1580158
/
automatic
mirror of https://github.com/vladmandic/automatic
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity
5,149 Commits
3 Branches
29 Tags
109 MiB
Commit Graph

188 Commits (0a49961baaaa594313baffba7a000e4e98712c71)

Author SHA1 Message Date
Disty0 6ea6f2448e Revert loss scale for ipex 2023-06-19 23:59:18 +03:00
Disty0 71b1532bb0 Scale loss for ipex 2023-06-15 23:55:23 +03:00
Disty0 618097dac2 GradScaler patch for IPEX 2023-06-15 01:19:35 +03:00
Disty0 a9f66cb33e loss / 2 for ipex train 2023-06-14 12:18:08 +03:00
Vladimir Mandic cb307399dd jumbo merge 2023-06-13 11:59:56 -04:00
Disty0 0d101b9def Revert xpu.optimize in training 2023-06-13 15:36:52 +03:00
Disty0 c9e95bec3f Fix loss=nan 2023-06-12 06:13:18 +03:00
Vladimir Mandic 1595c7a11c minor fixes 2023-06-11 21:49:48 -04:00
Disty0 ab255b732b Remove unnecessary ipex code 2023-06-12 04:45:29 +03:00
Disty0 c9e58c9604 Fix train for IPEX 2023-06-12 00:21:32 +03:00
Vladimir Mandic 409c9d4c9d upstream ports 2023-06-10 18:23:28 -04:00
Vladimir Mandic aaa0d46286 update installer and add sd_model_dict 2023-06-07 13:26:21 -04:00
Disty0 3bef3e3eee Train patches for IPEX 2023-06-07 17:25:11 +03:00
Vladimir Mandic 5f1fd7bd66 update common ui 2023-05-29 13:43:03 -04:00
Vladimir Mandic efd3810860 diffusers merge 2023-05-26 22:42:03 -04:00
Vladimir Mandic 9285326c6d fix tqdm 2023-05-25 07:53:25 -04:00
Vladimir Mandic a64bb4375a minor upadtes 2023-05-22 10:50:59 -04:00
Vladimir Mandic 42280ef804 add theme mode toggle 2023-05-19 13:24:40 -04:00
Vladimir Mandic 0ccda9bc8b jumbo patch 2023-05-17 14:15:55 -04:00
Vladimir Mandic fe496f4ebc add train preprocess options 2023-05-05 09:06:06 -04:00
Disty0 8171d57c36 Remove unnecessary IPEX imports 2023-05-04 02:34:34 +03:00
Vladimir Mandic 6f976c358f optimize model load 2023-05-02 21:30:34 -04:00
Vladimir Mandic 7a083d322b merge commits 2023-05-02 15:06:06 -04:00
Vladimir Mandic cb4cff3929 redesign logging 2023-05-02 13:57:16 -04:00
Vladimir Mandic 4dc5941912 fix embedding logging 2023-04-30 21:01:49 -04:00
Disty0 68fc95b2e1 Merge remote-tracking branch 'upstream/master' 2023-04-30 18:28:22 +03:00
Disty0 de8d0bef9f More patches and Import IPEX after Torch 2023-04-30 18:19:37 +03:00
Vladimir Mandic b23b6a6e2c update ti folders 2023-04-30 08:55:47 -04:00
Disty0 b075d3c8fd Intel ARC Support 2023-04-30 15:13:56 +03:00
David Pina 2220316920
Fix typo that prevents training Textual Inversion 
There was a small typo in line 529.

shared.ops.embeddings_train_log caused an attribute not found exception when training TIs.
 2023-04-29 22:41:24 +02:00
Vladimir Mandic 21ff7bad67 configurable train log 2023-04-28 09:42:19 -04:00
Vladimir Mandic 11fa3aff6d ti fixes 2023-04-25 09:21:38 -04:00
Vladimir Mandic cf277e7326 fix dtype logic 2023-04-21 15:04:05 -04:00
Vladimir Mandic fd51bb90d0 enable quick launch 2023-04-15 11:51:58 -04:00
Vladimir Mandic 2ece9782e4 handle duplicate extensions and redo exception handler 2023-04-14 09:57:53 -04:00
Vladimir Mandic 81b8294e93 switch cmdflags to settings 2023-04-12 10:40:11 -04:00
Vladimir Mandic f181885f0c
Merge pull request #57 from AUTOMATIC1111/master 
merge from upstream
 2023-03-25 08:47:00 -04:00
butaixianran 803d44c474
Fix None type error for TI module 
When user using model_name.png as a preview image, textural_inversion.py still treat it as an embeding, and didn't handle its error, just let python throw out an None type error like following:
```bash
  File "D:\Work\Dev\AI\stable-diffusion-webui\modules\textual_inversion\textual_inversion.py", line 155, in load_from_file
    name = data.get('name', name)
AttributeError: 'NoneType' object has no attribute 'get'
```

With just a simple `if data:` checking as following, there will be no error, breaks nothing, and now this module can works fine with user's preview images.
Old code:  
```python
                data = extract_image_data_embed(embed_image)
                name = data.get('name', name)
```
New code:  
```python
                data = extract_image_data_embed(embed_image)
                if data:
                    name = data.get('name', name)
                else:
                    # if data is None, means this is not an embeding, just a preview image
                    return
```

Also, since there is no more errors on textual inversion module, from now on, extra network can set "model_name.png" as preview image for embedings.
 2023-03-25 02:05:00 +08:00
Vladimir Mandic f275e43eb9 update torch load and external repos 2023-03-20 10:10:56 -04:00
Vladimir Mandic f6679fcc77 add global exception handler 2023-03-17 10:08:07 -04:00
Vladimir Mandic 54ab13502f
Merge pull request #37 from AUTOMATIC1111/master 
sync forks
 2023-02-19 10:11:25 -05:00
Shondoit edb10092de Add ability to choose using weighted loss or not 2023-02-15 10:03:59 +01:00
Shondoit bc50936745 Call weighted_forward during training 2023-02-15 10:03:59 +01:00
Vladimir Mandic e92b66e2ea update version 2023-01-29 13:44:33 -05:00
AUTOMATIC aa6e55e001 do not display the message for TI unless the list of loaded embeddings changed 2023-01-29 11:53:05 +03:00
Alex "mcmonkey" Goodwin e179b6098a allow symlinks in the textual inversion embeddings folder 2023-01-25 08:48:40 -08:00
AUTOMATIC 40ff6db532 extra networks UI 
rework of hypernets: rather than via settings, hypernets are added directly to prompt as <hypernet:name:weight>
 2023-01-21 08:36:07 +03:00
AUTOMATIC 924e222004 add option to show/hide warnings 
removed hiding warnings from LDSR
fixed/reworked few places that produced warnings
 2023-01-18 23:04:24 +03:00
AUTOMATIC d8b90ac121 big rework of progressbar/preview system to allow multiple users to prompts at the same time and do not get previews of each other 2023-01-15 18:51:04 +03:00
AUTOMATIC a95f135308 change hash to sha256 2023-01-14 09:56:59 +03:00