Upload folder using huggingface_hub

Abnormal-CT-Generation-MultiDisease/LICENSE

@@ -0,0 +1,674 @@
+                    GNU GENERAL PUBLIC LICENSE
+                       Version 3, 29 June 2007
+
+ 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.
+
+                            Preamble
+
+  The GNU General Public License is a free, copyleft license for
+software and other kinds of works.
+
+  The licenses for most software and other practical works are designed
+to take away your freedom to share and change the works.  By contrast,
+the GNU General Public License is 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.  We, the Free Software Foundation, use the
+GNU General Public License for most of our software; it applies also to
+any other work released this way by its authors.  You can apply it to
+your programs, too.
+
+  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.
+
+  To protect your rights, we need to prevent others from denying you
+these rights or asking you to surrender the rights.  Therefore, you have
+certain responsibilities if you distribute copies of the software, or if
+you modify it: responsibilities to respect the freedom of others.
+
+  For example, if you distribute copies of such a program, whether
+gratis or for a fee, you must pass on to the recipients the same
+freedoms that you received.  You must make sure that they, too, receive
+or can get the source code.  And you must show them these terms so they
+know their rights.
+
+  Developers that use the GNU GPL protect your rights with two steps:
+(1) assert copyright on the software, and (2) offer you this License
+giving you legal permission to copy, distribute and/or modify it.
+
+  For the developers' and authors' protection, the GPL clearly explains
+that there is no warranty for this free software.  For both users' and
+authors' sake, the GPL requires that modified versions be marked as
+changed, so that their problems will not be attributed erroneously to
+authors of previous versions.
+
+  Some devices are designed to deny users access to install or run
+modified versions of the software inside them, although the manufacturer
+can do so.  This is fundamentally incompatible with the aim of
+protecting users' freedom to change the software.  The systematic
+pattern of such abuse occurs in the area of products for individuals to
+use, which is precisely where it is most unacceptable.  Therefore, we
+have designed this version of the GPL to prohibit the practice for those
+products.  If such problems arise substantially in other domains, we
+stand ready to extend this provision to those domains in future versions
+of the GPL, as needed to protect the freedom of users.
+
+  Finally, every program is threatened constantly by software patents.
+States should not allow patents to restrict development and use of
+software on general-purpose computers, but in those that do, we wish to
+avoid the special danger that patents applied to a free program could
+make it effectively proprietary.  To prevent this, the GPL assures that
+patents cannot be used to render the program non-free.
+
+  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 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. Use with the GNU Affero General Public License.
+
+  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 Affero 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 special requirements of the GNU Affero General Public License,
+section 13, concerning interaction through a network will apply to the
+combination as such.
+
+  14. Revised Versions of this License.
+
+  The Free Software Foundation may publish revised and/or new versions of
+the GNU 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 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 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 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 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 General Public License for more details.
+
+    You should have received a copy of the GNU 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 the program does terminal interaction, make it output a short
+notice like this when it starts in an interactive mode:
+
+    <program>  Copyright (C) <year>  <name of author>
+    This program comes with ABSOLUTELY NO WARRANTY; for details type `show w'.
+    This is free software, and you are welcome to redistribute it
+    under certain conditions; type `show c' for details.
+
+The hypothetical commands `show w' and `show c' should show the appropriate
+parts of the General Public License.  Of course, your program's commands
+might be different; for a GUI interface, you would use an "about box".
+
+  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 GPL, see
+<https://www.gnu.org/licenses/>.
+
+  The GNU General Public License does not permit incorporating your program
+into proprietary programs.  If your program is a subroutine library, you
+may consider it more useful to permit linking proprietary applications with
+the library.  If this is what you want to do, use the GNU Lesser General
+Public License instead of this License.  But first, please read
+<https://www.gnu.org/licenses/why-not-lgpl.html>.

Abnormal-CT-Generation-MultiDisease/LeanVAE/__init__.py

@@ -0,0 +1,4 @@
+
+from .models.autoencoder import LeanVAE
+from .models.autoencoder_pl import AutoEncoderEngine 
+

Abnormal-CT-Generation-MultiDisease/LeanVAE/ckpts/LeanVAE-dim16.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1c1d65765e44ced040a43a0bb7084a936e5b3862d21df4a9fd13580508cd1ecb
+size 159199850

Abnormal-CT-Generation-MultiDisease/LeanVAE/data.py

@@ -0,0 +1,466 @@
+import os
+import os.path as osp
+import math
+import random
+import argparse
+import numpy as np
+from PIL import Image
+from torch.utils.data import BatchSampler, Dataset, Sampler
+import torch
+import torch.utils.data as data
+import torch.nn.functional as F
+import torch.distributed as dist
+from torchvision.datasets.video_utils import VideoClips
+import pytorch_lightning as pl
+from typing import TypeVar, Optional, Iterator, List
+from collections import Counter, defaultdict
+from decord import VideoReader
+from .utils.video_utils import VideoNorm
+
+try:
+    from torchvision.transforms import InterpolationMode
+
+    def _pil_interp(method):
+        if method == 'bicubic':
+            return InterpolationMode.BICUBIC
+        elif method == 'lanczos':
+            return InterpolationMode.LANCZOS
+        elif method == 'hamming':
+            return InterpolationMode.HAMMING
+        else:
+            # default bilinear, do we want to allow nearest?
+            return InterpolationMode.BILINEAR
+
+
+    import timm.data.transforms as timm_transforms
+
+    timm_transforms._pil_interp = _pil_interp
+except:
+    from timm.data.transforms import _pil_interp
+
+class MultiSizeVideoDataset(data.Dataset):
+    """ A flexible dataset for loading videos of different resolutions stored in a structured format.
+    This dataset reads video file paths from text files, where each file corresponds to a specific resolution (e.g., `256x256`).
+    Returns BCTHW videos in the range [-0.5, 0.5] """
+    def __init__(self, data_list, data_folder=None, sequence_length=17, train=True, sample_rate=1, dynamic_sample=False):
+        """
+        Args:
+            data_list (str): Path to the folder containing text files with video paths.
+            data_folder (Optional[str]): Root folder where videos are stored (if paths in data_list are relative).
+               
+            sequence_length: length of extracted video sequences
+        """
+        super().__init__()
+        self.train = train
+        self.data_folder = data_folder
+        self.sequence_length = sequence_length
+        self.dynamic_sample = dynamic_sample
+        self.sample_rate = sample_rate
+
+        lengths = []
+        annotations = []
+        for dir in os.listdir(data_list):
+            file_path = os.path.join(data_list, dir)
+            with open(file_path) as f:
+                annotation = [ann.strip() for ann in f.readlines()]
+                annotations.extend(annotation)
+                lengths.extend([dir] * len(annotation))
+                
+        self.annotations = annotations
+        self.lengths = lengths
+
+        self.norm = VideoNorm()
+
+    def __len__(self):
+        return len(self.annotations)
+    
+
+    def __getitem__(self, idx):
+        
+        video_path = self.annotations[idx] if self.data_folder is None else os.path.join(self.data_folder, self.annotations[idx])
+        try:
+            decord_vr = VideoReader(video_path)
+            total_frames = len(decord_vr)
+        except Exception as e:
+            raise RuntimeError(f"Failed to read video: {video_path}. Error: {e}")
+        
+        if self.dynamic_sample:
+            sample_rate = random.randint(1, self.sample_rate)
+        else:
+            sample_rate = self.sample_rate
+        
+        required_frames = self.sequence_length * sample_rate
+        if total_frames < self.sequence_length:
+            raise RuntimeError(f"Video {video_path} has only {total_frames} frames, but {self.sequence_length} are required.")
+        
+        if total_frames < required_frames:
+            sample_rate = 1
+            required_frames = self.sequence_length
+        
+        start_frame_ind = random.randint(0, max(0, total_frames - required_frames))
+        end_frame_ind = min(start_frame_ind + required_frames, total_frames)
+        frame_indice = np.linspace(
+            start_frame_ind, end_frame_ind - 1, self.sequence_length, dtype=int
+        )
+
+        video_data = decord_vr.get_batch(frame_indice).asnumpy()
+        video_data = torch.from_numpy(video_data).float()
+        video_data = video_data.permute(0, 3, 1, 2)
+
+        video = self.norm(video_data).permute(1, 0, 2, 3)
+        return {"video": video}
+
+class MultiFilesBatchVideoSampler(BatchSampler):
+    """A sampler wrapper for grouping videos within same folder into a same batch.
+    Args:
+        sampler (Sampler): Base sampler.
+        dataset (Dataset): Dataset providing data information.
+        batch_size (int): Size of mini-batch.
+        drop_last (bool): If ``True``, the sampler will drop the last batch if
+            its size would be less than ``batch_size``.
+        aspect_ratios (dict): The predefined aspect ratios.
+    """
+    def __init__(self,
+                 sampler: Sampler,
+                 dataset: Dataset,
+                 batch_size: int,
+                 train_folder: str = None,
+                 drop_last: bool = False
+                ) -> None:
+        if not isinstance(sampler, Sampler):
+            raise TypeError('sampler should be an instance of ``Sampler``, '
+                            f'but got {sampler}')
+        if not isinstance(batch_size, int) or batch_size <= 0:
+            raise ValueError('batch_size should be a positive integer value, '
+                             f'but got batch_size={batch_size}')
+        self.sampler = sampler
+        self.dataset = dataset
+        self.train_folder = train_folder
+        self.batch_size = batch_size
+        self.drop_last = drop_last
+        self.bucket = {file_name: [] for file_name in os.listdir(self.train_folder)}
+        
+        #{file_name: [list(os.listdir(os.path.join(self.train_folder, file_name)))] for file_name in os.listdir(self.train_folder)}
+        self.idx2file = []
+        
+
+    def __iter__(self):
+        for idx in self.sampler:          
+            file_name = self.idx2file[idx]
+            self.bucket[file_name].append(idx)
+            bucket = self.bucket[file_name]
+            bucket.append(idx)
+            # yield a batch of indices in the same aspect ratio group
+            if len(bucket) == self.batch_size:
+                yield bucket[:]
+                del bucket[:]
+
+
+def group_data_fun(lengths, generator=None):
+    # counter is decrease order
+    counter = Counter(lengths)  # counter {'1x256x256': 3, ''}   lengths ['1x256x256', '1x256x256', '1x256x256', ...]
+    grouped_indices = defaultdict(list)
+    for idx, item in enumerate(lengths):  # group idx to a list
+        grouped_indices[item].append(idx)
+
+    grouped_indices = dict(grouped_indices)  # {'1x256x256': [0, 1, 2], ...}
+    sorted_indices = [grouped_indices[item] for (item, _) in sorted(counter.items(), key=lambda x: x[1], reverse=True)]
+    
+    # shuffle in each group
+    shuffle_sorted_indices = []
+    for indice in sorted_indices:
+        shuffle_idx = torch.randperm(len(indice), generator=generator).tolist()
+        shuffle_sorted_indices.extend([indice[idx] for idx in shuffle_idx])
+    return shuffle_sorted_indices
+
+def last_group_data_fun(shuffled_megabatches, lengths):
+    # lengths ['1x256x256', '1x256x256', '1x256x256' ...]
+    re_shuffled_megabatches = []
+    # print('shuffled_megabatches', len(shuffled_megabatches))
+    for i_megabatch, megabatch in enumerate(shuffled_megabatches):
+        re_megabatch = []
+        for i_batch, batch in enumerate(megabatch):
+            assert len(batch) != 0
+                
+            len_each_batch = [lengths[i] for i in batch]  # ['1x256x256', '1x256x256']
+            idx_length_dict = dict([*zip(batch, len_each_batch)])  # {0: '1x256x256', 100: '1x256x256'}
+            count_dict = Counter(len_each_batch)  # {'1x256x256': 2} or {'1x256x256': 1, '1x768x256': 1}
+            if len(count_dict) != 1:
+                sorted_by_value = sorted(count_dict.items(), key=lambda item: item[1])  # {'1x256x256': 1, '1x768x256': 1}
+                # import ipdb;ipdb.set_trace()
+                # print(batch, idx_length_dict, count_dict, sorted_by_value)
+                pick_length = sorted_by_value[-1][0]  # the highest frequency
+                candidate_batch = [idx for idx, length in idx_length_dict.items() if length == pick_length]
+                random_select_batch = [random.choice(candidate_batch) for i in range(len(len_each_batch) - len(candidate_batch))]
+                # print(batch, idx_length_dict, count_dict, sorted_by_value, pick_length, candidate_batch, random_select_batch)
+                batch = candidate_batch + random_select_batch
+                # print(batch)
+
+            for i in range(1, len(batch)-1):
+                # if not lengths[batch[0]] == lengths[batch[i]]:
+                #     print(batch, [lengths[i] for i in batch])
+                #     import ipdb;ipdb.set_trace()
+                assert lengths[batch[0]] == lengths[batch[i]]
+            re_megabatch.append(batch)
+        re_shuffled_megabatches.append(re_megabatch)
+    
+    
+    # for megabatch, re_megabatch in zip(shuffled_megabatches, re_shuffled_megabatches):
+    #     for batch, re_batch in zip(megabatch, re_megabatch):
+    #         for i, re_i in zip(batch, re_batch):
+    #             if i != re_i:
+    #                 print(i, re_i)
+    return re_shuffled_megabatches
+                
+def split_to_even_chunks(megabatch, lengths, world_size, batch_size):
+    """
+    Split a list of indices into `chunks` chunks of roughly equal lengths.
+    """
+    # batch_size=2, world_size=2
+    # [1, 2, 3, 4] -> [[1, 2], [3, 4]]
+    # [1, 2, 3] -> [[1, 2], [3]]
+    # [1, 2] -> [[1], [2]]
+    # [1] -> [[1], []]
+    chunks = [megabatch[i::world_size] for i in range(world_size)]
+
+    pad_chunks = []
+    for idx, chunk in enumerate(chunks):
+        if batch_size != len(chunk):  
+            assert batch_size > len(chunk)
+            if len(chunk) != 0:  # [[1, 2], [3]] -> [[1, 2], [3, 3]]
+                chunk = chunk + [random.choice(chunk) for _ in range(batch_size - len(chunk))]
+            else:
+                chunk = random.choice(pad_chunks)  # [[1], []] -> [[1], [1]]
+                print(chunks[idx], '->', chunk)
+        pad_chunks.append(chunk)
+    return pad_chunks
+
+def get_length_grouped_indices(lengths, batch_size, world_size, gradient_accumulation_size, initial_global_step, generator=None, group_data=False, seed=42):
+    # We need to use torch for the random part as a distributed sampler will set the random seed for torch.
+    if generator is None:
+        generator = torch.Generator().manual_seed(seed)  # every rank will generate a fixed order but random index
+    # print('lengths', lengths)
+    
+    if group_data:
+        indices = group_data_fun(lengths, generator)
+    else:
+        indices = torch.randperm(len(lengths), generator=generator).tolist()
+
+    megabatch_size = world_size * batch_size
+    megabatches = [indices[i: i + megabatch_size] for i in range(0, len(lengths), megabatch_size)]
+
+    megabatches_len = [[lengths[i] for i in megabatch] for megabatch in megabatches]
+
+    megabatches = [split_to_even_chunks(megabatch, lengths, world_size, batch_size) for megabatch in megabatches]
+
+    split_to_even_chunks_len = [[[lengths[i] for i in batch] for batch in megabatch] for megabatch in megabatches]
+
+    indices_mega = torch.randperm(len(megabatches), generator=generator).tolist()
+    # print(f'rank {accelerator.process_index} seed {seed}, len(megabatches) {len(megabatches)}, indices_mega, {indices_mega[:50]}')
+    shuffled_megabatches = [megabatches[i] for i in indices_mega]
+    shuffled_megabatches_len = [[[lengths[i] for i in batch] for batch in megabatch] for megabatch in shuffled_megabatches]
+    # print(f'\nrank {accelerator.process_index} sorted shuffled_megabatches_len', shuffled_megabatches_len[0], shuffled_megabatches_len[1], shuffled_megabatches_len[-2], shuffled_megabatches_len[-1])
+
+    # import ipdb;ipdb.set_trace()
+    # print('shuffled_megabatches', len(shuffled_megabatches))
+    if group_data:
+        shuffled_megabatches = last_group_data_fun(shuffled_megabatches, lengths)
+        group_shuffled_megabatches_len = [[[lengths[i] for i in batch] for batch in megabatch] for megabatch in shuffled_megabatches]
+        # print(f'\nrank {accelerator.process_index} group_shuffled_megabatches_len', group_shuffled_megabatches_len[0], group_shuffled_megabatches_len[1], group_shuffled_megabatches_len[-2], group_shuffled_megabatches_len[-1])
+    
+ 
+    #initial_global_step = initial_global_step * gradient_accumulation_size #todo
+
+    shuffled_megabatches = shuffled_megabatches[initial_global_step:]
+    #print(f'Skip the data of {initial_global_step} step!')
+
+    return [batch for megabatch in shuffled_megabatches for batch in megabatch]
+
+class LengthGroupedSampler(Sampler):
+    r"""
+    Sampler that samples indices in a way that groups together features of the dataset of roughly the same length while
+    keeping a bit of randomness.
+    """
+
+    def __init__(
+        self,
+        batch_size: int,
+        world_size: int,
+        gradient_accumulation_size: int = 1, 
+        initial_global_step: int = 0, 
+        lengths: Optional[List[int]] = None, 
+        group_data=False, 
+        generator=None,
+        rank: Optional[int] = None, 
+        seed: int = 0, 
+    ):
+        if lengths is None:
+            raise ValueError("Lengths must be provided.")
+
+        self.batch_size = batch_size
+        self.world_size = world_size
+        self.initial_global_step = initial_global_step
+        self.gradient_accumulation_size = gradient_accumulation_size
+        self.lengths = lengths
+        self.group_data = group_data
+        self.generator = generator
+    
+        self.rank = rank
+        self.epoch = 0
+  
+        self.seed = seed
+        
+        megabatch_size = self.batch_size * self.world_size
+        self.num_samples =  ((len(lengths) + megabatch_size - 1) // megabatch_size ) * self.batch_size  #todo
+        #self.num_samples = self.num_samples #- self.initial_global_step * self.batch_size  * self.gradient_accumulation_size
+        # print('self.lengths, self.initial_global_step, self.batch_size, self.world_size, self.gradient_accumulation_size', 
+        #       len(self.lengths), self.initial_global_step, self.batch_size, self.world_size, self.gradient_accumulation_size)
+
+    def __len__(self):
+        return self.num_samples
+
+    def __iter__(self):
+        g = torch.Generator()
+        g.manual_seed(self.seed + self.epoch)
+        megabatch_indices = get_length_grouped_indices(self.lengths, self.batch_size, self.world_size, 
+                                             self.gradient_accumulation_size, self.initial_global_step, 
+                                             group_data=self.group_data, generator=g)
+   
+        # subsample
+        indices = [i for batch in megabatch_indices[self.rank::self.world_size] for i in batch]
+        assert len(indices) == self.num_samples
+        
+        return iter(indices)
+    
+    def set_epoch(self, epoch: int) -> None:
+        r"""
+        Set the epoch for this sampler.
+
+        When :attr:`shuffle=True`, this ensures all replicas
+        use a different random ordering for each epoch. Otherwise, the next iteration of this
+        sampler will yield the same ordering.
+
+        Args:
+            epoch (int): Epoch number.
+        """
+        self.epoch = epoch
+
+
+class VideoData(pl.LightningDataModule):
+    def __init__(self, args):
+        super().__init__()
+        self.args = args
+        
+    def _dataset(self, train):
+        datasets = []
+        for dataset_path, train_list, val_list in zip(self.args.data_path, self.args.train_datalist, self.args.val_datalist):
+           
+            dataset = MultiSizeVideoDataset(data_folder=dataset_path, data_list=train_list if train else val_list, sequence_length=self.args.sequence_length, 
+                            train=train, sample_rate=self.args.sample_rate, dynamic_sample=self.args.dynamic_sample)
+            datasets.append(dataset)  
+        return datasets
+
+    def _dataloader(self, train, steps = 0, batch_size = None):
+        dataset = self._dataset(train)
+        if isinstance(self.args.batch_size, int):
+            self.args.batch_size = [self.args.batch_size]
+        self.batch_size = self.args.batch_size if batch_size is None else batch_size
+        assert len(dataset) == len(self.args.batch_size)
+        dataloaders = []
+        for dset, d_batch_size in zip(dataset, self.batch_size):
+            if dist.is_initialized():
+                sampler = LengthGroupedSampler(
+                    batch_size=d_batch_size,
+                    world_size=dist.get_world_size(), 
+                    gradient_accumulation_size=1, 
+                    initial_global_step=steps if train else 0, 
+                    lengths=dset.lengths, 
+                    group_data=True, 
+                    rank = dist.get_rank()
+                )
+            else:
+                sampler = None
+            
+            dataloader = data.DataLoader(
+                dset,
+                batch_size=d_batch_size,
+                num_workers=self.args.num_workers if train else 0,
+                pin_memory=False,
+                sampler=sampler,
+            )
+
+            dataloaders.append(dataloader)
+        
+        return dataloaders
+
+    def train_dataloader(self):
+        return self._dataloader(True)
+
+    def val_dataloader(self):
+        return self._dataloader(False)[0]
+
+
+    @staticmethod
+    def add_data_specific_args(parent_parser):
+        parser = argparse.ArgumentParser(parents=[parent_parser], add_help=False)
+        parser.add_argument('--data_path', type=str, nargs="+", default=[''])
+        parser.add_argument('--train_datalist', type=str, nargs="+", default=['./video/kinetics-dataset/train/datapath'])
+        parser.add_argument('--val_datalist', type=str, nargs="+", default=['./video/kinetics-dataset/val/datapath'])
+
+        parser.add_argument('--sequence_length', type=int, default=17)
+        parser.add_argument('--sample_rate', type=int, default=1,
+                       help='Frame sampling rate')
+        parser.add_argument('--dynamic_sample', action='store_true',
+                       help='Enable dynamic sampling rate')
+
+        parser.add_argument('--batch_size', type=int, nargs="+", default=[5])
+        parser.add_argument('--num_workers', type=int, default=8)
+        return parser
+
+if __name__ == "__main__":
+    import os
+    def lines(file_path):
+        with open(file_path, 'r') as file:
+            return sum(1 for line in file)
+    train_folder ='./kinetics-dataset/datapath'
+    lengths_dict = {file_name: lines(os.path.join(train_folder, file_name)) for file_name in os.listdir(train_folder)}
+    lengths = []
+    for k, v in lengths_dict.items():
+        lengths += [k] * min(v, 50) #(v % 7)
+    world_size = 4
+    sampler = []
+    batch_size = 10
+    for rank in range(world_size):
+        sampler.append(LengthGroupedSampler(
+                    batch_size=batch_size,
+                    world_size=world_size, 
+                    gradient_accumulation_size=1, 
+                    initial_global_step=0, 
+                    lengths=lengths, 
+                    group_data=True, 
+                    rank = rank
+                ))
+    
+    
+    with open('./sampler.txt', 'w') as f:
+        for epoch in range(5):
+            rank_idx = {}
+            bk = []
+            print(f'epoch --------------------------------------  {epoch}  ----------------------------------------------------', file=f)
+            for rank in range(world_size):
+                sl = sampler[rank]
+                sl.set_epoch(epoch)
+                for i in iter(sl):
+                    bk.append(i)
+                    if len(bk) == batch_size:
+                        rank_idx.setdefault(f'rank_{rank}', [])
+                        rank_idx[f'rank_{rank}'].append(bk)
+                        bk = []
+            for num in range(5):
+                print('*'*5 + f'steps {num}' + '*'*5, file=f)
+                for rank, bk in rank_idx.items():
+                    print(f'rank {rank}: {bk[num]}', file=f)
+                    print([lengths[i] for i in bk[num]], file=f)
+
+
+    exit()

Abnormal-CT-Generation-MultiDisease/LeanVAE/models/autoencoder.py

@@ -0,0 +1,186 @@
+import argparse
+import torch
+import torch.distributed
+import torch.nn as nn
+import torch.nn.functional as F
+from einops import rearrange
+from ..modules import DiagonalGaussianDistribution, Encoder_Arch, Decoder_Arch, ISTA
+from ..utils.patcher_utils import Patcher, UnPatcher
+
+class LeanVAE(nn.Module):
+    def __init__(self, args):
+        super().__init__()
+        self.args = args
+        self.embedding_dim = args.embedding_dim
+        
+        self.latent_bottleneck = ISTA(points_num=args.embedding_dim, out_num=args.latent_dim, iter_num=args.ista_iter_num, layer_num=args.ista_layer_num)
+   
+        self.dwt = Patcher()
+        self.idwt = UnPatcher()
+    
+        self.encoder = Encoder_Arch(l_dim = args.l_dim, h_dim = args.h_dim, sep_num_layer = args.sep_num_layer, fusion_num_layer = args.fusion_num_layer)
+        self.decoder = Decoder_Arch(l_dim = args.l_dim, h_dim = args.h_dim, sep_num_layer = args.sep_num_layer, fusion_num_layer = args.fusion_num_layer) 
+        
+        self.std_layer = nn.Linear(args.embedding_dim, args.latent_dim)
+        
+        self.tile_inference = False
+        self.chunksize_enc = args.chunksize_enc if hasattr(args, 'chunksize_enc') and args.chunksize_enc else 5
+        self.chunksize_dec = args.chunksize_dec if hasattr(args, 'chunksize_dec') and args.chunksize_dec else 5
+        if args.use_tile_inference:
+            self.set_tile_inference(True)
+        else:
+            self.set_tile_inference(False)
+        
+    def _set_first_chunk(self, is_first_chunk=True):
+        for module in self.modules():
+            if hasattr(module, 'is_first_chunk'):
+                module.is_first_chunk = is_first_chunk
+    
+    def set_tile_inference(self, tile_inference=False):
+        for module in self.modules():
+            if hasattr(module, 'tile_inference'):
+                module.tile_inference = tile_inference
+    
+    def _build_chunk_index(self, T = 17, mtype = 'enc'):
+        start_end = []
+        if mtype == 'enc':
+            chunksize = self.chunksize_enc 
+        else:
+            chunksize = self.chunksize_dec
+        if T >= chunksize :
+            start_end.append((0, chunksize))
+            start_idx = chunksize 
+        else:
+            assert T < chunksize
+        
+        for i in range(start_idx, T, chunksize-1):
+            end_idx = min(i + chunksize -1, T)
+            start_end.append((i, end_idx))
+        return start_end
+    
+    def encode(self, x):
+        ndim = x.ndim
+        if ndim == 4:
+            x = x.unsqueeze(2)
+            self.set_tile_inference(False)
+    
+        if self.tile_inference:
+            z = []
+            chunk_indexs = self._build_chunk_index(T=x.shape[2], mtype='enc')
+            for idx, (start, end) in enumerate(chunk_indexs):
+                if idx == 0:
+                    self._set_first_chunk(True)
+                else:
+                    self._set_first_chunk(False)  
+        
+                x_dwt = self.dwt(x[:, :, start:end])
+                p = self.encoder.encode(x=x_dwt)
+                z.append(self.latent_bottleneck.sample(p))
+            z = torch.cat(z, dim = 1)
+        else:
+            x_dwt = self.dwt(x)
+            p = self.encoder.encode(x=x_dwt)
+            z = self.latent_bottleneck.sample(p)
+
+        z = rearrange(z, 'b t h w d -> b d t h w') 
+        return z
+    
+    def decode(self, z, is_image = False):
+        z = rearrange(z, 'b d t h w -> b t h w d')
+        if is_image:
+            self.set_tile_inference(False)
+        if self.tile_inference:
+            x_recon = []
+            chunk_indexs = self._build_chunk_index(T=z.shape[1], mtype='dec')
+            for idx, (start, end) in enumerate(chunk_indexs):
+                if idx == 0:
+                    self._set_first_chunk(True)
+                else:
+                    self._set_first_chunk(False)  
+                p_rec = self.latent_bottleneck.recon(z[:, start:end])
+                x_dwt_rec = self.decoder.decode(p_rec, is_image=is_image) 
+                
+                x_recon.append(self.idwt(x=x_dwt_rec))
+            x_recon = torch.cat(x_recon, dim = 2)
+        else:
+            p_rec = self.latent_bottleneck.recon(z)
+            x_dwt_rec = self.decoder.decode(p_rec, is_image=is_image) 
+            
+            x_recon = self.idwt(x=x_dwt_rec)
+        
+        return x_recon
+        
+
+
+    @torch.no_grad()
+    def inference(self, x):
+        if x.ndim == 4 : 
+            is_image = True
+        else:
+            is_image = False
+            assert x.shape[2] % 4 == 1, f"Expected frame_num % 4 == 1, but got {x.shape[2] % 4}"
+        
+        z = self.encode(x)
+        x_recon = self.decode(z, is_image=is_image)
+        
+        if is_image:
+            x = x.squeeze(2)
+        return x, x_recon
+        
+    def forward(self, x, log_image=False):
+        x_dwt = self.dwt(x) 
+        p = self.encoder(x=x_dwt)
+        z_mean = self.latent_bottleneck.sample(p)
+        z_std = self.std_layer(p)
+
+        posterior = DiagonalGaussianDistribution(parameters=(z_mean, z_std))
+        z = posterior.sample()
+        p_rec = self.latent_bottleneck.recon(z)
+        
+        x_dwt_rec = self.decoder(p_rec) #b c t h w
+        
+     
+        x_recon = self.idwt(x=x_dwt_rec)
+       
+        if log_image: 
+            return x, x_recon
+        
+        return x, x_recon, x_dwt, x_dwt_rec, posterior
+
+    
+    @classmethod
+    def load_from_checkpoint(cls, ckpt_path, device="cpu", strict=False):
+        """ Load model from checkpoint, initializing args and state_dict """
+        checkpoint = torch.load(ckpt_path, map_location=device)
+
+        if "args" not in checkpoint:
+            raise ValueError("Checkpoint does not contain 'args'. Ensure the checkpoint is saved correctly.")
+
+        args = argparse.Namespace(**checkpoint["args"]) 
+        
+        model = cls(args)
+        if "state_dict" in checkpoint:
+            msg = model.load_state_dict(checkpoint["state_dict"], strict=strict)
+            print(f"Successfully loaded weights from {ckpt_path}, {msg}")
+        return model
+    
+    @staticmethod
+    def add_model_specific_args(parent_parser):
+        parser = argparse.ArgumentParser(parents=[parent_parser], add_help=False)
+
+        # Model architecture parameters
+        parser.add_argument("--embedding_dim", type=int, default=512, help="Dimension of the embedding space.")
+        parser.add_argument("--latent_dim", type=int, default=4, help="Dimension of the latent channel.")
+        parser.add_argument("--ista_iter_num", type=int, default=2, help="Number of iterations in ISTA latent bottleneck.")
+        parser.add_argument("--ista_layer_num", type=int, default=2, help="Number of layers in ISTA latent bottleneck.")
+        
+        parser.add_argument("--l_dim", type=int, default=128)
+        parser.add_argument("--h_dim", type=int, default=384)
+        parser.add_argument("--sep_num_layer", type=int, default=2, help="Number of separate processing layers in encoder/decoder.")
+        parser.add_argument("--fusion_num_layer", type=int, default=4, help="Number of fusion layers in encoder/decoder.")
+
+        # Tiling inference (for memory-efficient processing)
+        parser.add_argument("--use_tile_inference", action="store_true", help="Enable tiling inference to process video in chunks.")
+        parser.add_argument("--chunksize_enc", type=int, default=9, help="Number of frames per chunk during tiled encoding.")
+        parser.add_argument("--chunksize_dec", type=int, default=5, help="Number of frames per chunk during tiled decoding.")
+        return parser

Abnormal-CT-Generation-MultiDisease/LeanVAE/models/autoencoder_pl.py

@@ -0,0 +1,232 @@
+import argparse
+import numpy as np
+from PIL import Image
+import pytorch_lightning as pl
+import torch
+import torch.distributed
+import torch.nn as nn
+import torch.nn.functional as F
+from timm.scheduler.cosine_lr import CosineLRScheduler
+from timm.models.layers import trunc_normal_
+from .autoencoder import LeanVAE
+from ..modules import LPIPS
+from ..utils.gan_loss import AdversarialLoss
+
+class AutoEncoderEngine(pl.LightningModule):
+    def __init__(self, args, data):
+        super().__init__()
+        self.args = args
+        self.video_data = data
+        
+        self.autoencoder = LeanVAE(args=args)
+
+        self.automatic_optimization = False
+        self.kl_weight = args.kl_weight
+        self.discriminator_iter_start = args.discriminator_iter_start
+ 
+        self.perceptual_weight = args.perceptual_weight
+        self.l1_weight = args.l1_weight
+        
+        self.automatic_optimization = False
+        self.grad_clip_val = args.grad_clip_val
+
+        if not hasattr(args, "grad_clip_val_disc"):
+            args.grad_clip_val_disc = 1.0
+        
+        self.grad_clip_val_disc = args.grad_clip_val_disc
+        
+        self.apply(self._init_weights)
+        self.perceptual_model = LPIPS().eval()
+        self.perceptual_model.requires_grad_(False)
+        self.gan_loss = AdversarialLoss(disc_weight=args.disc_weight)
+        self.save_hyperparameters()
+
+    def _init_weights(self, m):
+        if isinstance(m, nn.Linear):
+            trunc_normal_(m.weight, std=.02)
+            if isinstance(m, nn.Linear) and m.bias is not None:
+                nn.init.constant_(m.bias, 0)
+        elif isinstance(m, nn.LayerNorm):
+            if m.bias is not None:
+                nn.init.constant_(m.bias, 0)
+            if m.weight is not None:
+                nn.init.constant_(m.weight, 1.0)
+        
+        elif isinstance(m, nn.Conv3d) or isinstance(m, nn.Conv2d):
+            nn.init.xavier_uniform_(m.weight)
+            nn.init.zeros_(m.bias)
+
+
+    def forward(self, x, optimizer_idx=None, x_recon = None, log_image=False):
+        if log_image: 
+            return self.autoencoder(x, log_image)
+        
+        if optimizer_idx == 1:
+            discloss = self.gan_loss(inputs=x, reconstructions=x_recon, optimizer_idx=1)
+            self.log("train/discloss", discloss, prog_bar=True, logger=True, on_step=True, on_epoch=True)
+            return discloss
+        
+        elif optimizer_idx == 0:
+            assert x.ndim == 5
+            B, C, T, H, W = x.shape
+            x, x_recon, x_dwt, x_dwt_rec, posterior = self.autoencoder(x)
+            recon_loss = F.l1_loss(x_recon, x)* self.l1_weight
+            kl_loss = posterior.kl()
+            kl_loss = torch.sum(kl_loss) / kl_loss.shape[0] * self.kl_weight
+            
+            g_loss = 0.0
+            if self.global_step >= self.discriminator_iter_start:
+                g_loss = self.gan_loss(x, x_recon, optimizer_idx=0)
+                self.log("train/g_loss", g_loss, prog_bar=True, logger=True, on_step=True, on_epoch=True)
+            
+            recon_loss_low = (F.l1_loss(x_dwt_rec[0][:, :3], x_dwt[0][:, :3]) + F.l1_loss(x_dwt_rec[1][:, :3], x_dwt[1][:, :3])) * self.l1_weight * 0.05
+            recon_loss_high = (F.l1_loss(x_dwt_rec[0][:, 3:], x_dwt[0][:, 3:])+ F.l1_loss(x_dwt_rec[1][:, 3:], x_dwt[1][:, 3:])) * self.l1_weight * 0.1
+        
+            k = 4  
+            valid_start_indices = torch.tensor([x for x in range(T - k + 1) if x % 4 == 1])
+            start_idx = valid_start_indices[torch.randint(0, len(valid_start_indices), (B,))]
+            frame_idx = start_idx.unsqueeze(1) + torch.arange(k)
+            frame_idx = torch.cat((torch.zeros((B, 1), dtype=torch.int), frame_idx), dim=1).to(self.device) 
+            
+            frame_idx_selected = frame_idx.reshape(-1, 1, k+1, 1, 1).repeat(1, C, 1, H, W)
+            frames = torch.gather(x, 2, frame_idx_selected)
+            frames_recon = torch.gather(x_recon, 2, frame_idx_selected)
+            frames = frames.permute(0, 2, 1, 3, 4).contiguous().view(-1, 3, H, W)
+            frames_recon = frames_recon.permute(0, 2, 1, 3, 4).contiguous().view(-1, 3, H, W)
+            perceptual_loss = self.perceptual_model(frames, frames_recon).mean() * self.perceptual_weight 
+            
+            self.log("train/recon_loss", recon_loss, prog_bar=True, logger=True, on_step=True, on_epoch=True)
+            self.log("train/kl_loss", kl_loss, prog_bar=True, logger=True, on_step=True, on_epoch=True) 
+            self.log("train/recon_loss_low", recon_loss_low, prog_bar=True, logger=True, on_step=True, on_epoch=True)
+            self.log("train/recon_loss_high", recon_loss_high, prog_bar=True, logger=True, on_step=True, on_epoch=True)
+            self.log("train/perceptual_loss", perceptual_loss, prog_bar=True, logger=True, on_step=True, on_epoch=True)
+            return  perceptual_loss + recon_loss + recon_loss_low + recon_loss_high + kl_loss + g_loss, x_recon
+        
+        return perceptual_loss, recon_loss, kl_loss 
+
+  
+    def training_step(self, batch, batch_idx):
+  
+        x = batch[0]['video']
+        cur_global_step = self.global_step
+        
+        sch1, sch2 = self.lr_schedulers()
+        opt1, opt2 = self.optimizers()
+
+        cur_global_step = self.global_step
+
+        self.toggle_optimizer(opt1, optimizer_idx=0)
+        loss_generator, x_recon = self.forward(x, optimizer_idx=0)
+        opt1.zero_grad()
+        self.manual_backward(loss_generator)
+        if self.grad_clip_val is not None:
+            self.clip_gradients(opt1, gradient_clip_val=self.grad_clip_val) 
+        opt1.step()
+        sch1.step(cur_global_step)
+        self.untoggle_optimizer(optimizer_idx=0)
+        
+        if cur_global_step > self.discriminator_iter_start:
+            self.toggle_optimizer(opt2, optimizer_idx=1)
+            loss_discriminator = self.forward(x, optimizer_idx=1, x_recon=x_recon)
+            
+            opt2.zero_grad()
+            self.manual_backward(loss_discriminator)      
+          
+            if self.grad_clip_val_disc is not None:
+                self.clip_gradients(opt2, gradient_clip_val=self.grad_clip_val_disc)
+            opt2.step()
+            sch2.step(cur_global_step) 
+            self.untoggle_optimizer(optimizer_idx=1)   
+                
+      
+    def validation_step(self, batch, batch_idx):
+        x = batch['video'] 
+        perceptual_loss, recon_loss, kl_loss  = self.forward(x)
+        self.log('val/recon_loss', recon_loss, prog_bar=True)
+        self.log('val/perceptual_loss', perceptual_loss, prog_bar=True)
+        self.log("val/kl_loss", kl_loss, prog_bar=True)
+
+    def train_dataloader(self):
+        dataloaders = self.video_data._dataloader(train=True)
+        return dataloaders
+          
+    def val_dataloader(self):
+        return self.video_data._dataloader(False)[0]
+
+    def configure_optimizers(self):
+        opt_ae = torch.optim.Adam(self.autoencoder.parameters(),
+                            lr=self.args.lr, betas=(0.5, 0.9))
+        
+        opt_disc = torch.optim.Adam(
+                                    self.gan_loss.get_trainable_parameters(),
+                                    lr=self.args.lr_min, betas=(0.5, 0.9))
+        
+        lr_min = self.args.lr_min
+        train_iters = self.args.max_steps - self.discriminator_iter_start
+        warmup_steps = self.args.warmup_steps
+        warmup_lr_init = self.args.warmup_lr_init
+
+       
+        sch_ae = CosineLRScheduler(
+            opt_ae,
+            lr_min = lr_min,
+            t_initial = train_iters,
+            warmup_lr_init=warmup_lr_init,
+            warmup_t=warmup_steps,
+            cycle_mul = 1.,
+            cycle_limit=1,
+            t_in_epochs=True,
+        )
+
+        sch_disc = CosineLRScheduler(
+            opt_disc,
+            lr_min = lr_min ,
+            t_initial = train_iters,
+            warmup_lr_init=warmup_lr_init,
+            warmup_t= self.args.dis_warmup_steps,
+            cycle_mul = 1.,
+            cycle_limit=1,
+            t_in_epochs=True,
+        )
+        
+
+        return [opt_ae, opt_disc], [{"scheduler": sch_ae, "interval": "step"}, {"scheduler": sch_disc, "interval": "step"}]
+        
+  
+
+    def log_videos(self, batch, **kwargs):
+        log = dict()
+        if isinstance(batch, list):
+            batch = batch[0]
+        x = batch['video']
+        x, x_rec = self(x, log_image=True)
+        log["inputs"] = x
+        log["reconstructions"] = x_rec
+        return log
+
+    @staticmethod
+    def add_model_specific_args(parent_parser):
+        parser = argparse.ArgumentParser(parents=[parent_parser], add_help=False)
+        
+        # training configurations
+        parser.add_argument('--lr', type=float, default=5e-5)
+        parser.add_argument('--lr_min', type=float, default=1e-5)
+        parser.add_argument('--warmup_steps', type=int, default=5000)
+        parser.add_argument('--warmup_lr_init', type=float, default=0.)
+        parser.add_argument('--grad_clip_val', type=float, default=1.0)
+        parser.add_argument('--grad_clip_val_disc', type=float, default=1.0)
+
+
+        parser.add_argument('--kl_weight', type=float, default=1e-7)
+        parser.add_argument('--perceptual_weight', type=float, default=4.)
+        parser.add_argument('--l1_weight', type=float, default=4.)
+        parser.add_argument('--disc_weight', type=float, default=0.2)
+
+        # configuration for discriminator
+        parser.add_argument('--dis_warmup_steps', type=int, default=0)
+        parser.add_argument('--discriminator_iter_start', type=int, default=0)
+        parser.add_argument('--dis_lr_multiplier', type=float, default=1.)
+
+        return parser
+
+

Abnormal-CT-Generation-MultiDisease/LeanVAE/modules/__init__.py

@@ -0,0 +1,3 @@
+from .lpips import LPIPS
+from .backbones import *
+from .vae import DiagonalGaussianDistribution

Abnormal-CT-Generation-MultiDisease/LeanVAE/modules/backbones.py

@@ -0,0 +1,402 @@
+import torch
+import torch.nn as nn
+import torch.nn.functional as F
+from beartype import beartype
+from typing import Tuple
+from einops import rearrange, repeat
+from einops.layers.torch import Rearrange
+import numpy as np
+
+def exists(val):
+    return val is not None
+
+def default(val, d):
+    if exists(val):
+        return val
+    return d() if callable(d) else d
+
+class PEG3D(nn.Module):
+    def __init__(
+        self,
+        dim
+    ):
+        super().__init__()
+        self.ds_conv = nn.Conv3d(in_channels=dim, out_channels=dim, kernel_size=(3,3,3), groups = dim)
+        self.is_first_chunk = True
+        self.causal_cached = None
+        self.tile_inference = False
+    
+    def forward(self, x):
+        x = rearrange(x, 'b t h w d -> b d t h w')
+        if self.tile_inference:
+            if self.is_first_chunk:
+                x = F.pad(x, (1, 1, 1, 1, 2, 0), value=0.) 
+            else:
+                x = F.pad(x, (1, 1, 1, 1, 0, 0), value=0.) 
+                x = torch.concatenate((self.causal_cached, x), dim=2)
+            
+            self.causal_cached = x[:, :, -2:].clone()
+        else:
+            x = F.pad(x, (1, 1, 1, 1, 2, 0), value=0.)
+        x = self.ds_conv(x.contiguous())
+        x = rearrange(x, 'b d t h w -> b t h w d')
+        return x
+
+
+class GEGLU(nn.Module):
+    def forward(self, x):
+        x, gate = x.chunk(2, dim=-1)
+        return F.gelu(gate) * x
+
+
+def ffd(dim, mult=4, dropout=0.):
+    inner_dim = int(mult * (2 / 3) * dim)
+    return nn.Sequential(
+        nn.LayerNorm(dim),
+        nn.Linear(dim, inner_dim * 2, bias=False),
+        GEGLU(),
+        nn.Dropout(dropout),
+        nn.Linear(inner_dim, dim, bias=False)
+    )
+
+
+class NAF(nn.Module):
+    def __init__(self,
+                 num_layer, 
+                 dim,
+                 ): 
+        super(NAF, self).__init__()
+        self.num_layer = num_layer
+        self.dconv_layer = nn.Sequential()
+        self.ffd_layer = nn.Sequential()
+        for _ in range(num_layer):
+            self.ffd_layer.append(ffd(dim, 4)) 
+            self.dconv_layer.append(PEG3D(dim))
+   
+    def forward(self, x):
+        for i in range(self.num_layer):
+            x = self.dconv_layer[i](x)
+            x = self.ffd_layer[i](x) 
+        return x
+
+
+class ResNAF(nn.Module):
+    def __init__(self,
+                 num_layer, 
+                 dim,
+                 ): 
+        super(ResNAF, self).__init__()
+        self.num_layer = num_layer
+        self.dconv_layer = nn.Sequential()
+        self.ffd_layer = nn.Sequential()
+        for _ in range(num_layer):
+            self.ffd_layer.append(ffd(dim, 4)) 
+            self.dconv_layer.append(PEG3D(dim))
+   
+    def forward(self, x):
+        for i in range(self.num_layer):
+            x = x + self.dconv_layer[i](x)
+            x = x + self.ffd_layer[i](x) 
+        return x
+   
+
+class Encoder_Arch(nn.Module):
+    def __init__(self,
+                 l_dim = 128, 
+                 h_dim = 384, 
+                 sep_num_layer = 2,
+                 fusion_num_layer = 4,
+                 patch_size = (2,4,4),
+                 in_channel = 3
+                 ): 
+        super(Encoder_Arch, self).__init__()
+        
+        self.is_first_chunk = True
+        self.tile_inference = False
+
+        self.in_channel = in_channel
+        
+        self._build_linear_patch(in_channel=in_channel, out_channel_low=l_dim, out_channel_high=h_dim, pt=patch_size[0], ph=patch_size[1], pw=patch_size[2])
+
+        self.low_layer = ResNAF(num_layer=sep_num_layer, dim=l_dim) 
+        self.high_layer = ResNAF(num_layer=sep_num_layer, dim=h_dim)
+        self.fusion_layer = ResNAF(num_layer=fusion_num_layer, dim=l_dim + h_dim)
+       
+    def _build_linear_patch(self, in_channel = 3, out_channel_low = 128, out_channel_high = 384, pt = 2, ph = 4, pw = 4):
+        patch_config = {
+            'video_low': (pt, ph, pw),
+            'video_high': (pt, ph, pw),
+            'image_low': (1, ph, pw),
+            'image_high': (1, ph, pw)
+        }
+
+        for name, (t, h, w) in patch_config.items():
+            if 'low' in name:
+                in_dim = in_channel * t * h * w 
+                out_dim = out_channel_low 
+            else:
+                out_dim = out_channel_high
+                in_dim = in_channel * t * h * w * 7 if 'video' in name else in_channel * t * h * w * 3
+            proj = nn.Sequential(
+                Rearrange(f'b c (nt {t}) (nh {h}) (nw {w}) -> b nt nh nw (c {t} {h} {w})' if 'video' in name else f'b c (nh {h}) (nw {w}) -> b 1 nh nw (c {h} {w})'),
+                nn.Linear(in_dim, out_dim)
+            )
+            self.add_module(f"{name}_proj", proj)
+    
+    
+    def _linear_patch(self, x, proj_type):
+        low_comp, high_comp = x[:, :self.in_channel], x[:, self.in_channel:]
+        return getattr(self, f"{proj_type}_low_proj")(low_comp), getattr(self, f"{proj_type}_high_proj")(high_comp)
+    
+    def forward(self, x):
+        xi, xv = x
+        xi_low, xi_high = self._linear_patch(xi, 'image')
+        xv_low, xv_high = self._linear_patch(xv, 'video')
+       
+        low_x = torch.cat([xi_low, xv_low], dim=1)
+        high_x = torch.cat([xi_high, xv_high], dim=1)
+        
+        high_x = self.high_layer(high_x) 
+        low_x = self.low_layer(low_x)
+        x = torch.cat([low_x, high_x], dim=-1)
+        x = self.fusion_layer(x)
+        return x
+
+    
+    
+    def encode(self, x):
+        xi, xv = x
+        if xi is not None and xv is not None:
+            xi_low, xi_high = self._linear_patch(xi, 'image')
+            xv_low, xv_high = self._linear_patch(xv, 'video')
+        
+            low_x = torch.cat([xi_low, xv_low], dim=1)
+            high_x = torch.cat([xi_high, xv_high], dim=1)
+        elif xi is not None:
+            low_x, high_x = self._linear_patch(xi, 'image')
+        elif xv is not None:
+            low_x, high_x = self._linear_patch(xv, 'video')
+   
+        high_x = self.high_layer(high_x) 
+        low_x = self.low_layer(low_x)
+        x = torch.cat([low_x, high_x], dim=-1)
+        x = self.fusion_layer(x)
+        return x
+    
+    
+
+class Encoder_Arch(nn.Module):
+    def __init__(self,
+                 l_dim = 128, 
+                 h_dim = 384, 
+                 sep_num_layer = 2,
+                 fusion_num_layer = 4,
+                 patch_size = (2,4,4),
+                 in_channel = 3
+                 ): 
+        super(Encoder_Arch, self).__init__()
+        
+        self.is_first_chunk = True
+        self.tile_inference = False
+
+        self.in_channel = in_channel
+        
+        self._build_linear_patch(in_channel=in_channel, out_channel_low=l_dim, out_channel_high=h_dim, pt=patch_size[0], ph=patch_size[1], pw=patch_size[2])
+
+        self.low_layer = ResNAF(num_layer=sep_num_layer, dim=l_dim) 
+        self.high_layer = ResNAF(num_layer=sep_num_layer, dim=h_dim)
+        self.fusion_layer = ResNAF(num_layer=fusion_num_layer, dim=l_dim + h_dim)
+       
+    def _build_linear_patch(self, in_channel = 3, out_channel_low = 128, out_channel_high = 384, pt = 2, ph = 4, pw = 4):
+        patch_config = {
+            'video_low': (pt, ph, pw),
+            'video_high': (pt, ph, pw),
+            'image_low': (1, ph, pw),
+            'image_high': (1, ph, pw)
+        }
+
+        for name, (t, h, w) in patch_config.items():
+            if 'low' in name:
+                in_dim = in_channel * t * h * w 
+                out_dim = out_channel_low 
+            else:
+                out_dim = out_channel_high
+                in_dim = in_channel * t * h * w * 7 if 'video' in name else in_channel * t * h * w * 3
+            proj = nn.Sequential(
+                Rearrange('b c (nt pt) (nh ph) (nw pw) -> b nt nh nw (c pt ph pw)', pt=t, ph=h, pw=w),
+                nn.Linear(in_dim, out_dim)
+            )
+            self.add_module(f"{name}_proj", proj)
+    
+    
+    def _linear_patch(self, x, proj_type):
+        low_comp, high_comp = x[:, :self.in_channel], x[:, self.in_channel:]
+        return getattr(self, f"{proj_type}_low_proj")(low_comp), getattr(self, f"{proj_type}_high_proj")(high_comp)
+    
+    def _feature_transform(self, low_x, high_x):
+        low_x = self.low_layer(low_x)
+        high_x = self.high_layer(high_x) 
+        x = torch.cat([low_x, high_x], dim=-1)
+        x = self.fusion_layer(x)
+        return x
+    
+    def forward(self, x):
+        xi, xv = x
+        xi_low, xi_high = self._linear_patch(x=xi, proj_type='image')
+        xv_low, xv_high = self._linear_patch(x=xv, proj_type='video')
+       
+        low_x = torch.cat([xi_low, xv_low], dim=1)
+        high_x = torch.cat([xi_high, xv_high], dim=1)
+        
+        return self._feature_transform(low_x=low_x, high_x=high_x)
+
+    
+    
+    def encode(self, x):
+        xi, xv = x
+        if xi is not None and xv is not None:
+            xi_low, xi_high = self._linear_patch(x=xi, proj_type='image')
+            xv_low, xv_high = self._linear_patch(x=xv, proj_type='video')
+        
+            low_x = torch.cat([xi_low, xv_low], dim=1)
+            high_x = torch.cat([xi_high, xv_high], dim=1)
+        elif xi is not None:
+            low_x, high_x = self._linear_patch(x=xi, proj_type='image')
+        elif xv is not None:
+            low_x, high_x = self._linear_patch(x=xv, proj_type='video')
+   
+        return self._feature_transform(low_x=low_x, high_x=high_x)
+
+
+
+class Decoder_Arch(nn.Module):
+    def __init__(self,
+                 l_dim = 128, 
+                 h_dim = 384, 
+                 sep_num_layer = 2,
+                 fusion_num_layer = 4,
+                 patch_size = (2,4,4),
+                 in_channel = 3
+                 ): 
+        super(Decoder_Arch, self).__init__()
+        
+        self.l_dim = l_dim
+        self.is_first_chunk = True
+        self.tile_inference = False
+
+        self._build_linear_unpatch(in_channel=in_channel, out_channel_low=l_dim, out_channel_high=h_dim, pt=patch_size[0], ph=patch_size[1], pw=patch_size[2])
+
+        self.low_layer = ResNAF(num_layer=sep_num_layer, dim=l_dim) 
+        self.high_layer = ResNAF(num_layer=sep_num_layer, dim=h_dim)
+        self.fusion_layer = ResNAF(num_layer=fusion_num_layer, dim=l_dim + h_dim)
+       
+    
+    def _build_linear_unpatch(self, in_channel = 3, out_channel_low = 128, out_channel_high = 384, pt = 2, ph = 4, pw = 4):
+        patch_config = {
+            'video_low': (pt, ph, pw),
+            'video_high': (pt, ph, pw),
+            'image_low': (1, ph, pw),
+            'image_high': (1, ph, pw)
+        }
+
+        for name, (t, h, w) in patch_config.items():
+            if 'low' in name:
+                out_dim = in_channel * t * h * w 
+                in_dim = out_channel_low 
+            else:
+                in_dim = out_channel_high
+                out_dim = in_channel * t * h * w * 7 if 'video' in name else in_channel * t * h * w * 3
+            proj = nn.Sequential(
+                nn.Linear(in_dim, out_dim),
+                Rearrange('b nt nh nw (c pt ph pw) -> b c (nt pt) (nh ph) (nw pw)', pt=t, ph=h, pw=w),
+            )
+            self.add_module(f"{name}_proj", proj)
+    
+    def _linear_unpatch(self, x, proj_type):
+        low_comp, high_comp = getattr(self, f"{proj_type}_low_proj")(x[0]), getattr(self, f"{proj_type}_high_proj")(x[1])
+        return torch.cat([low_comp, high_comp], dim=1)
+    
+    def _feature_transform(self, x):
+        x = self.fusion_layer(x)
+        low_x = self.low_layer(x[:,:,:,:,:self.l_dim])
+        high_x = self.high_layer(x[:,:,:,:,self.l_dim:])
+
+        return low_x, high_x
+
+    
+    def decode(self, x, is_image = False):
+        low_x, high_x = self._feature_transform(x)
+        
+        if is_image:
+            xi = self._linear_unpatch(x=(low_x, high_x), proj_type='image')
+            return (xi, None)
+
+        else:
+            if self.tile_inference and not self.is_first_chunk:
+                xv = self._linear_unpatch(x=(low_x, high_x), proj_type='video')
+                return (None, xv)
+            else:
+                xi = self._linear_unpatch(x=(low_x[:, :1], high_x[:, :1]), proj_type='image')
+                xv = self._linear_unpatch(x=(low_x[:, 1:], high_x[:, 1:]), proj_type='video')
+                return (xi, xv)
+    
+    def forward(self, x):
+        low_x, high_x = self._feature_transform(x)
+        xi = self._linear_unpatch(x=(low_x[:, :1], high_x[:, :1]), proj_type='image')
+        xv = self._linear_unpatch(x=(low_x[:, 1:], high_x[:, 1:]), proj_type='video')
+        return (xi, xv) 
+
+class ISTA(nn.Module):
+    def __init__(self,
+                 points_num = 512,
+                 out_num = 4,
+                 iter_num = 2,
+                 layer_num = 2,
+                 ): 
+        super(ISTA, self).__init__()
+        phi_init = np.random.normal(0.0, (1 / points_num) ** 0.5, size=(out_num, points_num))
+        self.phi = nn.Parameter(torch.from_numpy(phi_init).float(), requires_grad=True)
+        self.Q = nn.Parameter(torch.from_numpy(np.transpose(phi_init)).float(), requires_grad=True)
+        self.iter_num = iter_num
+        self.forward_l = nn.ModuleList() 
+        self.backward_l = nn.ModuleList() 
+
+        for _ in range(self.iter_num):
+            self.forward_l.append(NAF(num_layer=layer_num, dim=points_num))
+            self.backward_l.append(NAF(num_layer=layer_num, dim=points_num))
+    
+        self.weights = nn.ParameterList()
+        self.etas = nn.ParameterList()
+        self.threshold = nn.ParameterList()
+        
+        for _ in range(self.iter_num):
+            self.threshold.append(nn.Parameter(torch.Tensor([0.01]), requires_grad=True))
+            self.weights.append(nn.Parameter(torch.tensor(1.), requires_grad=True))
+            
+    def sample(self, x):
+        b, t, h, w, d = x.shape
+        y = x.view(-1, d) @ self.phi.T    
+        return y.view(b, t, h, w, -1)
+    
+    def recon(self, y):
+        b, t, h, w, c = y.shape
+        y = y.reshape(-1, c)
+        recon = torch.mm(y, self.Q.t()) 
+        _, d = recon.shape
+        for i in range(self.iter_num):
+            recon_r = recon - self.weights[i] * torch.mm((torch.mm(recon, self.phi.t()) - y), self.phi)
+            recon = recon_r.reshape(b, t, h, w, -1)
+            recon = self.forward_l[i](recon)
+            recon = torch.mul(torch.sign(recon), F.relu(torch.abs(recon) - self.threshold[i]))
+            
+            recon = self.backward_l[i](recon).view(-1, d)
+            recon = recon_r + recon
+        return recon.view(b, t, h, w, -1)
+    
+    
+    def forward(self, x):
+        y = self.sample(x) 
+        recon = self.recon(y)
+        return recon
+    
+

Abnormal-CT-Generation-MultiDisease/LeanVAE/modules/cache/vgg.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a78928a0af1e5f0fcb1f3b9e8f8c3a2a5a3de244d830ad5c1feddc79b8432868
+size 7289

Abnormal-CT-Generation-MultiDisease/LeanVAE/modules/discriminator.py

@@ -0,0 +1,130 @@
+from typing import Any, Union
+import torch
+import torch.nn as nn
+import torch.nn.functional as F
+from einops import rearrange
+import functools
+
+class ActNorm(nn.Module):
+    def __init__(self, num_features, logdet=False, affine=True, allow_reverse_init=False):
+        assert affine
+        super().__init__()
+        self.logdet = logdet
+        self.loc = nn.Parameter(torch.zeros(1, num_features, 1, 1))
+        self.scale = nn.Parameter(torch.ones(1, num_features, 1, 1))
+        self.allow_reverse_init = allow_reverse_init
+
+        self.register_buffer("initialized", torch.tensor(0, dtype=torch.uint8))
+
+    def initialize(self, input):
+        with torch.no_grad():
+            flatten = input.permute(1, 0, 2, 3).contiguous().view(input.shape[1], -1)
+            mean = flatten.mean(1).unsqueeze(1).unsqueeze(2).unsqueeze(3).permute(1, 0, 2, 3)
+            std = flatten.std(1).unsqueeze(1).unsqueeze(2).unsqueeze(3).permute(1, 0, 2, 3)
+
+            self.loc.data.copy_(-mean)
+            self.scale.data.copy_(1 / (std + 1e-6))
+
+    def forward(self, input, reverse=False):
+        if reverse:
+            return self.reverse(input)
+        if len(input.shape) == 2:
+            input = input[:, :, None, None]
+            squeeze = True
+        else:
+            squeeze = False
+
+        _, _, height, width = input.shape
+
+        if self.training and self.initialized.item() == 0:
+            self.initialize(input)
+            self.initialized.fill_(1)
+
+        h = self.scale * (input + self.loc)
+
+        if squeeze:
+            h = h.squeeze(-1).squeeze(-1)
+
+        if self.logdet:
+            log_abs = torch.log(torch.abs(self.scale))
+            logdet = height * width * torch.sum(log_abs)
+            logdet = logdet * torch.ones(input.shape[0]).to(input)
+            return h, logdet
+
+        return h
+
+    def reverse(self, output):
+        if self.training and self.initialized.item() == 0:
+            if not self.allow_reverse_init:
+                raise RuntimeError(
+                    "Initializing ActNorm in reverse direction is "
+                    "disabled by default. Use allow_reverse_init=True to enable."
+                )
+            else:
+                self.initialize(output)
+                self.initialized.fill_(1)
+
+        if len(output.shape) == 2:
+            output = output[:, :, None, None]
+            squeeze = True
+        else:
+            squeeze = False
+
+        h = output / self.scale - self.loc
+
+        if squeeze:
+            h = h.squeeze(-1).squeeze(-1)
+        return h
+
+
+
+class NLayerDiscriminator(nn.Module):
+    """Defines a PatchGAN discriminator as in Pix2Pix."""
+    # https://github.com/junyanz/pytorch-CycleGAN-and-pix2pix/blob/master/models/networks.py
+    def __init__(self, input_nc=3, ndf=64, n_layers=3, use_actnorm=False):
+        """Construct a PatchGAN discriminator
+        Parameters:
+            input_nc (int)  -- the number of channels in input images
+            ndf (int)       -- the number of filters in the last conv layer
+            n_layers (int)  -- the number of conv layers in the discriminator
+        """
+        super(NLayerDiscriminator, self).__init__()
+        if not use_actnorm:
+            norm_layer = nn.BatchNorm2d
+        else:
+            norm_layer = ActNorm
+        if type(norm_layer) == functools.partial:  # no need to use bias as BatchNorm2d has affine parameters
+            use_bias = norm_layer.func != nn.BatchNorm2d
+        else:
+            use_bias = norm_layer != nn.BatchNorm2d
+
+        kw = 4
+        padw = 1
+        sequence = [nn.Conv2d(input_nc, ndf, kernel_size=kw, stride=2, padding=padw), nn.LeakyReLU(0.2, True)]
+        nf_mult = 1
+        nf_mult_prev = 1
+        for n in range(1, n_layers):  # gradually increase the number of filters
+            nf_mult_prev = nf_mult
+            nf_mult = min(2**n, 8)
+            sequence += [
+                nn.Conv2d(ndf * nf_mult_prev, ndf * nf_mult, kernel_size=kw, stride=2, padding=padw, bias=use_bias),
+                norm_layer(ndf * nf_mult),
+                nn.LeakyReLU(0.2, True),
+            ]
+
+        nf_mult_prev = nf_mult
+        nf_mult = min(2**n_layers, 8)
+        sequence += [
+            nn.Conv2d(ndf * nf_mult_prev, ndf * nf_mult, kernel_size=kw, stride=1, padding=padw, bias=use_bias),
+            norm_layer(ndf * nf_mult),
+            nn.LeakyReLU(0.2, True),
+        ]
+
+        sequence += [
+            nn.Conv2d(ndf * nf_mult, 1, kernel_size=kw, stride=1, padding=padw)
+        ]  # output 1 channel prediction map
+        self.main = nn.Sequential(*sequence)
+
+    def forward(self, input):
+        """Standard forward."""
+        return self.main(input)

Abnormal-CT-Generation-MultiDisease/LeanVAE/modules/lpips.py

@@ -0,0 +1,230 @@
+"""Stripped version of https://github.com/richzhang/PerceptualSimilarity/tree/master/models"""
+
+import os, hashlib
+import requests
+from tqdm import tqdm
+
+import torch
+import torch.nn as nn
+from torchvision import models
+from collections import namedtuple
+import torchvision
+URL_MAP = {
+    "vgg_lpips": "https://heibox.uni-heidelberg.de/f/607503859c864bc1b30b/?dl=1"
+}
+
+CKPT_MAP = {
+    "vgg_lpips": "vgg.pth"
+}
+
+MD5_MAP = {
+    "vgg_lpips": "d507d7349b931f0638a25a48a722f98a"
+}
+
+def download(url, local_path, chunk_size=1024):
+    os.makedirs(os.path.split(local_path)[0], exist_ok=True)
+    with requests.get(url, stream=True) as r:
+        total_size = int(r.headers.get("content-length", 0))
+        with tqdm(total=total_size, unit="B", unit_scale=True) as pbar:
+            with open(local_path, "wb") as f:
+                for data in r.iter_content(chunk_size=chunk_size):
+                    if data:
+                        f.write(data)
+                        pbar.update(chunk_size)
+
+
+def md5_hash(path):
+    with open(path, "rb") as f:
+        content = f.read()
+    return hashlib.md5(content).hexdigest()
+
+
+def get_ckpt_path(name, root, check=False):
+    assert name in URL_MAP
+    path = os.path.join(root, CKPT_MAP[name])
+    if not os.path.exists(path) or (check and not md5_hash(path) == MD5_MAP[name]):
+        print("Downloading {} model from {} to {}".format(name, URL_MAP[name], path))
+        download(URL_MAP[name], path)
+        md5 = md5_hash(path)
+        assert md5 == MD5_MAP[name], md5
+    return path
+
+
+class LPIPS(nn.Module):
+    # Learned perceptual metric
+    def __init__(self, use_dropout=True):
+        super().__init__()
+        self.scaling_layer = ScalingLayer()
+        self.chns = [64, 128, 256, 512, 512]  # vg16 features
+        self.net = vgg16(pretrained=True, requires_grad=False)
+        self.lin0 = NetLinLayer(self.chns[0], use_dropout=use_dropout)
+        self.lin1 = NetLinLayer(self.chns[1], use_dropout=use_dropout)
+        self.lin2 = NetLinLayer(self.chns[2], use_dropout=use_dropout)
+        self.lin3 = NetLinLayer(self.chns[3], use_dropout=use_dropout)
+        self.lin4 = NetLinLayer(self.chns[4], use_dropout=use_dropout)
+        self.load_from_pretrained()
+        for param in self.parameters():
+            param.requires_grad = False
+
+    def load_from_pretrained(self, name="vgg_lpips"):
+        ckpt = get_ckpt_path(name, os.path.join(os.path.dirname(os.path.abspath(__file__)), "cache"))
+        self.load_state_dict(torch.load(ckpt, map_location=torch.device("cpu")), strict=False)
+        print("loaded pretrained LPIPS loss from {}".format(ckpt))
+
+    @classmethod
+    def from_pretrained(cls, name="vgg_lpips"):
+        if name is not "vgg_lpips":
+            raise NotImplementedError
+        model = cls()
+        ckpt = get_ckpt_path(name, os.path.join(os.path.dirname(os.path.abspath(__file__)), "cache"))
+        r = model.load_state_dict(torch.load(ckpt, map_location=torch.device("cpu")), strict=False)
+        print(r)
+        return model
+
+    def forward(self, input, target):
+        in0_input, in1_input = (self.scaling_layer(input), self.scaling_layer(target))
+        outs0, outs1 = self.net(in0_input), self.net(in1_input)
+        feats0, feats1, diffs = {}, {}, {}
+        lins = [self.lin0, self.lin1, self.lin2, self.lin3, self.lin4]
+        for kk in range(len(self.chns)):
+            feats0[kk], feats1[kk] = normalize_tensor(outs0[kk]), normalize_tensor(outs1[kk])
+            diffs[kk] = (feats0[kk] - feats1[kk]) ** 2
+            
+        res = [spatial_average(lins[kk].model(diffs[kk]), keepdim=True) for kk in range(len(self.chns))]
+        val = res[0]
+        for l in range(1, len(self.chns)):
+            # print(res[l].shape)
+            val += res[l]
+        
+        return val
+
+
+
+
+
+class ScalingLayer(nn.Module):
+    def __init__(self):
+        super(ScalingLayer, self).__init__()
+        self.register_buffer('shift', torch.Tensor([-.030, -.088, -.188])[None, :, None, None])
+        self.register_buffer('scale', torch.Tensor([.458, .448, .450])[None, :, None, None])
+
+    def forward(self, inp):
+        return (inp - self.shift) / self.scale
+
+
+class NetLinLayer(nn.Module):
+    """ A single linear layer which does a 1x1 conv """
+    def __init__(self, chn_in, chn_out=1, use_dropout=False):
+        super(NetLinLayer, self).__init__()
+        layers = [nn.Dropout(), ] if (use_dropout) else []
+        layers += [nn.Conv2d(chn_in, chn_out, 1, stride=1, padding=0, bias=False), ]
+        self.model = nn.Sequential(*layers)
+
+
+class vgg16(torch.nn.Module):
+    def __init__(self, requires_grad=False, pretrained=True):
+        super(vgg16, self).__init__()
+        vgg_pretrained_features = models.vgg16(pretrained=pretrained).features
+        self.slice1 = torch.nn.Sequential()
+        self.slice2 = torch.nn.Sequential()
+        self.slice3 = torch.nn.Sequential()
+        self.slice4 = torch.nn.Sequential()
+        self.slice5 = torch.nn.Sequential()
+        self.N_slices = 5
+        for x in range(4):
+            self.slice1.add_module(str(x), vgg_pretrained_features[x])
+        for x in range(4, 9):
+            self.slice2.add_module(str(x), vgg_pretrained_features[x])
+        for x in range(9, 16):
+            self.slice3.add_module(str(x), vgg_pretrained_features[x])
+        for x in range(16, 23):
+            self.slice4.add_module(str(x), vgg_pretrained_features[x])
+        for x in range(23, 30):
+            self.slice5.add_module(str(x), vgg_pretrained_features[x])
+        if not requires_grad:
+            for param in self.parameters():
+                param.requires_grad = False
+
+    def forward(self, X):
+        h = self.slice1(X)
+        h_relu1_2 = h
+        h = self.slice2(h)
+        h_relu2_2 = h
+        h = self.slice3(h)
+        h_relu3_3 = h
+        h = self.slice4(h)
+        h_relu4_3 = h
+        h = self.slice5(h)
+        h_relu5_3 = h
+        vgg_outputs = namedtuple("VggOutputs", ['relu1_2', 'relu2_2', 'relu3_3', 'relu4_3', 'relu5_3'])
+        out = vgg_outputs(h_relu1_2, h_relu2_2, h_relu3_3, h_relu4_3, h_relu5_3)
+        return out
+
+
+def normalize_tensor(x,eps=1e-10):
+    norm_factor = torch.sqrt(torch.sum(x**2,dim=1,keepdim=True))
+    return x/(norm_factor+eps)
+
+
+def spatial_average(x, keepdim=True):
+    return x.mean([2,3],keepdim=keepdim)
+
+
+class ResNetLPIPS(nn.Module):
+    # Learned perceptual metric
+    def __init__(self, use_dropout=True):
+        super().__init__()
+        net, _ = clip.load(device='cpu', name='RN50')
+        self.net = net.visual
+        self.net.attnpool = nn.Identity()
+        for param in self.parameters():
+            param.requires_grad = False
+
+    def forward(self, input, target):
+        
+        outs0, outs1 = self.net(input), self.net(target)
+        #feats0, feats1= normalize_tensor(outs0), normalize_tensor(outs1)
+        diffs = ((outs0 - outs1) ** 2 )     #(feats0 - feats1) ** 2
+        
+        return diffs
+
+
+class MeanShift(nn.Conv2d):
+    def __init__(self, data_mean, data_std, data_range=1, norm=True):
+        c = len(data_mean)
+        super(MeanShift, self).__init__(c, c, kernel_size=1)
+        std = torch.Tensor(data_std)
+        self.weight.data = torch.eye(c).view(c, c, 1, 1)
+        if norm:
+            self.weight.data.div_(std.view(c, 1, 1, 1))
+            self.bias.data = -1 * data_range * torch.Tensor(data_mean)
+            self.bias.data.div_(std)
+        else:
+            self.weight.data.mul_(std.view(c, 1, 1, 1))
+            self.bias.data = data_range * torch.Tensor(data_mean)
+        self.requires_grad = False
+            
+class VGGPerceptualLoss(torch.nn.Module):
+    def __init__(self, rank):
+        super(VGGPerceptualLoss, self).__init__()
+        blocks = []
+        pretrained = True
+        self.vgg_pretrained_features = models.vgg19(pretrained=pretrained).features
+        self.normalize = MeanShift([0.485, 0.456, 0.406], [0.229, 0.224, 0.225], norm=True).to(rank)
+        for param in self.parameters():
+            param.requires_grad = False
+
+    def forward(self, Y, X, indices=None):
+        X = self.normalize(X)
+        Y = self.normalize(Y)
+        indices = [2, 7, 12, 21, 30]
+        weights = [1.0/2.6, 1.0/4.8, 1.0/3.7, 1.0/5.6, 10/1.5]
+        k = 0
+        loss = 0
+        for i in range(indices[-1]):
+            X = self.vgg_pretrained_features[i](X)
+            Y = self.vgg_pretrained_features[i](Y)
+            if (i+1) in indices:
+                loss += weights[k] * (X - Y.detach()).abs().mean() * 0.1
+                k += 1
+        return loss

Abnormal-CT-Generation-MultiDisease/LeanVAE/modules/vae.py

@@ -0,0 +1,73 @@
+import torch
+import numpy as np
+
+class DiagonalGaussianDistribution(object):
+    def __init__(self, parameters, deterministic=False):
+        self.parameters = parameters
+        self.mean, self.logvar = parameters #torch.chunk(parameters, 2, dim=1)
+        self.logvar = torch.clamp(self.logvar, -30.0, 20.0)
+        self.deterministic = deterministic
+        self.std = torch.exp(0.5 * self.logvar)
+        self.var = torch.exp(self.logvar)
+        if self.deterministic:
+            self.var = self.std = torch.zeros_like(self.mean).to(device=self.mean.device)
+
+    def sample(self):
+        x = self.mean + self.std * torch.randn(self.mean.shape).to(device=self.mean.device)
+        return x
+
+    def kl(self, other=None):
+        if self.deterministic:
+            return torch.Tensor([0.])
+        else:
+            if other is None:
+                return 0.5 * torch.sum(torch.pow(self.mean, 2)
+                                       + self.var - 1.0 - self.logvar,
+                                       dim=[1, 2, 3])
+            else:
+                return 0.5 * torch.sum(
+                    torch.pow(self.mean - other.mean, 2) / other.var
+                    + self.var / other.var - 1.0 - self.logvar + other.logvar,
+                    dim=[1, 2, 3])
+
+    def nll(self, sample, dims=[1,2,3]):
+        if self.deterministic:
+            return torch.Tensor([0.])
+        logtwopi = np.log(2.0 * np.pi)
+        return 0.5 * torch.sum(
+            logtwopi + self.logvar + torch.pow(sample - self.mean, 2) / self.var,
+            dim=dims)
+
+    def mode(self):
+        return self.mean
+
+
+
+def normal_kl(mean1, logvar1, mean2, logvar2):
+    """
+    source: https://github.com/openai/guided-diffusion/blob/27c20a8fab9cb472df5d6bdd6c8d11c8f430b924/guided_diffusion/losses.py#L12
+    Compute the KL divergence between two gaussians.
+    Shapes are automatically broadcasted, so batches can be compared to
+    scalars, among other use cases.
+    """
+    tensor = None
+    for obj in (mean1, logvar1, mean2, logvar2):
+        if isinstance(obj, torch.Tensor):
+            tensor = obj
+            break
+    assert tensor is not None, "at least one argument must be a Tensor"
+
+    # Force variances to be Tensors. Broadcasting helps convert scalars to
+    # Tensors, but it does not work for torch.exp().
+    logvar1, logvar2 = [
+        x if isinstance(x, torch.Tensor) else torch.tensor(x).to(tensor)
+        for x in (logvar1, logvar2)
+    ]
+
+    return 0.5 * (
+        -1.0
+        + logvar2
+        - logvar1
+        + torch.exp(logvar1 - logvar2)
+        + ((mean1 - mean2) ** 2) * torch.exp(-logvar2)
+    )

Abnormal-CT-Generation-MultiDisease/LeanVAE/utils/callbacks.py

@@ -0,0 +1,120 @@
+import os
+import numpy as np
+from PIL import Image
+
+import torch
+import torchvision
+from pytorch_lightning.callbacks import Callback
+from pytorch_lightning.utilities.distributed import rank_zero_only 
+
+import random
+from .utils import save_video_grid
+
+
+
+class VideoLogger(Callback):
+    def __init__(self, batch_frequency, max_videos, clamp=True, increase_log_steps=True):
+        super().__init__()
+        self.batch_freq = batch_frequency
+        self.max_videos = max_videos
+        self.log_steps = [2 ** n for n in range(int(np.log2(self.batch_freq)) + 1)]
+        if not increase_log_steps:
+            self.log_steps = [self.batch_freq]
+        self.clamp = clamp
+
+
+    @rank_zero_only
+    def log_local(self, save_dir, split, videos,
+                  global_step, current_epoch, batch_idx):
+        root = os.path.join(save_dir, "videos", split)
+        for k in videos:
+            grid = videos[k] + 0.5
+            filename = "gs-{:06}_e-{:06}_b-{:06}_{}.mp4".format(
+                global_step,
+                current_epoch,
+                batch_idx,
+                k)
+            path = os.path.join(root, filename)
+            os.makedirs(os.path.split(path)[0], exist_ok=True)
+            save_video_grid(grid, path)
+
+    def log_vid(self, pl_module, batch, batch_idx, split="train"):
+        # print(batch_idx, self.batch_freq, self.check_frequency(batch_idx) and hasattr(pl_module, "log_videos") and callable(pl_module.log_videos) and self.max_videos > 0)
+        if (self.check_frequency(batch_idx) and  # batch_idx % self.batch_freq == 0
+                hasattr(pl_module, "log_videos") and
+                callable(pl_module.log_videos) and
+                self.max_videos > 0):
+            # print(batch_idx, self.batch_freq,  self.check_frequency(batch_idx))
+            logger = type(pl_module.logger)
+
+            is_train = pl_module.training
+            if is_train:
+                pl_module.eval()
+
+            with torch.no_grad():
+                videos = pl_module.log_videos(batch, split=split, batch_idx=batch_idx)
+
+            for k in videos:
+                N = min(videos[k].shape[0], self.max_videos)
+                videos[k] = videos[k][:N]
+                if isinstance(videos[k], torch.Tensor):
+                    videos[k] = videos[k].detach().cpu()
+                    if self.clamp:
+                        videos[k] = torch.clamp(videos[k], -0.5, 0.5)
+
+            self.log_local(pl_module.logger.save_dir, split, videos,
+                           pl_module.global_step, pl_module.current_epoch, batch_idx)
+
+            if is_train:
+                pl_module.train()
+
+    def check_frequency(self, batch_idx):
+        if (batch_idx % self.batch_freq) == 0 or (batch_idx in self.log_steps):
+            try:
+                self.log_steps.pop(0)
+            except IndexError:
+                pass
+            return True
+        return False
+
+    def on_train_batch_end(self, trainer, pl_module, outputs, batch, batch_idx, dataloader_idx):
+        if batch[0]['video'].ndim == 4:
+            return
+        self.log_vid(pl_module, batch, batch_idx, split="train")
+
+    def on_validation_batch_end(self, trainer, pl_module, outputs, batch, batch_idx, dataloader_idx):
+        self.log_vid(pl_module, batch, batch_idx, split="val")
+
+
+class DatasetCallback(Callback):
+    def __init__(self, initial_batch_size, new_batch_size, step_threshold):
+        self.initial_batch_size = initial_batch_size
+        self.new_batch_size = new_batch_size
+        self.step_threshold = step_threshold
+
+    def on_train_batch_start(self, trainer, pl_module, batch, batch_idx, dataloader_idx):
+        if trainer.global_step == self.step_threshold:
+            # 更新 DataLoader 的 batch_size
+            trainer.train_dataloader = trainer.video_data._dataloader(train=True, batch_size=self.new_batch_size) #  self.new_batch_size
+            print(f'Batch size changed to {self.new_batch_size} at step {self.step_threshold}')
+    # def __init__(self):
+    #     self.seqlen_list = [17, 21, 17, 25, 21, 29, 33, 17, 21, 17]
+        
+#     #临时取消
+
+#     def on_batch_start(self, trainer, pl_module):
+#         seq_len = random.randint(0, 9)
+#         trainer.train_dataloader.dataset.datasets[0].sequence_length = self.seqlen_list[seq_len]
+       
+#     def on_train_batch_start(self, trainer, pl_module, outputs, batch, batch_idx, dataloader_idx):
+#         seq_len = batch_idx % 10
+#         trainer.train_dataloader.dataset.datasets[dataloader_idx].sequence_length = self.seqlen_list[seq_len]
+#         return
+#     #为啥上面那个没报错 下边那个报错说是list呢？？
+#     def on_epoch_end(self, trainer, pl_module):
+#         with open('tmp_shape2.txt', 'a') as f:
+#             print(trainer.current_epoch, file=f )
+#         #trainer.train_dataloader[0].sampler.set_epoch(trainer.current_epoch)
+#         return 
+        
+