2211275
/
kgr
1
0
Fork 0
Code Issues Pull Requests Packages Projects 1 Releases Wiki Activity
kgr/lego/lego_graph_rebrickable.ipynb

699 lines
21 KiB
Plaintext
Raw Blame History

{
"cells": [
{
"cell_type": "markdown",
"id": "747b245f",
"metadata": {},
"source": [
"Build the Lego Knowledge Graph using the sources in `/data`."
]
},
{
"cell_type": "code",
"execution_count": 45,
"id": "90209948",
"metadata": {},
"outputs": [],
"source": [
"from rdflib import Graph, Namespace, XSD, OWL, RDF, RDFS, SKOS, URIRef, Literal\n",
"import pandas as pd\n",
"from datetime import datetime\n",
"import os"
]
},
{
"cell_type": "markdown",
"id": "fe91fa67",
"metadata": {},
"source": [
"Setup the requirements for building a knowledge graph"
]
},
{
"cell_type": "code",
"execution_count": 46,
"id": "8e573135",
"metadata": {},
"outputs": [],
"source": [
"g = Graph()\n",
"thm = Namespace(\"https://thm.de/\")\n",
"THM = Namespace(\"https://thm.de/ont/\")"
]
},
{
"cell_type": "markdown",
"id": "d56199d5",
"metadata": {},
"source": [
"# Rebrickable"
]
},
{
"cell_type": "markdown",
"id": "d1e1abb0",
"metadata": {},
"source": [
"![Rebrickable](\\data\\rebrickable\\downloads_schema_v3.png)"
]
},
{
"cell_type": "code",
"execution_count": 47,
"id": "d8a1fe84",
"metadata": {},
"outputs": [],
"source": [
"re_colors = pd.read_csv(\"data/rebrickable/colors.csv\")\n",
"re_elements = pd.read_csv(\"data/rebrickable/elements.csv\")\n",
"re_inventories = pd.read_csv(\"data/rebrickable/inventories.csv\")\n",
"re_inventory_minifigs = pd.read_csv(\"data/rebrickable/inventory_minifigs.csv\")\n",
"re_inventory_parts = pd.read_csv(\"data/rebrickable/inventory_parts.csv\")\n",
"re_inventory_sets = pd.read_csv(\"data/rebrickable/inventory_sets.csv\")\n",
"re_minifigs = pd.read_csv(\"data/rebrickable/minifigs.csv\")\n",
"re_part_categories = pd.read_csv(\"data/rebrickable/part_categories.csv\")\n",
"re_part_relationships = pd.read_csv(\"data/rebrickable/part_relationships.csv\")\n",
"re_parts = pd.read_csv(\"data/rebrickable/parts.csv\")\n",
"re_sets = pd.read_csv(\"data/rebrickable/sets.csv\")\n",
"re_themes = pd.read_csv(\"data/rebrickable/themes.csv\")"
]
},
{
"cell_type": "markdown",
"id": "f3677416",
"metadata": {},
"source": [
"Colors"
]
},
{
"cell_type": "code",
"execution_count": 48,
"id": "ae505704",
"metadata": {},
"outputs": [],
"source": [
"for color in re_colors.itertuples(index=False):\n",
" color_ref = thm[f\"color/{color.id}\"]\n",
"\n",
" g.add((color_ref, RDFS.label, Literal(color.name, lang=\"en\")))\n",
" g.add((color_ref, THM.color, Literal(color.rgb)))\n",
" g.add((color_ref, THM.is_transparent, Literal(color.is_trans, datatype=XSD.boolean)))\n",
" \n",
" if not pd.isna(color.y1):\n",
" # First appearance\n",
" g.add((color_ref, THM.first_year, Literal(datetime(year = int(color.y1), month=1, day=1))))\n",
" if not pd.isna(color.y2):\n",
" # Last appearance\n",
" g.add((color_ref, THM.last_year, Literal(datetime(year = int(color.y2), month=1, day=1))))\n"
]
},
{
"cell_type": "markdown",
"id": "e27b2bc4",
"metadata": {},
"source": [
"Part Categories"
]
},
{
"cell_type": "code",
"execution_count": 49,
"id": "fb9e17d6",
"metadata": {},
"outputs": [],
"source": [
"for part_category in re_part_categories.itertuples(index=False):\n",
" part_category_ref = thm[f\"part_category/{part_category.id}\"]\n",
"\n",
" g.add((part_category_ref, RDFS.label, Literal(part_category_ref, lang=\"en\")))"
]
},
{
"cell_type": "markdown",
"id": "ea32849b",
"metadata": {},
"source": [
"Parts"
]
},
{
"cell_type": "code",
"execution_count": 50,
"id": "8fdb080e",
"metadata": {},
"outputs": [],
"source": [
"for part in re_parts.itertuples(index=False):\n",
" part_ref = thm[f\"part/{part.part_num}\"]\n",
"\n",
" g.add((part_ref, RDFS.label, Literal(part.name, lang=\"en\")))\n",
" g.add((part_ref, THM.part_category, thm[f\"part_category/{part.part_cat_id}\"]))\n",
" g.add((part_ref, THM.part_material, Literal(part.part_material)))"
]
},
{
"cell_type": "markdown",
"id": "fcaadd84",
"metadata": {},
"source": [
"Elements"
]
},
{
"cell_type": "code",
"execution_count": 51,
"id": "579b1d67",
"metadata": {},
"outputs": [],
"source": [
"for element in re_elements.itertuples(index=False):\n",
" part_ref = thm[f\"part/{element.part_num}\"]\n",
" color_ref = thm[f\"color/{element.color_id}\"]\n",
"\n",
" g.add((part_ref, THM.has_color, color_ref))"
]
},
{
"cell_type": "markdown",
"id": "44dae336",
"metadata": {},
"source": [
"Part Relationships"
]
},
{
"cell_type": "code",
"execution_count": 52,
"id": "00db079a",
"metadata": {},
"outputs": [],
"source": [
"for part_relationship in re_part_relationships.itertuples(index=False):\n",
" part_ref_parent = thm[f\"part/{part_relationship.parent_part_num}\"]\n",
" part_ref_child = thm[f\"part/{part_relationship.child_part_num}\"]\n",
"\n",
" g.add((part_ref_parent, THM.has_child, part_ref_child))"
]
},
{
"cell_type": "markdown",
"id": "19dc64b8",
"metadata": {},
"source": [
"Themes"
]
},
{
"cell_type": "code",
"execution_count": 53,
"id": "1a529aae",
"metadata": {},
"outputs": [],
"source": [
"for theme in re_themes.itertuples(index=False):\n",
" theme_ref = thm[f\"theme/{int(theme.id)}\"]\n",
"\n",
" g.add((theme_ref, RDFS.label, Literal(theme.name, lang=\"en\")))\n",
"\n",
" if not pd.isna(theme.parent_id):\n",
" g.add((theme_ref, THM.parent_theme, thm[f\"theme/{int(theme.parent_id)}\"]))"
]
},
{
"cell_type": "markdown",
"id": "3f72c2e9",
"metadata": {},
"source": [
"Sets"
]
},
{
"cell_type": "code",
"execution_count": 54,
"id": "29b357ef",
"metadata": {},
"outputs": [],
"source": [
"for lego_set in re_sets.itertuples(index=False):\n",
" set_ref = thm[f\"set/lego/{lego_set.set_num}\"]\n",
"\n",
" g.add((set_ref, RDFS.label, Literal(lego_set.name, lang=\"en\")))\n",
" g.add((set_ref, THM.year, Literal(datetime(int(lego_set.year), 1, 1))))\n",
" g.add((set_ref, THM.theme, thm[f\"theme/{int(lego_set.theme_id)}\"]))\n",
" g.add((set_ref, THM.num_parts, Literal(int(lego_set.num_parts), datatype=XSD.integer)))\n",
" g.add((set_ref, THM.brand, Literal(\"Lego\")))"
]
},
{
"cell_type": "markdown",
"id": "d2616476",
"metadata": {},
"source": [
"Minifigures"
]
},
{
"cell_type": "code",
"execution_count": 55,
"id": "a67b3e70",
"metadata": {},
"outputs": [],
"source": [
"for minifig in re_minifigs.itertuples(index=False):\n",
" minifig_ref = thm[f\"minifig/{minifig.fig_num}\"]\n",
"\n",
" g.add((set_ref, RDFS.label, Literal(minifig.name, lang=\"en\")))\n",
" g.add((set_ref, THM.num_parts, Literal(int(minifig.num_parts), datatype=XSD.integer)))"
]
},
{
"cell_type": "markdown",
"id": "2e9baff1",
"metadata": {},
"source": [
"Now the ugly part: Inventories"
]
},
{
"cell_type": "code",
"execution_count": 56,
"id": "0c97dc4d",
"metadata": {},
"outputs": [],
"source": [
"for inventory in re_inventories.itertuples(index=False):\n",
" inventory_ref = thm[f\"inventory/{inventory.id}\"]\n",
"\n",
" g.add((inventory_ref, THM.set, thm[f\"set/lego/{inventory.set_num}\"]))"
]
},
{
"cell_type": "markdown",
"id": "7c962cf0",
"metadata": {},
"source": [
"Inventories relate sets, minifigures and parts to each other, creating a kind of \"top level set\" \n",
"(this takes a lot of time)"
]
},
{
"cell_type": "code",
"execution_count": 57,
"id": "dc2ba03e",
"metadata": {},
"outputs": [],
"source": [
"for inventory_part in re_inventory_parts.itertuples(index=False):\n",
" inventory_part_ref = thm[f\"inventory_part/{inventory_part.inventory_id}/{inventory_part.part_num}\"]\n",
" \n",
" inventory_ref = thm[f\"inventory/{inventory_part.inventory_id}\"]\n",
" part_ref = thm[f\"part/{inventory_part.part_num}\"]\n",
"\n",
" g.add((inventory_part_ref, RDFS.domain, inventory_ref))\n",
" g.add((inventory_part_ref, RDFS.range, part_ref))\n",
" g.add((inventory_part_ref, RDF.type, RDF.Property))\n",
" \n",
" g.add((inventory_part_ref, THM.quantity, Literal(int(inventory_part.quantity), datatype=XSD.integer)))\n",
" g.add((inventory_part_ref, THM.is_spare, Literal(inventory_part.is_spare, datatype=XSD.boolean)))\n",
" g.add((inventory_part_ref, THM.color, thm[f\"color/{inventory_part.color_id}\"]))"
]
},
{
"cell_type": "code",
"execution_count": 58,
"id": "8715a1cf",
"metadata": {},
"outputs": [],
"source": [
"for inventory_set in re_inventory_sets.itertuples(index=False):\n",
" inventory_set_ref = thm[f\"inventory_set/{inventory_set.inventory_id}/{inventory_set.set_num}\"]\n",
"\n",
" inventory_ref = thm[f\"inventory/{inventory_set.inventory_id}\"]\n",
" set_ref = thm[f\"set/lego/{inventory_set.set_num}\"]\n",
"\n",
" g.add((inventory_set_ref, RDFS.domain, inventory_ref))\n",
" g.add((inventory_set_ref, RDFS.range, set_ref))\n",
" g.add((inventory_set_ref, RDF.type, RDF.Property))\n",
"\n",
" g.add((inventory_set_ref, THM.quantity, Literal(int(inventory_set.quantity), datatype=XSD.integer)))"
]
},
{
"cell_type": "code",
"execution_count": 59,
"id": "08c2c580",
"metadata": {},
"outputs": [],
"source": [
"for inventory_minifig in re_inventory_minifigs.itertuples(index=False):\n",
" inventory_minifig_ref = thm[f\"inventory_minifig/{inventory_minifig.inventory_id}/{inventory_minifig.fig_num}\"]\n",
"\n",
" inventory_ref = thm[f\"inventory/{inventory_minifig.inventory_id}\"]\n",
" minifig_ref = thm[f\"minifig/lego/{inventory_minifig.fig_num}\"]\n",
"\n",
" g.add((inventory_minifig_ref, RDFS.domain, inventory_ref))\n",
" g.add((inventory_minifig_ref, RDFS.range, minifig_ref))\n",
" g.add((inventory_minifig_ref, RDF.type, RDF.Property))\n",
"\n",
" g.add((inventory_minifig_ref, THM.quantity, Literal(int(inventory_minifig.quantity), datatype=XSD.integer)))"
]
},
{
"cell_type": "markdown",
"id": "dcbab237",
"metadata": {},
"source": [
"# Brickset"
]
},
{
"cell_type": "markdown",
"id": "d8fb5374",
"metadata": {},
"source": [
"add for prices"
]
},
{
"cell_type": "code",
"execution_count": 60,
"id": "1e0ac437",
"metadata": {},
"outputs": [],
"source": [
"bs_sets = pd.read_csv(\"./data/brickset/sets.csv\")"
]
},
{
"cell_type": "code",
"execution_count": 61,
"id": "fd944ccb",
"metadata": {},
"outputs": [],
"source": [
"for bs_set in bs_sets.itertuples(index=False):\n",
" num = f\"{str(bs_set.Number).strip()}-{str(bs_set.Variant)}\" #Error for Set 853357\n",
" set_ref = thm[f\"set/lego/{num}\"]\n",
"\n",
" if (set_ref, None, None) in g:\n",
" if not pd.isna(bs_set.USRetailPrice):\n",
" g.add((set_ref, THM.us_retail_price, Literal(bs_set.USRetailPrice, datatype=XSD.float)))\n",
" if not pd.isna(bs_set.DERetailPrice):\n",
" g.add((set_ref, THM.de_retail_price, Literal(bs_set.DERetailPrice, datatype=XSD.float)))\n",
" if not pd.isna(bs_set.UKRetailPrice):\n",
" g.add((set_ref, THM.us_retail_price, Literal(bs_set.UKRetailPrice, datatype=XSD.float)))\n",
" if not pd.isna(bs_set.CARetailPrice):\n",
" g.add((set_ref, THM.ca_retail_price, Literal(bs_set.CARetailPrice, datatype=XSD.float)))"
]
},
{
"cell_type": "markdown",
"id": "a16fd51a",
"metadata": {},
"source": [
"# Bricklink\n",
"\n",
"Add missing minifigs, parts and sets using bricklink data"
]
},
{
"cell_type": "code",
"execution_count": 62,
"id": "a8beb593",
"metadata": {},
"outputs": [],
"source": [
"bl_minifigs = pd.read_csv(\"./data/bricklink/minifigs.csv\")\n",
"bl_sets = pd.read_csv(\"./data/bricklink/sets.csv\")\n",
"bl_parts = pd.read_csv(\"./data/bricklink/parts.csv\")"
]
},
{
"cell_type": "code",
"execution_count": 63,
"id": "b14e6e89",
"metadata": {},
"outputs": [],
"source": [
"additional_entries = 0"
]
},
{
"cell_type": "code",
"execution_count": 64,
"id": "bbf5462b",
"metadata": {},
"outputs": [],
"source": [
"for bl_set in bl_sets.itertuples(index=False):\n",
" set_ref = thm[f\"set/lego/{bl_set.set_id}\"]\n",
"\n",
" if not (set_ref, None, None) in g:\n",
" additional_entries += 1\n",
" g.add((set_ref, RDFS.label, Literal(lego_set.name, lang=\"en\")))\n",
" if not pd.isna(bl_set.year) and str(bl_set.year).isnumeric():\n",
" g.add((set_ref, THM.first_year, Literal(datetime(int(bl_set.year), 1, 1))))\n",
" if not pd.isna(bl_set.parts) and str(bl_set.parts).isnumeric():\n",
" g.add((set_ref, THM.num_parts, Literal(int(bl_set.parts), datatype=XSD.integer)))\n",
" g.add((set_ref, THM.brand, Literal(\"Lego\")))"
]
},
{
"cell_type": "code",
"execution_count": 65,
"id": "ef52582e",
"metadata": {},
"outputs": [],
"source": [
"for bl_part in bl_parts.itertuples(index=False):\n",
" part_ref = thm[f\"part/{bl_part.part_id}\"]\n",
"\n",
" if not (part_ref, None, None) in g:\n",
" additional_entries += 1\n",
" g.add((part_ref, RDFS.label, Literal(bl_part.part_name, lang=\"en\")))"
]
},
{
"cell_type": "code",
"execution_count": 66,
"id": "8bf0ffeb",
"metadata": {},
"outputs": [],
"source": [
"for bl_minifig in bl_minifigs.itertuples(index=False):\n",
" minifig_ref = thm[f\"minfig/{bl_minifig.minifig_id}\"]\n",
"\n",
" if not (minifig_ref, None, None) in g:\n",
" additional_entries += 1\n",
" g.add((minifig_ref, RDFS.label, Literal(bl_minifig.minifig_name, lang=\"en\")))"
]
},
{
"cell_type": "code",
"execution_count": 67,
"id": "3491b032",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Added 107748 items\n"
]
}
],
"source": [
"print(f\"Added {additional_entries} items\")"
]
},
{
"cell_type": "markdown",
"id": "6eca7f24",
"metadata": {},
"source": [
"# Merlin"
]
},
{
"cell_type": "code",
"execution_count": 68,
"id": "c1e9ff32",
"metadata": {},
"outputs": [],
"source": [
"merlin_sets = pd.read_csv(\"./data/merlin/others.csv\")"
]
},
{
"cell_type": "code",
"execution_count": 69,
"id": "217dc4d2",
"metadata": {},
"outputs": [],
"source": [
"additional_entries = 0 \n",
"\n",
"for merlin_set in merlin_sets.itertuples(index=False):\n",
" set_ref = thm[f\"set/lego/{merlin_set.id}\"]\n",
" producer_ref = thm[f\"producer/{merlin_set.producer}\"]\n",
"\n",
" #add producer\n",
" if not (producer_ref, None, None) in g:\n",
" g.add((producer_ref, RDFS.label, Literal(merlin_set.producer)))\n",
"\n",
" #add the set\n",
" if not (set_ref, None, None) in g:\n",
" additional_entries += 1\n",
" g.add((set_ref, RDFS.label, Literal(merlin_set.name)))\n",
" g.add((set_ref, THM.brand, producer_ref))\n",
" if not pd.isna(merlin_set.year):\n",
" g.add((set_ref, THM.first_year, Literal(int(merlin_set.year), datatype=XSD.integer)))\n",
" if not pd.isna(merlin_set.parts):\n",
" g.add((set_ref, THM.num_parts, Literal(int(merlin_set.parts), datatype=XSD.integer)))\n",
" g.add((set_ref, THM.size_word, Literal(merlin_set.size)))\n"
]
},
{
"cell_type": "code",
"execution_count": 70,
"id": "54eaa79e",
"metadata": {},
"outputs": [],
"source": [
"merlin_prices = pd.read_csv(\"./data/merlin/prices.csv\")\n",
"\n",
"for merlin_price in merlin_prices.itertuples(index=False):\n",
" set_ref = thm[f\"set/lego/{merlin_price.id}\"]\n",
" if merlin_price.price_eur not in [None, \"_\"]:\n",
" g.add((set_ref, THM.eur_price, Literal(merlin_price.price_eur, datatype=XSD.float)))\n",
" #usd (data src not clean)\n",
" if merlin_price._3 not in [None, \"_\"]:\n",
" g.add((set_ref, THM.usd_price, Literal(merlin_price._3, datatype=XSD.float)))\n",
" # cn \n",
" if merlin_price._2 not in [None, \"_\"]:\n",
" g.add((set_ref, THM.cn_price, Literal(merlin_price._2, datatype=XSD.float)))\n",
" # best price euro \n",
" if merlin_price._4 not in [None, \"_\"]:\n",
" g.add((set_ref, THM.eur_bestprice, Literal(merlin_price._4, datatype=XSD.float)))\n",
" # best price cn\n",
" if merlin_price._5 not in [None, \"_\"]:\n",
" g.add((set_ref, THM.cn_bestprice, Literal(merlin_price._5, datatype=XSD.float)))\n",
" # best price usd\n",
" if merlin_price._6 not in [None, \"_\"]:\n",
" g.add((set_ref, THM.usd_bestprice, Literal(merlin_price._6, datatype=XSD.float)))\n"
]
},
{
"cell_type": "markdown",
"id": "bfab0c73",
"metadata": {},
"source": [
"Serialize the graph in turtle format"
]
},
{
"cell_type": "markdown",
"id": "2abd6894",
"metadata": {},
"source": [
"```\n",
" ___-------___\n",
" _-~~ ~~-_\n",
" _-~ /~-_\n",
" /^\\__/^\\ /~ \\ / \\\n",
" /| O|| O| / \\_______________/ \\\n",
"| |___||__| / / \\ \\\n",
"| \\ / / \\ \\\n",
"| (_______) /______/ \\_________ \\\n",
"| / / \\ / \\\n",
" \\ \\^\\\\ \\ / \\ /\n",
" \\ || \\______________/ _-_ //\\__//\n",
" \\ ||------_-~~-_ ------------- \\ --/~ ~\\ || __/\n",
" ~-----||====/~ |==================| |/~~~~~\n",
" (_(__/ ./ / \\_\\ \\.\n",
" (_(___/ \\_____)_)\n",
"```"
]
},
{
"cell_type": "code",
"execution_count": 71,
"id": "1a30bff8",
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"<Graph identifier=N20b690213c5c45319ad250ee399204d9 (<class 'rdflib.graph.Graph'>)>"
]
},
"execution_count": 71,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"g.bind(\"thmont\", THM)\n",
"\n",
"g.serialize(\"lego_graph_rebrickable.ttl\", format=\"turtle\")"
]
},
{
"cell_type": "code",
"execution_count": 100,
"id": "7b9c3bcf",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"(rdflib.term.URIRef('https://thm.de/inventory/102116'), rdflib.term.URIRef('https://thm.de/inventory_part/102116/3626cpr3530'), rdflib.term.Literal('1', datatype=rdflib.term.URIRef('http://www.w3.org/2001/XMLSchema#integer')))\n",
"(rdflib.term.URIRef('https://thm.de/inventory/102116'), rdflib.term.URIRef('https://thm.de/inventory_part/102116/970c12'), rdflib.term.Literal('1', datatype=rdflib.term.URIRef('http://www.w3.org/2001/XMLSchema#integer')))\n",
"(rdflib.term.URIRef('https://thm.de/inventory/102116'), rdflib.term.URIRef('https://thm.de/inventory_part/102116/973c14h14pr5739'), rdflib.term.Literal('1', datatype=rdflib.term.URIRef('http://www.w3.org/2001/XMLSchema#integer')))\n"
]
}
],
"source": [
"import rdflib\n",
" \n",
"query = \"\"\"\n",
"SELECT ?inventory ?inventory_part ?quantity\n",
"WHERE {\n",
" ?inventory thmont:set <https://thm.de/set/lego/fig-011397> .\n",
" ?inventory_part rdfs:domain ?inventory .\n",
" ?inventory_part thmont:quantity ?quantity \n",
"}\n",
"\"\"\"\n",
"\n",
"qres = g.query(query)\n",
"\n",
"for row in qres :\n",
" print(row)"
]
}
],
"metadata": {
"kernelspec": {
"display_name": "kgr",
"language": "python",
"name": "python3"
},
"language_info": {
"codemirror_mode": {
"name": "ipython",
"version": 3
},
"file_extension": ".py",
"mimetype": "text/x-python",
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
"version": "3.11.2"
}
},
"nbformat": 4,
"nbformat_minor": 5
}
Reference in New Issue View Git Blame Copy Permalink