Repo initialized

.gitattributes

@@ -33,3 +33,5 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+calibration_reliability_diagrams.png filter=lfs diff=lfs merge=lfs -text
+training_plots.png filter=lfs diff=lfs merge=lfs -text

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>.

README.md

@@ -1,5 +1,457 @@
----
-license: other
-license_name: gpl-3.0
-license_link: LICENSE
----
+---
+language:
+- pl
+license: gpl-3.0
+tags:
+- text-classification
+- emotion-classification
+- sentiment-analysis
+- polish
+- multi-label-classification
+- twitter
+datasets:
+- yazoniak/TwitterEmo-PL-Refined
+base_model: PKOBP/polish-roberta-8k
+metrics:
+- f1
+- accuracy
+pipeline_tag: text-classification
+model-index:
+  - name: twitter-emotion-pl-classifier
+    results:
+      - task:
+          type: text-classification
+          name: Multi-Label Emotion Classification
+        dataset:
+          type: yazoniak/TwitterEmo-PL-Refined
+          name: TwitterEmo-PL-Refined
+          split: validation
+        metrics:
+          - type: f1
+            value: 0.8500
+            name: F1 Macro
+            verified: true
+            args:
+              average: macro
+          - type: f1
+            value: 0.8900
+            name: F1 Micro
+            verified: true
+            args:
+              average: micro
+          - type: f1
+            value: 0.8895
+            name: F1 Weighted
+            verified: true
+            args:
+              average: weighted
+          - type: accuracy
+            value: 0.5125
+            name: Exact Match Accuracy
+            verified: true
+          - type: accuracy
+            value: 0.8900
+            name: Subset Accuracy
+            verified: true
+---
+
+# Polish Twitter Emotion Classifier (RoBERTa-8k)
+
+## Model Description
+
+This model is a fine-tuned version of [PKOBP/polish-roberta-8k](https://huggingface.co/PKOBP/polish-roberta-8k) for multi-label emotion and sentiment classification in Polish. It was trained on the [TwitterEmo-PL-Refined](https://huggingface.co/datasets/yazoniak/TwitterEmo-PL-Refined) dataset.
+
+The model predicts 8 emotion and sentiment labels simultaneously:
+
+- **Emotions**: `radość` (joy), `wstręt` (disgust), `gniew` (anger), `przeczuwanie` (anticipation)
+- **Sentiment**: `pozytywny` (positive), `negatywny` (negative), `neutralny` (neutral)
+- **Special**: `sarkazm` (sarcasm)
+
+### Model Details
+
+- **Model type**: RoBERTa (Polish)
+- **Language**: Polish
+- **Base model**: [PKOBP/polish-roberta-8k](https://huggingface.co/PKOBP/polish-roberta-8k)
+- **Task**: Multi-label text classification (emotion & sentiment)
+- **Training data**: 35,921 Polish tweets from TwitterEmo-PL-Refined
+- **License**: GPL-3.0
+- **Context window**: 8,192 tokens (max; for tweet-length texts you can use a smaller tokenizer `max_length`, e.g., 256-1024)
+
+## Intended Use
+
+### Primary Use Cases
+
+- **Social media monitoring**: Analyze emotions and sentiment in Polish tweets and social media posts
+- **Customer feedback analysis**: Understand emotional responses in Polish customer reviews
+- **Research**: Study emotion expression patterns in Polish language social media
+- **Multi-label sentiment analysis**: Capture nuanced emotional states beyond binary positive/negative
+
+### Out-of-Scope Use
+
+- This model is specifically trained on Polish Twitter data and may not generalize well to:
+  - Formal Polish text (news articles, academic writing)
+  - Other languages
+  - Very long documents (optimal for tweet-length texts)
+
+## Performance
+
+### Overall Metrics
+
+| Metric | Score |
+|--------|-------|
+| **F1 Macro** | **0.8500** |
+| **F1 Micro** | **0.8900** |
+| **F1 Weighted** | **0.8895** |
+| **Exact Match Accuracy** | **0.5125** |
+| **Subset Accuracy** | **0.8900** |
+| **Validation Loss** | **0.2761** |
+
+### Per-Label Performance
+
+| Label | F1 Score | Coverage |
+|-------|----------|----------|
+| **negatywny** (negative) | **0.8553** | 42.4% |
+| **neutralny** (neutral) | **0.8172** | 41.0% |
+| **pozytywny** (positive) | **0.7814** | 17.4% |
+| **gniew** (anger) | **0.7693** | 25.8% |
+| **radość** (joy) | **0.7476** | 11.9% |
+| **wstręt** (disgust) | **0.7337** | 20.4% |
+| **przeczuwanie** (anticipation) | **0.7220** | 21.6% |
+| **sarkazm** (sarcasm) | **0.5337** | 16.0% |
+
+## Training Details
+
+### Training Data
+
+The model was trained on [TwitterEmo-PL-Refined](https://huggingface.co/datasets/yazoniak/TwitterEmo-PL-Refined), which contains:
+
+- **Total samples**: 35,921 Polish tweets
+- **Label distribution**:
+  - `negatywny`: 15,231 samples (42.4%)
+  - `neutralny`: 14,720 samples (41.0%)
+  - `gniew`: 9,252 samples (25.8%)
+  - `przeczuwanie`: 7,776 samples (21.6%)
+  - `wstręt`: 7,337 samples (20.4%)
+  - `pozytywny`: 6,248 samples (17.4%)
+  - `sarkazm`: 5,756 samples (16.0%)
+  - `radość`: 4,283 samples (11.9%)
+
+### Training Configuration
+
+```python
+Model: PKOBP/polish-roberta-8k
+Training samples: 28,737 (80%)
+Validation samples: 7,184 (20%)
+
+Hyperparameters:
+- Learning rate: 1e-5
+- Batch size: 32 (train), 32 (eval)
+- Epochs: 4
+- Weight decay: 0.03
+- Warmup ratio: 0.1
+- Dropout rate: 0.2
+- Max gradient norm: 1.0
+- Optimizer: AdamW
+- LR scheduler: Cosine with warmup
+- Early stopping patience: 3
+- Mixed precision: BF16
+
+Training strategy:
+- Save strategy: Every 200 steps
+- Evaluation strategy: Every 200 steps
+- Best model selection: F1 Macro
+- Total training steps: 3,600
+- Best checkpoint: 3,400
+```
+
+### Training Process
+
+Training was conducted on single NVIDIA RTX 3090 GPU using a stratified 80/20 train-validation split with the following progression:
+
+![Training Progress](training_plots.png)
+
+## Calibration
+
+The model's predictions can be improved using **temperature scaling** and **optimized thresholds**. Calibration analysis shows:
+
+### Temperature Scaling Results
+
+Per-label temperature scaling reduces calibration error (Expected Calibration Error - ECE):
+
+| Label | Temperature | ECE Before | ECE After | Improvement |
+|-------|------------|------------|-----------|-------------|
+| `radość` | 1.066 | 0.0163 | 0.0166 | -1.8% |
+| `wstręt` | 1.117 | 0.0211 | 0.0152 | **+27.9%** |
+| `gniew` | 1.186 | 0.0308 | 0.0194 | **+37.0%** |
+| `przeczuwanie` | 1.102 | 0.0228 | 0.0237 | -3.9% |
+| `pozytywny` | 1.181 | 0.0280 | 0.0293 | -4.6% |
+| `negatywny` | 1.437 | 0.0594 | 0.0345 | **+41.9%** |
+| `neutralny` | 1.472 | 0.0696 | 0.0390 | **+44.0%** |
+| `sarkazm` | 1.078 | 0.0202 | 0.0202 | 0.0% |
+
+**Key findings:**
+
+- `neutralny`, `negatywny`, and `gniew` benefit most from temperature scaling
+- Some labels (`radość`, `przeczuwanie`, `pozytywny`) show minor degradation
+- Overall, calibration significantly improves probability reliability
+
+### Optimized Decision Thresholds
+
+Per-label F1-optimized thresholds (vs. default 0.5):
+
+| Label | Optimal Threshold | F1 @ Optimal | F1 @ 0.5 | Improvement |
+|-------|------------------|--------------|----------|-------------|
+| `neutralny` | **0.330** | **0.8211** | 0.8110 | **+1.00%** |
+| `sarkazm` | **0.330** | **0.5766** | 0.5256 | **+5.10%** |
+| `przeczuwanie` | 0.410 | 0.7276 | 0.7187 | +0.89% |
+| `gniew` | 0.440 | 0.7692 | 0.7676 | +0.16% |
+| `negatywny` | 0.450 | 0.8516 | 0.8511 | +0.05% |
+| `wstręt` | 0.460 | 0.7477 | 0.7464 | +0.13% |
+| `pozytywny` | 0.510 | 0.7864 | 0.7859 | +0.04% |
+| `radość` | 0.560 | 0.7572 | 0.7558 | +0.14% |
+
+**Key findings:**
+
+- `sarkazm` shows the largest improvement (+5.10%) with a lower threshold (0.33)
+- `neutralny` also benefits significantly (+1.00%) from a lower threshold (0.33)
+- Most labels perform optimally near the default 0.5 threshold
+- Total improvement with optimized thresholds: **~0.5-1.0% F1 Macro**
+
+### Calibration Files
+
+The model repository includes:
+
+- **Base model**: `model.safetensors` - Use with default threshold (0.5)
+- **Calibration artifacts**: `calibration_artifacts.json` - Contains temperature parameters and optimal thresholds
+
+![Reliability diagrams*](calibration_reliability_diagrams.png)
+
+**Recommendation**: For production use, apply both temperature scaling and optimized thresholds for best performance.
+
+## Model Files
+
+This repository contains:
+
+- **Model weights**: `model.safetensors` - Fine-tuned RoBERTa model
+- **Tokenizer**: `tokenizer.json`, `tokenizer_config.json` - Polish RoBERTa tokenizer
+- **Configuration**: `config.json` - Model configuration
+- **Calibration**: `calibration_artifacts.json` - Temperature scaling parameters and optimal thresholds
+- **Inference scripts**:
+  - `predict.py` - Basic inference (threshold: 0.5)
+  - `predict_calibrated.py` - Calibrated inference (recommended)
+- **Training artifacts**: `training_plots`, `calibration_reliability_diagrams`
+- **Requirements**: `requirements.txt` - Python dependencies
+- **License**: `LICENSE` - Full GPL-3.0 license text
+
+### Installation
+
+```bash
+pip install -r requirements.txt
+```
+
+Or install dependencies manually:
+
+```bash
+pip install transformers torch numpy
+```
+
+## Usage
+
+### Important: Text Preprocessing
+
+**The model expects @mentions to be anonymized**, as they were during training. Both inference scripts automatically replace all `@username` mentions with `@anonymized_account` to match the training data distribution.
+
+### Quick Start (Basic Inference)
+
+Use the `predict.py` script for basic inference with default threshold (0.5):
+
+```bash
+# From Hugging Face (default) - mentions are automatically anonymized
+python predict.py "Uwielbiam czekać na peronie 3 godziny! Gratulacje dla #zgp"
+
+# Example with mentions
+python predict.py "@zgp_intervillage Uwielbiam czekać na peronie 3 godziny! Gratulacje dla #zgp"
+# Preprocessed internally: "@anonymized_account Uwielbiam czekać..."
+
+# From local model
+python predict.py "Uwielbiam czekać na peronie 3 godziny! Gratulacje dla #zgp" --model-path ./
+
+# With custom threshold
+python predict.py "Uwielbiam czekać na peronie 3 godziny! Gratulacje dla #zgp" --model-path ./ --threshold 0.3
+```
+
+**Example Output:**
+
+```
+Loading model from: yazoniak/twitter-emotion-pl-classifier
+
+Input text: Uwielbiam czekać na peronie 3 godziny! Gratulacje dla #zgp
+
+Assigned Labels:
+----------------------------------------
+  radość
+  pozytywny
+  sarkazm
+
+All Labels (with probabilities):
+----------------------------------------
+✓ radość         : 0.9574
+  wstręt         : 0.0566
+  gniew          : 0.0516
+  przeczuwanie   : 0.0347
+✓ pozytywny      : 0.9782
+  negatywny      : 0.0602
+  neutralny      : 0.0336
+✓ sarkazm        : 0.5404
+```
+
+### With Calibration
+
+Use the `predict_calibrated.py` script for calibrated inference with temperature scaling and optimized thresholds:
+
+```bash
+# From Hugging Face with calibration (requires calibration_artifacts.json)
+python predict_calibrated.py "Uwielbiam czekać na peronie 3 godziny! Gratulacje dla #zgp"
+```
+
+### Python API Usage
+
+```python
+from transformers import AutoTokenizer, AutoModelForSequenceClassification
+import torch
+import numpy as np
+import re
+
+def preprocess_text(text):
+    """Preprocess text to match training data format."""
+    # Anonymize @mentions (IMPORTANT for best performance)
+    text = re.sub(r'@\w+', '@anonymized_account', text)
+    return text
+
+# Load model
+model_name = "yazoniak/twitter-emotion-pl-classifier"
+tokenizer = AutoTokenizer.from_pretrained(model_name)
+model = AutoModelForSequenceClassification.from_pretrained(model_name)
+model.eval()
+
+# Get labels from model config
+labels = [model.config.id2label[i] for i in range(model.config.num_labels)]
+
+# Prepare input with preprocessing
+text = "@jan_kowalski To jest wspaniały dzień!"
+preprocessed_text = preprocess_text(text)  # "@anonymized_account To jest wspaniały dzień!"
+inputs = tokenizer(preprocessed_text, return_tensors="pt", truncation=True, max_length=8192)
+
+# Inference
+with torch.no_grad():
+    outputs = model(**inputs)
+    logits = outputs.logits
+
+# Get probabilities
+probabilities = torch.sigmoid(logits).squeeze().numpy()
+
+# Apply threshold
+threshold = 0.5
+predictions = {
+    label: float(prob) 
+    for label, prob in zip(labels, probabilities) 
+    if prob > threshold
+}
+
+print(predictions)
+# Output: {'radość': 0.8734, 'pozytywny': 0.9156}
+```
+
+### Interpretation
+
+The model outputs logits for each of the 8 labels. To get predictions:
+
+1. **Without calibration**: Apply sigmoid, threshold at 0.5
+1. **With calibration**:
+   - Apply sigmoid
+   - Apply temperature scaling (divide logits by temperature before sigmoid)
+   - Apply per-label optimized thresholds
+
+## Limitations and Biases
+
+### Known Limitations
+
+1. **Preprocessing required**: The model expects `@mentions` to be anonymized as `@anonymized_account` (matching training data). The provided inference scripts handle this automatically, but custom implementations must include this preprocessing step for optimal performance.
+
+1. **Sarcasm detection**: The model struggles with Polish sarcasm (F1: 0.53), which is inherently difficult to detect in text for BERT models without additional context.
+
+1. **Class imbalance**: Performance varies with label frequency:
+
+   - High-frequency labels (`negatywny`, `neutralny`) perform best
+   - Low-frequency labels (`radość`, `sarkazm`) show lower F1 scores
+
+1. **Twitter-specific**: The model is optimized for tweet-length texts (up to 8,192 tokens) with informal language, hashtags, and mentions.
+
+## Citation
+
+If you use this model in your research or applications, please cite:
+
+```bibtex
+@model{yazoniak2025twitteremotionpl,
+  title={Polish Twitter Emotion Classifier (RoBERTa-8k)},
+  author={yazoniak},
+  year={2025},
+  publisher={Hugging Face},
+  url={https://huggingface.co/yazoniak/twitter-emotion-pl-classifier}
+}
+```
+
+Also cite the base model and dataset:
+
+```bibtex
+@dataset{yazoniak_twitteremo_pl_refined_2025,
+  title   = {TwitterEmo-PL-Refined: Polish Twitter Emotions (8 labels, refined)},
+  author  = {yazoniak},
+  year    = {2025},
+  url     = {https://huggingface.co/datasets/yazoniak/TwitterEmo-PL-Refined}
+}
+
+@inproceedings{bogdanowicz2023twitteremo,
+  title     = {TwitterEmo: Annotating Emotions and Sentiment in Polish Twitter},
+  author    = {Bogdanowicz, S. and Cwynar, H. and Zwierzchowska, A. and Klamra, C. and Kiera{\'s}, W. and Kobyli{\'n}ski, {\L}.},
+  booktitle = {Computational Science -- ICCS 2023},
+  series    = {Lecture Notes in Computer Science},
+  volume    = {14074},
+  publisher = {Springer, Cham},
+  year      = {2023},
+  doi       = {10.1007/978-3-031-36021-3_20}
+}
+```
+
+## Acknowledgments
+
+- **Base model**: [PKOBP/polish-roberta-8k](https://huggingface.co/PKOBP/polish-roberta-8k)
+- **Original dataset**: [CLARIN-PL TwitterEmo](https://huggingface.co/datasets/clarin-pl/twitteremo)
+- **Label cleaning**: Cleanlab library for noise detection
+- **LLM assistance**: Gemini-2.5-Flash and GPT-4.1 for label review
+
+## License
+
+### License Terms
+
+This model is released under the **GNU General Public License v3.0 (GPL-3.0)**, inherited from the training dataset.
+
+**License Chain:**
+
+- **Base Model** ([PKOBP/polish-roberta-8k](https://huggingface.co/PKOBP/polish-roberta-8k)): Apache-2.0
+- **Training Dataset** ([TwitterEmo-PL-Refined](https://huggingface.co/datasets/yazoniak/TwitterEmo-PL-Refined)): GPL-3.0
+- **Original Dataset** ([clarin-pl/twitteremo](https://huggingface.co/datasets/clarin-pl/twitteremo)): GPL-3.0
+- **This Fine-tuned Model**: **GPL-3.0** (inherited from training data)
+
+### Full License Text
+
+The complete GPL-3.0 license text is available in the [LICENSE](LICENSE) file in this repository, or at: https://www.gnu.org/licenses/gpl-3.0.html
+
+## Model Card Contact
+
+For questions, issues, or feedback about this model, please open an issue in the model repository or contact the author through Hugging Face.
+
+______________________________________________________________________
+
+**Model Version**: v1.0
+**Last Updated**: 2025-10-10

added_tokens.json

@@ -0,0 +1,66 @@
+{
+  "<mask>": 128000,
+  "<user_token_10>": 128010,
+  "<user_token_11>": 128011,
+  "<user_token_12>": 128012,
+  "<user_token_13>": 128013,
+  "<user_token_14>": 128014,
+  "<user_token_15>": 128015,
+  "<user_token_16>": 128016,
+  "<user_token_17>": 128017,
+  "<user_token_18>": 128018,
+  "<user_token_19>": 128019,
+  "<user_token_1>": 128001,
+  "<user_token_20>": 128020,
+  "<user_token_21>": 128021,
+  "<user_token_22>": 128022,
+  "<user_token_23>": 128023,
+  "<user_token_24>": 128024,
+  "<user_token_25>": 128025,
+  "<user_token_26>": 128026,
+  "<user_token_27>": 128027,
+  "<user_token_28>": 128028,
+  "<user_token_29>": 128029,
+  "<user_token_2>": 128002,
+  "<user_token_30>": 128030,
+  "<user_token_31>": 128031,
+  "<user_token_32>": 128032,
+  "<user_token_33>": 128033,
+  "<user_token_34>": 128034,
+  "<user_token_35>": 128035,
+  "<user_token_36>": 128036,
+  "<user_token_37>": 128037,
+  "<user_token_38>": 128038,
+  "<user_token_39>": 128039,
+  "<user_token_3>": 128003,
+  "<user_token_40>": 128040,
+  "<user_token_41>": 128041,
+  "<user_token_42>": 128042,
+  "<user_token_43>": 128043,
+  "<user_token_44>": 128044,
+  "<user_token_45>": 128045,
+  "<user_token_46>": 128046,
+  "<user_token_47>": 128047,
+  "<user_token_48>": 128048,
+  "<user_token_49>": 128049,
+  "<user_token_4>": 128004,
+  "<user_token_50>": 128050,
+  "<user_token_51>": 128051,
+  "<user_token_52>": 128052,
+  "<user_token_53>": 128053,
+  "<user_token_54>": 128054,
+  "<user_token_55>": 128055,
+  "<user_token_56>": 128056,
+  "<user_token_57>": 128057,
+  "<user_token_58>": 128058,
+  "<user_token_59>": 128059,
+  "<user_token_5>": 128005,
+  "<user_token_60>": 128060,
+  "<user_token_61>": 128061,
+  "<user_token_62>": 128062,
+  "<user_token_63>": 128063,
+  "<user_token_6>": 128006,
+  "<user_token_7>": 128007,
+  "<user_token_8>": 128008,
+  "<user_token_9>": 128009
+}

calibration_artifacts.json

@@ -0,0 +1,23 @@
+{
+  "method": "temperature",
+  "optimal_thresholds": {
+    "rado\u015b\u0107": 0.5599999999999999,
+    "wstr\u0119t": 0.4599999999999999,
+    "gniew": 0.43999999999999995,
+    "przeczuwanie": 0.4099999999999999,
+    "pozytywny": 0.5099999999999999,
+    "negatywny": 0.4499999999999999,
+    "neutralny": 0.32999999999999996,
+    "sarkazm": 0.32999999999999996
+  },
+  "temperatures": {
+    "rado\u015b\u0107": 1.0663778214928685,
+    "wstr\u0119t": 1.1170095549267465,
+    "gniew": 1.1862162654090356,
+    "przeczuwanie": 1.1020299680100392,
+    "pozytywny": 1.1814368991677284,
+    "negatywny": 1.437031537754527,
+    "neutralny": 1.4719093876932696,
+    "sarkazm": 1.0780899936871722
+  }
+}

config.json

@@ -0,0 +1,57 @@
+{
+  "architectures": [
+    "RobertaForSequenceClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "auto_map": {
+    "AutoConfig": "configuration_roberta.RobertaConfig",
+    "AutoModel": "modeling_roberta.RobertaModel",
+    "AutoModelForCausalLM": "modeling_roberta.RobertaForCausalLM",
+    "AutoModelForMaskedLM": "modeling_roberta.RobertaForMaskedLM",
+    "AutoModelForMultipleChoice": "modeling_roberta.RobertaForMultipleChoice",
+    "AutoModelForQuestionAnswering": "modeling_roberta.RobertaForQuestionAnswering",
+    "AutoModelForSequenceClassification": "modeling_roberta.RobertaForSequenceClassification",
+    "AutoModelForTokenClassification": "modeling_roberta.RobertaForTokenClassification"
+  },
+  "bos_token_id": 0,
+  "classifier_dropout": null,
+  "dtype": "float32",
+  "eos_token_id": 2,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 1024,
+  "id2label": {
+    "0": "rado\u015b\u0107",
+    "1": "wstr\u0119t",
+    "2": "gniew",
+    "3": "przeczuwanie",
+    "4": "pozytywny",
+    "5": "negatywny",
+    "6": "neutralny",
+    "7": "sarkazm"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 4096,
+  "label2id": {
+    "gniew": 2,
+    "negatywny": 5,
+    "neutralny": 6,
+    "pozytywny": 4,
+    "przeczuwanie": 3,
+    "rado\u015b\u0107": 0,
+    "sarkazm": 7,
+    "wstr\u0119t": 1
+  },
+  "layer_norm_eps": 1e-05,
+  "max_position_embeddings": 8194,
+  "model_type": "roberta",
+  "num_attention_heads": 16,
+  "num_hidden_layers": 24,
+  "pad_token_id": 1,
+  "position_embedding_type": "absolute",
+  "problem_type": "multi_label_classification",
+  "transformers_version": "4.56.2",
+  "type_vocab_size": 1,
+  "use_cache": true,
+  "vocab_size": 128064
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:72f8732f18c3f668da8b1cee2d079c447b888b34852e71bb61e5e98deb996d54
+size 1771642216

predict.py

@@ -0,0 +1,68 @@
+import argparse
+import torch
+import numpy as np
+import re
+from transformers import AutoTokenizer, AutoModelForSequenceClassification
+
+DEFAULT_THRESHOLD = 0.5
+
+def preprocess_text(text, anonymize_mentions=True):
+    if anonymize_mentions:
+        text = re.sub(r'@\w+', '@anonymized_account', text)
+    return text
+
+def main():
+    parser = argparse.ArgumentParser()
+    parser.add_argument("text", type=str, help="Text to classify")
+    parser.add_argument("--model-path", type=str, default="yazoniak/twitter-emotion-pl-classifier", 
+                        help="Path to model or HF model ID")
+    parser.add_argument("--threshold", type=float, default=DEFAULT_THRESHOLD, 
+                        help="Classification threshold (default: 0.5)")
+    parser.add_argument("--no-anonymize", action="store_true",
+                        help="Disable mention anonymization (not recommended)")
+    args = parser.parse_args()
+
+    print(f"Loading model from: {args.model_path}")
+    tokenizer = AutoTokenizer.from_pretrained(args.model_path)
+    model = AutoModelForSequenceClassification.from_pretrained(args.model_path)
+    model.eval()
+    
+    labels = [model.config.id2label[i] for i in range(model.config.num_labels)]
+
+    anonymize = not args.no_anonymize
+    processed_text = preprocess_text(args.text, anonymize_mentions=anonymize)
+    
+    if anonymize and processed_text != args.text:
+        print(f"Preprocessed text: {processed_text}")
+    
+    print(f"\nInput text: {args.text}\n")
+
+    inputs = tokenizer(processed_text, return_tensors="pt", truncation=True, max_length=8192)
+    
+    with torch.no_grad():
+        outputs = model(**inputs)
+        logits = outputs.logits.squeeze().numpy()
+    
+    probabilities = 1 / (1 + np.exp(-logits))
+    predictions = probabilities > args.threshold
+
+    assigned_labels = [labels[i] for i in range(len(labels)) if predictions[i]]
+    
+    if assigned_labels:
+        print("Assigned Labels:")
+        print("-" * 40)
+        for label in assigned_labels:
+            print(f"  {label}")
+        print()
+    else:
+        print("No labels assigned (all below threshold)\n")
+    
+    print("All Labels (with probabilities):")
+    print("-" * 40)
+    for i, label in enumerate(labels):
+        status = "✓" if predictions[i] else " "
+        print(f"{status} {label:15s}: {probabilities[i]:.4f}")
+
+if __name__ == "__main__":
+    main()
+

predict_calibrated.py

@@ -0,0 +1,99 @@
+import argparse
+import json
+import os
+import re
+import torch
+import numpy as np
+from transformers import AutoTokenizer, AutoModelForSequenceClassification
+
+def preprocess_text(text, anonymize_mentions=True):
+    if anonymize_mentions:
+        text = re.sub(r'@\w+', '@anonymized_account', text)
+    return text
+
+def load_calibration_artifacts(calib_path):
+    if not os.path.exists(calib_path):
+        raise FileNotFoundError(f"Calibration artifacts not found at: {calib_path}")
+    
+    with open(calib_path, 'r') as f:
+        return json.load(f)
+
+def main():
+    parser = argparse.ArgumentParser()
+    parser.add_argument("text", type=str, help="Text to classify")
+    parser.add_argument("--model-path", type=str, default="yazoniak/twitter-emotion-pl-classifier", 
+                        help="Path to model or HF model ID")
+    parser.add_argument("--calibration-path", type=str, default=None,
+                        help="Path to calibration_artifacts.json (default: auto-detect)")
+    parser.add_argument("--no-anonymize", action="store_true",
+                        help="Disable mention anonymization (not recommended)")
+    args = parser.parse_args()
+
+    print(f"Loading model from: {args.model_path}")
+    tokenizer = AutoTokenizer.from_pretrained(args.model_path)
+    model = AutoModelForSequenceClassification.from_pretrained(args.model_path)
+    model.eval()
+    
+    labels = [model.config.id2label[i] for i in range(model.config.num_labels)]
+
+    anonymize = not args.no_anonymize
+    processed_text = preprocess_text(args.text, anonymize_mentions=anonymize)
+    
+    if anonymize and processed_text != args.text:
+        print(f"Preprocessed text: {processed_text}")
+
+    if args.calibration_path:
+        calib_path = args.calibration_path
+    elif os.path.isdir(args.model_path):
+        calib_path = os.path.join(args.model_path, "calibration_artifacts.json")
+    else:
+        calib_path = os.path.join(os.path.dirname(__file__), "calibration_artifacts.json")
+    
+    print(f"Loading calibration from: {calib_path}")
+    calib_artifacts = load_calibration_artifacts(calib_path)
+    
+    temperatures = calib_artifacts["temperatures"]
+    optimal_thresholds = calib_artifacts["optimal_thresholds"]
+
+    print(f"\nInput text: {args.text}\n")
+
+    inputs = tokenizer(processed_text, return_tensors="pt", truncation=True, max_length=8192)
+    
+    with torch.no_grad():
+        outputs = model(**inputs)
+        logits = outputs.logits.squeeze().numpy()
+    
+    calibrated_probs = []
+    predictions = []
+    
+    for i, label in enumerate(labels):
+        temp = temperatures[label]
+        threshold = optimal_thresholds[label]
+        
+        calibrated_logit = logits[i] / temp
+        prob = 1 / (1 + np.exp(-calibrated_logit))
+        
+        calibrated_probs.append(prob)
+        predictions.append(prob > threshold)
+
+    assigned_labels = [labels[i] for i in range(len(labels)) if predictions[i]]
+    
+    if assigned_labels:
+        print("Assigned Labels (Calibrated):")
+        print("-" * 40)
+        for label in assigned_labels:
+            print(f"  {label}")
+        print()
+    else:
+        print("No labels assigned (all below optimal thresholds)\n")
+    
+    print("All Labels (with calibrated probabilities):")
+    print("-" * 40)
+    for i, label in enumerate(labels):
+        status = "✓" if predictions[i] else " "
+        threshold = optimal_thresholds[label]
+        print(f"{status} {label:15s}: {calibrated_probs[i]:.4f} (threshold: {threshold:.2f})")
+
+if __name__ == "__main__":
+    main()
+

requirements.txt

@@ -0,0 +1,4 @@
+transformers>=4.36.0
+torch>=2.0.0
+numpy>=1.24.0
+

special_tokens_map.json

@@ -0,0 +1,51 @@
+{
+  "bos_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "cls_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "mask_token": {
+    "content": "<mask>",
+    "lstrip": true,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<pad>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "sep_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer_config.json

@@ -0,0 +1,562 @@
+{
+  "add_prefix_space": true,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128000": {
+      "content": "<mask>",
+      "lstrip": true,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128001": {
+      "content": "<user_token_1>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128002": {
+      "content": "<user_token_2>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128003": {
+      "content": "<user_token_3>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128004": {
+      "content": "<user_token_4>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128005": {
+      "content": "<user_token_5>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128006": {
+      "content": "<user_token_6>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128007": {
+      "content": "<user_token_7>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128008": {
+      "content": "<user_token_8>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128009": {
+      "content": "<user_token_9>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128010": {
+      "content": "<user_token_10>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128011": {
+      "content": "<user_token_11>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128012": {
+      "content": "<user_token_12>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128013": {
+      "content": "<user_token_13>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128014": {
+      "content": "<user_token_14>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128015": {
+      "content": "<user_token_15>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128016": {
+      "content": "<user_token_16>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128017": {
+      "content": "<user_token_17>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128018": {
+      "content": "<user_token_18>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128019": {
+      "content": "<user_token_19>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128020": {
+      "content": "<user_token_20>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128021": {
+      "content": "<user_token_21>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128022": {
+      "content": "<user_token_22>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128023": {
+      "content": "<user_token_23>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128024": {
+      "content": "<user_token_24>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128025": {
+      "content": "<user_token_25>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128026": {
+      "content": "<user_token_26>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128027": {
+      "content": "<user_token_27>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128028": {
+      "content": "<user_token_28>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128029": {
+      "content": "<user_token_29>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128030": {
+      "content": "<user_token_30>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128031": {
+      "content": "<user_token_31>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128032": {
+      "content": "<user_token_32>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128033": {
+      "content": "<user_token_33>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128034": {
+      "content": "<user_token_34>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128035": {
+      "content": "<user_token_35>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128036": {
+      "content": "<user_token_36>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128037": {
+      "content": "<user_token_37>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128038": {
+      "content": "<user_token_38>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128039": {
+      "content": "<user_token_39>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128040": {
+      "content": "<user_token_40>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128041": {
+      "content": "<user_token_41>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128042": {
+      "content": "<user_token_42>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128043": {
+      "content": "<user_token_43>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128044": {
+      "content": "<user_token_44>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128045": {
+      "content": "<user_token_45>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128046": {
+      "content": "<user_token_46>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128047": {
+      "content": "<user_token_47>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128048": {
+      "content": "<user_token_48>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128049": {
+      "content": "<user_token_49>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128050": {
+      "content": "<user_token_50>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128051": {
+      "content": "<user_token_51>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128052": {
+      "content": "<user_token_52>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128053": {
+      "content": "<user_token_53>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128054": {
+      "content": "<user_token_54>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128055": {
+      "content": "<user_token_55>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128056": {
+      "content": "<user_token_56>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128057": {
+      "content": "<user_token_57>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128058": {
+      "content": "<user_token_58>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128059": {
+      "content": "<user_token_59>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128060": {
+      "content": "<user_token_60>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128061": {
+      "content": "<user_token_61>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128062": {
+      "content": "<user_token_62>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128063": {
+      "content": "<user_token_63>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "bos_token": "<s>",
+  "clean_up_tokenization_spaces": false,
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "errors": "replace",
+  "extra_special_tokens": {},
+  "mask_token": "<mask>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<pad>",
+  "sep_token": "</s>",
+  "tokenizer_class": "RobertaTokenizer",
+  "trim_offsets": true,
+  "unk_token": "<unk>"
+}