relicense
This commit is contained in:
harubaru
parent
bc626e80e1
commit
1f5b671b67
|
|
@ -1,4 +0,0 @@
|
||||||
./venv
|
|
||||||
./danbooru-aesthetic
|
|
||||||
./logs
|
|
||||||
*.ckpt
|
|
||||||
|
|
@ -1,59 +0,0 @@
|
||||||
# OS-generated
|
|
||||||
# ------------
|
|
||||||
.DS_Store*
|
|
||||||
[Tt]humbs.db
|
|
||||||
[Dd]esktop.ini
|
|
||||||
|
|
||||||
# Programming - general
|
|
||||||
*.log
|
|
||||||
example.png
|
|
||||||
scores.json
|
|
||||||
danbooru-aesthetic
|
|
||||||
logs
|
|
||||||
*.tar
|
|
||||||
|
|
||||||
# =========================================================================== #
|
|
||||||
# Python-related
|
|
||||||
# =========================================================================== #
|
|
||||||
# src: https://github.com/github/gitignore/blob/master/Python.gitignore
|
|
||||||
|
|
||||||
# JetBrains PyCharm / Rider
|
|
||||||
.idea/
|
|
||||||
|
|
||||||
# Byte-compiled / optimized / DLL files
|
|
||||||
__pycache__/
|
|
||||||
*.py[cod]
|
|
||||||
*$py.class
|
|
||||||
|
|
||||||
# C extensions
|
|
||||||
*.so
|
|
||||||
|
|
||||||
# Distribution / packaging
|
|
||||||
.Python
|
|
||||||
build/
|
|
||||||
develop-eggs/
|
|
||||||
dist/
|
|
||||||
downloads/
|
|
||||||
eggs/
|
|
||||||
.eggs/
|
|
||||||
lib/
|
|
||||||
lib64/
|
|
||||||
parts/
|
|
||||||
sdist/
|
|
||||||
var/
|
|
||||||
venv/
|
|
||||||
wheels/
|
|
||||||
share/python-wheels/
|
|
||||||
*.egg-info/
|
|
||||||
.installed.cfg
|
|
||||||
*.egg
|
|
||||||
MANIFEST
|
|
||||||
|
|
||||||
|
|
||||||
# =========================================================================== #
|
|
||||||
# Repo-specific
|
|
||||||
# =========================================================================== #
|
|
||||||
/src/
|
|
||||||
|
|
||||||
#Obsidian
|
|
||||||
.obsidian/
|
|
||||||
|
|
@ -0,0 +1,3 @@
|
||||||
|
[submodule "dataset/aesthetic"]
|
||||||
|
path = dataset/aesthetic
|
||||||
|
url = https://github.com/waifu-diffusion/aesthetic
|
||||||
|
|
@ -1,10 +1,6 @@
|
||||||
FROM pytorch/pytorch:latest
|
FROM pytorch/pytorch:latest
|
||||||
|
|
||||||
RUN apt update && \
|
|
||||||
apt install -y git curl unzip vim && \
|
|
||||||
pip install git+https://github.com/derfred/lightning.git@waifu-1.6.0#egg=pytorch-lightning
|
|
||||||
RUN mkdir /waifu
|
RUN mkdir /waifu
|
||||||
COPY . /waifu/
|
COPY . /waifu/
|
||||||
WORKDIR /waifu
|
WORKDIR /waifu
|
||||||
RUN grep -v pytorch-lightning requirements.txt > requirements-waifu.txt && \
|
RUN pip install -r requirement.txt
|
||||||
pip install -r requirements-waifu.txt
|
|
||||||
|
|
|
||||||
671
LICENSE
671
LICENSE
|
|
@ -1,14 +1,661 @@
|
||||||
All rights reserved by the authors.
|
GNU AFFERO GENERAL PUBLIC LICENSE
|
||||||
You must not distribute the weights provided to you directly or indirectly without explicit consent of the authors.
|
Version 3, 19 November 2007
|
||||||
You must not distribute harmful, offensive, dehumanizing content or otherwise harmful representations of people or their environments, cultures, religions, etc. produced with the model weights
|
|
||||||
or other generated content described in the "Misuse and Malicious Use" section in the model card.
|
|
||||||
The model weights are provided for research purposes only.
|
|
||||||
|
|
||||||
|
Copyright (C) 2007 Free Software Foundation, Inc. <https://fsf.org/>
|
||||||
|
Everyone is permitted to copy and distribute verbatim copies
|
||||||
|
of this license document, but changing it is not allowed.
|
||||||
|
|
||||||
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
|
Preamble
|
||||||
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
|
|
||||||
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
|
The GNU Affero General Public License is a free, copyleft license for
|
||||||
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
|
software and other kinds of works, specifically designed to ensure
|
||||||
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
|
cooperation with the community in the case of network server software.
|
||||||
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
|
|
||||||
SOFTWARE.
|
The licenses for most software and other practical works are designed
|
||||||
|
to take away your freedom to share and change the works. By contrast,
|
||||||
|
our General Public Licenses are intended to guarantee your freedom to
|
||||||
|
share and change all versions of a program--to make sure it remains free
|
||||||
|
software for all its users.
|
||||||
|
|
||||||
|
When we speak of free software, we are referring to freedom, not
|
||||||
|
price. Our General Public Licenses are designed to make sure that you
|
||||||
|
have the freedom to distribute copies of free software (and charge for
|
||||||
|
them if you wish), that you receive source code or can get it if you
|
||||||
|
want it, that you can change the software or use pieces of it in new
|
||||||
|
free programs, and that you know you can do these things.
|
||||||
|
|
||||||
|
Developers that use our General Public Licenses protect your rights
|
||||||
|
with two steps: (1) assert copyright on the software, and (2) offer
|
||||||
|
you this License which gives you legal permission to copy, distribute
|
||||||
|
and/or modify the software.
|
||||||
|
|
||||||
|
A secondary benefit of defending all users' freedom is that
|
||||||
|
improvements made in alternate versions of the program, if they
|
||||||
|
receive widespread use, become available for other developers to
|
||||||
|
incorporate. Many developers of free software are heartened and
|
||||||
|
encouraged by the resulting cooperation. However, in the case of
|
||||||
|
software used on network servers, this result may fail to come about.
|
||||||
|
The GNU General Public License permits making a modified version and
|
||||||
|
letting the public access it on a server without ever releasing its
|
||||||
|
source code to the public.
|
||||||
|
|
||||||
|
The GNU Affero General Public License is designed specifically to
|
||||||
|
ensure that, in such cases, the modified source code becomes available
|
||||||
|
to the community. It requires the operator of a network server to
|
||||||
|
provide the source code of the modified version running there to the
|
||||||
|
users of that server. Therefore, public use of a modified version, on
|
||||||
|
a publicly accessible server, gives the public access to the source
|
||||||
|
code of the modified version.
|
||||||
|
|
||||||
|
An older license, called the Affero General Public License and
|
||||||
|
published by Affero, was designed to accomplish similar goals. This is
|
||||||
|
a different license, not a version of the Affero GPL, but Affero has
|
||||||
|
released a new version of the Affero GPL which permits relicensing under
|
||||||
|
this license.
|
||||||
|
|
||||||
|
The precise terms and conditions for copying, distribution and
|
||||||
|
modification follow.
|
||||||
|
|
||||||
|
TERMS AND CONDITIONS
|
||||||
|
|
||||||
|
0. Definitions.
|
||||||
|
|
||||||
|
"This License" refers to version 3 of the GNU Affero General Public License.
|
||||||
|
|
||||||
|
"Copyright" also means copyright-like laws that apply to other kinds of
|
||||||
|
works, such as semiconductor masks.
|
||||||
|
|
||||||
|
"The Program" refers to any copyrightable work licensed under this
|
||||||
|
License. Each licensee is addressed as "you". "Licensees" and
|
||||||
|
"recipients" may be individuals or organizations.
|
||||||
|
|
||||||
|
To "modify" a work means to copy from or adapt all or part of the work
|
||||||
|
in a fashion requiring copyright permission, other than the making of an
|
||||||
|
exact copy. The resulting work is called a "modified version" of the
|
||||||
|
earlier work or a work "based on" the earlier work.
|
||||||
|
|
||||||
|
A "covered work" means either the unmodified Program or a work based
|
||||||
|
on the Program.
|
||||||
|
|
||||||
|
To "propagate" a work means to do anything with it that, without
|
||||||
|
permission, would make you directly or secondarily liable for
|
||||||
|
infringement under applicable copyright law, except executing it on a
|
||||||
|
computer or modifying a private copy. Propagation includes copying,
|
||||||
|
distribution (with or without modification), making available to the
|
||||||
|
public, and in some countries other activities as well.
|
||||||
|
|
||||||
|
To "convey" a work means any kind of propagation that enables other
|
||||||
|
parties to make or receive copies. Mere interaction with a user through
|
||||||
|
a computer network, with no transfer of a copy, is not conveying.
|
||||||
|
|
||||||
|
An interactive user interface displays "Appropriate Legal Notices"
|
||||||
|
to the extent that it includes a convenient and prominently visible
|
||||||
|
feature that (1) displays an appropriate copyright notice, and (2)
|
||||||
|
tells the user that there is no warranty for the work (except to the
|
||||||
|
extent that warranties are provided), that licensees may convey the
|
||||||
|
work under this License, and how to view a copy of this License. If
|
||||||
|
the interface presents a list of user commands or options, such as a
|
||||||
|
menu, a prominent item in the list meets this criterion.
|
||||||
|
|
||||||
|
1. Source Code.
|
||||||
|
|
||||||
|
The "source code" for a work means the preferred form of the work
|
||||||
|
for making modifications to it. "Object code" means any non-source
|
||||||
|
form of a work.
|
||||||
|
|
||||||
|
A "Standard Interface" means an interface that either is an official
|
||||||
|
standard defined by a recognized standards body, or, in the case of
|
||||||
|
interfaces specified for a particular programming language, one that
|
||||||
|
is widely used among developers working in that language.
|
||||||
|
|
||||||
|
The "System Libraries" of an executable work include anything, other
|
||||||
|
than the work as a whole, that (a) is included in the normal form of
|
||||||
|
packaging a Major Component, but which is not part of that Major
|
||||||
|
Component, and (b) serves only to enable use of the work with that
|
||||||
|
Major Component, or to implement a Standard Interface for which an
|
||||||
|
implementation is available to the public in source code form. A
|
||||||
|
"Major Component", in this context, means a major essential component
|
||||||
|
(kernel, window system, and so on) of the specific operating system
|
||||||
|
(if any) on which the executable work runs, or a compiler used to
|
||||||
|
produce the work, or an object code interpreter used to run it.
|
||||||
|
|
||||||
|
The "Corresponding Source" for a work in object code form means all
|
||||||
|
the source code needed to generate, install, and (for an executable
|
||||||
|
work) run the object code and to modify the work, including scripts to
|
||||||
|
control those activities. However, it does not include the work's
|
||||||
|
System Libraries, or general-purpose tools or generally available free
|
||||||
|
programs which are used unmodified in performing those activities but
|
||||||
|
which are not part of the work. For example, Corresponding Source
|
||||||
|
includes interface definition files associated with source files for
|
||||||
|
the work, and the source code for shared libraries and dynamically
|
||||||
|
linked subprograms that the work is specifically designed to require,
|
||||||
|
such as by intimate data communication or control flow between those
|
||||||
|
subprograms and other parts of the work.
|
||||||
|
|
||||||
|
The Corresponding Source need not include anything that users
|
||||||
|
can regenerate automatically from other parts of the Corresponding
|
||||||
|
Source.
|
||||||
|
|
||||||
|
The Corresponding Source for a work in source code form is that
|
||||||
|
same work.
|
||||||
|
|
||||||
|
2. Basic Permissions.
|
||||||
|
|
||||||
|
All rights granted under this License are granted for the term of
|
||||||
|
copyright on the Program, and are irrevocable provided the stated
|
||||||
|
conditions are met. This License explicitly affirms your unlimited
|
||||||
|
permission to run the unmodified Program. The output from running a
|
||||||
|
covered work is covered by this License only if the output, given its
|
||||||
|
content, constitutes a covered work. This License acknowledges your
|
||||||
|
rights of fair use or other equivalent, as provided by copyright law.
|
||||||
|
|
||||||
|
You may make, run and propagate covered works that you do not
|
||||||
|
convey, without conditions so long as your license otherwise remains
|
||||||
|
in force. You may convey covered works to others for the sole purpose
|
||||||
|
of having them make modifications exclusively for you, or provide you
|
||||||
|
with facilities for running those works, provided that you comply with
|
||||||
|
the terms of this License in conveying all material for which you do
|
||||||
|
not control copyright. Those thus making or running the covered works
|
||||||
|
for you must do so exclusively on your behalf, under your direction
|
||||||
|
and control, on terms that prohibit them from making any copies of
|
||||||
|
your copyrighted material outside their relationship with you.
|
||||||
|
|
||||||
|
Conveying under any other circumstances is permitted solely under
|
||||||
|
the conditions stated below. Sublicensing is not allowed; section 10
|
||||||
|
makes it unnecessary.
|
||||||
|
|
||||||
|
3. Protecting Users' Legal Rights From Anti-Circumvention Law.
|
||||||
|
|
||||||
|
No covered work shall be deemed part of an effective technological
|
||||||
|
measure under any applicable law fulfilling obligations under article
|
||||||
|
11 of the WIPO copyright treaty adopted on 20 December 1996, or
|
||||||
|
similar laws prohibiting or restricting circumvention of such
|
||||||
|
measures.
|
||||||
|
|
||||||
|
When you convey a covered work, you waive any legal power to forbid
|
||||||
|
circumvention of technological measures to the extent such circumvention
|
||||||
|
is effected by exercising rights under this License with respect to
|
||||||
|
the covered work, and you disclaim any intention to limit operation or
|
||||||
|
modification of the work as a means of enforcing, against the work's
|
||||||
|
users, your or third parties' legal rights to forbid circumvention of
|
||||||
|
technological measures.
|
||||||
|
|
||||||
|
4. Conveying Verbatim Copies.
|
||||||
|
|
||||||
|
You may convey verbatim copies of the Program's source code as you
|
||||||
|
receive it, in any medium, provided that you conspicuously and
|
||||||
|
appropriately publish on each copy an appropriate copyright notice;
|
||||||
|
keep intact all notices stating that this License and any
|
||||||
|
non-permissive terms added in accord with section 7 apply to the code;
|
||||||
|
keep intact all notices of the absence of any warranty; and give all
|
||||||
|
recipients a copy of this License along with the Program.
|
||||||
|
|
||||||
|
You may charge any price or no price for each copy that you convey,
|
||||||
|
and you may offer support or warranty protection for a fee.
|
||||||
|
|
||||||
|
5. Conveying Modified Source Versions.
|
||||||
|
|
||||||
|
You may convey a work based on the Program, or the modifications to
|
||||||
|
produce it from the Program, in the form of source code under the
|
||||||
|
terms of section 4, provided that you also meet all of these conditions:
|
||||||
|
|
||||||
|
a) The work must carry prominent notices stating that you modified
|
||||||
|
it, and giving a relevant date.
|
||||||
|
|
||||||
|
b) The work must carry prominent notices stating that it is
|
||||||
|
released under this License and any conditions added under section
|
||||||
|
7. This requirement modifies the requirement in section 4 to
|
||||||
|
"keep intact all notices".
|
||||||
|
|
||||||
|
c) You must license the entire work, as a whole, under this
|
||||||
|
License to anyone who comes into possession of a copy. This
|
||||||
|
License will therefore apply, along with any applicable section 7
|
||||||
|
additional terms, to the whole of the work, and all its parts,
|
||||||
|
regardless of how they are packaged. This License gives no
|
||||||
|
permission to license the work in any other way, but it does not
|
||||||
|
invalidate such permission if you have separately received it.
|
||||||
|
|
||||||
|
d) If the work has interactive user interfaces, each must display
|
||||||
|
Appropriate Legal Notices; however, if the Program has interactive
|
||||||
|
interfaces that do not display Appropriate Legal Notices, your
|
||||||
|
work need not make them do so.
|
||||||
|
|
||||||
|
A compilation of a covered work with other separate and independent
|
||||||
|
works, which are not by their nature extensions of the covered work,
|
||||||
|
and which are not combined with it such as to form a larger program,
|
||||||
|
in or on a volume of a storage or distribution medium, is called an
|
||||||
|
"aggregate" if the compilation and its resulting copyright are not
|
||||||
|
used to limit the access or legal rights of the compilation's users
|
||||||
|
beyond what the individual works permit. Inclusion of a covered work
|
||||||
|
in an aggregate does not cause this License to apply to the other
|
||||||
|
parts of the aggregate.
|
||||||
|
|
||||||
|
6. Conveying Non-Source Forms.
|
||||||
|
|
||||||
|
You may convey a covered work in object code form under the terms
|
||||||
|
of sections 4 and 5, provided that you also convey the
|
||||||
|
machine-readable Corresponding Source under the terms of this License,
|
||||||
|
in one of these ways:
|
||||||
|
|
||||||
|
a) Convey the object code in, or embodied in, a physical product
|
||||||
|
(including a physical distribution medium), accompanied by the
|
||||||
|
Corresponding Source fixed on a durable physical medium
|
||||||
|
customarily used for software interchange.
|
||||||
|
|
||||||
|
b) Convey the object code in, or embodied in, a physical product
|
||||||
|
(including a physical distribution medium), accompanied by a
|
||||||
|
written offer, valid for at least three years and valid for as
|
||||||
|
long as you offer spare parts or customer support for that product
|
||||||
|
model, to give anyone who possesses the object code either (1) a
|
||||||
|
copy of the Corresponding Source for all the software in the
|
||||||
|
product that is covered by this License, on a durable physical
|
||||||
|
medium customarily used for software interchange, for a price no
|
||||||
|
more than your reasonable cost of physically performing this
|
||||||
|
conveying of source, or (2) access to copy the
|
||||||
|
Corresponding Source from a network server at no charge.
|
||||||
|
|
||||||
|
c) Convey individual copies of the object code with a copy of the
|
||||||
|
written offer to provide the Corresponding Source. This
|
||||||
|
alternative is allowed only occasionally and noncommercially, and
|
||||||
|
only if you received the object code with such an offer, in accord
|
||||||
|
with subsection 6b.
|
||||||
|
|
||||||
|
d) Convey the object code by offering access from a designated
|
||||||
|
place (gratis or for a charge), and offer equivalent access to the
|
||||||
|
Corresponding Source in the same way through the same place at no
|
||||||
|
further charge. You need not require recipients to copy the
|
||||||
|
Corresponding Source along with the object code. If the place to
|
||||||
|
copy the object code is a network server, the Corresponding Source
|
||||||
|
may be on a different server (operated by you or a third party)
|
||||||
|
that supports equivalent copying facilities, provided you maintain
|
||||||
|
clear directions next to the object code saying where to find the
|
||||||
|
Corresponding Source. Regardless of what server hosts the
|
||||||
|
Corresponding Source, you remain obligated to ensure that it is
|
||||||
|
available for as long as needed to satisfy these requirements.
|
||||||
|
|
||||||
|
e) Convey the object code using peer-to-peer transmission, provided
|
||||||
|
you inform other peers where the object code and Corresponding
|
||||||
|
Source of the work are being offered to the general public at no
|
||||||
|
charge under subsection 6d.
|
||||||
|
|
||||||
|
A separable portion of the object code, whose source code is excluded
|
||||||
|
from the Corresponding Source as a System Library, need not be
|
||||||
|
included in conveying the object code work.
|
||||||
|
|
||||||
|
A "User Product" is either (1) a "consumer product", which means any
|
||||||
|
tangible personal property which is normally used for personal, family,
|
||||||
|
or household purposes, or (2) anything designed or sold for incorporation
|
||||||
|
into a dwelling. In determining whether a product is a consumer product,
|
||||||
|
doubtful cases shall be resolved in favor of coverage. For a particular
|
||||||
|
product received by a particular user, "normally used" refers to a
|
||||||
|
typical or common use of that class of product, regardless of the status
|
||||||
|
of the particular user or of the way in which the particular user
|
||||||
|
actually uses, or expects or is expected to use, the product. A product
|
||||||
|
is a consumer product regardless of whether the product has substantial
|
||||||
|
commercial, industrial or non-consumer uses, unless such uses represent
|
||||||
|
the only significant mode of use of the product.
|
||||||
|
|
||||||
|
"Installation Information" for a User Product means any methods,
|
||||||
|
procedures, authorization keys, or other information required to install
|
||||||
|
and execute modified versions of a covered work in that User Product from
|
||||||
|
a modified version of its Corresponding Source. The information must
|
||||||
|
suffice to ensure that the continued functioning of the modified object
|
||||||
|
code is in no case prevented or interfered with solely because
|
||||||
|
modification has been made.
|
||||||
|
|
||||||
|
If you convey an object code work under this section in, or with, or
|
||||||
|
specifically for use in, a User Product, and the conveying occurs as
|
||||||
|
part of a transaction in which the right of possession and use of the
|
||||||
|
User Product is transferred to the recipient in perpetuity or for a
|
||||||
|
fixed term (regardless of how the transaction is characterized), the
|
||||||
|
Corresponding Source conveyed under this section must be accompanied
|
||||||
|
by the Installation Information. But this requirement does not apply
|
||||||
|
if neither you nor any third party retains the ability to install
|
||||||
|
modified object code on the User Product (for example, the work has
|
||||||
|
been installed in ROM).
|
||||||
|
|
||||||
|
The requirement to provide Installation Information does not include a
|
||||||
|
requirement to continue to provide support service, warranty, or updates
|
||||||
|
for a work that has been modified or installed by the recipient, or for
|
||||||
|
the User Product in which it has been modified or installed. Access to a
|
||||||
|
network may be denied when the modification itself materially and
|
||||||
|
adversely affects the operation of the network or violates the rules and
|
||||||
|
protocols for communication across the network.
|
||||||
|
|
||||||
|
Corresponding Source conveyed, and Installation Information provided,
|
||||||
|
in accord with this section must be in a format that is publicly
|
||||||
|
documented (and with an implementation available to the public in
|
||||||
|
source code form), and must require no special password or key for
|
||||||
|
unpacking, reading or copying.
|
||||||
|
|
||||||
|
7. Additional Terms.
|
||||||
|
|
||||||
|
"Additional permissions" are terms that supplement the terms of this
|
||||||
|
License by making exceptions from one or more of its conditions.
|
||||||
|
Additional permissions that are applicable to the entire Program shall
|
||||||
|
be treated as though they were included in this License, to the extent
|
||||||
|
that they are valid under applicable law. If additional permissions
|
||||||
|
apply only to part of the Program, that part may be used separately
|
||||||
|
under those permissions, but the entire Program remains governed by
|
||||||
|
this License without regard to the additional permissions.
|
||||||
|
|
||||||
|
When you convey a copy of a covered work, you may at your option
|
||||||
|
remove any additional permissions from that copy, or from any part of
|
||||||
|
it. (Additional permissions may be written to require their own
|
||||||
|
removal in certain cases when you modify the work.) You may place
|
||||||
|
additional permissions on material, added by you to a covered work,
|
||||||
|
for which you have or can give appropriate copyright permission.
|
||||||
|
|
||||||
|
Notwithstanding any other provision of this License, for material you
|
||||||
|
add to a covered work, you may (if authorized by the copyright holders of
|
||||||
|
that material) supplement the terms of this License with terms:
|
||||||
|
|
||||||
|
a) Disclaiming warranty or limiting liability differently from the
|
||||||
|
terms of sections 15 and 16 of this License; or
|
||||||
|
|
||||||
|
b) Requiring preservation of specified reasonable legal notices or
|
||||||
|
author attributions in that material or in the Appropriate Legal
|
||||||
|
Notices displayed by works containing it; or
|
||||||
|
|
||||||
|
c) Prohibiting misrepresentation of the origin of that material, or
|
||||||
|
requiring that modified versions of such material be marked in
|
||||||
|
reasonable ways as different from the original version; or
|
||||||
|
|
||||||
|
d) Limiting the use for publicity purposes of names of licensors or
|
||||||
|
authors of the material; or
|
||||||
|
|
||||||
|
e) Declining to grant rights under trademark law for use of some
|
||||||
|
trade names, trademarks, or service marks; or
|
||||||
|
|
||||||
|
f) Requiring indemnification of licensors and authors of that
|
||||||
|
material by anyone who conveys the material (or modified versions of
|
||||||
|
it) with contractual assumptions of liability to the recipient, for
|
||||||
|
any liability that these contractual assumptions directly impose on
|
||||||
|
those licensors and authors.
|
||||||
|
|
||||||
|
All other non-permissive additional terms are considered "further
|
||||||
|
restrictions" within the meaning of section 10. If the Program as you
|
||||||
|
received it, or any part of it, contains a notice stating that it is
|
||||||
|
governed by this License along with a term that is a further
|
||||||
|
restriction, you may remove that term. If a license document contains
|
||||||
|
a further restriction but permits relicensing or conveying under this
|
||||||
|
License, you may add to a covered work material governed by the terms
|
||||||
|
of that license document, provided that the further restriction does
|
||||||
|
not survive such relicensing or conveying.
|
||||||
|
|
||||||
|
If you add terms to a covered work in accord with this section, you
|
||||||
|
must place, in the relevant source files, a statement of the
|
||||||
|
additional terms that apply to those files, or a notice indicating
|
||||||
|
where to find the applicable terms.
|
||||||
|
|
||||||
|
Additional terms, permissive or non-permissive, may be stated in the
|
||||||
|
form of a separately written license, or stated as exceptions;
|
||||||
|
the above requirements apply either way.
|
||||||
|
|
||||||
|
8. Termination.
|
||||||
|
|
||||||
|
You may not propagate or modify a covered work except as expressly
|
||||||
|
provided under this License. Any attempt otherwise to propagate or
|
||||||
|
modify it is void, and will automatically terminate your rights under
|
||||||
|
this License (including any patent licenses granted under the third
|
||||||
|
paragraph of section 11).
|
||||||
|
|
||||||
|
However, if you cease all violation of this License, then your
|
||||||
|
license from a particular copyright holder is reinstated (a)
|
||||||
|
provisionally, unless and until the copyright holder explicitly and
|
||||||
|
finally terminates your license, and (b) permanently, if the copyright
|
||||||
|
holder fails to notify you of the violation by some reasonable means
|
||||||
|
prior to 60 days after the cessation.
|
||||||
|
|
||||||
|
Moreover, your license from a particular copyright holder is
|
||||||
|
reinstated permanently if the copyright holder notifies you of the
|
||||||
|
violation by some reasonable means, this is the first time you have
|
||||||
|
received notice of violation of this License (for any work) from that
|
||||||
|
copyright holder, and you cure the violation prior to 30 days after
|
||||||
|
your receipt of the notice.
|
||||||
|
|
||||||
|
Termination of your rights under this section does not terminate the
|
||||||
|
licenses of parties who have received copies or rights from you under
|
||||||
|
this License. If your rights have been terminated and not permanently
|
||||||
|
reinstated, you do not qualify to receive new licenses for the same
|
||||||
|
material under section 10.
|
||||||
|
|
||||||
|
9. Acceptance Not Required for Having Copies.
|
||||||
|
|
||||||
|
You are not required to accept this License in order to receive or
|
||||||
|
run a copy of the Program. Ancillary propagation of a covered work
|
||||||
|
occurring solely as a consequence of using peer-to-peer transmission
|
||||||
|
to receive a copy likewise does not require acceptance. However,
|
||||||
|
nothing other than this License grants you permission to propagate or
|
||||||
|
modify any covered work. These actions infringe copyright if you do
|
||||||
|
not accept this License. Therefore, by modifying or propagating a
|
||||||
|
covered work, you indicate your acceptance of this License to do so.
|
||||||
|
|
||||||
|
10. Automatic Licensing of Downstream Recipients.
|
||||||
|
|
||||||
|
Each time you convey a covered work, the recipient automatically
|
||||||
|
receives a license from the original licensors, to run, modify and
|
||||||
|
propagate that work, subject to this License. You are not responsible
|
||||||
|
for enforcing compliance by third parties with this License.
|
||||||
|
|
||||||
|
An "entity transaction" is a transaction transferring control of an
|
||||||
|
organization, or substantially all assets of one, or subdividing an
|
||||||
|
organization, or merging organizations. If propagation of a covered
|
||||||
|
work results from an entity transaction, each party to that
|
||||||
|
transaction who receives a copy of the work also receives whatever
|
||||||
|
licenses to the work the party's predecessor in interest had or could
|
||||||
|
give under the previous paragraph, plus a right to possession of the
|
||||||
|
Corresponding Source of the work from the predecessor in interest, if
|
||||||
|
the predecessor has it or can get it with reasonable efforts.
|
||||||
|
|
||||||
|
You may not impose any further restrictions on the exercise of the
|
||||||
|
rights granted or affirmed under this License. For example, you may
|
||||||
|
not impose a license fee, royalty, or other charge for exercise of
|
||||||
|
rights granted under this License, and you may not initiate litigation
|
||||||
|
(including a cross-claim or counterclaim in a lawsuit) alleging that
|
||||||
|
any patent claim is infringed by making, using, selling, offering for
|
||||||
|
sale, or importing the Program or any portion of it.
|
||||||
|
|
||||||
|
11. Patents.
|
||||||
|
|
||||||
|
A "contributor" is a copyright holder who authorizes use under this
|
||||||
|
License of the Program or a work on which the Program is based. The
|
||||||
|
work thus licensed is called the contributor's "contributor version".
|
||||||
|
|
||||||
|
A contributor's "essential patent claims" are all patent claims
|
||||||
|
owned or controlled by the contributor, whether already acquired or
|
||||||
|
hereafter acquired, that would be infringed by some manner, permitted
|
||||||
|
by this License, of making, using, or selling its contributor version,
|
||||||
|
but do not include claims that would be infringed only as a
|
||||||
|
consequence of further modification of the contributor version. For
|
||||||
|
purposes of this definition, "control" includes the right to grant
|
||||||
|
patent sublicenses in a manner consistent with the requirements of
|
||||||
|
this License.
|
||||||
|
|
||||||
|
Each contributor grants you a non-exclusive, worldwide, royalty-free
|
||||||
|
patent license under the contributor's essential patent claims, to
|
||||||
|
make, use, sell, offer for sale, import and otherwise run, modify and
|
||||||
|
propagate the contents of its contributor version.
|
||||||
|
|
||||||
|
In the following three paragraphs, a "patent license" is any express
|
||||||
|
agreement or commitment, however denominated, not to enforce a patent
|
||||||
|
(such as an express permission to practice a patent or covenant not to
|
||||||
|
sue for patent infringement). To "grant" such a patent license to a
|
||||||
|
party means to make such an agreement or commitment not to enforce a
|
||||||
|
patent against the party.
|
||||||
|
|
||||||
|
If you convey a covered work, knowingly relying on a patent license,
|
||||||
|
and the Corresponding Source of the work is not available for anyone
|
||||||
|
to copy, free of charge and under the terms of this License, through a
|
||||||
|
publicly available network server or other readily accessible means,
|
||||||
|
then you must either (1) cause the Corresponding Source to be so
|
||||||
|
available, or (2) arrange to deprive yourself of the benefit of the
|
||||||
|
patent license for this particular work, or (3) arrange, in a manner
|
||||||
|
consistent with the requirements of this License, to extend the patent
|
||||||
|
license to downstream recipients. "Knowingly relying" means you have
|
||||||
|
actual knowledge that, but for the patent license, your conveying the
|
||||||
|
covered work in a country, or your recipient's use of the covered work
|
||||||
|
in a country, would infringe one or more identifiable patents in that
|
||||||
|
country that you have reason to believe are valid.
|
||||||
|
|
||||||
|
If, pursuant to or in connection with a single transaction or
|
||||||
|
arrangement, you convey, or propagate by procuring conveyance of, a
|
||||||
|
covered work, and grant a patent license to some of the parties
|
||||||
|
receiving the covered work authorizing them to use, propagate, modify
|
||||||
|
or convey a specific copy of the covered work, then the patent license
|
||||||
|
you grant is automatically extended to all recipients of the covered
|
||||||
|
work and works based on it.
|
||||||
|
|
||||||
|
A patent license is "discriminatory" if it does not include within
|
||||||
|
the scope of its coverage, prohibits the exercise of, or is
|
||||||
|
conditioned on the non-exercise of one or more of the rights that are
|
||||||
|
specifically granted under this License. You may not convey a covered
|
||||||
|
work if you are a party to an arrangement with a third party that is
|
||||||
|
in the business of distributing software, under which you make payment
|
||||||
|
to the third party based on the extent of your activity of conveying
|
||||||
|
the work, and under which the third party grants, to any of the
|
||||||
|
parties who would receive the covered work from you, a discriminatory
|
||||||
|
patent license (a) in connection with copies of the covered work
|
||||||
|
conveyed by you (or copies made from those copies), or (b) primarily
|
||||||
|
for and in connection with specific products or compilations that
|
||||||
|
contain the covered work, unless you entered into that arrangement,
|
||||||
|
or that patent license was granted, prior to 28 March 2007.
|
||||||
|
|
||||||
|
Nothing in this License shall be construed as excluding or limiting
|
||||||
|
any implied license or other defenses to infringement that may
|
||||||
|
otherwise be available to you under applicable patent law.
|
||||||
|
|
||||||
|
12. No Surrender of Others' Freedom.
|
||||||
|
|
||||||
|
If conditions are imposed on you (whether by court order, agreement or
|
||||||
|
otherwise) that contradict the conditions of this License, they do not
|
||||||
|
excuse you from the conditions of this License. If you cannot convey a
|
||||||
|
covered work so as to satisfy simultaneously your obligations under this
|
||||||
|
License and any other pertinent obligations, then as a consequence you may
|
||||||
|
not convey it at all. For example, if you agree to terms that obligate you
|
||||||
|
to collect a royalty for further conveying from those to whom you convey
|
||||||
|
the Program, the only way you could satisfy both those terms and this
|
||||||
|
License would be to refrain entirely from conveying the Program.
|
||||||
|
|
||||||
|
13. Remote Network Interaction; Use with the GNU General Public License.
|
||||||
|
|
||||||
|
Notwithstanding any other provision of this License, if you modify the
|
||||||
|
Program, your modified version must prominently offer all users
|
||||||
|
interacting with it remotely through a computer network (if your version
|
||||||
|
supports such interaction) an opportunity to receive the Corresponding
|
||||||
|
Source of your version by providing access to the Corresponding Source
|
||||||
|
from a network server at no charge, through some standard or customary
|
||||||
|
means of facilitating copying of software. This Corresponding Source
|
||||||
|
shall include the Corresponding Source for any work covered by version 3
|
||||||
|
of the GNU General Public License that is incorporated pursuant to the
|
||||||
|
following paragraph.
|
||||||
|
|
||||||
|
Notwithstanding any other provision of this License, you have
|
||||||
|
permission to link or combine any covered work with a work licensed
|
||||||
|
under version 3 of the GNU General Public License into a single
|
||||||
|
combined work, and to convey the resulting work. The terms of this
|
||||||
|
License will continue to apply to the part which is the covered work,
|
||||||
|
but the work with which it is combined will remain governed by version
|
||||||
|
3 of the GNU General Public License.
|
||||||
|
|
||||||
|
14. Revised Versions of this License.
|
||||||
|
|
||||||
|
The Free Software Foundation may publish revised and/or new versions of
|
||||||
|
the GNU Affero General Public License from time to time. Such new versions
|
||||||
|
will be similar in spirit to the present version, but may differ in detail to
|
||||||
|
address new problems or concerns.
|
||||||
|
|
||||||
|
Each version is given a distinguishing version number. If the
|
||||||
|
Program specifies that a certain numbered version of the GNU Affero General
|
||||||
|
Public License "or any later version" applies to it, you have the
|
||||||
|
option of following the terms and conditions either of that numbered
|
||||||
|
version or of any later version published by the Free Software
|
||||||
|
Foundation. If the Program does not specify a version number of the
|
||||||
|
GNU Affero General Public License, you may choose any version ever published
|
||||||
|
by the Free Software Foundation.
|
||||||
|
|
||||||
|
If the Program specifies that a proxy can decide which future
|
||||||
|
versions of the GNU Affero General Public License can be used, that proxy's
|
||||||
|
public statement of acceptance of a version permanently authorizes you
|
||||||
|
to choose that version for the Program.
|
||||||
|
|
||||||
|
Later license versions may give you additional or different
|
||||||
|
permissions. However, no additional obligations are imposed on any
|
||||||
|
author or copyright holder as a result of your choosing to follow a
|
||||||
|
later version.
|
||||||
|
|
||||||
|
15. Disclaimer of Warranty.
|
||||||
|
|
||||||
|
THERE IS NO WARRANTY FOR THE PROGRAM, TO THE EXTENT PERMITTED BY
|
||||||
|
APPLICABLE LAW. EXCEPT WHEN OTHERWISE STATED IN WRITING THE COPYRIGHT
|
||||||
|
HOLDERS AND/OR OTHER PARTIES PROVIDE THE PROGRAM "AS IS" WITHOUT WARRANTY
|
||||||
|
OF ANY KIND, EITHER EXPRESSED OR IMPLIED, INCLUDING, BUT NOT LIMITED TO,
|
||||||
|
THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
|
||||||
|
PURPOSE. THE ENTIRE RISK AS TO THE QUALITY AND PERFORMANCE OF THE PROGRAM
|
||||||
|
IS WITH YOU. SHOULD THE PROGRAM PROVE DEFECTIVE, YOU ASSUME THE COST OF
|
||||||
|
ALL NECESSARY SERVICING, REPAIR OR CORRECTION.
|
||||||
|
|
||||||
|
16. Limitation of Liability.
|
||||||
|
|
||||||
|
IN NO EVENT UNLESS REQUIRED BY APPLICABLE LAW OR AGREED TO IN WRITING
|
||||||
|
WILL ANY COPYRIGHT HOLDER, OR ANY OTHER PARTY WHO MODIFIES AND/OR CONVEYS
|
||||||
|
THE PROGRAM AS PERMITTED ABOVE, BE LIABLE TO YOU FOR DAMAGES, INCLUDING ANY
|
||||||
|
GENERAL, SPECIAL, INCIDENTAL OR CONSEQUENTIAL DAMAGES ARISING OUT OF THE
|
||||||
|
USE OR INABILITY TO USE THE PROGRAM (INCLUDING BUT NOT LIMITED TO LOSS OF
|
||||||
|
DATA OR DATA BEING RENDERED INACCURATE OR LOSSES SUSTAINED BY YOU OR THIRD
|
||||||
|
PARTIES OR A FAILURE OF THE PROGRAM TO OPERATE WITH ANY OTHER PROGRAMS),
|
||||||
|
EVEN IF SUCH HOLDER OR OTHER PARTY HAS BEEN ADVISED OF THE POSSIBILITY OF
|
||||||
|
SUCH DAMAGES.
|
||||||
|
|
||||||
|
17. Interpretation of Sections 15 and 16.
|
||||||
|
|
||||||
|
If the disclaimer of warranty and limitation of liability provided
|
||||||
|
above cannot be given local legal effect according to their terms,
|
||||||
|
reviewing courts shall apply local law that most closely approximates
|
||||||
|
an absolute waiver of all civil liability in connection with the
|
||||||
|
Program, unless a warranty or assumption of liability accompanies a
|
||||||
|
copy of the Program in return for a fee.
|
||||||
|
|
||||||
|
END OF TERMS AND CONDITIONS
|
||||||
|
|
||||||
|
How to Apply These Terms to Your New Programs
|
||||||
|
|
||||||
|
If you develop a new program, and you want it to be of the greatest
|
||||||
|
possible use to the public, the best way to achieve this is to make it
|
||||||
|
free software which everyone can redistribute and change under these terms.
|
||||||
|
|
||||||
|
To do so, attach the following notices to the program. It is safest
|
||||||
|
to attach them to the start of each source file to most effectively
|
||||||
|
state the exclusion of warranty; and each file should have at least
|
||||||
|
the "copyright" line and a pointer to where the full notice is found.
|
||||||
|
|
||||||
|
<one line to give the program's name and a brief idea of what it does.>
|
||||||
|
Copyright (C) <year> <name of author>
|
||||||
|
|
||||||
|
This program is free software: you can redistribute it and/or modify
|
||||||
|
it under the terms of the GNU Affero General Public License as published by
|
||||||
|
the Free Software Foundation, either version 3 of the License, or
|
||||||
|
(at your option) any later version.
|
||||||
|
|
||||||
|
This program is distributed in the hope that it will be useful,
|
||||||
|
but WITHOUT ANY WARRANTY; without even the implied warranty of
|
||||||
|
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
||||||
|
GNU Affero General Public License for more details.
|
||||||
|
|
||||||
|
You should have received a copy of the GNU Affero General Public License
|
||||||
|
along with this program. If not, see <https://www.gnu.org/licenses/>.
|
||||||
|
|
||||||
|
Also add information on how to contact you by electronic and paper mail.
|
||||||
|
|
||||||
|
If your software can interact with users remotely through a computer
|
||||||
|
network, you should also make sure that it provides a way for users to
|
||||||
|
get its source. For example, if your program is a web application, its
|
||||||
|
interface could display a "Source" link that leads users to an archive
|
||||||
|
of the code. There are many ways you could offer source, and different
|
||||||
|
solutions will be better for different programs; see section 13 for the
|
||||||
|
specific requirements.
|
||||||
|
|
||||||
|
You should also get your employer (if you work as a programmer) or school,
|
||||||
|
if any, to sign a "copyright disclaimer" for the program, if necessary.
|
||||||
|
For more information on this, and how to apply and follow the GNU AGPL, see
|
||||||
|
<https://www.gnu.org/licenses/>.
|
||||||
61
README.md
61
README.md
|
|
@ -1,55 +1,28 @@
|
||||||
|
|
||||||
|
|
||||||
# Waifu Diffusion
|
# Waifu Diffusion
|
||||||
|
|
||||||
Waifu Diffusion is the name for this project of finetuning Stable Diffusion on images and captions downloaded through Danbooru
|
[Waifu Diffusion](https://huggingface.co/hakurei/waifu-diffusion) is the name for this project of finetuning [Stable Diffusion](https://huggingface.co/runwayml/stable-diffusion-v1-5) on anime-styled images.
|
||||||
|
|
||||||
(**Note:** This project has **no affiliation with Danbooru.**)
|
<img src=https://user-images.githubusercontent.com/26317155/194690196-8da73f2a-039d-4349-8b08-e24e8fd20959.png width=40% height=40%>
|
||||||
|
|
||||||
<img src=https://cdn.discordapp.com/attachments/872361510133981234/1016022078635388979/unknown.png?3867929 width=40% height=40%>
|
<sub>1girl, aqua eyes, baseball cap, blonde hair, closed mouth, earrings, green background, hat, hoop earrings, jewelry, looking at viewer, shirt, short hair, simple background, solo, upper body, yellow shirt</sub>
|
||||||
|
|
||||||
<sub>Prompt: touhou 1girl komeiji_koishi portrait</sub>
|
## Setup
|
||||||
|
|
||||||
## Documentation
|
```shell
|
||||||
|
pip install -r requirements.txt
|
||||||
|
```
|
||||||
|
|
||||||
[Index](./docs/en/README.md)
|
## Project Structure
|
||||||
|
|
||||||
[Weights](./docs/en/weights/README.md)
|
```
|
||||||
|
├── dataset: Dataset preparation and utilities
|
||||||
|
│ ├── aesthetic: Aesthetic ranking
|
||||||
|
│ └── download: Downloading utilities
|
||||||
|
└── trainer: The actual training code
|
||||||
|
```
|
||||||
|
|
||||||
[Training Guide](./docs/en/training/README.md)
|
## License
|
||||||
|
Training Code: [AGPL-3.0](LICENSE)
|
||||||
All thanks goes to CompVis and Stability AI for releasing this codebase!
|
Model Weights: [CreativeML Open RAIL-M](https://huggingface.co/spaces/CompVis/stable-diffusion-license)
|
||||||
|
|
||||||
Model Link: https://huggingface.co/hakurei/waifu-diffusion
|
|
||||||
|
|
||||||
### Any questions? Come hop on by to our Discord server!
|
|
||||||
|
|
||||||
[](https://discord.gg/Sx6Spmsgx7)
|
[](https://discord.gg/Sx6Spmsgx7)
|
||||||
|
|
||||||
# Stable Diffusion
|
|
||||||
*Stable Diffusion was made possible thanks to a collaboration with [Stability AI](https://stability.ai/) and [Runway](https://runwayml.com/) and builds upon our previous work:*
|
|
||||||
|
|
||||||
## Comments
|
|
||||||
|
|
||||||
- Our codebase for the diffusion models builds heavily on [OpenAI's ADM codebase](https://github.com/openai/guided-diffusion)
|
|
||||||
and [https://github.com/lucidrains/denoising-diffusion-pytorch](https://github.com/lucidrains/denoising-diffusion-pytorch).
|
|
||||||
Thanks for open-sourcing!
|
|
||||||
|
|
||||||
- The implementation of the transformer encoder is from [x-transformers](https://github.com/lucidrains/x-transformers) by [lucidrains](https://github.com/lucidrains?tab=repositories).
|
|
||||||
|
|
||||||
|
|
||||||
## BibTeX
|
|
||||||
|
|
||||||
```
|
|
||||||
@misc{rombach2021highresolution,
|
|
||||||
title={High-Resolution Image Synthesis with Latent Diffusion Models},
|
|
||||||
author={Robin Rombach and Andreas Blattmann and Dominik Lorenz and Patrick Esser and Björn Ommer},
|
|
||||||
year={2021},
|
|
||||||
eprint={2112.10752},
|
|
||||||
archivePrefix={arXiv},
|
|
||||||
primaryClass={cs.CV}
|
|
||||||
}
|
|
||||||
|
|
||||||
```
|
|
||||||
|
|
||||||
|
|
||||||
|
|
|
||||||
|
|
@ -1,140 +0,0 @@
|
||||||
# Stable Diffusion v1 Model Card
|
|
||||||
This model card focuses on the model associated with the Stable Diffusion model, available [here](https://github.com/CompVis/stable-diffusion).
|
|
||||||
|
|
||||||
## Model Details
|
|
||||||
- **Developed by:** Robin Rombach, Patrick Esser
|
|
||||||
- **Model type:** Diffusion-based text-to-image generation model
|
|
||||||
- **Language(s):** English
|
|
||||||
- **License:** [Proprietary](LICENSE)
|
|
||||||
- **Model Description:** This is a model that can be used to generate and modify images based on text prompts. It is a [Latent Diffusion Model](https://arxiv.org/abs/2112.10752) that uses a fixed, pretrained text encoder ([CLIP ViT-L/14](https://arxiv.org/abs/2103.00020)) as suggested in the [Imagen paper](https://arxiv.org/abs/2205.11487).
|
|
||||||
- **Resources for more information:** [GitHub Repository](https://github.com/CompVis/stable-diffusion), [Paper](https://arxiv.org/abs/2112.10752).
|
|
||||||
- **Cite as:**
|
|
||||||
|
|
||||||
@InProceedings{Rombach_2022_CVPR,
|
|
||||||
author = {Rombach, Robin and Blattmann, Andreas and Lorenz, Dominik and Esser, Patrick and Ommer, Bj\"orn},
|
|
||||||
title = {High-Resolution Image Synthesis With Latent Diffusion Models},
|
|
||||||
booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
|
|
||||||
month = {June},
|
|
||||||
year = {2022},
|
|
||||||
pages = {10684-10695}
|
|
||||||
}
|
|
||||||
|
|
||||||
# Uses
|
|
||||||
|
|
||||||
## Direct Use
|
|
||||||
The model is intended for research purposes only. Possible research areas and
|
|
||||||
tasks include
|
|
||||||
|
|
||||||
- Safe deployment of models which have the potential to generate harmful content.
|
|
||||||
- Probing and understanding the limitations and biases of generative models.
|
|
||||||
- Generation of artworks and use in design and other artistic processes.
|
|
||||||
- Applications in educational or creative tools.
|
|
||||||
- Research on generative models.
|
|
||||||
|
|
||||||
Excluded uses are described below.
|
|
||||||
|
|
||||||
### Misuse, Malicious Use, and Out-of-Scope Use
|
|
||||||
_Note: This section is taken from the [DALLE-MINI model card](https://huggingface.co/dalle-mini/dalle-mini), but applies in the same way to Stable Diffusion v1_.
|
|
||||||
|
|
||||||
|
|
||||||
The model should not be used to intentionally create or disseminate images that create hostile or alienating environments for people. This includes generating images that people would foreseeably find disturbing, distressing, or offensive; or content that propagates historical or current stereotypes.
|
|
||||||
#### Out-of-Scope Use
|
|
||||||
The model was not trained to be factual or true representations of people or events, and therefore using the model to generate such content is out-of-scope for the abilities of this model.
|
|
||||||
#### Misuse and Malicious Use
|
|
||||||
Using the model to generate content that is cruel to individuals is a misuse of this model. This includes, but is not limited to:
|
|
||||||
|
|
||||||
- Generating demeaning, dehumanizing, or otherwise harmful representations of people or their environments, cultures, religions, etc.
|
|
||||||
- Intentionally promoting or propagating discriminatory content or harmful stereotypes.
|
|
||||||
- Impersonating individuals without their consent.
|
|
||||||
- Sexual content without consent of the people who might see it.
|
|
||||||
- Mis- and disinformation
|
|
||||||
- Representations of egregious violence and gore
|
|
||||||
- Sharing of copyrighted or licensed material in violation of its terms of use.
|
|
||||||
- Sharing content that is an alteration of copyrighted or licensed material in violation of its terms of use.
|
|
||||||
|
|
||||||
## Limitations and Bias
|
|
||||||
|
|
||||||
### Limitations
|
|
||||||
|
|
||||||
- The model does not achieve perfect photorealism
|
|
||||||
- The model cannot render legible text
|
|
||||||
- The model does not perform well on more difficult tasks which involve compositionality, such as rendering an image corresponding to “A red cube on top of a blue sphere”
|
|
||||||
- Faces and people in general may not be generated properly.
|
|
||||||
- The model was trained mainly with English captions and will not work as well in other languages.
|
|
||||||
- The autoencoding part of the model is lossy
|
|
||||||
- The model was trained on a large-scale dataset
|
|
||||||
[LAION-5B](https://laion.ai/blog/laion-5b/) which contains adult material
|
|
||||||
and is not fit for product use without additional safety mechanisms and
|
|
||||||
considerations.
|
|
||||||
|
|
||||||
### Bias
|
|
||||||
While the capabilities of image generation models are impressive, they can also reinforce or exacerbate social biases.
|
|
||||||
Stable Diffusion v1 was trained on subsets of [LAION-2B(en)](https://laion.ai/blog/laion-5b/),
|
|
||||||
which consists of images that are primarily limited to English descriptions.
|
|
||||||
Texts and images from communities and cultures that use other languages are likely to be insufficiently accounted for.
|
|
||||||
This affects the overall output of the model, as white and western cultures are often set as the default. Further, the
|
|
||||||
ability of the model to generate content with non-English prompts is significantly worse than with English-language prompts.
|
|
||||||
|
|
||||||
|
|
||||||
## Training
|
|
||||||
|
|
||||||
**Training Data**
|
|
||||||
The model developers used the following dataset for training the model:
|
|
||||||
|
|
||||||
- LAION-2B (en) and subsets thereof (see next section)
|
|
||||||
|
|
||||||
**Training Procedure**
|
|
||||||
Stable Diffusion v1 is a latent diffusion model which combines an autoencoder with a diffusion model that is trained in the latent space of the autoencoder. During training,
|
|
||||||
|
|
||||||
- Images are encoded through an encoder, which turns images into latent representations. The autoencoder uses a relative downsampling factor of 8 and maps images of shape H x W x 3 to latents of shape H/f x W/f x 4
|
|
||||||
- Text prompts are encoded through a ViT-L/14 text-encoder.
|
|
||||||
- The non-pooled output of the text encoder is fed into the UNet backbone of the latent diffusion model via cross-attention.
|
|
||||||
- The loss is a reconstruction objective between the noise that was added to the latent and the prediction made by the UNet.
|
|
||||||
|
|
||||||
We currently provide three checkpoints, `sd-v1-1.ckpt`, `sd-v1-2.ckpt` and `sd-v1-3.ckpt`,
|
|
||||||
which were trained as follows,
|
|
||||||
|
|
||||||
- `sd-v1-1.ckpt`: 237k steps at resolution `256x256` on [laion2B-en](https://huggingface.co/datasets/laion/laion2B-en).
|
|
||||||
194k steps at resolution `512x512` on [laion-high-resolution](https://huggingface.co/datasets/laion/laion-high-resolution) (170M examples from LAION-5B with resolution `>= 1024x1024`).
|
|
||||||
- `sd-v1-2.ckpt`: Resumed from `sd-v1-1.ckpt`.
|
|
||||||
515k steps at resolution `512x512` on "laion-improved-aesthetics" (a subset of laion2B-en,
|
|
||||||
filtered to images with an original size `>= 512x512`, estimated aesthetics score `> 5.0`, and an estimated watermark probability `< 0.5`. The watermark estimate is from the LAION-5B metadata, the aesthetics score is estimated using an [improved aesthetics estimator](https://github.com/christophschuhmann/improved-aesthetic-predictor)).
|
|
||||||
- `sd-v1-3.ckpt`: Resumed from `sd-v1-2.ckpt`. 195k steps at resolution `512x512` on "laion-improved-aesthetics" and 10\% dropping of the text-conditioning to improve [classifier-free guidance sampling](https://arxiv.org/abs/2207.12598).
|
|
||||||
|
|
||||||
|
|
||||||
- **Hardware:** 32 x 8 x A100 GPUs
|
|
||||||
- **Optimizer:** AdamW
|
|
||||||
- **Gradient Accumulations**: 2
|
|
||||||
- **Batch:** 32 x 8 x 2 x 4 = 2048
|
|
||||||
- **Learning rate:** warmup to 0.0001 for 10,000 steps and then kept constant
|
|
||||||
|
|
||||||
## Evaluation Results
|
|
||||||
Evaluations with different classifier-free guidance scales (1.5, 2.0, 3.0, 4.0,
|
|
||||||
5.0, 6.0, 7.0, 8.0) and 50 PLMS sampling
|
|
||||||
steps show the relative improvements of the checkpoints:
|
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
Evaluated using 50 PLMS steps and 10000 random prompts from the COCO2017 validation set, evaluated at 512x512 resolution. Not optimized for FID scores.
|
|
||||||
## Environmental Impact
|
|
||||||
|
|
||||||
**Stable Diffusion v1** **Estimated Emissions**
|
|
||||||
Based on that information, we estimate the following CO2 emissions using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700). The hardware, runtime, cloud provider, and compute region were utilized to estimate the carbon impact.
|
|
||||||
|
|
||||||
- **Hardware Type:** A100 PCIe 40GB
|
|
||||||
- **Hours used:** 150000
|
|
||||||
- **Cloud Provider:** AWS
|
|
||||||
- **Compute Region:** US-east
|
|
||||||
- **Carbon Emitted (Power consumption x Time x Carbon produced based on location of power grid):** 11250 kg CO2 eq.
|
|
||||||
## Citation
|
|
||||||
@InProceedings{Rombach_2022_CVPR,
|
|
||||||
author = {Rombach, Robin and Blattmann, Andreas and Lorenz, Dominik and Esser, Patrick and Ommer, Bj\"orn},
|
|
||||||
title = {High-Resolution Image Synthesis With Latent Diffusion Models},
|
|
||||||
booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
|
|
||||||
month = {June},
|
|
||||||
year = {2022},
|
|
||||||
pages = {10684-10695}
|
|
||||||
}
|
|
||||||
|
|
||||||
*This model card was written by: Robin Rombach and Patrick Esser and is based on the [DALL-E Mini model card](https://huggingface.co/dalle-mini/dalle-mini).*
|
|
||||||
|
|
||||||
|
|
@ -1,7 +0,0 @@
|
||||||
@echo off
|
|
||||||
IF NOT EXIST CONDA umamba create -r conda -f environment.yaml -y
|
|
||||||
call conda\condabin\activate.bat ldm
|
|
||||||
cls
|
|
||||||
|
|
||||||
:PROMPT
|
|
||||||
python scripts/txt2img_gradio.py
|
|
||||||
|
|
@ -1,142 +0,0 @@
|
||||||
import webdataset as wds
|
|
||||||
from PIL import Image
|
|
||||||
import io
|
|
||||||
import matplotlib.pyplot as plt
|
|
||||||
import os
|
|
||||||
import json
|
|
||||||
|
|
||||||
from warnings import filterwarnings
|
|
||||||
|
|
||||||
|
|
||||||
os.environ["CUDA_VISIBLE_DEVICES"] = "1" # choose GPU if you are on a multi GPU server
|
|
||||||
import numpy as np
|
|
||||||
import torch
|
|
||||||
import pytorch_lightning as pl
|
|
||||||
import torch.nn as nn
|
|
||||||
from torchvision import datasets, transforms
|
|
||||||
import tqdm
|
|
||||||
|
|
||||||
from os.path import join
|
|
||||||
from datasets import load_dataset
|
|
||||||
import pandas as pd
|
|
||||||
from torch.utils.data import Dataset, DataLoader
|
|
||||||
import json
|
|
||||||
|
|
||||||
import clip
|
|
||||||
|
|
||||||
|
|
||||||
from PIL import Image, ImageFile
|
|
||||||
|
|
||||||
|
|
||||||
##### This script will predict the aesthetic score for this image file:
|
|
||||||
|
|
||||||
img_path = "../250k_data-0/img/000baa665498e7a61130d7662f81e698.jpg"
|
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
# if you changed the MLP architecture during training, change it also here:
|
|
||||||
class MLP(pl.LightningModule):
|
|
||||||
def __init__(self, input_size, xcol='emb', ycol='avg_rating'):
|
|
||||||
super().__init__()
|
|
||||||
self.input_size = input_size
|
|
||||||
self.xcol = xcol
|
|
||||||
self.ycol = ycol
|
|
||||||
self.layers = nn.Sequential(
|
|
||||||
nn.Linear(self.input_size, 1024),
|
|
||||||
#nn.ReLU(),
|
|
||||||
nn.Dropout(0.2),
|
|
||||||
nn.Linear(1024, 128),
|
|
||||||
#nn.ReLU(),
|
|
||||||
nn.Dropout(0.2),
|
|
||||||
nn.Linear(128, 64),
|
|
||||||
#nn.ReLU(),
|
|
||||||
nn.Dropout(0.1),
|
|
||||||
|
|
||||||
nn.Linear(64, 16),
|
|
||||||
#nn.ReLU(),
|
|
||||||
|
|
||||||
nn.Linear(16, 1)
|
|
||||||
)
|
|
||||||
|
|
||||||
def forward(self, x):
|
|
||||||
return self.layers(x)
|
|
||||||
|
|
||||||
def training_step(self, batch, batch_idx):
|
|
||||||
x = batch[self.xcol]
|
|
||||||
y = batch[self.ycol].reshape(-1, 1)
|
|
||||||
x_hat = self.layers(x)
|
|
||||||
loss = F.mse_loss(x_hat, y)
|
|
||||||
return loss
|
|
||||||
|
|
||||||
def validation_step(self, batch, batch_idx):
|
|
||||||
x = batch[self.xcol]
|
|
||||||
y = batch[self.ycol].reshape(-1, 1)
|
|
||||||
x_hat = self.layers(x)
|
|
||||||
loss = F.mse_loss(x_hat, y)
|
|
||||||
return loss
|
|
||||||
|
|
||||||
def configure_optimizers(self):
|
|
||||||
optimizer = torch.optim.Adam(self.parameters(), lr=1e-3)
|
|
||||||
return optimizer
|
|
||||||
|
|
||||||
def normalized(a, axis=-1, order=2):
|
|
||||||
import numpy as np # pylint: disable=import-outside-toplevel
|
|
||||||
|
|
||||||
l2 = np.atleast_1d(np.linalg.norm(a, order, axis))
|
|
||||||
l2[l2 == 0] = 1
|
|
||||||
return a / np.expand_dims(l2, axis)
|
|
||||||
|
|
||||||
|
|
||||||
model = MLP(768) # CLIP embedding dim is 768 for CLIP ViT L 14
|
|
||||||
|
|
||||||
s = torch.load("sac+logos+ava1-l14-linearMSE.pth") # load the model you trained previously or the model available in this repo
|
|
||||||
|
|
||||||
model.load_state_dict(s)
|
|
||||||
|
|
||||||
model.to("cuda")
|
|
||||||
model.eval()
|
|
||||||
|
|
||||||
|
|
||||||
device = "cuda" if torch.cuda.is_available() else "cpu"
|
|
||||||
model2, preprocess = clip.load("ViT-L/14", device=device) #RN50x64
|
|
||||||
|
|
||||||
@torch.inference_mode()
|
|
||||||
def aesthetic(img_path):
|
|
||||||
pil_image = Image.open(img_path)
|
|
||||||
image = preprocess(pil_image).unsqueeze(0).to(device)
|
|
||||||
with torch.no_grad():
|
|
||||||
image_features = model2.encode_image(image)
|
|
||||||
im_emb_arr = normalized(image_features.cpu().detach().numpy())
|
|
||||||
prediction = model(torch.from_numpy(im_emb_arr).to(device).type(torch.cuda.FloatTensor))
|
|
||||||
return prediction.item()
|
|
||||||
|
|
||||||
import json
|
|
||||||
import glob
|
|
||||||
import shutil
|
|
||||||
|
|
||||||
imdir = '../250k_data-0/img/'
|
|
||||||
ext = ['png', 'jpg', 'jpeg', 'bmp']
|
|
||||||
images = []
|
|
||||||
[images.extend(glob.glob(imdir + '*.' + e)) for e in ext]
|
|
||||||
|
|
||||||
aesthetic_scores = {}
|
|
||||||
|
|
||||||
try:
|
|
||||||
for i in tqdm.tqdm(images):
|
|
||||||
try:
|
|
||||||
score = aesthetic(i)
|
|
||||||
except:
|
|
||||||
print(f'skipping {i}')
|
|
||||||
continue
|
|
||||||
if score < 5.0:
|
|
||||||
shutil.move(i, i.replace('img', 'nonaesthetic'))
|
|
||||||
elif score > 6.0:
|
|
||||||
shutil.move(i, i.replace('img', 'aesthetic'))
|
|
||||||
aesthetic_scores[i] = score
|
|
||||||
except KeyboardInterrupt:
|
|
||||||
pass
|
|
||||||
finally:
|
|
||||||
with open('scores.json', 'w') as f:
|
|
||||||
f.write(json.dumps(aesthetic_scores))
|
|
||||||
Binary file not shown.
Binary file not shown.
|
Before Width: | Height: | Size: 651 KiB |
Binary file not shown.
|
Before Width: | Height: | Size: 596 KiB |
Binary file not shown.
|
Before Width: | Height: | Size: 609 KiB |
Binary file not shown.
|
Before Width: | Height: | Size: 548 KiB |
Binary file not shown.
|
Before Width: | Height: | Size: 705 KiB |
Binary file not shown.
|
Before Width: | Height: | Size: 757 KiB |
BIN
assets/fire.png
BIN
assets/fire.png
Binary file not shown.
|
Before Width: | Height: | Size: 612 KiB |
Binary file not shown.
|
Before Width: | Height: | Size: 312 KiB |
Binary file not shown.
|
Before Width: | Height: | Size: 72 KiB |
Binary file not shown.
|
Before Width: | Height: | Size: 319 KiB |
Binary file not shown.
|
Before Width: | Height: | Size: 788 KiB |
Binary file not shown.
|
Before Width: | Height: | Size: 958 KiB |
Binary file not shown.
|
Before Width: | Height: | Size: 9.4 MiB |
Binary file not shown.
|
Before Width: | Height: | Size: 610 KiB |
Binary file not shown.
|
Before Width: | Height: | Size: 643 KiB |
Binary file not shown.
|
Before Width: | Height: | Size: 641 KiB |
Binary file not shown.
|
Before Width: | Height: | Size: 174 KiB |
Binary file not shown.
|
Before Width: | Height: | Size: 1.1 MiB |
Binary file not shown.
|
Before Width: | Height: | Size: 1.3 MiB |
Binary file not shown.
|
Before Width: | Height: | Size: 945 KiB |
Binary file not shown.
|
Before Width: | Height: | Size: 972 KiB |
Binary file not shown.
|
Before Width: | Height: | Size: 2.5 MiB |
Binary file not shown.
|
Before Width: | Height: | Size: 2.5 MiB |
Binary file not shown.
|
Before Width: | Height: | Size: 2.3 MiB |
Binary file not shown.
|
Before Width: | Height: | Size: 662 KiB |
Binary file not shown.
|
Before Width: | Height: | Size: 302 KiB |
Binary file not shown.
|
Before Width: | Height: | Size: 2.2 MiB |
Binary file not shown.
|
Before Width: | Height: | Size: 70 KiB |
|
|
@ -1,54 +0,0 @@
|
||||||
model:
|
|
||||||
base_learning_rate: 4.5e-6
|
|
||||||
target: ldm.models.autoencoder.AutoencoderKL
|
|
||||||
params:
|
|
||||||
monitor: "val/rec_loss"
|
|
||||||
embed_dim: 16
|
|
||||||
lossconfig:
|
|
||||||
target: ldm.modules.losses.LPIPSWithDiscriminator
|
|
||||||
params:
|
|
||||||
disc_start: 50001
|
|
||||||
kl_weight: 0.000001
|
|
||||||
disc_weight: 0.5
|
|
||||||
|
|
||||||
ddconfig:
|
|
||||||
double_z: True
|
|
||||||
z_channels: 16
|
|
||||||
resolution: 256
|
|
||||||
in_channels: 3
|
|
||||||
out_ch: 3
|
|
||||||
ch: 128
|
|
||||||
ch_mult: [ 1,1,2,2,4] # num_down = len(ch_mult)-1
|
|
||||||
num_res_blocks: 2
|
|
||||||
attn_resolutions: [16]
|
|
||||||
dropout: 0.0
|
|
||||||
|
|
||||||
|
|
||||||
data:
|
|
||||||
target: main.DataModuleFromConfig
|
|
||||||
params:
|
|
||||||
batch_size: 12
|
|
||||||
wrap: True
|
|
||||||
train:
|
|
||||||
target: ldm.data.imagenet.ImageNetSRTrain
|
|
||||||
params:
|
|
||||||
size: 256
|
|
||||||
degradation: pil_nearest
|
|
||||||
validation:
|
|
||||||
target: ldm.data.imagenet.ImageNetSRValidation
|
|
||||||
params:
|
|
||||||
size: 256
|
|
||||||
degradation: pil_nearest
|
|
||||||
|
|
||||||
lightning:
|
|
||||||
callbacks:
|
|
||||||
image_logger:
|
|
||||||
target: main.ImageLogger
|
|
||||||
params:
|
|
||||||
batch_frequency: 1000
|
|
||||||
max_images: 8
|
|
||||||
increase_log_steps: True
|
|
||||||
|
|
||||||
trainer:
|
|
||||||
benchmark: True
|
|
||||||
accumulate_grad_batches: 2
|
|
||||||
|
|
@ -1,53 +0,0 @@
|
||||||
model:
|
|
||||||
base_learning_rate: 4.5e-6
|
|
||||||
target: ldm.models.autoencoder.AutoencoderKL
|
|
||||||
params:
|
|
||||||
monitor: "val/rec_loss"
|
|
||||||
embed_dim: 4
|
|
||||||
lossconfig:
|
|
||||||
target: ldm.modules.losses.LPIPSWithDiscriminator
|
|
||||||
params:
|
|
||||||
disc_start: 50001
|
|
||||||
kl_weight: 0.000001
|
|
||||||
disc_weight: 0.5
|
|
||||||
|
|
||||||
ddconfig:
|
|
||||||
double_z: True
|
|
||||||
z_channels: 4
|
|
||||||
resolution: 256
|
|
||||||
in_channels: 3
|
|
||||||
out_ch: 3
|
|
||||||
ch: 128
|
|
||||||
ch_mult: [ 1,2,4,4 ] # num_down = len(ch_mult)-1
|
|
||||||
num_res_blocks: 2
|
|
||||||
attn_resolutions: [ ]
|
|
||||||
dropout: 0.0
|
|
||||||
|
|
||||||
data:
|
|
||||||
target: main.DataModuleFromConfig
|
|
||||||
params:
|
|
||||||
batch_size: 12
|
|
||||||
wrap: True
|
|
||||||
train:
|
|
||||||
target: ldm.data.imagenet.ImageNetSRTrain
|
|
||||||
params:
|
|
||||||
size: 256
|
|
||||||
degradation: pil_nearest
|
|
||||||
validation:
|
|
||||||
target: ldm.data.imagenet.ImageNetSRValidation
|
|
||||||
params:
|
|
||||||
size: 256
|
|
||||||
degradation: pil_nearest
|
|
||||||
|
|
||||||
lightning:
|
|
||||||
callbacks:
|
|
||||||
image_logger:
|
|
||||||
target: main.ImageLogger
|
|
||||||
params:
|
|
||||||
batch_frequency: 1000
|
|
||||||
max_images: 8
|
|
||||||
increase_log_steps: True
|
|
||||||
|
|
||||||
trainer:
|
|
||||||
benchmark: True
|
|
||||||
accumulate_grad_batches: 2
|
|
||||||
|
|
@ -1,54 +0,0 @@
|
||||||
model:
|
|
||||||
base_learning_rate: 4.5e-6
|
|
||||||
target: ldm.models.autoencoder.AutoencoderKL
|
|
||||||
params:
|
|
||||||
monitor: "val/rec_loss"
|
|
||||||
embed_dim: 3
|
|
||||||
lossconfig:
|
|
||||||
target: ldm.modules.losses.LPIPSWithDiscriminator
|
|
||||||
params:
|
|
||||||
disc_start: 50001
|
|
||||||
kl_weight: 0.000001
|
|
||||||
disc_weight: 0.5
|
|
||||||
|
|
||||||
ddconfig:
|
|
||||||
double_z: True
|
|
||||||
z_channels: 3
|
|
||||||
resolution: 256
|
|
||||||
in_channels: 3
|
|
||||||
out_ch: 3
|
|
||||||
ch: 128
|
|
||||||
ch_mult: [ 1,2,4 ] # num_down = len(ch_mult)-1
|
|
||||||
num_res_blocks: 2
|
|
||||||
attn_resolutions: [ ]
|
|
||||||
dropout: 0.0
|
|
||||||
|
|
||||||
|
|
||||||
data:
|
|
||||||
target: main.DataModuleFromConfig
|
|
||||||
params:
|
|
||||||
batch_size: 12
|
|
||||||
wrap: True
|
|
||||||
train:
|
|
||||||
target: ldm.data.imagenet.ImageNetSRTrain
|
|
||||||
params:
|
|
||||||
size: 256
|
|
||||||
degradation: pil_nearest
|
|
||||||
validation:
|
|
||||||
target: ldm.data.imagenet.ImageNetSRValidation
|
|
||||||
params:
|
|
||||||
size: 256
|
|
||||||
degradation: pil_nearest
|
|
||||||
|
|
||||||
lightning:
|
|
||||||
callbacks:
|
|
||||||
image_logger:
|
|
||||||
target: main.ImageLogger
|
|
||||||
params:
|
|
||||||
batch_frequency: 1000
|
|
||||||
max_images: 8
|
|
||||||
increase_log_steps: True
|
|
||||||
|
|
||||||
trainer:
|
|
||||||
benchmark: True
|
|
||||||
accumulate_grad_batches: 2
|
|
||||||
|
|
@ -1,53 +0,0 @@
|
||||||
model:
|
|
||||||
base_learning_rate: 4.5e-6
|
|
||||||
target: ldm.models.autoencoder.AutoencoderKL
|
|
||||||
params:
|
|
||||||
monitor: "val/rec_loss"
|
|
||||||
embed_dim: 64
|
|
||||||
lossconfig:
|
|
||||||
target: ldm.modules.losses.LPIPSWithDiscriminator
|
|
||||||
params:
|
|
||||||
disc_start: 50001
|
|
||||||
kl_weight: 0.000001
|
|
||||||
disc_weight: 0.5
|
|
||||||
|
|
||||||
ddconfig:
|
|
||||||
double_z: True
|
|
||||||
z_channels: 64
|
|
||||||
resolution: 256
|
|
||||||
in_channels: 3
|
|
||||||
out_ch: 3
|
|
||||||
ch: 128
|
|
||||||
ch_mult: [ 1,1,2,2,4,4] # num_down = len(ch_mult)-1
|
|
||||||
num_res_blocks: 2
|
|
||||||
attn_resolutions: [16,8]
|
|
||||||
dropout: 0.0
|
|
||||||
|
|
||||||
data:
|
|
||||||
target: main.DataModuleFromConfig
|
|
||||||
params:
|
|
||||||
batch_size: 12
|
|
||||||
wrap: True
|
|
||||||
train:
|
|
||||||
target: ldm.data.imagenet.ImageNetSRTrain
|
|
||||||
params:
|
|
||||||
size: 256
|
|
||||||
degradation: pil_nearest
|
|
||||||
validation:
|
|
||||||
target: ldm.data.imagenet.ImageNetSRValidation
|
|
||||||
params:
|
|
||||||
size: 256
|
|
||||||
degradation: pil_nearest
|
|
||||||
|
|
||||||
lightning:
|
|
||||||
callbacks:
|
|
||||||
image_logger:
|
|
||||||
target: main.ImageLogger
|
|
||||||
params:
|
|
||||||
batch_frequency: 1000
|
|
||||||
max_images: 8
|
|
||||||
increase_log_steps: True
|
|
||||||
|
|
||||||
trainer:
|
|
||||||
benchmark: True
|
|
||||||
accumulate_grad_batches: 2
|
|
||||||
|
|
@ -1,86 +0,0 @@
|
||||||
model:
|
|
||||||
base_learning_rate: 2.0e-06
|
|
||||||
target: ldm.models.diffusion.ddpm.LatentDiffusion
|
|
||||||
params:
|
|
||||||
linear_start: 0.0015
|
|
||||||
linear_end: 0.0195
|
|
||||||
num_timesteps_cond: 1
|
|
||||||
log_every_t: 200
|
|
||||||
timesteps: 1000
|
|
||||||
first_stage_key: image
|
|
||||||
image_size: 64
|
|
||||||
channels: 3
|
|
||||||
monitor: val/loss_simple_ema
|
|
||||||
|
|
||||||
unet_config:
|
|
||||||
target: ldm.modules.diffusionmodules.openaimodel.UNetModel
|
|
||||||
params:
|
|
||||||
image_size: 64
|
|
||||||
in_channels: 3
|
|
||||||
out_channels: 3
|
|
||||||
model_channels: 224
|
|
||||||
attention_resolutions:
|
|
||||||
# note: this isn\t actually the resolution but
|
|
||||||
# the downsampling factor, i.e. this corresnponds to
|
|
||||||
# attention on spatial resolution 8,16,32, as the
|
|
||||||
# spatial reolution of the latents is 64 for f4
|
|
||||||
- 8
|
|
||||||
- 4
|
|
||||||
- 2
|
|
||||||
num_res_blocks: 2
|
|
||||||
channel_mult:
|
|
||||||
- 1
|
|
||||||
- 2
|
|
||||||
- 3
|
|
||||||
- 4
|
|
||||||
num_head_channels: 32
|
|
||||||
first_stage_config:
|
|
||||||
target: ldm.models.autoencoder.VQModelInterface
|
|
||||||
params:
|
|
||||||
embed_dim: 3
|
|
||||||
n_embed: 8192
|
|
||||||
ckpt_path: models/first_stage_models/vq-f4/model.ckpt
|
|
||||||
ddconfig:
|
|
||||||
double_z: false
|
|
||||||
z_channels: 3
|
|
||||||
resolution: 256
|
|
||||||
in_channels: 3
|
|
||||||
out_ch: 3
|
|
||||||
ch: 128
|
|
||||||
ch_mult:
|
|
||||||
- 1
|
|
||||||
- 2
|
|
||||||
- 4
|
|
||||||
num_res_blocks: 2
|
|
||||||
attn_resolutions: []
|
|
||||||
dropout: 0.0
|
|
||||||
lossconfig:
|
|
||||||
target: torch.nn.Identity
|
|
||||||
cond_stage_config: __is_unconditional__
|
|
||||||
data:
|
|
||||||
target: main.DataModuleFromConfig
|
|
||||||
params:
|
|
||||||
batch_size: 48
|
|
||||||
num_workers: 5
|
|
||||||
wrap: false
|
|
||||||
train:
|
|
||||||
target: taming.data.faceshq.CelebAHQTrain
|
|
||||||
params:
|
|
||||||
size: 256
|
|
||||||
validation:
|
|
||||||
target: taming.data.faceshq.CelebAHQValidation
|
|
||||||
params:
|
|
||||||
size: 256
|
|
||||||
|
|
||||||
|
|
||||||
lightning:
|
|
||||||
callbacks:
|
|
||||||
image_logger:
|
|
||||||
target: main.ImageLogger
|
|
||||||
params:
|
|
||||||
batch_frequency: 5000
|
|
||||||
max_images: 8
|
|
||||||
increase_log_steps: False
|
|
||||||
|
|
||||||
trainer:
|
|
||||||
benchmark: True
|
|
||||||
|
|
@ -1,98 +0,0 @@
|
||||||
model:
|
|
||||||
base_learning_rate: 1.0e-06
|
|
||||||
target: ldm.models.diffusion.ddpm.LatentDiffusion
|
|
||||||
params:
|
|
||||||
linear_start: 0.0015
|
|
||||||
linear_end: 0.0195
|
|
||||||
num_timesteps_cond: 1
|
|
||||||
log_every_t: 200
|
|
||||||
timesteps: 1000
|
|
||||||
first_stage_key: image
|
|
||||||
cond_stage_key: class_label
|
|
||||||
image_size: 32
|
|
||||||
channels: 4
|
|
||||||
cond_stage_trainable: true
|
|
||||||
conditioning_key: crossattn
|
|
||||||
monitor: val/loss_simple_ema
|
|
||||||
unet_config:
|
|
||||||
target: ldm.modules.diffusionmodules.openaimodel.UNetModel
|
|
||||||
params:
|
|
||||||
image_size: 32
|
|
||||||
in_channels: 4
|
|
||||||
out_channels: 4
|
|
||||||
model_channels: 256
|
|
||||||
attention_resolutions:
|
|
||||||
#note: this isn\t actually the resolution but
|
|
||||||
# the downsampling factor, i.e. this corresnponds to
|
|
||||||
# attention on spatial resolution 8,16,32, as the
|
|
||||||
# spatial reolution of the latents is 32 for f8
|
|
||||||
- 4
|
|
||||||
- 2
|
|
||||||
- 1
|
|
||||||
num_res_blocks: 2
|
|
||||||
channel_mult:
|
|
||||||
- 1
|
|
||||||
- 2
|
|
||||||
- 4
|
|
||||||
num_head_channels: 32
|
|
||||||
use_spatial_transformer: true
|
|
||||||
transformer_depth: 1
|
|
||||||
context_dim: 512
|
|
||||||
first_stage_config:
|
|
||||||
target: ldm.models.autoencoder.VQModelInterface
|
|
||||||
params:
|
|
||||||
embed_dim: 4
|
|
||||||
n_embed: 16384
|
|
||||||
ckpt_path: configs/first_stage_models/vq-f8/model.yaml
|
|
||||||
ddconfig:
|
|
||||||
double_z: false
|
|
||||||
z_channels: 4
|
|
||||||
resolution: 256
|
|
||||||
in_channels: 3
|
|
||||||
out_ch: 3
|
|
||||||
ch: 128
|
|
||||||
ch_mult:
|
|
||||||
- 1
|
|
||||||
- 2
|
|
||||||
- 2
|
|
||||||
- 4
|
|
||||||
num_res_blocks: 2
|
|
||||||
attn_resolutions:
|
|
||||||
- 32
|
|
||||||
dropout: 0.0
|
|
||||||
lossconfig:
|
|
||||||
target: torch.nn.Identity
|
|
||||||
cond_stage_config:
|
|
||||||
target: ldm.modules.encoders.modules.ClassEmbedder
|
|
||||||
params:
|
|
||||||
embed_dim: 512
|
|
||||||
key: class_label
|
|
||||||
data:
|
|
||||||
target: main.DataModuleFromConfig
|
|
||||||
params:
|
|
||||||
batch_size: 64
|
|
||||||
num_workers: 12
|
|
||||||
wrap: false
|
|
||||||
train:
|
|
||||||
target: ldm.data.imagenet.ImageNetTrain
|
|
||||||
params:
|
|
||||||
config:
|
|
||||||
size: 256
|
|
||||||
validation:
|
|
||||||
target: ldm.data.imagenet.ImageNetValidation
|
|
||||||
params:
|
|
||||||
config:
|
|
||||||
size: 256
|
|
||||||
|
|
||||||
|
|
||||||
lightning:
|
|
||||||
callbacks:
|
|
||||||
image_logger:
|
|
||||||
target: main.ImageLogger
|
|
||||||
params:
|
|
||||||
batch_frequency: 5000
|
|
||||||
max_images: 8
|
|
||||||
increase_log_steps: False
|
|
||||||
|
|
||||||
trainer:
|
|
||||||
benchmark: True
|
|
||||||
|
|
@ -1,68 +0,0 @@
|
||||||
model:
|
|
||||||
base_learning_rate: 0.0001
|
|
||||||
target: ldm.models.diffusion.ddpm.LatentDiffusion
|
|
||||||
params:
|
|
||||||
linear_start: 0.0015
|
|
||||||
linear_end: 0.0195
|
|
||||||
num_timesteps_cond: 1
|
|
||||||
log_every_t: 200
|
|
||||||
timesteps: 1000
|
|
||||||
first_stage_key: image
|
|
||||||
cond_stage_key: class_label
|
|
||||||
image_size: 64
|
|
||||||
channels: 3
|
|
||||||
cond_stage_trainable: true
|
|
||||||
conditioning_key: crossattn
|
|
||||||
monitor: val/loss
|
|
||||||
use_ema: False
|
|
||||||
|
|
||||||
unet_config:
|
|
||||||
target: ldm.modules.diffusionmodules.openaimodel.UNetModel
|
|
||||||
params:
|
|
||||||
image_size: 64
|
|
||||||
in_channels: 3
|
|
||||||
out_channels: 3
|
|
||||||
model_channels: 192
|
|
||||||
attention_resolutions:
|
|
||||||
- 8
|
|
||||||
- 4
|
|
||||||
- 2
|
|
||||||
num_res_blocks: 2
|
|
||||||
channel_mult:
|
|
||||||
- 1
|
|
||||||
- 2
|
|
||||||
- 3
|
|
||||||
- 5
|
|
||||||
num_heads: 1
|
|
||||||
use_spatial_transformer: true
|
|
||||||
transformer_depth: 1
|
|
||||||
context_dim: 512
|
|
||||||
|
|
||||||
first_stage_config:
|
|
||||||
target: ldm.models.autoencoder.VQModelInterface
|
|
||||||
params:
|
|
||||||
embed_dim: 3
|
|
||||||
n_embed: 8192
|
|
||||||
ddconfig:
|
|
||||||
double_z: false
|
|
||||||
z_channels: 3
|
|
||||||
resolution: 256
|
|
||||||
in_channels: 3
|
|
||||||
out_ch: 3
|
|
||||||
ch: 128
|
|
||||||
ch_mult:
|
|
||||||
- 1
|
|
||||||
- 2
|
|
||||||
- 4
|
|
||||||
num_res_blocks: 2
|
|
||||||
attn_resolutions: []
|
|
||||||
dropout: 0.0
|
|
||||||
lossconfig:
|
|
||||||
target: torch.nn.Identity
|
|
||||||
|
|
||||||
cond_stage_config:
|
|
||||||
target: ldm.modules.encoders.modules.ClassEmbedder
|
|
||||||
params:
|
|
||||||
n_classes: 1001
|
|
||||||
embed_dim: 512
|
|
||||||
key: class_label
|
|
||||||
|
|
@ -1,85 +0,0 @@
|
||||||
model:
|
|
||||||
base_learning_rate: 2.0e-06
|
|
||||||
target: ldm.models.diffusion.ddpm.LatentDiffusion
|
|
||||||
params:
|
|
||||||
linear_start: 0.0015
|
|
||||||
linear_end: 0.0195
|
|
||||||
num_timesteps_cond: 1
|
|
||||||
log_every_t: 200
|
|
||||||
timesteps: 1000
|
|
||||||
first_stage_key: image
|
|
||||||
image_size: 64
|
|
||||||
channels: 3
|
|
||||||
monitor: val/loss_simple_ema
|
|
||||||
unet_config:
|
|
||||||
target: ldm.modules.diffusionmodules.openaimodel.UNetModel
|
|
||||||
params:
|
|
||||||
image_size: 64
|
|
||||||
in_channels: 3
|
|
||||||
out_channels: 3
|
|
||||||
model_channels: 224
|
|
||||||
attention_resolutions:
|
|
||||||
# note: this isn\t actually the resolution but
|
|
||||||
# the downsampling factor, i.e. this corresnponds to
|
|
||||||
# attention on spatial resolution 8,16,32, as the
|
|
||||||
# spatial reolution of the latents is 64 for f4
|
|
||||||
- 8
|
|
||||||
- 4
|
|
||||||
- 2
|
|
||||||
num_res_blocks: 2
|
|
||||||
channel_mult:
|
|
||||||
- 1
|
|
||||||
- 2
|
|
||||||
- 3
|
|
||||||
- 4
|
|
||||||
num_head_channels: 32
|
|
||||||
first_stage_config:
|
|
||||||
target: ldm.models.autoencoder.VQModelInterface
|
|
||||||
params:
|
|
||||||
embed_dim: 3
|
|
||||||
n_embed: 8192
|
|
||||||
ckpt_path: configs/first_stage_models/vq-f4/model.yaml
|
|
||||||
ddconfig:
|
|
||||||
double_z: false
|
|
||||||
z_channels: 3
|
|
||||||
resolution: 256
|
|
||||||
in_channels: 3
|
|
||||||
out_ch: 3
|
|
||||||
ch: 128
|
|
||||||
ch_mult:
|
|
||||||
- 1
|
|
||||||
- 2
|
|
||||||
- 4
|
|
||||||
num_res_blocks: 2
|
|
||||||
attn_resolutions: []
|
|
||||||
dropout: 0.0
|
|
||||||
lossconfig:
|
|
||||||
target: torch.nn.Identity
|
|
||||||
cond_stage_config: __is_unconditional__
|
|
||||||
data:
|
|
||||||
target: main.DataModuleFromConfig
|
|
||||||
params:
|
|
||||||
batch_size: 42
|
|
||||||
num_workers: 5
|
|
||||||
wrap: false
|
|
||||||
train:
|
|
||||||
target: taming.data.faceshq.FFHQTrain
|
|
||||||
params:
|
|
||||||
size: 256
|
|
||||||
validation:
|
|
||||||
target: taming.data.faceshq.FFHQValidation
|
|
||||||
params:
|
|
||||||
size: 256
|
|
||||||
|
|
||||||
|
|
||||||
lightning:
|
|
||||||
callbacks:
|
|
||||||
image_logger:
|
|
||||||
target: main.ImageLogger
|
|
||||||
params:
|
|
||||||
batch_frequency: 5000
|
|
||||||
max_images: 8
|
|
||||||
increase_log_steps: False
|
|
||||||
|
|
||||||
trainer:
|
|
||||||
benchmark: True
|
|
||||||
|
|
@ -1,85 +0,0 @@
|
||||||
model:
|
|
||||||
base_learning_rate: 2.0e-06
|
|
||||||
target: ldm.models.diffusion.ddpm.LatentDiffusion
|
|
||||||
params:
|
|
||||||
linear_start: 0.0015
|
|
||||||
linear_end: 0.0195
|
|
||||||
num_timesteps_cond: 1
|
|
||||||
log_every_t: 200
|
|
||||||
timesteps: 1000
|
|
||||||
first_stage_key: image
|
|
||||||
image_size: 64
|
|
||||||
channels: 3
|
|
||||||
monitor: val/loss_simple_ema
|
|
||||||
unet_config:
|
|
||||||
target: ldm.modules.diffusionmodules.openaimodel.UNetModel
|
|
||||||
params:
|
|
||||||
image_size: 64
|
|
||||||
in_channels: 3
|
|
||||||
out_channels: 3
|
|
||||||
model_channels: 224
|
|
||||||
attention_resolutions:
|
|
||||||
# note: this isn\t actually the resolution but
|
|
||||||
# the downsampling factor, i.e. this corresnponds to
|
|
||||||
# attention on spatial resolution 8,16,32, as the
|
|
||||||
# spatial reolution of the latents is 64 for f4
|
|
||||||
- 8
|
|
||||||
- 4
|
|
||||||
- 2
|
|
||||||
num_res_blocks: 2
|
|
||||||
channel_mult:
|
|
||||||
- 1
|
|
||||||
- 2
|
|
||||||
- 3
|
|
||||||
- 4
|
|
||||||
num_head_channels: 32
|
|
||||||
first_stage_config:
|
|
||||||
target: ldm.models.autoencoder.VQModelInterface
|
|
||||||
params:
|
|
||||||
ckpt_path: configs/first_stage_models/vq-f4/model.yaml
|
|
||||||
embed_dim: 3
|
|
||||||
n_embed: 8192
|
|
||||||
ddconfig:
|
|
||||||
double_z: false
|
|
||||||
z_channels: 3
|
|
||||||
resolution: 256
|
|
||||||
in_channels: 3
|
|
||||||
out_ch: 3
|
|
||||||
ch: 128
|
|
||||||
ch_mult:
|
|
||||||
- 1
|
|
||||||
- 2
|
|
||||||
- 4
|
|
||||||
num_res_blocks: 2
|
|
||||||
attn_resolutions: []
|
|
||||||
dropout: 0.0
|
|
||||||
lossconfig:
|
|
||||||
target: torch.nn.Identity
|
|
||||||
cond_stage_config: __is_unconditional__
|
|
||||||
data:
|
|
||||||
target: main.DataModuleFromConfig
|
|
||||||
params:
|
|
||||||
batch_size: 48
|
|
||||||
num_workers: 5
|
|
||||||
wrap: false
|
|
||||||
train:
|
|
||||||
target: ldm.data.lsun.LSUNBedroomsTrain
|
|
||||||
params:
|
|
||||||
size: 256
|
|
||||||
validation:
|
|
||||||
target: ldm.data.lsun.LSUNBedroomsValidation
|
|
||||||
params:
|
|
||||||
size: 256
|
|
||||||
|
|
||||||
|
|
||||||
lightning:
|
|
||||||
callbacks:
|
|
||||||
image_logger:
|
|
||||||
target: main.ImageLogger
|
|
||||||
params:
|
|
||||||
batch_frequency: 5000
|
|
||||||
max_images: 8
|
|
||||||
increase_log_steps: False
|
|
||||||
|
|
||||||
trainer:
|
|
||||||
benchmark: True
|
|
||||||
|
|
@ -1,91 +0,0 @@
|
||||||
model:
|
|
||||||
base_learning_rate: 5.0e-5 # set to target_lr by starting main.py with '--scale_lr False'
|
|
||||||
target: ldm.models.diffusion.ddpm.LatentDiffusion
|
|
||||||
params:
|
|
||||||
linear_start: 0.0015
|
|
||||||
linear_end: 0.0155
|
|
||||||
num_timesteps_cond: 1
|
|
||||||
log_every_t: 200
|
|
||||||
timesteps: 1000
|
|
||||||
loss_type: l1
|
|
||||||
first_stage_key: "image"
|
|
||||||
cond_stage_key: "image"
|
|
||||||
image_size: 32
|
|
||||||
channels: 4
|
|
||||||
cond_stage_trainable: False
|
|
||||||
concat_mode: False
|
|
||||||
scale_by_std: True
|
|
||||||
monitor: 'val/loss_simple_ema'
|
|
||||||
|
|
||||||
scheduler_config: # 10000 warmup steps
|
|
||||||
target: ldm.lr_scheduler.LambdaLinearScheduler
|
|
||||||
params:
|
|
||||||
warm_up_steps: [10000]
|
|
||||||
cycle_lengths: [10000000000000]
|
|
||||||
f_start: [1.e-6]
|
|
||||||
f_max: [1.]
|
|
||||||
f_min: [ 1.]
|
|
||||||
|
|
||||||
unet_config:
|
|
||||||
target: ldm.modules.diffusionmodules.openaimodel.UNetModel
|
|
||||||
params:
|
|
||||||
image_size: 32
|
|
||||||
in_channels: 4
|
|
||||||
out_channels: 4
|
|
||||||
model_channels: 192
|
|
||||||
attention_resolutions: [ 1, 2, 4, 8 ] # 32, 16, 8, 4
|
|
||||||
num_res_blocks: 2
|
|
||||||
channel_mult: [ 1,2,2,4,4 ] # 32, 16, 8, 4, 2
|
|
||||||
num_heads: 8
|
|
||||||
use_scale_shift_norm: True
|
|
||||||
resblock_updown: True
|
|
||||||
|
|
||||||
first_stage_config:
|
|
||||||
target: ldm.models.autoencoder.AutoencoderKL
|
|
||||||
params:
|
|
||||||
embed_dim: 4
|
|
||||||
monitor: "val/rec_loss"
|
|
||||||
ckpt_path: "models/first_stage_models/kl-f8/model.ckpt"
|
|
||||||
ddconfig:
|
|
||||||
double_z: True
|
|
||||||
z_channels: 4
|
|
||||||
resolution: 256
|
|
||||||
in_channels: 3
|
|
||||||
out_ch: 3
|
|
||||||
ch: 128
|
|
||||||
ch_mult: [ 1,2,4,4 ] # num_down = len(ch_mult)-1
|
|
||||||
num_res_blocks: 2
|
|
||||||
attn_resolutions: [ ]
|
|
||||||
dropout: 0.0
|
|
||||||
lossconfig:
|
|
||||||
target: torch.nn.Identity
|
|
||||||
|
|
||||||
cond_stage_config: "__is_unconditional__"
|
|
||||||
|
|
||||||
data:
|
|
||||||
target: main.DataModuleFromConfig
|
|
||||||
params:
|
|
||||||
batch_size: 96
|
|
||||||
num_workers: 5
|
|
||||||
wrap: False
|
|
||||||
train:
|
|
||||||
target: ldm.data.lsun.LSUNChurchesTrain
|
|
||||||
params:
|
|
||||||
size: 256
|
|
||||||
validation:
|
|
||||||
target: ldm.data.lsun.LSUNChurchesValidation
|
|
||||||
params:
|
|
||||||
size: 256
|
|
||||||
|
|
||||||
lightning:
|
|
||||||
callbacks:
|
|
||||||
image_logger:
|
|
||||||
target: main.ImageLogger
|
|
||||||
params:
|
|
||||||
batch_frequency: 5000
|
|
||||||
max_images: 8
|
|
||||||
increase_log_steps: False
|
|
||||||
|
|
||||||
|
|
||||||
trainer:
|
|
||||||
benchmark: True
|
|
||||||
|
|
@ -1,71 +0,0 @@
|
||||||
model:
|
|
||||||
base_learning_rate: 5.0e-05
|
|
||||||
target: ldm.models.diffusion.ddpm.LatentDiffusion
|
|
||||||
params:
|
|
||||||
linear_start: 0.00085
|
|
||||||
linear_end: 0.012
|
|
||||||
num_timesteps_cond: 1
|
|
||||||
log_every_t: 200
|
|
||||||
timesteps: 1000
|
|
||||||
first_stage_key: image
|
|
||||||
cond_stage_key: caption
|
|
||||||
image_size: 32
|
|
||||||
channels: 4
|
|
||||||
cond_stage_trainable: true
|
|
||||||
conditioning_key: crossattn
|
|
||||||
monitor: val/loss_simple_ema
|
|
||||||
scale_factor: 0.18215
|
|
||||||
use_ema: False
|
|
||||||
|
|
||||||
unet_config:
|
|
||||||
target: ldm.modules.diffusionmodules.openaimodel.UNetModel
|
|
||||||
params:
|
|
||||||
image_size: 32
|
|
||||||
in_channels: 4
|
|
||||||
out_channels: 4
|
|
||||||
model_channels: 320
|
|
||||||
attention_resolutions:
|
|
||||||
- 4
|
|
||||||
- 2
|
|
||||||
- 1
|
|
||||||
num_res_blocks: 2
|
|
||||||
channel_mult:
|
|
||||||
- 1
|
|
||||||
- 2
|
|
||||||
- 4
|
|
||||||
- 4
|
|
||||||
num_heads: 8
|
|
||||||
use_spatial_transformer: true
|
|
||||||
transformer_depth: 1
|
|
||||||
context_dim: 1280
|
|
||||||
use_checkpoint: true
|
|
||||||
legacy: False
|
|
||||||
|
|
||||||
first_stage_config:
|
|
||||||
target: ldm.models.autoencoder.AutoencoderKL
|
|
||||||
params:
|
|
||||||
embed_dim: 4
|
|
||||||
monitor: val/rec_loss
|
|
||||||
ddconfig:
|
|
||||||
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
|
|
||||||
|
|
||||||
cond_stage_config:
|
|
||||||
target: ldm.modules.encoders.modules.BERTEmbedder
|
|
||||||
params:
|
|
||||||
n_embed: 1280
|
|
||||||
n_layer: 32
|
|
||||||
|
|
@ -1,68 +0,0 @@
|
||||||
model:
|
|
||||||
base_learning_rate: 0.0001
|
|
||||||
target: ldm.models.diffusion.ddpm.LatentDiffusion
|
|
||||||
params:
|
|
||||||
linear_start: 0.0015
|
|
||||||
linear_end: 0.015
|
|
||||||
num_timesteps_cond: 1
|
|
||||||
log_every_t: 200
|
|
||||||
timesteps: 1000
|
|
||||||
first_stage_key: jpg
|
|
||||||
cond_stage_key: nix
|
|
||||||
image_size: 48
|
|
||||||
channels: 16
|
|
||||||
cond_stage_trainable: false
|
|
||||||
conditioning_key: crossattn
|
|
||||||
monitor: val/loss_simple_ema
|
|
||||||
scale_by_std: false
|
|
||||||
scale_factor: 0.22765929
|
|
||||||
unet_config:
|
|
||||||
target: ldm.modules.diffusionmodules.openaimodel.UNetModel
|
|
||||||
params:
|
|
||||||
image_size: 48
|
|
||||||
in_channels: 16
|
|
||||||
out_channels: 16
|
|
||||||
model_channels: 448
|
|
||||||
attention_resolutions:
|
|
||||||
- 4
|
|
||||||
- 2
|
|
||||||
- 1
|
|
||||||
num_res_blocks: 2
|
|
||||||
channel_mult:
|
|
||||||
- 1
|
|
||||||
- 2
|
|
||||||
- 3
|
|
||||||
- 4
|
|
||||||
use_scale_shift_norm: false
|
|
||||||
resblock_updown: false
|
|
||||||
num_head_channels: 32
|
|
||||||
use_spatial_transformer: true
|
|
||||||
transformer_depth: 1
|
|
||||||
context_dim: 768
|
|
||||||
use_checkpoint: true
|
|
||||||
first_stage_config:
|
|
||||||
target: ldm.models.autoencoder.AutoencoderKL
|
|
||||||
params:
|
|
||||||
monitor: val/rec_loss
|
|
||||||
embed_dim: 16
|
|
||||||
ddconfig:
|
|
||||||
double_z: true
|
|
||||||
z_channels: 16
|
|
||||||
resolution: 256
|
|
||||||
in_channels: 3
|
|
||||||
out_ch: 3
|
|
||||||
ch: 128
|
|
||||||
ch_mult:
|
|
||||||
- 1
|
|
||||||
- 1
|
|
||||||
- 2
|
|
||||||
- 2
|
|
||||||
- 4
|
|
||||||
num_res_blocks: 2
|
|
||||||
attn_resolutions:
|
|
||||||
- 16
|
|
||||||
dropout: 0.0
|
|
||||||
lossconfig:
|
|
||||||
target: torch.nn.Identity
|
|
||||||
cond_stage_config:
|
|
||||||
target: torch.nn.Identity
|
|
||||||
|
|
@ -1,123 +0,0 @@
|
||||||
model:
|
|
||||||
base_learning_rate: 7.5e-06
|
|
||||||
target: ldm.models.diffusion.ddpm.LatentDiffusion
|
|
||||||
params:
|
|
||||||
linear_start: 0.00085
|
|
||||||
linear_end: 0.0120
|
|
||||||
num_timesteps_cond: 1
|
|
||||||
log_every_t: 200
|
|
||||||
timesteps: 1000
|
|
||||||
first_stage_key: image
|
|
||||||
cond_stage_key: caption
|
|
||||||
image_size: 64
|
|
||||||
channels: 4
|
|
||||||
cond_stage_trainable: false # Note: different from the one we trained before
|
|
||||||
conditioning_key: crossattn
|
|
||||||
monitor: val/loss_simple_ema
|
|
||||||
scale_factor: 0.18215
|
|
||||||
|
|
||||||
scheduler_config: # 10000 warmup steps
|
|
||||||
target: ldm.lr_scheduler.LambdaLinearScheduler
|
|
||||||
params:
|
|
||||||
warm_up_steps: [ 1 ] # NOTE for resuming. use 10000 if starting from scratch
|
|
||||||
cycle_lengths: [ 10000000000000 ] # incredibly large number to prevent corner cases
|
|
||||||
f_start: [ 1.e-6 ]
|
|
||||||
f_max: [ 1. ]
|
|
||||||
f_min: [ 1. ]
|
|
||||||
|
|
||||||
unet_config:
|
|
||||||
target: ldm.modules.diffusionmodules.openaimodel.UNetModel
|
|
||||||
params:
|
|
||||||
image_size: 32 # unused
|
|
||||||
in_channels: 4
|
|
||||||
out_channels: 4
|
|
||||||
model_channels: 320
|
|
||||||
attention_resolutions: [ 4, 2, 1 ]
|
|
||||||
num_res_blocks: 2
|
|
||||||
channel_mult: [ 1, 2, 4, 4 ]
|
|
||||||
num_heads: 8
|
|
||||||
use_spatial_transformer: True
|
|
||||||
transformer_depth: 1
|
|
||||||
context_dim: 768
|
|
||||||
use_checkpoint: True
|
|
||||||
legacy: False
|
|
||||||
|
|
||||||
first_stage_config:
|
|
||||||
target: ldm.models.autoencoder.AutoencoderKL
|
|
||||||
params:
|
|
||||||
embed_dim: 4
|
|
||||||
monitor: val/rec_loss
|
|
||||||
ckpt_path: "../latent-diffusion/logs/original/checkpoints/last.ckpt"
|
|
||||||
ddconfig:
|
|
||||||
double_z: true
|
|
||||||
z_channels: 4
|
|
||||||
resolution: 512
|
|
||||||
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
|
|
||||||
|
|
||||||
cond_stage_config:
|
|
||||||
target: ldm.modules.encoders.modules.FrozenCLIPEmbedder
|
|
||||||
params:
|
|
||||||
penultimate: true # use 2nd last layer - https://arxiv.org/pdf/2205.11487.pdf D.1
|
|
||||||
extended_mode: 3 # extend clip context to 225 tokens - as per NAI blogpost
|
|
||||||
|
|
||||||
data:
|
|
||||||
target: main.DataModuleFromConfig
|
|
||||||
params:
|
|
||||||
batch_size: 2
|
|
||||||
num_workers: 2
|
|
||||||
wrap: false
|
|
||||||
train:
|
|
||||||
target: ldm.data.localdanboorubase.LocalDanbooruBase
|
|
||||||
params:
|
|
||||||
data_root: '../dataset'
|
|
||||||
size: 512
|
|
||||||
mode: "train"
|
|
||||||
ucg: 0.1 # unconditional guidance training
|
|
||||||
validation:
|
|
||||||
target: ldm.data.localdanboorubase.LocalDanbooruBase
|
|
||||||
params:
|
|
||||||
data_root: '../dataset'
|
|
||||||
size: 512
|
|
||||||
mode: "val"
|
|
||||||
val_split: 64
|
|
||||||
ucg: 0.1
|
|
||||||
|
|
||||||
lightning:
|
|
||||||
modelcheckpoint:
|
|
||||||
params:
|
|
||||||
every_n_train_steps: 500
|
|
||||||
callbacks:
|
|
||||||
image_logger:
|
|
||||||
target: main.ImageLogger
|
|
||||||
params:
|
|
||||||
batch_frequency: 500
|
|
||||||
max_images: 4
|
|
||||||
increase_log_steps: False
|
|
||||||
log_first_step: False
|
|
||||||
log_images_kwargs:
|
|
||||||
use_ema_scope: False
|
|
||||||
inpaint: False
|
|
||||||
plot_progressive_rows: False
|
|
||||||
plot_diffusion_rows: False
|
|
||||||
N: 4
|
|
||||||
ddim_steps: 50
|
|
||||||
trainer:
|
|
||||||
precision: 16
|
|
||||||
amp_backend: "native"
|
|
||||||
strategy: "fsdp"
|
|
||||||
benchmark: True
|
|
||||||
limit_val_batches: 0
|
|
||||||
num_sanity_val_steps: 0
|
|
||||||
accumulate_grad_batches: 1
|
|
||||||
|
|
@ -1,62 +0,0 @@
|
||||||
model:
|
|
||||||
base_learning_rate: 1.5e-7
|
|
||||||
target: ldm.models.autoencoder.AutoencoderKL
|
|
||||||
params:
|
|
||||||
monitor: "val/rec_loss"
|
|
||||||
embed_dim: 4
|
|
||||||
lossconfig:
|
|
||||||
target: ldm.modules.losses.LPIPSWithDiscriminator
|
|
||||||
params:
|
|
||||||
disc_start: 50001
|
|
||||||
kl_weight: 0.000001
|
|
||||||
disc_weight: 0.5
|
|
||||||
|
|
||||||
ddconfig:
|
|
||||||
double_z: True
|
|
||||||
z_channels: 4
|
|
||||||
resolution: 256
|
|
||||||
in_channels: 3
|
|
||||||
out_ch: 3
|
|
||||||
ch: 128
|
|
||||||
ch_mult: [ 1,2,4,4 ] # num_down = len(ch_mult)-1
|
|
||||||
num_res_blocks: 2
|
|
||||||
attn_resolutions: [ ]
|
|
||||||
dropout: 0.0
|
|
||||||
|
|
||||||
data:
|
|
||||||
target: main.DataModuleFromConfig
|
|
||||||
params:
|
|
||||||
num_workers: 16
|
|
||||||
batch_size: 16
|
|
||||||
wrap: True
|
|
||||||
train:
|
|
||||||
target: ldm.data.localdanbooruvae.LocalDanbooruBaseVAE
|
|
||||||
params:
|
|
||||||
data_root: "../dataset"
|
|
||||||
size: 256
|
|
||||||
mode: "train"
|
|
||||||
downscale_f: 8
|
|
||||||
validation:
|
|
||||||
target: ldm.data.localdanbooruvae.LocalDanbooruBaseVAE
|
|
||||||
params:
|
|
||||||
data_root: "../dataset"
|
|
||||||
size: 256
|
|
||||||
mode: "val"
|
|
||||||
val_split: 64
|
|
||||||
downscale_f: 8
|
|
||||||
|
|
||||||
lightning:
|
|
||||||
callbacks:
|
|
||||||
image_logger:
|
|
||||||
target: main.ImageLogger
|
|
||||||
params:
|
|
||||||
batch_frequency: 200
|
|
||||||
max_images: 4
|
|
||||||
increase_log_steps: True
|
|
||||||
|
|
||||||
trainer:
|
|
||||||
find_unused_parameters: True
|
|
||||||
benchmark: True
|
|
||||||
limit_val_batches: 0
|
|
||||||
num_sanity_val_steps: 0
|
|
||||||
accumulate_grad_batches: 1
|
|
||||||
|
|
@ -1,117 +0,0 @@
|
||||||
model:
|
|
||||||
base_learning_rate: 5.0e-06
|
|
||||||
target: ldm.models.diffusion.ddpm.LatentDiffusion
|
|
||||||
params:
|
|
||||||
linear_start: 0.00085
|
|
||||||
linear_end: 0.0120
|
|
||||||
num_timesteps_cond: 1
|
|
||||||
log_every_t: 200
|
|
||||||
timesteps: 1000
|
|
||||||
first_stage_key: image
|
|
||||||
cond_stage_key: caption
|
|
||||||
image_size: 64
|
|
||||||
channels: 4
|
|
||||||
cond_stage_trainable: false # Note: different from the one we trained before
|
|
||||||
conditioning_key: crossattn
|
|
||||||
monitor: val/loss_simple_ema
|
|
||||||
scale_factor: 0.18215
|
|
||||||
|
|
||||||
scheduler_config: # 10000 warmup steps
|
|
||||||
target: ldm.lr_scheduler.LambdaLinearScheduler
|
|
||||||
params:
|
|
||||||
warm_up_steps: [ 1 ] # NOTE for resuming. use 10000 if starting from scratch
|
|
||||||
cycle_lengths: [ 10000000000000 ] # incredibly large number to prevent corner cases
|
|
||||||
f_start: [ 1.e-6 ]
|
|
||||||
f_max: [ 1. ]
|
|
||||||
f_min: [ 1. ]
|
|
||||||
|
|
||||||
unet_config:
|
|
||||||
target: ldm.modules.diffusionmodules.openaimodel.UNetModel
|
|
||||||
params:
|
|
||||||
image_size: 32 # unused
|
|
||||||
in_channels: 4
|
|
||||||
out_channels: 4
|
|
||||||
model_channels: 320
|
|
||||||
attention_resolutions: [ 4, 2, 1 ]
|
|
||||||
num_res_blocks: 2
|
|
||||||
channel_mult: [ 1, 2, 4, 4 ]
|
|
||||||
num_heads: 8
|
|
||||||
use_spatial_transformer: True
|
|
||||||
transformer_depth: 1
|
|
||||||
context_dim: 768
|
|
||||||
use_checkpoint: True
|
|
||||||
legacy: False
|
|
||||||
|
|
||||||
first_stage_config:
|
|
||||||
target: ldm.models.autoencoder.AutoencoderKL
|
|
||||||
params:
|
|
||||||
embed_dim: 4
|
|
||||||
monitor: val/rec_loss
|
|
||||||
ddconfig:
|
|
||||||
double_z: true
|
|
||||||
z_channels: 4
|
|
||||||
resolution: 512
|
|
||||||
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
|
|
||||||
|
|
||||||
cond_stage_config:
|
|
||||||
target: ldm.modules.encoders.modules.FrozenCLIPEmbedder
|
|
||||||
params:
|
|
||||||
penultimate: True
|
|
||||||
extended_mode: True
|
|
||||||
max_chunks: 3
|
|
||||||
|
|
||||||
data:
|
|
||||||
target: main.DataModuleFromConfig
|
|
||||||
params:
|
|
||||||
batch_size: 4
|
|
||||||
num_workers: 4
|
|
||||||
wrap: false
|
|
||||||
train:
|
|
||||||
target: ldm.data.local.LocalBase
|
|
||||||
params:
|
|
||||||
size: 512
|
|
||||||
mode: "train"
|
|
||||||
validation:
|
|
||||||
target: ldm.data.local.LocalBase
|
|
||||||
params:
|
|
||||||
size: 512
|
|
||||||
mode: "val"
|
|
||||||
val_split: 64
|
|
||||||
|
|
||||||
lightning:
|
|
||||||
modelcheckpoint:
|
|
||||||
params:
|
|
||||||
every_n_train_steps: 500
|
|
||||||
callbacks:
|
|
||||||
image_logger:
|
|
||||||
target: main.ImageLogger
|
|
||||||
params:
|
|
||||||
batch_frequency: 500
|
|
||||||
max_images: 4
|
|
||||||
increase_log_steps: False
|
|
||||||
log_first_step: False
|
|
||||||
log_images_kwargs:
|
|
||||||
use_ema_scope: False
|
|
||||||
inpaint: False
|
|
||||||
plot_progressive_rows: False
|
|
||||||
plot_diffusion_rows: False
|
|
||||||
N: 4
|
|
||||||
ddim_steps: 50
|
|
||||||
|
|
||||||
trainer:
|
|
||||||
benchmark: True
|
|
||||||
val_check_interval: 5000000
|
|
||||||
num_sanity_val_steps: 0
|
|
||||||
accumulate_grad_batches: 1
|
|
||||||
|
|
@ -1,113 +0,0 @@
|
||||||
model:
|
|
||||||
base_learning_rate: 1.5e-06
|
|
||||||
target: ldm.models.diffusion.ddpm.LatentDiffusion
|
|
||||||
params:
|
|
||||||
linear_start: 0.00085
|
|
||||||
linear_end: 0.0120
|
|
||||||
num_timesteps_cond: 1
|
|
||||||
log_every_t: 200
|
|
||||||
timesteps: 1000
|
|
||||||
first_stage_key: image
|
|
||||||
cond_stage_key: caption
|
|
||||||
image_size: 64
|
|
||||||
channels: 4
|
|
||||||
cond_stage_trainable: false # Note: different from the one we trained before
|
|
||||||
conditioning_key: crossattn
|
|
||||||
monitor: val/loss_simple_ema
|
|
||||||
scale_factor: 0.18215
|
|
||||||
|
|
||||||
scheduler_config: # 10000 warmup steps
|
|
||||||
target: ldm.lr_scheduler.LambdaLinearScheduler
|
|
||||||
params:
|
|
||||||
warm_up_steps: [ 1 ] # NOTE for resuming. use 10000 if starting from scratch
|
|
||||||
cycle_lengths: [ 10000000000000 ] # incredibly large number to prevent corner cases
|
|
||||||
f_start: [ 1.e-6 ]
|
|
||||||
f_max: [ 1. ]
|
|
||||||
f_min: [ 1. ]
|
|
||||||
|
|
||||||
unet_config:
|
|
||||||
target: ldm.modules.diffusionmodules.openaimodel.UNetModel
|
|
||||||
params:
|
|
||||||
image_size: 32 # unused
|
|
||||||
in_channels: 4
|
|
||||||
out_channels: 4
|
|
||||||
model_channels: 320
|
|
||||||
attention_resolutions: [ 4, 2, 1 ]
|
|
||||||
num_res_blocks: 2
|
|
||||||
channel_mult: [ 1, 2, 4, 4 ]
|
|
||||||
num_heads: 8
|
|
||||||
use_spatial_transformer: True
|
|
||||||
transformer_depth: 1
|
|
||||||
context_dim: 768
|
|
||||||
use_checkpoint: True
|
|
||||||
legacy: False
|
|
||||||
|
|
||||||
first_stage_config:
|
|
||||||
target: ldm.models.autoencoder.AutoencoderKL
|
|
||||||
params:
|
|
||||||
embed_dim: 4
|
|
||||||
monitor: val/rec_loss
|
|
||||||
ddconfig:
|
|
||||||
double_z: true
|
|
||||||
z_channels: 4
|
|
||||||
resolution: 512
|
|
||||||
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
|
|
||||||
|
|
||||||
cond_stage_config:
|
|
||||||
target: ldm.modules.encoders.modules.FrozenCLIPEmbedder
|
|
||||||
|
|
||||||
data:
|
|
||||||
target: main.DataModuleFromConfig
|
|
||||||
params:
|
|
||||||
batch_size: 4
|
|
||||||
num_workers: 4
|
|
||||||
wrap: false
|
|
||||||
train:
|
|
||||||
target: ldm.data.local.LocalBase
|
|
||||||
params:
|
|
||||||
size: 512
|
|
||||||
mode: "train"
|
|
||||||
validation:
|
|
||||||
target: ldm.data.local.LocalBase
|
|
||||||
params:
|
|
||||||
size: 512
|
|
||||||
mode: "val"
|
|
||||||
val_split: 64
|
|
||||||
|
|
||||||
lightning:
|
|
||||||
modelcheckpoint:
|
|
||||||
params:
|
|
||||||
every_n_train_steps: 500
|
|
||||||
callbacks:
|
|
||||||
image_logger:
|
|
||||||
target: main.ImageLogger
|
|
||||||
params:
|
|
||||||
batch_frequency: 500
|
|
||||||
max_images: 4
|
|
||||||
increase_log_steps: False
|
|
||||||
log_first_step: False
|
|
||||||
log_images_kwargs:
|
|
||||||
use_ema_scope: False
|
|
||||||
inpaint: False
|
|
||||||
plot_progressive_rows: False
|
|
||||||
plot_diffusion_rows: False
|
|
||||||
N: 4
|
|
||||||
ddim_steps: 50
|
|
||||||
|
|
||||||
trainer:
|
|
||||||
benchmark: True
|
|
||||||
val_check_interval: 5000000
|
|
||||||
num_sanity_val_steps: 0
|
|
||||||
accumulate_grad_batches: 1
|
|
||||||
|
|
@ -1,113 +0,0 @@
|
||||||
model:
|
|
||||||
base_learning_rate: 1.5e-06
|
|
||||||
target: ldm.models.diffusion.ddpm.LatentDiffusion
|
|
||||||
params:
|
|
||||||
linear_start: 0.00085
|
|
||||||
linear_end: 0.0120
|
|
||||||
num_timesteps_cond: 1
|
|
||||||
log_every_t: 200
|
|
||||||
timesteps: 1000
|
|
||||||
first_stage_key: image
|
|
||||||
cond_stage_key: caption
|
|
||||||
image_size: 64
|
|
||||||
channels: 4
|
|
||||||
cond_stage_trainable: false # Note: different from the one we trained before
|
|
||||||
conditioning_key: crossattn
|
|
||||||
monitor: val/loss_simple_ema
|
|
||||||
scale_factor: 0.18215
|
|
||||||
|
|
||||||
scheduler_config: # 10000 warmup steps
|
|
||||||
target: ldm.lr_scheduler.LambdaLinearScheduler
|
|
||||||
params:
|
|
||||||
warm_up_steps: [ 1 ] # NOTE for resuming. use 10000 if starting from scratch
|
|
||||||
cycle_lengths: [ 10000000000000 ] # incredibly large number to prevent corner cases
|
|
||||||
f_start: [ 1.e-6 ]
|
|
||||||
f_max: [ 1. ]
|
|
||||||
f_min: [ 1. ]
|
|
||||||
|
|
||||||
unet_config:
|
|
||||||
target: ldm.modules.diffusionmodules.openaimodel.UNetModel
|
|
||||||
params:
|
|
||||||
image_size: 32 # unused
|
|
||||||
in_channels: 4
|
|
||||||
out_channels: 4
|
|
||||||
model_channels: 320
|
|
||||||
attention_resolutions: [ 4, 2, 1 ]
|
|
||||||
num_res_blocks: 2
|
|
||||||
channel_mult: [ 1, 2, 4, 4 ]
|
|
||||||
num_heads: 8
|
|
||||||
use_spatial_transformer: True
|
|
||||||
transformer_depth: 1
|
|
||||||
context_dim: 768
|
|
||||||
use_checkpoint: True
|
|
||||||
legacy: False
|
|
||||||
|
|
||||||
first_stage_config:
|
|
||||||
target: ldm.models.autoencoder.AutoencoderKL
|
|
||||||
params:
|
|
||||||
embed_dim: 4
|
|
||||||
monitor: val/rec_loss
|
|
||||||
ddconfig:
|
|
||||||
double_z: true
|
|
||||||
z_channels: 4
|
|
||||||
resolution: 512
|
|
||||||
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
|
|
||||||
|
|
||||||
cond_stage_config:
|
|
||||||
target: ldm.modules.encoders.modules.FrozenCLIPEmbedder
|
|
||||||
|
|
||||||
data:
|
|
||||||
target: ldm.data.localdanbooru.DanbooruWebDataModuleFromConfig
|
|
||||||
params:
|
|
||||||
tar_base: "links.tar"
|
|
||||||
batch_size: 1
|
|
||||||
num_workers: 1
|
|
||||||
max_size: 768
|
|
||||||
resize: false
|
|
||||||
flip_p: 0.5
|
|
||||||
image_key: "image"
|
|
||||||
copyright_rate: 1.0
|
|
||||||
character_rate: 1.0
|
|
||||||
general_rate: 1.0
|
|
||||||
artist_rate: 1.0
|
|
||||||
normalize: true
|
|
||||||
caption_shuffle: true
|
|
||||||
random_order: true
|
|
||||||
|
|
||||||
lightning:
|
|
||||||
modelcheckpoint:
|
|
||||||
params:
|
|
||||||
every_n_train_steps: 500
|
|
||||||
callbacks:
|
|
||||||
image_logger:
|
|
||||||
target: main.ImageLogger
|
|
||||||
params:
|
|
||||||
batch_frequency: 500
|
|
||||||
max_images: 4
|
|
||||||
increase_log_steps: False
|
|
||||||
log_first_step: False
|
|
||||||
log_images_kwargs:
|
|
||||||
use_ema_scope: False
|
|
||||||
inpaint: False
|
|
||||||
plot_progressive_rows: False
|
|
||||||
plot_diffusion_rows: False
|
|
||||||
N: 4
|
|
||||||
ddim_steps: 50
|
|
||||||
|
|
||||||
trainer:
|
|
||||||
benchmark: True
|
|
||||||
val_check_interval: 5000000
|
|
||||||
num_sanity_val_steps: 0
|
|
||||||
accumulate_grad_batches: 1
|
|
||||||
|
|
@ -1,116 +0,0 @@
|
||||||
model:
|
|
||||||
base_learning_rate: 1.5e-06
|
|
||||||
target: ldm.models.diffusion.ddpm.LatentDiffusion
|
|
||||||
params:
|
|
||||||
linear_start: 0.00085
|
|
||||||
linear_end: 0.0120
|
|
||||||
num_timesteps_cond: 1
|
|
||||||
log_every_t: 200
|
|
||||||
timesteps: 1000
|
|
||||||
first_stage_key: image
|
|
||||||
cond_stage_key: caption
|
|
||||||
image_size: 64
|
|
||||||
channels: 4
|
|
||||||
cond_stage_trainable: false # Note: different from the one we trained before
|
|
||||||
conditioning_key: crossattn
|
|
||||||
monitor: val/loss_simple_ema
|
|
||||||
scale_factor: 0.18215
|
|
||||||
|
|
||||||
scheduler_config: # 10000 warmup steps
|
|
||||||
target: ldm.lr_scheduler.LambdaLinearScheduler
|
|
||||||
params:
|
|
||||||
warm_up_steps: [ 1 ] # NOTE for resuming. use 10000 if starting from scratch
|
|
||||||
cycle_lengths: [ 10000000000000 ] # incredibly large number to prevent corner cases
|
|
||||||
f_start: [ 1.e-6 ]
|
|
||||||
f_max: [ 1. ]
|
|
||||||
f_min: [ 1. ]
|
|
||||||
|
|
||||||
unet_config:
|
|
||||||
target: ldm.modules.diffusionmodules.openaimodel.UNetModel
|
|
||||||
params:
|
|
||||||
image_size: 32 # unused
|
|
||||||
in_channels: 4
|
|
||||||
out_channels: 4
|
|
||||||
model_channels: 320
|
|
||||||
attention_resolutions: [ 4, 2, 1 ]
|
|
||||||
num_res_blocks: 2
|
|
||||||
channel_mult: [ 1, 2, 4, 4 ]
|
|
||||||
num_heads: 8
|
|
||||||
use_spatial_transformer: True
|
|
||||||
transformer_depth: 1
|
|
||||||
context_dim: 768
|
|
||||||
use_checkpoint: True
|
|
||||||
legacy: False
|
|
||||||
|
|
||||||
first_stage_config:
|
|
||||||
target: ldm.models.autoencoder.AutoencoderKL
|
|
||||||
params:
|
|
||||||
embed_dim: 4
|
|
||||||
monitor: val/rec_loss
|
|
||||||
ddconfig:
|
|
||||||
double_z: true
|
|
||||||
z_channels: 4
|
|
||||||
resolution: 512
|
|
||||||
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
|
|
||||||
|
|
||||||
cond_stage_config:
|
|
||||||
target: ldm.modules.encoders.modules.FrozenCLIPEmbedder
|
|
||||||
|
|
||||||
data:
|
|
||||||
target: main.DataModuleFromConfig
|
|
||||||
params:
|
|
||||||
batch_size: 1
|
|
||||||
num_workers: 1
|
|
||||||
wrap: false
|
|
||||||
train:
|
|
||||||
target: ldm.data.localdanboorubase.LocalDanbooruBase
|
|
||||||
params:
|
|
||||||
data_root: "./dataset"
|
|
||||||
size: 768
|
|
||||||
mode: "train"
|
|
||||||
validation:
|
|
||||||
target: ldm.data.localdanboorubase.LocalDanbooruBase
|
|
||||||
params:
|
|
||||||
data_root: "./dataset"
|
|
||||||
size: 768
|
|
||||||
mode: "val"
|
|
||||||
val_split: 64
|
|
||||||
|
|
||||||
lightning:
|
|
||||||
find_unused_parameters: False
|
|
||||||
modelcheckpoint:
|
|
||||||
params:
|
|
||||||
every_n_train_steps: 2000
|
|
||||||
callbacks:
|
|
||||||
image_logger:
|
|
||||||
target: main.ImageLogger
|
|
||||||
params:
|
|
||||||
batch_frequency: 2000
|
|
||||||
max_images: 2
|
|
||||||
increase_log_steps: False
|
|
||||||
log_first_step: False
|
|
||||||
log_images_kwargs:
|
|
||||||
use_ema_scope: False
|
|
||||||
inpaint: False
|
|
||||||
plot_progressive_rows: False
|
|
||||||
plot_diffusion_rows: False
|
|
||||||
N: 4
|
|
||||||
ddim_steps: 50
|
|
||||||
|
|
||||||
trainer:
|
|
||||||
benchmark: True
|
|
||||||
val_check_interval: 5000000
|
|
||||||
num_sanity_val_steps: 0
|
|
||||||
accumulate_grad_batches: 1
|
|
||||||
|
|
@ -1,100 +0,0 @@
|
||||||
model:
|
|
||||||
base_learning_rate: 1.0e-04
|
|
||||||
target: ldm.models.diffusion.ddpm.LatentDiffusion
|
|
||||||
params:
|
|
||||||
linear_start: 0.00085
|
|
||||||
linear_end: 0.0120
|
|
||||||
num_timesteps_cond: 1
|
|
||||||
log_every_t: 50
|
|
||||||
timesteps: 1000
|
|
||||||
first_stage_key: image
|
|
||||||
cond_stage_key: caption
|
|
||||||
image_size: 64
|
|
||||||
channels: 4
|
|
||||||
cond_stage_trainable: true # Note: different from the one we trained before
|
|
||||||
conditioning_key: crossattn
|
|
||||||
monitor: val/loss_simple_ema
|
|
||||||
scale_factor: 0.18215
|
|
||||||
|
|
||||||
scheduler_config: # 10000 warmup steps
|
|
||||||
target: ldm.lr_scheduler.LambdaLinearScheduler
|
|
||||||
params:
|
|
||||||
warm_up_steps: [ 1 ] # NOTE for resuming. use 10000 if starting from scratch
|
|
||||||
cycle_lengths: [ 10000000000000 ] # incredibly large number to prevent corner cases
|
|
||||||
f_start: [ 1.e-6 ]
|
|
||||||
f_max: [ 1. ]
|
|
||||||
f_min: [ 1. ]
|
|
||||||
|
|
||||||
unet_config:
|
|
||||||
target: ldm.modules.diffusionmodules.openaimodel.UNetModel
|
|
||||||
params:
|
|
||||||
image_size: 32 # unused
|
|
||||||
in_channels: 4
|
|
||||||
out_channels: 4
|
|
||||||
model_channels: 320
|
|
||||||
attention_resolutions: [ 4, 2, 1 ]
|
|
||||||
num_res_blocks: 2
|
|
||||||
channel_mult: [ 1, 2, 4, 4 ]
|
|
||||||
num_heads: 8
|
|
||||||
use_spatial_transformer: True
|
|
||||||
transformer_depth: 1
|
|
||||||
context_dim: 768
|
|
||||||
use_checkpoint: True
|
|
||||||
legacy: False
|
|
||||||
|
|
||||||
first_stage_config:
|
|
||||||
target: ldm.models.autoencoder.AutoencoderKL
|
|
||||||
params:
|
|
||||||
embed_dim: 4
|
|
||||||
monitor: val/rec_loss
|
|
||||||
ddconfig:
|
|
||||||
double_z: true
|
|
||||||
z_channels: 4
|
|
||||||
resolution: 512
|
|
||||||
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
|
|
||||||
|
|
||||||
cond_stage_config:
|
|
||||||
target: ldm.modules.encoders.modules.FrozenCLIPEmbedder
|
|
||||||
|
|
||||||
data:
|
|
||||||
target: main.DataModuleFromConfig
|
|
||||||
params:
|
|
||||||
batch_size: 1
|
|
||||||
num_workers: 1
|
|
||||||
wrap: false
|
|
||||||
train:
|
|
||||||
target: ldm.data.local.LocalBase
|
|
||||||
params:
|
|
||||||
size: 512
|
|
||||||
validation:
|
|
||||||
target: ldm.data.local.LocalBase
|
|
||||||
params:
|
|
||||||
size: 512
|
|
||||||
|
|
||||||
lightning:
|
|
||||||
modelcheckpoint:
|
|
||||||
params:
|
|
||||||
every_n_train_steps: 500
|
|
||||||
callbacks:
|
|
||||||
image_logger:
|
|
||||||
target: main.ImageLogger
|
|
||||||
params:
|
|
||||||
batch_frequency: 500
|
|
||||||
max_images: 4
|
|
||||||
increase_log_steps: False
|
|
||||||
|
|
||||||
trainer:
|
|
||||||
benchmark: True
|
|
||||||
max_steps: 6100
|
|
||||||
|
|
@ -1,73 +0,0 @@
|
||||||
model:
|
|
||||||
base_learning_rate: 1.0e-04
|
|
||||||
target: ldm.models.diffusion.ddpm.LatentDiffusion
|
|
||||||
params:
|
|
||||||
linear_start: 0.00085
|
|
||||||
linear_end: 0.0120
|
|
||||||
num_timesteps_cond: 1
|
|
||||||
log_every_t: 200
|
|
||||||
timesteps: 1000
|
|
||||||
first_stage_key: "jpg"
|
|
||||||
cond_stage_key: "txt"
|
|
||||||
image_size: 64
|
|
||||||
channels: 4
|
|
||||||
cond_stage_trainable: false # Note: different from the one we trained before
|
|
||||||
conditioning_key: crossattn
|
|
||||||
monitor: val/loss_simple_ema
|
|
||||||
scale_factor: 0.18215
|
|
||||||
use_ema: False
|
|
||||||
|
|
||||||
scheduler_config: # 10000 warmup steps
|
|
||||||
target: ldm.lr_scheduler.LambdaLinearScheduler
|
|
||||||
params:
|
|
||||||
warm_up_steps: [ 10000 ]
|
|
||||||
cycle_lengths: [ 10000000000000 ] # incredibly large number to prevent corner cases
|
|
||||||
f_start: [ 1.e-6 ]
|
|
||||||
f_max: [ 1. ]
|
|
||||||
f_min: [ 1. ]
|
|
||||||
|
|
||||||
unet_config:
|
|
||||||
target: ldm.modules.diffusionmodules.openaimodel.UNetModel
|
|
||||||
params:
|
|
||||||
image_size: 32 # unused
|
|
||||||
in_channels: 4
|
|
||||||
out_channels: 4
|
|
||||||
model_channels: 320
|
|
||||||
attention_resolutions: [ 4, 2, 1 ]
|
|
||||||
num_res_blocks: 2
|
|
||||||
channel_mult: [ 1, 2, 4, 4 ]
|
|
||||||
num_heads: 8
|
|
||||||
use_spatial_transformer: True
|
|
||||||
transformer_depth: 1
|
|
||||||
context_dim: 768
|
|
||||||
use_checkpoint: True
|
|
||||||
legacy: False
|
|
||||||
|
|
||||||
first_stage_config:
|
|
||||||
target: ldm.models.autoencoder.AutoencoderKL
|
|
||||||
params:
|
|
||||||
embed_dim: 4
|
|
||||||
monitor: val/rec_loss
|
|
||||||
ddconfig:
|
|
||||||
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
|
|
||||||
|
|
||||||
cond_stage_config:
|
|
||||||
target: ldm.modules.encoders.modules.FrozenCLIPEmbedder
|
|
||||||
params:
|
|
||||||
penultimate: True
|
|
||||||
extended_mode: 3
|
|
||||||
Binary file not shown.
Binary file not shown.
|
Before Width: | Height: | Size: 14 KiB |
|
|
@ -1 +0,0 @@
|
||||||
A basket of cerries
|
|
||||||
File diff suppressed because it is too large
Load Diff
Binary file not shown.
Binary file not shown.
File diff suppressed because it is too large
Load Diff
Binary file not shown.
|
Before Width: | Height: | Size: 466 KiB |
Binary file not shown.
|
Before Width: | Height: | Size: 7.4 KiB |
Binary file not shown.
|
Before Width: | Height: | Size: 539 KiB |
Binary file not shown.
|
Before Width: | Height: | Size: 7.6 KiB |
Binary file not shown.
|
Before Width: | Height: | Size: 450 KiB |
Binary file not shown.
|
Before Width: | Height: | Size: 12 KiB |
Binary file not shown.
|
Before Width: | Height: | Size: 553 KiB |
Binary file not shown.
|
Before Width: | Height: | Size: 12 KiB |
Binary file not shown.
|
Before Width: | Height: | Size: 418 KiB |
Binary file not shown.
|
Before Width: | Height: | Size: 6.1 KiB |
Binary file not shown.
|
Before Width: | Height: | Size: 542 KiB |
Binary file not shown.
|
Before Width: | Height: | Size: 9.5 KiB |
Binary file not shown.
|
Before Width: | Height: | Size: 395 KiB |
Binary file not shown.
|
Before Width: | Height: | Size: 12 KiB |
Binary file not shown.
|
Before Width: | Height: | Size: 465 KiB |
Binary file not shown.
|
Before Width: | Height: | Size: 7.8 KiB |
|
|
@ -1,7 +0,0 @@
|
||||||
# Documentation
|
|
||||||
|
|
||||||
Waifu Diffusion is a project based off CompVis/Stable-Diffusion.
|
|
||||||
|
|
||||||
For guidance on how to start training, see [training](./training/README.md).
|
|
||||||
|
|
||||||
For a list of trained weights, see [weights](./weights/README.md).
|
|
||||||
|
|
@ -1,8 +0,0 @@
|
||||||
# Training documentation
|
|
||||||
Training is available with waifu-diffusion. Before starting, we remind you that, at this moment at least 30GB of VRAM is needed, along with at least 30gb of storage if you don't mind cleaning up every so often.
|
|
||||||
## Contents
|
|
||||||
1. [Dataset](./dataset.md)
|
|
||||||
2. [Configuration](./configuration.md)
|
|
||||||
3. [Executing](./executing.md)
|
|
||||||
4. Recommendations
|
|
||||||
5. FAQ
|
|
||||||
|
|
@ -1,3 +0,0 @@
|
||||||
# 2. Configuration
|
|
||||||
This section is to be done on the machine where you are going to train.
|
|
||||||
Soon because my instance is on maintenance
|
|
||||||
|
|
@ -1,120 +0,0 @@
|
||||||
# 1. Dataset
|
|
||||||
|
|
||||||
In this guide we are going to use the Danbooru2021 dataset by Gwern.net. You are free to use any other dataset as long as you know how to convert it to the right format.
|
|
||||||
|
|
||||||
## Contents
|
|
||||||
1. Dataset requirements
|
|
||||||
2. Downloading the dataset
|
|
||||||
3. Organizing the dataset
|
|
||||||
4. Packaging the dataset
|
|
||||||
|
|
||||||
## Dataset requirements
|
|
||||||
|
|
||||||
The dataset needs to be in the following format
|
|
||||||
|
|
||||||
/dataset/ : Root dataset folder, can be any name
|
|
||||||
|
|
||||||
/dataset/img/ : Folder for images
|
|
||||||
|
|
||||||
/dataset/txt/ : Folder for text files
|
|
||||||
|
|
||||||
It is recommended to have the images in 512x512 resolution and in JPG format. While the text files need to have the same name as the images it refers to.
|
|
||||||
|
|
||||||
Foe example:
|
|
||||||
````
|
|
||||||
mydataset
|
|
||||||
├── img
|
|
||||||
│ └── image001.jpg
|
|
||||||
└── txt
|
|
||||||
└── image001.txt
|
|
||||||
````
|
|
||||||
Where image001.txt has the tags (prompt) to be used for image001.jpg
|
|
||||||
|
|
||||||
## Downloading the dataset
|
|
||||||
This is optional; If you have your own dataset skip this part.
|
|
||||||
|
|
||||||
### Downloading Rsync
|
|
||||||
Danbooru2021 is available for download through rsync.
|
|
||||||
#### Linux
|
|
||||||
On Linux, you should be able to install rsync via your package manager.
|
|
||||||
````bash
|
|
||||||
apt install rsync
|
|
||||||
````
|
|
||||||
#### Windows
|
|
||||||
On Windows, you are going to need to install Cygwin, a posix runtime for Windows which allows the usage of many linux-only programs inside windows.
|
|
||||||
|
|
||||||
[Cygwin Installer for x86](https://www.cygwin.com/setup-x86_64.exe)
|
|
||||||
|
|
||||||
On the installer, select mirrors.kernel.org for Download Site:
|
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
Next, search for "rsync" on the search bar, change "View: Pending" to "View: Full", and select on the "New" tab the latest version. Do the same for "zip".
|
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
GIF explaining the entire process:
|
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
Once the installation is finished, you should see "Cygwin64 Terminal" on your Start Menu. Launch it and you should be greated by the following window:
|
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
You may now follow the intructions
|
|
||||||
|
|
||||||
### Downloading the dataset
|
|
||||||
Remember that instructions here apply universally, both on Linux and Windows (If you are using Cygwin that is).
|
|
||||||
|
|
||||||
The entire dataset weights about 5TB. You are not going to download everything, instead, you are only going to download two kinds of files:
|
|
||||||
|
|
||||||
1. The images
|
|
||||||
2. The JSON files (metadata)
|
|
||||||
|
|
||||||
If you want to see the entire file list, you can refer to the [Danbooru2021 information site](https://www.gwern.net/Danbooru2021).
|
|
||||||
|
|
||||||
We are going to extract the images from the 512px folder for convinience, since this folder already has the images resized to 512x512 resolution in JPG format. It only has safe rated images, for NSFW refer to [gwern.net](https://www.gwern.net/Danbooru2021#samples).
|
|
||||||
|
|
||||||
Folders from 0000 to 0009.
|
|
||||||
> The folders are named according to the last 3 digits of the image ID on danbooru. Images on folder 0001 will have its ID end on 001.
|
|
||||||
|
|
||||||
We are also going to download the only the first JSON batch. If you want to train on more data you should download more JSON batches.
|
|
||||||
|
|
||||||
Download the 512px folders from 0000 to 0009 (3.86GB):
|
|
||||||
```bash
|
|
||||||
rsync -r rsync://176.9.41.242:873/danbooru2021/512px/000* ./512px/
|
|
||||||
```
|
|
||||||
Download the first batch of metadata, posts000000000000.json (800MB):
|
|
||||||
``` shell
|
|
||||||
rsync rsync://176.9.41.242:873/danbooru2021/metadata/posts000000000000.json ./metadata/
|
|
||||||
```
|
|
||||||
You should now have two folders named: 512px and metadata.
|
|
||||||
|
|
||||||
## Organizing the dataset
|
|
||||||
Although we have the dataset, the metadata that explains what the image is, is inside the JSON file. In order to extract the data into individual txt files, we are going to use the script inside ``danbooru_data/local/extractfromjson_danboo21.py``
|
|
||||||
|
|
||||||
Assuming you are in the same directory as metadata and 512px folder:
|
|
||||||
````bash
|
|
||||||
python danbooru_data/local/extractfromjson_danboo21.py -J metadata/posts000000000000.json -E danbooru-aesthetic
|
|
||||||
````
|
|
||||||
|
|
||||||
Once the script has finished, you should have a "danbooru-aesthetic" folder, whose insides look like this:
|
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
## Packaging the dataset
|
|
||||||
Next we need to put the extracted data into the format required in the section "Dataset requirements". Run the following commands:
|
|
||||||
``` shell
|
|
||||||
mkdir danbooru-aesthetic/img danbooru-aesthetic/txt
|
|
||||||
mv danbooru-aesthetic/*.jpg danbooru-aesthetic/img
|
|
||||||
mv danbooru-aesthetic/*.txt danbooru-aesthetic/txt
|
|
||||||
```
|
|
||||||
|
|
||||||
In order to reduce size, zip the contents of labeled_data:
|
|
||||||
``` shell
|
|
||||||
zip -r danbooru-aesthetic.zip danbooru-aesthetic
|
|
||||||
```
|
|
||||||
This will package the entire danbooru-aesthetic folder into a zip file. This command DOES NOT output any information in the terminal, so be patient.
|
|
||||||
|
|
||||||
## Finish
|
|
||||||
You can now continue to Configure
|
|
||||||
|
|
@ -1,51 +0,0 @@
|
||||||
# 3. Executing
|
|
||||||
|
|
||||||
There are two modes of executing the training:
|
|
||||||
1. Using docker image. This is the fastest way to get started.
|
|
||||||
2. Using system python install. Allows more customization.
|
|
||||||
|
|
||||||
Note: You will need to provide the initial checkpoint for resuming the training. This must be a version with the full EMA. Otherwise you will get this error:
|
|
||||||
```
|
|
||||||
RuntimeError: Error(s) in loading state_dict for LatentDiffusion:
|
|
||||||
Missing key(s) in state_dict: "model_ema.diffusion_modeltime_embed0weight", "model_ema.diffusion_modeltime_embed0bias".... (Many lines of similar outputs)
|
|
||||||
```
|
|
||||||
|
|
||||||
## 1. Using docker image
|
|
||||||
|
|
||||||
An image is provided at `ghcr.io/derfred/waifu-diffusion`. Execute it using by adjusting the NUM_GPU variable:
|
|
||||||
```
|
|
||||||
docker run -it -e NUM_GPU=x ghcr.io/derfred/waifu-diffusion
|
|
||||||
```
|
|
||||||
|
|
||||||
Next you will want to download the starting checkpoint into the file `model.ckpt` and copy the training data in the directory `/waifu/danbooru-aesthetic`.
|
|
||||||
|
|
||||||
Finally execute the training using:
|
|
||||||
```
|
|
||||||
sh train.sh -t -n "aesthetic" --resume_from_checkpoint model.ckpt --base ./configs/stable-diffusion/v1-finetune-4gpu.yaml --no-test --seed 25 --scale_lr False --data_root "./danbooru-aesthetic"
|
|
||||||
```
|
|
||||||
|
|
||||||
## 2. system python install
|
|
||||||
|
|
||||||
First install the dependencies:
|
|
||||||
```bash
|
|
||||||
pip install -r requirements.txt
|
|
||||||
```
|
|
||||||
|
|
||||||
Next you will want to download the starting checkpoint into the file `model.ckpt` and copy the training data in the directory `/waifu/danbooru-aesthetic`.
|
|
||||||
|
|
||||||
Also you will need to edit the configuration in `./configs/stable-diffusion/v1-finetune-4gpu.yaml`. In the `data` section (around line 70) change the `batch_size` and `num_workers` to the number of GPUs you are using:
|
|
||||||
```
|
|
||||||
data:
|
|
||||||
target: main.DataModuleFromConfig
|
|
||||||
params:
|
|
||||||
batch_size: 4
|
|
||||||
num_workers: 4
|
|
||||||
wrap: false
|
|
||||||
```
|
|
||||||
|
|
||||||
Finally execute the training using the following command. You need to adjust the `--gpu` parameter according to your GPU settings.
|
|
||||||
```bash
|
|
||||||
sh train.sh -t -n "aesthetic" --resume_from_checkpoint model.ckpt --base ./configs/stable-diffusion/v1-finetune-4gpu.yaml --no-test --seed 25 --scale_lr False --data_root "./danbooru-aesthetic" --gpu=0,1,2,3,
|
|
||||||
```
|
|
||||||
|
|
||||||
In case you get an error stating `KeyError: 'Trying to restore optimizer state but checkpoint contains only the model. This is probably due to ModelCheckpoint.save_weights_only being set to True.'` follow these instructions: https://discord.com/channels/930499730843250783/953132470528798811/1018668937052962908
|
|
||||||
Binary file not shown.
|
Before Width: | Height: | Size: 83 MiB |
Binary file not shown.
|
Before Width: | Height: | Size: 4.6 KiB |
Binary file not shown.
|
Before Width: | Height: | Size: 20 KiB |
Binary file not shown.
|
Before Width: | Height: | Size: 153 KiB |
Binary file not shown.
|
Before Width: | Height: | Size: 173 KiB |
|
|
@ -1,40 +0,0 @@
|
||||||
# Waifu Diffusion v1.3
|
|
||||||
|
|
||||||
Waifu Diffusion is a latent text-to-image diffusion model that has been conditioned on high-quality anime images through fine-tuning.
|
|
||||||
|
|
||||||
- [Float 16 EMA Pruned](https://huggingface.co/hakurei/waifu-diffusion-v1-3/blob/main/wd-v1-3-float16.ckpt)
|
|
||||||
- [Float 32 EMA Pruned](https://huggingface.co/hakurei/waifu-diffusion-v1-3/blob/main/wd-v1-3-float32.ckpt)
|
|
||||||
- [Float 32 Full Weights](https://huggingface.co/hakurei/waifu-diffusion-v1-3/blob/main/wd-v1-3-full.ckpt)
|
|
||||||
- [Float 32 Full Weights + Optimizer Weights (For Training)](https://huggingface.co/hakurei/waifu-diffusion-v1-3/blob/main/wd-v1-3-full-opt.ckpt)
|
|
||||||
|
|
||||||
## Model Description
|
|
||||||
|
|
||||||
The model originally used for fine-tuning is [Stable Diffusion 1.4](https://huggingface.co/CompVis/stable-diffusion-v1-4), which is a latent image diffusion model trained on [LAION2B-en](https://huggingface.co/datasets/laion/laion2B-en). The current model has been fine-tuned with a learning rate of 5.0e-6 for 10 epochs on 680k anime-styled images.
|
|
||||||
|
|
||||||
[See here for an in-depth overview of Waifu Diffusion 1.3.](https://gist.github.com/harubaru/f727cedacae336d1f7877c4bbe2196e1)
|
|
||||||
|
|
||||||
## License
|
|
||||||
|
|
||||||
This model is open access and available to all, with a CreativeML OpenRAIL-M license further specifying rights and usage.
|
|
||||||
The CreativeML OpenRAIL License specifies:
|
|
||||||
|
|
||||||
1. You can't use the model to deliberately produce nor share illegal or harmful outputs or content
|
|
||||||
2. The authors claims no rights on the outputs you generate, you are free to use them and are accountable for their use which must not go against the provisions set in the license
|
|
||||||
3. You may re-distribute the weights and use the model commercially and/or as a service. If you do, please be aware you have to include the same use restrictions as the ones in the license and share a copy of the CreativeML OpenRAIL-M to all your users (please read the license entirely and carefully)
|
|
||||||
[Please read the full license here](https://huggingface.co/spaces/CompVis/stable-diffusion-license)
|
|
||||||
|
|
||||||
## Downstream Uses
|
|
||||||
|
|
||||||
This model can be used for entertainment purposes and as a generative art assistant.
|
|
||||||
|
|
||||||
## Team Members and Acknowledgements
|
|
||||||
|
|
||||||
This project would not have been possible without the incredible work by the [CompVis Researchers](https://ommer-lab.com/).
|
|
||||||
|
|
||||||
- [Anthony Mercurio](https://github.com/harubaru)
|
|
||||||
- [Salt](https://github.com/sALTaccount/)
|
|
||||||
- [Cafe](https://twitter.com/cafeai_labs)
|
|
||||||
|
|
||||||
In order to reach us, you can join our [Discord server](https://discord.gg/touhouai).
|
|
||||||
|
|
||||||
[](https://discord.gg/touhouai)
|
|
||||||
|
|
@ -1,19 +0,0 @@
|
||||||
Waifu Diffusion v1.2
|
|
||||||
|
|
||||||
Release Date: 07/09/2022
|
|
||||||
|
|
||||||
Steps/Epochs/Images: 5 Epochs, 56,000 Images
|
|
||||||
|
|
||||||
License: None
|
|
||||||
|
|
||||||
Authors: Haru (haru#1367@discord)
|
|
||||||
|
|
||||||
Mirrors:
|
|
||||||
|
|
||||||
Google Drive (rate limit): https://drive.google.com/file/d/1XeoFCILTcc9kn_5uS-G0uqWS5XVANpha
|
|
||||||
|
|
||||||
Magnet Link: magnet:?xt=urn:btih:INEYUMLLBBMZF22IIP4AEXLUK6XQKCSD&dn=wd-v1-2-full-ema.ckpt&xl=7703810927&tr=udp%3A%2F%2Ftracker.opentrackr.org%3A1337%2Fannounce
|
|
||||||
|
|
||||||
HTTPS mirror: https://thisanimedoesnotexist.ai/downloads/wd-v1-2-full-ema.ckpt (Fastest)
|
|
||||||
|
|
||||||
HTTP mirror: http://wd.links.sd:8880/wd-v1-2-full-ema.ckpt
|
|
||||||
|
|
@ -1,32 +0,0 @@
|
||||||
name: ldm
|
|
||||||
channels:
|
|
||||||
- pytorch
|
|
||||||
- defaults
|
|
||||||
dependencies:
|
|
||||||
- git
|
|
||||||
- python=3.8.5
|
|
||||||
- pip=20.3
|
|
||||||
- cudatoolkit=11.3
|
|
||||||
- pytorch=1.11.0
|
|
||||||
- torchvision=0.12.0
|
|
||||||
- numpy=1.19.2
|
|
||||||
- pip:
|
|
||||||
- albumentations==0.4.3
|
|
||||||
- opencv-python==4.1.2.30
|
|
||||||
- pudb==2019.2
|
|
||||||
- imageio==2.9.0
|
|
||||||
- imageio-ffmpeg==0.4.2
|
|
||||||
- pytorch-lightning==1.4.2
|
|
||||||
- omegaconf==2.1.1
|
|
||||||
- test-tube>=0.7.5
|
|
||||||
- streamlit>=0.73.1
|
|
||||||
- einops==0.3.0
|
|
||||||
- torch-fidelity==0.3.0
|
|
||||||
- transformers==4.19.2
|
|
||||||
- torchmetrics==0.6.0
|
|
||||||
- kornia==0.6
|
|
||||||
- gradio==3.1.6
|
|
||||||
- -e git+https://github.com/CompVis/taming-transformers.git@master#egg=taming-transformers
|
|
||||||
- -e git+https://github.com/openai/CLIP.git@main#egg=clip
|
|
||||||
- -e git+https://github.com/hlky/k-diffusion-sd#egg=k_diffusion
|
|
||||||
- -e .
|
|
||||||
Some files were not shown because too many files have changed in this diff Show More
Loading…
Reference in New Issue