asagi4
eec4bc4da9
Split old code into its own file for easy removal later
2025-06-08 00:55:43 +03:00
asagi4
3d3218e831
Tests for verifying refactor
2025-06-08 00:55:43 +03:00
asagi4
f4e57ec514
Switch on new implementation by default
2025-06-08 00:55:40 +03:00
asagi4
74f65c1b31
Use old from_masked batching for now
...
I can't figure out why from_masked works differently from down_weight
which also used that batching function but could be replaced with a simple
torch.cat
2025-06-08 00:50:26 +03:00
asagi4
4d94cca88f
Restore adv_encoding to original implementation to compare them
2025-06-08 00:44:56 +03:00
asagi4
4569ecccf9
Manual testing...
2025-06-08 00:44:31 +03:00
asagi4
192e6d30d4
refactor adv_encode
2025-06-07 21:41:13 +03:00
asagi4
67112f11e0
T5 makes STYLE(perp) return NaNs. Just replace them with 0
2025-06-07 21:41:13 +03:00
asagi4
f761dfac86
Fix comfy++ with more than one weight, see #115
...
I'm not sure if this is correct, but it at least doesn't fail.
2025-06-07 14:09:32 +03:00
asagi4
278e733835
Fix normalization validity check
2025-06-07 14:09:32 +03:00
asagi4
2bf65720eb
Fix STYLE(perp) exception, see #115
2025-06-07 14:09:32 +03:00
asagi4
99f3af92b7
Add tests for weighting and a way to run manual testing
2025-06-07 14:09:14 +03:00
asagi4
2437cd4daf
Merge pull request #116 from pamparamm/pooled_none_check
...
Partially resolve #115
2025-06-07 13:30:01 +03:00
asagi4
d262a7dc7a
Remove an extra clone.
2025-06-07 13:23:06 +03:00
Pam
6c319ad5b4
Partially resolve #115
2025-06-07 09:38:13 +05:00
asagi4
34056cac19
v2.0.0-rc.5
2025-06-06 19:52:02 +03:00
asagi4
8ae436abf1
Merge pull request #114 from pamparamm/negpip_option
...
Use ppm_negpip option to detect NegPiP
2025-06-06 16:49:20 +03:00
Pam
c2ce2ce023
Use ppm_negpip option to detect NegPiP
2025-06-06 16:24:15 +05:00
asagi4
7a9e69ec31
Fix errors found in testing
...
Who knew tests could be useful, too?
2025-06-05 23:49:02 +03:00
asagi4
aaff8dc7da
Encoding tests
2025-06-05 23:49:02 +03:00
asagi4
3e4722278a
v2.0.0-rc.4
2025-06-05 21:39:47 +03:00
asagi4
11cb430396
Support NegPiP without requiring a monkeypatch.
2025-06-05 21:38:31 +03:00
asagi4
9f1cbfd11c
Add a test suite for text encoding
...
Can only be run manually and imports ComfyUI main to configure search paths, but
at least it works...
2025-06-05 21:05:39 +03:00
asagi4
5b3a914f1d
Try not to fail in cases where masks have irregular shapes
2025-06-04 22:34:19 +03:00
asagi4
a5ffa1acd7
Documentation
2025-06-04 21:23:52 +03:00
asagi4
7f8783147b
I keep forgetting expand only works on batch size 1, see #108
2025-06-04 21:12:52 +03:00
asagi4
3c9b806e5f
Remove useless import
2025-06-04 21:04:15 +03:00
asagi4
a485c2655a
Fix case where negative prompt size changes lcm of cond size
...
Also don't mutate existing hooks on input negative prompts if they exist.
2025-06-04 21:02:48 +03:00
asagi4
f888e69b00
Merge pull request #112 from pamparamm/attn_couple_batch
...
Add PCAttentionCoupleBatchNegative
2025-06-04 20:59:52 +03:00
asagi4
b4b0858214
Fix multi-TE failure
...
See #113
2025-06-04 08:48:40 +03:00
Pam
7a1cb2cf51
Fix latent masking
2025-06-04 00:30:23 +05:00
Pam
fbb6b5c8fa
Fix some uncond edgecases
2025-06-03 07:50:45 +05:00
Pam
6115c095cb
Fix attn couple batching with multiple positive schedules
2025-06-03 07:42:43 +05:00
Pam
04d3d2e959
Missing space
2025-06-03 03:46:56 +05:00
Pam
e4f64837ef
Add PCAttentionCoupleBatchNegative;
...
Revert some optimizations in AttentionCoupleHook
2025-06-03 03:17:47 +05:00
asagi4
4a785b294b
Avoid hardcoding length in adv encode
...
Makes these not explode on T5 at least. They seems to produce the same results still.
2025-06-03 00:47:57 +03:00
asagi4
b3195a6297
Stop using batched_clip_encode, it doesn't do anything?
2025-06-02 23:32:19 +03:00
asagi4
5c52bffc9d
Docs
2025-06-02 22:16:42 +03:00
asagi4
ac2d275dfb
Clarify TE lookups
2025-06-02 21:37:27 +03:00
asagi4
399f992a26
Fix BREAK while maintaining old behaviour, add CAT instead
2025-06-02 20:36:27 +03:00
asagi4
672a2a09bb
BREAK is now ConditioningConcat
2025-06-02 19:36:29 +03:00
asagi4
e3629961ce
Add AVG(weight) to work as ConditioningAverage
2025-06-02 19:14:20 +03:00
asagi4
ddac624ad1
Add NBREAK (Warning: unstable. Name is likely to change)
...
NBREAK should have the same behaviour as ComfyUI's ConditioningConcat
See #111
2025-06-02 17:39:09 +03:00
asagi4
99ddfe357e
DEF docs
2025-06-01 22:44:06 +03:00
asagi4
110d5248a0
DEF tests
2025-06-01 22:34:39 +03:00
asagi4
61b1ecc88e
Clean up tests a bit
2025-06-01 22:34:39 +03:00
asagi4
05b0b2ad26
Set the default value of $1 to empty with DEF(MACRO()=)
2025-06-01 22:34:34 +03:00
asagi4
5831608c4e
Add a macro expansion node
2025-06-01 21:41:03 +03:00
asagi4
c815bb44f1
Reduce logging verbosity
2025-05-31 01:19:37 +03:00
asagi4
e55c50e9d7
Fix the case with more than one coupled conditioning
2025-05-31 01:09:29 +03:00
asagi4