entity nodes and ent nodes

.jupyter/desktop-settings.json

@@ -0,0 +1,3 @@
+{
+ "uiMode": "multi-document"
+}

.jupyter/desktop-workspaces/default-37a8.jupyterlab-workspace

@@ -0,0 +1 @@
+{"data":{"layout-restorer:data":{"main":{"dock":{"type":"tab-area","currentIndex":0,"widgets":["notebook:programs/entities.ipynb","notebook:tutorial/entities.ipynb"]},"current":"notebook:programs/entities.ipynb"},"down":{"size":0,"widgets":[]},"left":{"collapsed":false,"visible":true,"current":"filebrowser","widgets":["filebrowser","running-sessions","@jupyterlab/toc:plugin","extensionmanager.main-view"],"widgetStates":{"jp-running-sessions":{"sizes":[0.25,0.25,0.25,0.25],"expansionStates":[false,false,false,false]},"extensionmanager.main-view":{"sizes":[0.3333333333333333,0.3333333333333333,0.3333333333333333],"expansionStates":[false,false,false]}}},"right":{"collapsed":true,"visible":true,"widgets":["jp-property-inspector","debugger-sidebar"],"widgetStates":{"jp-debugger-sidebar":{"sizes":[0.2,0.2,0.2,0.2,0.2],"expansionStates":[false,false,false,false,false]}}},"relativeSizes":[0.2115921787709497,0.7884078212290503,0],"top":{"simpleVisibility":true}},"file-browser-filebrowser:cwd":{"path":"tutorial"},"notebook:programs/entities.ipynb":{"data":{"path":"programs/entities.ipynb","factory":"Notebook"}},"notebook:tutorial/entities.ipynb":{"data":{"path":"tutorial/entities.ipynb","factory":"Notebook"}}},"metadata":{"id":"default"}}

Untitled.ipynb

@@ -0,0 +1,33 @@
+{
+ "cells": [
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "id": "1b4e1830-3934-49b5-b08a-88d40e9c4c03",
+ "metadata": {},
+ "outputs": [],
+ "source": []
+ }
+ ],
+ "metadata": {
+ "kernelspec": {
+ "display_name": "Python 3 (ipykernel)",
+ "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": "3.12.2"
+ }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}

app/__checkout__.txt

@@ -1,2 +1,2 @@
 v1.1
-g8666face5331ed80d6b88c942fa450d9f82d4959
+g61b0cb1b6bb6e9c4549a53aa5db557ffe37c1946

ner/__checkout__.txt

@@ -1,2 +1,2 @@
 v1.1
-g8666face5331ed80d6b88c942fa450d9f82d4959
+g61b0cb1b6bb6e9c4549a53aa5db557ffe37c1946

programs/addEntities.py

@@ -1,3 +1,4 @@
+from itertools import chain
 from tf.app import use
 from tf.core.helpers import console
 from tf.dataset import modify
@@ -90,19 +91,36 @@ def prepareData(self):
  slotLink = {}
  kindFeature = {}
  eidFeature = {}
+ occEdge = {}
+ entities = {}
 
- for i, streak in enumerate(streaks):
- n = i + 1
+ n = 0
+
+ for streak in streaks:
+ n += 1
  refS = streak[0]
  slotLink[n] = streak
- kindFeature[n] = kindv(refS)
+ ekind = kindv(refS)
+ kindFeature[n] = ekind
  eid = toId(T.text(streak))
  eidFeature[n] = eid
+ entities.setdefault((eid, ekind), []).append(n)
+
+ nStreaks = len(streaks)
+
+ for ((eid, ekind), ms) in entities.items():
+ n += 1
+ occEdge[n] = set(ms)
+ slotLink[n] = tuple(chain.from_iterable(slotLink[m] for m in ms))
+ kindFeature[n] = ekind
+ eidFeature[n] = eid
 
  console(f"{len(streaks):>5} entity nodes")
  console(f"{len(set(eidFeature.values())):>5} distinct eids")
+ console(f"{len(entities):>5} distinct entities")
 
- features = dict(kind=kindFeature, eid=eidFeature)
+ nodeFeatures = dict(kind=kindFeature, eid=eidFeature)
+ edgeFeatures = dict(eoccs=occEdge)
 
  featureMeta = dict(
  eid=dict(
@@ -113,18 +131,31 @@ def prepareData(self):
  valueType="str",
  description="entity kind",
  ),
+ eoccs=dict(
+ valueType="str",
+ description="entity occurrences",
+ )
  )
 
  self.addTypes = dict(
  ent=dict(
  nodeFrom=1,
- nodeTo=len(streaks),
+ nodeTo=nStreaks,
  nodeSlots=slotLink,
- nodeFeatures=features,
+ nodeFeatures=nodeFeatures,
+ ),
+ entity=dict(
+ nodeFrom=nStreaks + 1,
+ nodeTo=nStreaks + len(entities),
+ nodeSlots=slotLink,
+ nodeFeatures=nodeFeatures,
+ edgeFeatures=edgeFeatures,
  ),
  )
  self.featureMeta = featureMeta
 
+ return True
+
  def modify(self):
  A = self.A
  TF = A.TF
@@ -143,7 +174,7 @@ def modify(self):
  f for f in TF.features if f.startswith("omap@")
  ]
 
- modify(
+ return modify(
  origTf,
  newTf,
  targetVersion=newVersion,
@@ -157,7 +188,7 @@ def tweakApp(self):
 
  config = f"{A.repoLocation}/app/config.yaml"
  oldVersion = A.version
- newVersion = f"{oldVersion}e"
+ newVersion = f"{oldVersion}ent"
 
  with open(config) as fh:
  text = fh.read()
@@ -180,7 +211,11 @@ def run(self):
  if not self.checkStreaks():
  return
 
- self.prepareData()
- self.modify()
+ if not self.prepareData():
+ return
+
+ if not self.modify():
+ return
+
  self.tweakApp()
  self.loadNew()