com/invoke-ai/InvokeAI synced 2024-08-30 20:32:17 +00:00
Code Issues Packages Projects Releases Wiki Activity
7,647 Commits 274 Branches 239 Tags 397 MiB
bdfdf854fc
Commit Graph

609 Commits

Author SHA1 Message Date
Sergey Borisov
bf0dfcac2f Add inapint mask field class 2023-08-17 19:19:07 +03:00
psychedelicious
43b30355e4 feat: make primitive node titles consistent 2023-08-16 23:04:36 +10:00
psychedelicious
2b7dd3e236 feat: add missing primitive collections 
- add missing primitive collections
- remove `Seed` and `LoRAField` (they don't exist)
 2023-08-16 09:54:38 +10:00
psychedelicious
fa884134d9 feat: rename ui_type_hint to ui_type 
Just a bit more succinct while not losing any clarity.
 2023-08-16 09:54:38 +10:00
psychedelicious
c48fd9c083 feat(nodes): refactor parameter/primitive nodes 
Refine concept of "parameter" nodes to "primitives":
- integer
- float
- string
- boolean
- image
- latents
- conditioning
- color

Each primitive has:
- A field definition, if it is not already python primitive value. The field is how this primitive value is passed between nodes. Collections are lists of the field in node definitions. ex: `ImageField` & `list[ImageField]`
- A single output class. ex: `ImageOutput`
- A collection output class. ex: `ImageCollectionOutput`
- A node, which functions to load or pass on the primitive value. ex: `ImageInvocation` (in this case, `ImageInvocation` replaces `LoadImage`)

Plus a number of related changes:
- Reorganize these into `primitives.py`
- Update all nodes and logic to use primitives
- Consolidate "prompt" outputs into "string" & "mask" into "image" (there's no reason for these to be different, the function identically)
- Update default graphs & tests
- Regen frontend types & minor frontend tidy related to changes
 2023-08-16 09:54:38 +10:00
psychedelicious
f49fc7fb55 feat: node editor 
squashed rebase on main after backendd refactor
 2023-08-16 09:54:38 +10:00
psychedelicious
9d3cd85bdd chore: black 2023-08-14 13:02:33 +10:00
Sergey Borisov
409e5d01ba Fix cpu_only schedulers(unipc) 2023-08-14 05:14:05 +03:00
Sergey Borisov
d63bb39475 Make dpmpp_sde(_k) use not random seed 2023-08-14 00:24:38 +03:00
Sergey Borisov
59ba9fc0f6 Flip bits in seed for sde/ancestral schedulers to have different noise from initial 2023-08-13 19:50:16 +03:00
Sergey Borisov
6e0beb1ed4 Fixes for second order scheduler timesteps 2023-08-13 19:31:47 +03:00
Sergey Borisov
94636ddb03 Fix empty prompt handling 2023-08-13 19:31:14 +03:00
blessedcoolant
499e89d6f6 feat: Add SDXL Negative Aesthetic Score 2023-08-14 04:02:36 +12:00
blessedcoolant
561951ad98 chore: Black linting 2023-08-13 21:28:39 +12:00
blessedcoolant
f296e5c41e wip: Remove MaskBlur / Adjust color correction 2023-08-12 20:54:30 +12:00
blessedcoolant
6034fa12de feat: Add Mask Blur node 2023-08-12 16:20:58 +12:00
Sergey Borisov
ce3675fc14 Apply denoising_start/end according on timestep value 2023-08-12 03:19:49 +03:00
blessedcoolant
f343ab0302 wip: Port Outpainting to new backend 2023-08-12 06:15:59 +12:00
Sergey Borisov
f3ae52ff97 Fix error at high denoising_start, fix unipc(cpu_only) 2023-08-11 15:46:16 +03:00
blessedcoolant
7c0023ad9e feat: Remove TextToLatents / Rename Latents To Latents -> DenoiseLatents 2023-08-11 22:20:37 +12:00
blessedcoolant
231e665675 Merge branch 'main' into feat/refactor_generation_backend 2023-08-11 20:53:38 +12:00
Sergey Borisov
2ef6a8995b Temporary force set vae to same precision as unet 2023-08-10 18:01:58 -04:00
psychedelicious
49cce1eec6 feat: add app_version to image metadata 2023-08-10 14:22:39 +10:00
Sergey Borisov
e9ec5ab85c Apply requested changes 
Co-Authored-By: psychedelicious <4822129+psychedelicious@users.noreply.github.com>
 2023-08-10 06:19:22 +03:00
Sergey Borisov
17fed1c870 Fix merge conflict errors 2023-08-10 05:03:33 +03:00
Sergey Borisov
ade78b9591 Merge branch 'main' into feat/refactor_generation_backend 2023-08-10 04:32:16 +03:00
Martin Kristiansen
c8fbaf54b6 Add self.min, not self.max 2023-08-10 09:59:22 +10:00
Sergey Borisov
b4a74f6523 Add MaskEdge and ColorCorrect nodes 
Co-Authored-By: Kent Keirsey <31807370+hipsterusername@users.noreply.github.com>
 2023-08-08 23:57:02 +03:00
Sergey Borisov
f7aec3b934 Move conditioning class to backend 2023-08-08 23:33:52 +03:00
Sergey Borisov
a7e44678fb Remove legacy/unused code 2023-08-08 20:49:01 +03:00
Sergey Borisov
da0184a786 Invert mask, fix l2l on no mask conntected, remove zeroing latents on zero start 2023-08-08 20:01:49 +03:00
Sergey Borisov
96b7248051 Add mask to l2l 2023-08-08 18:50:36 +03:00
Sergey Borisov
5f29526a8e Add seed to latents field 2023-08-08 04:00:33 +03:00
Sergey Borisov
492bfe002a Remove sdxl t2l/l2l nodes 2023-08-08 03:38:42 +03:00
Kevin Turner
25c669b1d6 Merge remote-tracking branch 'origin/main' into refactor/remove_unused_pipeline_methods 2023-08-07 13:03:10 -07:00
Sergey Borisov
1db2c93f75 Fix preview, inpaint 2023-08-07 21:27:32 +03:00
Sergey Borisov
2539e26c18 Apply denoising_start/end, add torch-sdp to memory effictiend attention func 2023-08-07 19:57:11 +03:00
Sergey Borisov
b0738b7f70 Fixes, zero tensor for empty negative prompt, remove raw prompt node 2023-08-07 18:37:06 +03:00
Jonathan
ae17d01e1d
Fix hue adjustment (#4182) 
* Fix hue adjustment

Hue adjustment wasn't working correctly because color channels got swapped. This has now been fixed and we're using PIL rather than cv2 to do the RGBA->HSV->RGBA conversion. The range of hue adjustment is also the more typical 0..360 degrees.
 2023-08-06 23:23:51 +00:00
Kevin Turner
6487e7d906 refactor(diffusers_pipeline): remove unused ModelGroup 🚮 
orphaned since #3550 removed the LazilyLoadedModelGroup code, probably unused since ModelCache took over responsibility for sequential offload somewhere around #3335.
 2023-08-05 21:50:52 -07:00
Kevin Turner
77033eabd3 refactor(diffusers_pipeline): remove unused precision 🚮 2023-08-05 21:50:52 -07:00
Sergey Borisov
9aaf67c5b4 wip 2023-08-06 05:05:25 +03:00
psychedelicious
b8b589c150 fix(nodes): fix hsl nodes rebase conflict 2023-08-06 09:57:49 +10:00
Kent Keirsey
d93900a8de Added HSL Nodes 2023-08-06 09:57:49 +10:00
Lincoln Stein
6ad565d84c folded in changes from 4099 2023-08-04 18:24:47 -04:00
Sergey Borisov
04229082d6 Provide ti name from model manager, not from ti itself 2023-08-04 18:24:47 -04:00
StAlKeR7779
0d3c27f46c Fix typo 
Co-authored-by: Ryan Dick <ryanjdick3@gmail.com>
 2023-08-04 11:44:56 -04:00
Sergey Borisov
cff91f06d3 Add lora apply in sdxl l2l node 2023-08-04 11:44:56 -04:00
Lincoln Stein
1d5d187ba1 model probe detects sdxl lora models 2023-08-04 11:44:56 -04:00
Sergey Borisov
1ac14a1e43 add sdxl lora support 2023-08-04 11:44:56 -04:00
psychedelicious
bf94412d14 feat: add multi-select to gallery 
multi-select actions include:
- drag to board to move all to that board
- right click to add all to board or delete all

backend changes:
- add routes for changing board for list of image names, deleting list of images
- change image-specific routes to `images/i/{image_name}` to not clobber other routes (like `images/upload`, `images/delete`)
- subclass pydantic `BaseModel` as `BaseModelExcludeNull`, which excludes null values when calling `dict()` on the model. this fixes inconsistent types related to JSON parsing null values into `null` instead of `undefined`
- remove `board_id` from `remove_image_from_board`

frontend changes:
- multi-selection stuff uses `ImageDTO[]` as payloads, for dnd and other mutations. this gives us access to image `board_id`s when hitting routes, and enables efficient cache updates.
- consolidate change board and delete image modals to handle single and multiples
- board totals are now re-fetched on mutation and not kept in sync manually - was way too tedious to do this
- fixed warning about nested `<p>` elements
- closes #4088 , need to handle case when `autoAddBoardId` is `"none"`
- add option to show gallery image delete button on every gallery image

frontend refactors/organisation:
- make typegen script js instead of ts
- enable `noUncheckedIndexedAccess` to help avoid bugs when indexing into arrays, many small changes needed to satisfy TS after this
- move all image-related endpoints into `endpoints/images.ts`, its a big file now, but this fixes a number of circular dependency issues that were otherwise felt impossible to resolve
 2023-08-03 11:46:59 +10:00
Lincoln Stein
e7d9e552a7
Merge branch 'main' into feat_compel_and 2023-08-01 07:20:25 -04:00
Brandon Rising
f5ac73b091 Merge branch 'main' into feat/onnx 2023-07-31 10:58:40 -04:00
Damian Stewart
d2c55dc011 enable .and() syntax and long prompts 2023-07-30 14:20:59 +02:00
blessedcoolant
6d82a1019a fix: Black linting 2023-07-29 17:34:43 +12:00
blessedcoolant
bee678fdd1 fix: Prompt Node using incorrect output type 2023-07-29 17:12:25 +12:00
blessedcoolant
aae1670080 fix: Incorrect Prompt Node output type 2023-07-29 16:04:19 +12:00
blessedcoolant
8a0e1b6cfc feat: Create Prompt Input Node 2023-07-29 15:52:37 +12:00
Brandon Rising
6ca0c38ee3 Merge branch 'main' into feat/onnx 2023-07-28 22:06:28 -04:00
ZachNagengast
2164674b01 Black format 2023-07-28 07:49:29 -07:00
Lincoln Stein
e342ca872f fix to work on non-MPS systems 2023-07-28 10:27:49 -04:00
Brandon Rising
da751da3dd Merge branch 'main' into feat/onnx 2023-07-28 09:59:35 -04:00
Brandon Rising
2b7b3dd4ba Run python black 2023-07-28 09:46:44 -04:00
ZachNagengast
ddaf753f7b Merge branch 'set-timestep-mps-fix' of ssh://github.com/ZachNagengast/InvokeAI into set-timestep-mps-fix 2023-07-27 23:40:44 -07:00
ZachNagengast
e6d14c708c Fix variable name 2023-07-27 23:40:23 -07:00
Millun Atluri
7f81a95b20
Merge branch 'main' into set-timestep-mps-fix 2023-07-28 16:12:07 +10:00
ZachNagengast
aa1f827271 Fix unet_info location, can have no device prop 2023-07-27 14:47:09 -07:00
ZachNagengast
6edeb4e072 Pass device to set_timestep to avoid float64 error 2023-07-27 12:52:18 -07:00
Brandon Rising
1ea9ba84f5 Release session if applying ti or lora 2023-07-27 15:20:38 -04:00
Lincoln Stein
0d8f9cbe55 resolved conflicts with main 2023-07-27 15:11:25 -04:00
Lincoln Stein
fd75a1dd10 reformat with black 2023-07-27 15:01:00 -04:00
Brandon Rising
59716938bf Remove TensorRT support at the current time until we validate it works, remove time step recorder 2023-07-27 11:18:50 -04:00
Martin Kristiansen
218b6d0546 Apply black 2023-07-27 10:54:01 -04:00
Brandon Rising
f7bb4c3f05 Remove more files no longer needed in main 2023-07-27 10:49:43 -04:00
Brandon Rising
c16da75ac7 Merge branch 'main' into feat/onnx 2023-07-26 10:42:31 -04:00
Lincoln Stein
85ad5ef204 refactored code; added watermark and nsfw facilities to app config route 2023-07-26 15:27:04 +10:00
Lincoln Stein
5beb11f4e2 tweaks in response to psychedelicious review of PR 2023-07-26 15:27:04 +10:00
Lincoln Stein
bd43751323 update linear graphs to perform safety checking and watermarking 2023-07-26 15:27:04 +10:00
Lincoln Stein
e32cd794f7 add safetychecker and watermark nodes 2023-07-26 15:26:45 +10:00
psychedelicious
75b0507434 feat(nodes): change denoising start/end min/max to 0/1 2023-07-26 14:46:38 +10:00
psychedelicious
61291ea105 feat: sdxl metadata 
- update `CoreMetadata` class & `MetadataAccumulator` with fields for SDXL-specific metadata
- update the linear UI graphs to populate this metadata
 2023-07-26 14:46:38 +10:00
psychedelicious
840205496a feat(nodes): fix model load events on sdxl nodes 
they need the `context` to be provided to emit socket events
 2023-07-26 14:46:38 +10:00
psychedelicious
8e90f9024d feat(ui): remove isRefinerAvailable state, update refiner node 
We can derive `isRefinerAvailable` from the query result (eg are there any refiner models installed). This is a piece of server state, so by using the list models response directly, we can avoid needing to manually keep the client in sync with the server.

Created a `useIsRefinerAvailable()` hook to return this boolean wherever it is needed.

Also updated the main models & refiner models endpoints to only return the appropriate models. Now we don't need to filter the data on these endpoints.
 2023-07-26 14:46:38 +10:00
Lincoln Stein
fc4e104c61 tested on 3.11 and 3.10 2023-07-24 17:13:32 -04:00
psychedelicious
66cdeba8a1 fix(nodes): fix seed modulus operation 
This was incorect and resulted in the max seed being one less than intended.
 2023-07-24 16:44:32 +10:00
Brandon Rising
78750042f5 Pass in dim overrides 2023-07-21 12:16:24 -04:00
Brandon Rising
ce08aa350c Allow controlnet passthrough for now 2023-07-20 14:14:04 -04:00
Brandon Rising
4e90376d11 Allow passing in of precision, use available providers if none provided 2023-07-20 13:15:45 -04:00
Sergey Borisov
4a0774b260 Use scale from vae 2023-07-20 18:54:51 +03:00
Sergey Borisov
29a590cced Add sdxl generation preview 2023-07-20 18:45:54 +03:00
user1
909f538fb5 Switching over to controlnet_utils prepare_control_image(), with added resize_mode. 2023-07-20 00:41:49 -07:00
user1
f2f49bd8d0 Added resize_mode param to ControlNet node 2023-07-20 00:41:49 -07:00
Martin Kristiansen
07c48b2fd1 Moving detected precision to DEFAULT_PRECISION constant 2023-07-19 11:55:37 -04:00
Martin Kristiansen
fface339ae Same fix for ImageToLatentsInvocation 2023-07-19 11:38:13 -04:00
Martin Kristiansen
2ec9dab595 Changing ImageToLatentsInvocation node to default to detected precision instead of fp16 2023-07-19 11:16:00 -04:00
mickr777
d8db618de0
import choose_torch_device from ...backend.util.devices 2023-07-19 16:43:02 +10:00
mickr777
19d67b29e7
Remove not needed text 2023-07-19 15:20:40 +10:00
mickr777
52e7e0b31b
Missing def choose_torch_device 2023-07-19 15:15:55 +10:00
Brandon Rising
ee7b36cea5 Merge branch 'main' into onnx-testing 2023-07-18 22:56:41 -04:00
Brandon Rising
487455ef2e Add model_type to the model state object 2023-07-18 22:40:27 -04:00
Lincoln Stein
893e199677
Merge branch 'main' into feat/ui/upscale 2023-07-18 19:18:55 -04:00
Brandon Rising
869f418b03 Setup onnx on linear text2image 2023-07-18 14:27:54 -04:00
Brandon Rising
35d5ef9118 Emit step completions 2023-07-18 12:35:07 -04:00
Sergey Borisov
3240f98f4e Rename clip1 to clip 2023-07-18 18:58:17 +03:00
blessedcoolant
7fde1f93ea fix: Missing context on string param node 2023-07-19 02:49:09 +12:00
blessedcoolant
3f1d5000c0 Merge branch 'main' into nodes-stuff 2023-07-19 02:37:50 +12:00
blessedcoolant
0c18c5d603 feat: Add titles and tags to all Nodes 2023-07-19 02:26:45 +12:00
blessedcoolant
7d49c727a0 feat: Add String Param & types to other params 2023-07-19 02:26:33 +12:00
StAlKeR7779
889b77d3d6
Merge branch 'main' into save_vram 2023-07-18 16:55:48 +03:00
Sergey Borisov
fbbc4b3f69 Fixes 2023-07-18 16:51:16 +03:00
Sergey Borisov
bc11296a5e Disable lazy offloading on disabled vram cache, move resulted tensors to cpu(to not stack vram tensors in cache), fix - text encoder not freed(detach) 2023-07-18 16:20:25 +03:00
psychedelicious
c19d48abd0 fix(nodes): fix inpaint cond logic for new compel version 
thanks @StAlKeR7779
 2023-07-18 22:39:34 +10:00
psychedelicious
42c440c73f
Merge branch 'main' into feat/ui/upscale 2023-07-18 22:08:02 +10:00
Lincoln Stein
9c3c393b84 merge with main 2023-07-18 07:00:55 -04:00
psychedelicious
56098f370c feat(nodes): add RealESRGAN_x2plus.pth, update upscale nodes 
- add `RealESRGAN_x2plus.pth` model to installer
- add `RealESRGAN_x2plus.pth` to `realesrgan` node
- rename `RealESRGAN` to `ESRGAN` in nodes
- make `scale_factor` optional in `img_scale` node
 2023-07-18 14:55:18 +10:00
Lincoln Stein
ef31837167 fix caption on sdxl raw prompt 2023-07-17 23:49:23 -04:00
blessedcoolant
13da881953 Merge branch 'main' into sdxl-support 2023-07-18 13:34:07 +12:00
Sergey Borisov
ada9b06e48 Implement compel prompt nodes for sdxl 2023-07-18 01:49:45 +03:00
Lincoln Stein
6ae10798b0
Merge branch 'main' into feat/model-events 2023-07-17 17:15:12 -04:00
Brandon Rising
bcce70fca6 Testing different session opts, added timings for testing 2023-07-17 16:27:33 -04:00
skunkworxdark
f767bf2330 use FileNotFoundError insted of "File path not found" 2023-07-17 05:49:09 -04:00
skunkworxdark
b1008af696 apply changes as suggested @psychedelicious in PR comments. 
- filename -> file_path
- pre and post prompt changed to optional
- clearer pre and post prompt descriptions
- handle pre and post prompt passed as None
- max_prompts defaults to 1 isted of 0 to avoid accidentally processing large prompt files with it set to 0 when adding a new node.
 2023-07-17 05:49:09 -04:00
skunkworxdark
956011066d Added class PromptsFromFileInvocation to prompt.py. A new PromptFromFile Custom node that reads prompts from a file one line per prompt and outputs them as a prompt collection. With inputs for filename, pre_prompt, post_prompt, start line number, and max_prompts 2023-07-17 05:49:09 -04:00
blessedcoolant
e039771d07 fix: Incorrect type on SDXL Model Loader 2023-07-17 21:47:41 +12:00
psychedelicious
3e2a948007
Merge branch 'main' into feat/model-events 2023-07-17 17:36:20 +10:00
psychedelicious
ba12849685 fix(nodes): fix some model load events not emitting 
Missed adding the `context` arg to them initially
 2023-07-17 17:16:55 +10:00
Lincoln Stein
cad3f96831 add model input to refiner 2023-07-16 12:38:04 -04:00
Lincoln Stein
6534288b75 refiner only has clip2 not clip 2023-07-16 12:36:38 -04:00
Lincoln Stein
0a2964d8c0 add differentiated sdxl and sdxl_refiner model loaders 2023-07-16 12:17:56 -04:00
Brandon Rising
91112167b1 Fix syntax err 2023-07-15 23:56:48 -04:00
Lincoln Stein
5206ddf9b2 truncate long prompts to avoid a crash with controlnet 2023-07-15 23:49:25 -04:00
Sergey Borisov
fe78a08e37 Fix sd1/2 models conditionings 2023-07-16 06:24:24 +03:00
Sergey Borisov
c9c2229917 Separate prompt to sdxl and sdxl-refiner, add denoising start-end fields, add l2l node(supports both sdxl and sdxl-refiner), add fp32 to vae encode 2023-07-16 06:00:37 +03:00
psychedelicious
5d59dd4b97 feat(nodes): use correctly-typed configuration service in upscale node 2023-07-16 10:54:52 +10:00
Lincoln Stein
ccbfa5d862 resolve conflicts 2023-07-15 19:47:50 -04:00
psychedelicious
7b6159f8d6 feat(nodes): emit model loading events 
- remove dependency on having access to a `node` during emits, would need a bit of additional args passed through the system and I don't think its necessary at this point. this also allowed us to drop an extraneous fetching/parsing of the session from db.
- provide the invocation context to all `get_model()` calls, so the events are able to be emitted
- test all model loading events in the app and confirm socket events are received
 2023-07-16 02:12:01 +10:00
psychedelicious
c7b547ea3e feat(nodes): remove references to restoration services 
- remove restoration services
- remove the restore faces nodes
- update tests
 2023-07-16 01:12:39 +10:00
psychedelicious
74ca87ac9e feat(nodes): add realesrgan node 2023-07-16 01:06:50 +10:00
psychedelicious
d270f21c85 feat(nodes): valid controlnet weights are -1 to 2 2023-07-15 19:56:44 +10:00
psychedelicious
ae72f372be fix(nodes): do not use hardcoded controlnet model 2023-07-15 19:56:44 +10:00
psychedelicious
29b2e59e65 fix(nodes): fix ref to ctx mgr service, missing import 2023-07-15 19:56:44 +10:00
psychedelicious
82fa39b531 feat(nodes): add controlnet nodes type hint 2023-07-15 19:56:44 +10:00
psychedelicious
788dcbde70 fix(nodes): add missing import 2023-07-15 19:56:44 +10:00
Sergey Borisov
6ab9a5e108 Draft 2023-07-15 19:56:44 +10:00
Sergey Borisov
7093e5d033 Pad conditionings using zeros and encoder_attention_mask 2023-07-15 00:52:54 +03:00
Brandon Rising
bd7b59910d Testing onnx in new ui updates 2023-07-14 14:24:15 -04:00
Jonathan
9960d7ca2a
Allow ImageResizeInvocation w/h to take inputs from other nodes (#3765) 2023-07-15 05:34:13 +12:00
psychedelicious
23c1a6b9d5 fix(nodes): make ResizeLatents w/h optional 
now you can connect to them in node editor
 2023-07-14 21:42:42 +10:00
Sergey Borisov
9fb0b0959f Make sde and ancestral schedulers reproducible 2023-07-14 05:25:09 +03:00
Sergey Borisov
d8f88c09ea Fix pink output on a lot of samplers 2023-07-14 05:00:33 +03:00
Brandon Rising
524888bf3b Merge branch 'main' into feat/onnx 2023-07-13 14:23:57 -04:00
blessedcoolant
430b9c291f fix: Loras not working correctly with Inpainting 2023-07-13 22:59:38 +12:00
psychedelicious
50bef87da7 feat(db,nodes,api): refactor metadata 
Metadata for the Linear UI is now sneakily provided via a `MetadataAccumulator` node, which the client populates / hooks up while building the graph.

Additionally, we provide the unexpanded graph with the metadata API response.

Both of these are embedded into the PNGs.

- Remove `metadata` from `ImageDTO`
- Split up the `images/` routes to accomodate this; metadata is only retrieved per-image
- `images/{image_name}` now gets the DTO
- `images/{image_name}/metadata` gets the new metadata
- `images/{image_name}/full` gets the full-sized image file
- Remove old metadata service
- Add `MetadataAccumulator` node, `CoreMetadataField`, hook up to `LatentsToImage` node
- Add `get_raw()` method to `ItemStorage`, retrieves the row from DB as a string, no pydantic parsing
- Update `images`related services to handle storing and retrieving the new metadata
- Add `get_metadata_graph_from_raw_session` which extracts the `graph` from `session` without needing to hydrate the session in pydantic, in preparation for providing it as metadata; also removes all references to the `MetadataAccumulator` node
 2023-07-13 15:40:05 +10:00
Lincoln Stein
75c5ce46bc merged SDXLModelLoader into ModelLoader invocation 2023-07-11 16:33:08 -04:00
Sergey Borisov
358ced6bab SDXL Prompt and t2l nodes draft, add fp32 to vae decode 2023-07-11 18:19:36 +03:00
Lincoln Stein
8e42502dfd partial implementation of SDXL model loader 2023-07-10 20:18:30 -04:00
Lincoln Stein
10d3bccf32
Mac MPS FP16 fixes (#3641) 
This PR is to allow FP16 precision to work on Macs with MPS. In
addition, it centralizes the torch fixes/workarounds required for MPS
into a new backend utility `mps_fixes.py`. This is conditionally
imported in `api_app.py`/`cli_app.py`.

Many MANY thanks to @StAlKeR7779 for patiently working to debug and fix
these issues.
 2023-07-07 17:43:23 -04:00
Eugene Brodsky
97b2ec58e2
Merge branch 'main' into release/invokeai-3-0-alpha 2023-07-07 14:18:12 -04:00
Sergey Borisov
a9e77675a8 Move clip skip to separate node 2023-07-06 17:39:49 +03:00
Sergey Borisov
04b57c408f Add clip skip option to prompt node 2023-07-06 16:09:40 +03:00
Lincoln Stein
83d3a043da merge latest changes from main 2023-07-05 19:15:53 -04:00
Lincoln Stein
021e1eca8e
Merge branch 'main' into mps-fp16-fixes 2023-07-05 13:19:52 -04:00
Sergey Borisov
0ac9dca926 Fix loading diffusers ti 2023-07-05 19:46:00 +03:00
Lincoln Stein
9edf78dd2e merge with main 2023-07-05 09:12:54 -04:00
gogurtenjoyer
ba7345deb4
Merge branch 'main' into mps-fp16-fixes 2023-07-05 07:38:41 -04:00
blessedcoolant
639d88afd6 revert: inference_mode to no_grad 2023-07-05 16:39:15 +12:00
blessedcoolant
c0501ed5c2 fix: Slow loading of Loras 
Co-Authored-By: StAlKeR7779 <7768370+StAlKeR7779@users.noreply.github.com>
 2023-07-05 12:47:34 +10:00
psychedelicious
08d428a5e7 feat(nodes): add lora field, update lora loader 2023-07-05 12:47:34 +10:00
gogurtenjoyer
233869b56a Mac MPS FP16 fixes 
This PR is to allow FP16 precision to work on Macs with MPS. In addition, it centralizes the torch fixes/workarounds
required for MPS into a new backend utility file `mps_fixes.py`. This is conditionally imported in `api_app.py`/`cli_app.py`.

Many MANY thanks to StAlKeR7779 for patiently working to debug and fix these issues.
 2023-07-04 18:10:53 -04:00
blessedcoolant
7e18814dd0 Add standard names for Model Loader Nodes 2023-07-04 14:35:06 +10:00
Lincoln Stein
a8a2209560 VAE loader is loading proper VAE. Unclear if it is changing the image 2023-07-04 14:35:06 +10:00
Lincoln Stein
fa8a5838d3 add vae lodaer 2023-07-04 14:35:06 +10:00
blessedcoolant
6c62f41f2e chore: Change PipelineModels to MainModels 2023-07-04 14:33:56 +10:00
Lincoln Stein
4d2c7806fc quash memory leak when compel invocation called 2023-07-03 14:12:35 -04:00
Lincoln Stein
3937428563 Merge branch 'release/invokeai-3-0-alpha' of github.com:invoke-ai/InvokeAI into release/invokeai-3-0-alpha 2023-07-03 14:11:28 -04:00
Lincoln Stein
ed86d0b708 Union[foo, None]=>Optional[foo] 2023-07-03 12:17:45 -04:00
psychedelicious
fa169b5517 feat(nodes): add ImageCollection node in prep for batch processing 2023-07-04 00:18:27 +10:00
Lincoln Stein
aae60b6142 quash memory leak when compel invocation called 2023-07-03 10:08:10 -04:00
Lincoln Stein
b79740d61d back out torch.no_grad() 2023-07-02 23:03:24 -04:00
Lincoln Stein
fa8ccd2a94 add no_grad() to compel node invoke() method 2023-07-02 18:20:16 -04:00
StAlKeR7779
ac46b129bf
Merge branch 'main' into feat/lora_model_patch 2023-06-28 22:43:58 +03:00
Lincoln Stein
72209d0cc3
Merge branch 'main' into lstein/installer-for-new-model-layout 2023-06-28 14:49:37 -04:00
blessedcoolant
75614bbba3
Merge branch 'main' into feat/nodes/cpu-noise 2023-06-28 18:22:08 +12:00
blessedcoolant
32883adf6e
Merge branch 'main' into feat/controlnet_extras 2023-06-28 17:36:21 +12:00
Kent Keirsey
fc322aa9f7 Update controlnet-aux to 0.0.6 and add LeReS 2023-06-27 23:45:47 -04:00
psychedelicious
73f63853ba fix(nodes): use context for logger in param_easing 2023-06-27 23:30:10 -04:00
psychedelicious
2e14528e4c feat(nodes): default to CPU noise 2023-06-27 13:57:31 +10:00
user1
10d2d85c83 Started to add ControlNet resize_crop and resize_fill options, but commented out, not ready to deploy yet. 2023-06-26 12:03:05 -07:00
user1
af566adf56 For MediapipeFace ControlNet preprocessor, if input image is RGBA format then convert to RGB (otherwise MediapipeFace image processing throws an error) 2023-06-26 04:29:43 -07:00
user1
873c18bc4b Added TileResampler ControlNet preprocessor node. 
Also fixes to SegmentAnything ControlNet preprocessor node.
 2023-06-26 04:27:26 -07:00
Eugene Brodsky
7b97639961
Merge branch 'main' into lstein/installer-for-new-model-layout 2023-06-26 01:24:30 -04:00
Sergey Borisov
5cebf67ee4 Apply lora by patching lora instead of hooks 2023-06-26 03:57:33 +03:00
user1
10e8389fa4 Commenting out LeReS ControlNet image preprocessor until release of controlnet_aux v0.0.6 (supported on controlnet_aux current main, but not on latest release v0.0.5) 2023-06-25 14:25:14 -07:00
user1
414a04774c Added LeReS ControlNet image preprocessor. 2023-06-25 14:19:55 -07:00
user1
b872e7a5e0 Simplifying ControlNet SAM preprocessor segmentation color mapping. 2023-06-25 12:54:48 -07:00
user1
de4064bdac Fixed problem with with non-reproducible results from ControlNet SegmentAnything preprocessor. Cause was controlnet_aux randomization of segmentation coloring, which seems to lead to some randomization of resulting images using ControlNet seg model. Switched to using deterministic ADE20K color palette instead, which solved the problem. 2023-06-25 12:38:17 -07:00
user1
10c3753d7f Added SAM preprocessor 2023-06-25 11:16:39 -07:00
user1
c5faffc18b Merge branch 'main' of github.com:invoke-ai/InvokeAI into feat/controlnet-control-modes 
Only "real" conflicts were in:
     invokeai/frontend/web/src/features/controlNet/components/ControlNet.tsx
     invokeai/frontend/web/src/features/controlNet/store/controlNetSlice.ts
 2023-06-24 17:05:57 -07:00
Lincoln Stein
d5f742620f Merge branch 'main' into lstein/installer-for-new-model-layout 2023-06-24 11:58:06 -04:00
Lincoln Stein
ba1371a88f rename ModelType.Pipeline to ModelType.Main 2023-06-24 11:45:49 -04:00