cboulanger

uv run scripts/evaluate_llm.py --max-items 10 --show-annotations --output-file .local/evaluate-llm.log --match-mode overlap --provider kisski


────────────────────────────────────────────────────────────────
  Provider  : Gemini 2.0 Flash
  Gold file : tests/fixtures/blbl-examples.tei.xml
  Records   : 10   match-mode: overlap
  GLiNER    : disabled
────────────────────────────────────────────────────────────────
  ────────────────────────────────────────────────────────────

cboulanger

<?xml version="1.0" encoding="UTF-8"?>
<?xml-model href="https://relaxng.org/relaxng.rng" type="application/xml" schematypens="http://relaxng.org/ns/structure/1.0"?>

<!--
RNG Schema for validating GROBID *.training.segmentation.tei.xml files
-->

<grammar xmlns="http://relaxng.org/ns/structure/1.0"
  xmlns:xml="http://www.w3.org/XML/1998/namespace"
  ns="http://www.tei-c.org/ns/1.0">

cboulanger

from datetime import datetime, timedelta
from collections import defaultdict

def solve_advent_calendar():
    persons = ['T', 'C', 'K']
    c_absent_days = {1, 2, 5, 6, 7, 8, 9, 10, 11, 15, 16, 17}
    
    # Jede Person muss jedem anderen genau 4 Geschenke geben
    # T->C: 4, T->K: 4, C->T: 4, C->K: 4, K->T: 4, K->C: 4
    gifts_needed = {

cboulanger

Crossref to RIS Converter

This Python script converts Crossref API bibliographic data into RIS format, specifically designed to handle books with chapters. It enriches chapter records with complete book information (title, editors, ISBN) that is typically missing in the Crossref API response.

Features

• Fetches bibliographic data from Crossref API using ISBN
• Generates complete RIS records for both the book and all chapters
• Includes full book metadata (title, editors, publisher, ISBNs) in each chapter entry
• Can work with pre-downloaded JSON files when API access is restricted

cboulanger

<?xml version="1.0" encoding="UTF-8"?>
<?xml-model href="https://relaxng.org/relaxng.rng" type="application/xml" schematypens="http://relaxng.org/ns/structure/1.0"?>
<!--
To validate TEI documents against this schema, add this processing instruction
to the beginning of your TEI document (after the XML declaration):
<?xml-model href="http://localhost:3001/api/files/872ab" type="application/xml" schematypens="http://relaxng.org/ns/structure/1.0"?>
-->
<grammar xmlns="http://relaxng.org/ns/structure/1.0"
  xmlns:xml="http://www.w3.org/XML/1998/namespace"
  ns="http://www.tei-c.org/ns/1.0"

cboulanger

/**
 * patched version of https://www.npmjs.com/package/js-plugin
 */

var _plugins = [];
var _byName = {};
var _cache = {};

// Only support debug mode on browser and node, not web workers.
var isBrowser = typeof window !== 'undefined' && typeof window.document !== 'undefined';

cboulanger

    /**
     * Displays Countdown clock
     * Usage: <div class="countdown" style="--duration-seconds: 15;"></div>
     */
    @property --total-seconds-remaining {
      syntax: "<number>";
      initial-value: 0;
      inherits: true;
    }

cboulanger

Lorem ipsum dolor sit amet

Lorem

• dolor sit amet
• consectetur adipiscing
• sed do eiusmod tempor

cboulanger

OpenAI reference extraction via API: Code/Result

December 2022

Code

import openai
import json
import time

cboulanger

<?xml version="1.0" ?>
<TEI xmlns="http://www.tei-c.org/ns/1.0">
  <teiHeader>
    <fileDesc>
      <titleStmt>
        <title>10.1111_1467-6478.00057</title>
      </titleStmt>
      <publicationStmt>
        <publisher>mpilhlt</publisher>
      </publicationStmt>