mapdata 3.20.0__tar.gz → 3.22.0__tar.gz

{mapdata-3.20.0/mapdata.egg-info → mapdata-3.22.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: mapdata
-Version: 3.20.0
+Version: 3.22.0
 Summary: An interactive map and table explorer for geographic coordinates in a spreadsheet, CSV file, or database that includes data plotting and statistical summaries
 Home-page: https:/hg.sr.ht/~rdnielsen/mapdata
 Author: Dreas Nielsen
@@ -24,21 +24,6 @@ Classifier: Topic :: Scientific/Engineering
 Requires-Python: >=3.8
 Description-Content-Type: text/markdown
 License-File: LICENSE.txt
-Requires-Dist: tkintermapview
-Requires-Dist: pyproj
-Requires-Dist: jenkspy
-Requires-Dist: odfpy
-Requires-Dist: openpyxl
-Requires-Dist: xlrd
-Requires-Dist: matplotlib
-Requires-Dist: seaborn
-Requires-Dist: loess
-Requires-Dist: statsmodels
-Requires-Dist: scipy
-Requires-Dist: scikit-learn
-Requires-Dist: pymannkendall
-Requires-Dist: umap-learn
-Requires-Dist: pynndescent
 
 
 *mapdata.py* is a data explorer for data sets containing geographic coordinates.  Data can be read from a CSV file, spreadsheet, or database.  Both a map and a data

{mapdata-3.20.0 → mapdata-3.22.0}/mapdata/mapdata.py

@@ -24,8 +24,8 @@
 #
 # ==================================================================
 
-version = "3.20.0"
-vdate = "2026-03-09"
+version = "3.22.0"
+vdate = "2026-03-31"
 
 copyright = "2023-2026"
 
@@ -99,6 +99,7 @@ try:
 	mk_available = True
 except:
 	mk_available = False
+import kneeliverse
 
 
 # Default name of configuration file.  Files with other names may be read.
@@ -2361,6 +2362,41 @@ def list_mse(y):
     # This computes the MSE for differences between elements of the list and 1.0
     return sum([(s-1)**2 for s in y])/len(y)
 
+def elbow_kneeliverse(rmses):
+	# Returns the x value of the elbow
+	pts = np.array(list(zip(list(range(1, len(rmses)+1)), rmses)))
+	return kneeliverse.kneedle.knee(pts)
+
+class LinePoints(object):
+	# For calculation of elbow inflection using `kneedle` algorithm:
+	# maximum distance of point from line.
+	def __init__(self, x1, y1, x2, y2):
+		self.yd = y2 - y1
+		self.xd = x2 - x1
+		self.xyterm = x2 * y1 - y2 * x1
+		self.denom = math.sqrt(self.yd**2 + self.xd**2)
+	def point_dist(self, xx, yy):
+		return abs(self.yd * xx - self.xd * yy + self.xyterm) / self.denom
+	def signed_point_dist(self, xx, yy):
+		# Positive values are below the line, negative values above.
+		return (self.yd * xx - self.xd * yy + self.xyterm) / self.denom
+
+def kneedle_maxdist(rmses):
+	xs = list(range(1, len(rmses)+1))
+	lp = LinePoints(xs[0], rmses[0], xs[-1], rmses[-1])
+	dists = [lp.signed_point_dist(xs[i], rmses[i]) for i in range(1,len(rmses)-1)]
+	return dists.index(max(dists)) + 2
+
+def elbow(rmses, eucdists):
+	# Returns the elbow in the list of RMSE or Euclidean distance values.
+	if len(rmses) < 3:
+		return len(rmses)
+	kd_rmse_em = max(kneedle_maxdist(rmses), kneedle_maxdist(eucdists))
+	if kd_rmse_em < 5:
+		return kd_rmse_em
+	return elbow_kneeliverse(rmses)
+
+
 def iterate_nmf(D, n_em, max_iterations=200, mse_goal=0.00001):
 	# Performs NMF on D until the row sums of the W matrix are all (approximately) 1.0.
     nmf_model1 = NMF(n_components=n_em, solver='mu', max_iter=500)
@@ -6745,6 +6781,15 @@ class UniqueValuesDialog(Dialog):
 		self.data_labels = None
 		self.dlg.bind("<Control-s>")
 
+		# Top controls
+		self.sel_only_var, self.sel_only_ck = add_sel_only(self.ctrl_frame, 0, 0, self.q_recalc)
+		self.indiv_var = tk.StringVar(self.ctrl_frame, "0")
+		indiv_ck = ttk.Checkbutton(self.ctrl_frame, text="Individually", command=self.q_recalc, variable=self.indiv_var,
+				onvalue="1", offvalue="0")
+		indiv_ck.grid(row=0, column=1, sticky=tk.W, padx=(6,3), pady=(3,3), columnspan=None)
+		self.ctrl_frame.columnconfigure(0, weight=0)
+		self.ctrl_frame.columnconfigure(1, weight=1)
+
 		# The content_frame encompasses the two panes of the variable frame and the content frame
 		frame_panes = ttk.PanedWindow(self.content_frame, orient=tk.HORIZONTAL, width=600)
 		frame_panes.grid(row=0, column=0, sticky=tk.NSEW)
@@ -6753,7 +6798,7 @@ class UniqueValuesDialog(Dialog):
 		var_frame = tk.Frame(frame_panes, width=250, borderwidth=2, relief=tk.RIDGE)
 		var_frame.grid(row=0, column=0, sticky=tk.NSEW)
 		var_frame.rowconfigure(0, weight=1)
-		var_frame.columnconfigure(0, weight=1)
+		var_frame.columnconfigure(0, weight=1, minsize=200)
 		frame_panes.add(var_frame, weight=1)
 		# Add multi-select list of variables to the leftmost pane
 		varvals = [[v] for v in self.variables]
@@ -6787,20 +6832,48 @@ class UniqueValuesDialog(Dialog):
 
 	def q_recalc(self, *event, get_data=True):
 		self.clear_output()
-		if len(self.column_table.selection()) > 0 and get_data:
-			self.recalc()
+		if len(self.column_table.selection()) > 0:
+			self.get_data()
+			if self.rawdataset is not None and len(self.rawdataset[0]) > 0:
+				self.recalc()
 
-	def recalc(self):
-		column_list = []
+	def get_data(self):
+		# Get the selected data into 'dataset'
+		self.rawdataset = None
+		self.colnames = []
 		for sel_row in self.column_table.selection():
 			datarow = self.column_table.item(sel_row)["values"]
-			column_list.append(datarow[0])
-		dbcols = db_colnames(column_list)
-		colstr = ",".join(dbcols)
-		sqlcmd = f"select distinct {colstr}, count(*) as nrows from mapdata group by {colstr} order by {colstr};"
-		cur = data_db.cursor()
-		self.dataset = cur.execute(sqlcmd).fetchall()
-		self.data_labels = column_list + ["Count"]
+			self.colnames.append(datarow[0])
+		if len(self.colnames) > 0:
+			# Get either only the selected data or all data.
+			dataset = chosen_dataset(self.parent, self.colnames, self.sel_only_var.get()=='1')
+			if dataset is None or len(dataset[0]) == 0:
+				self.rawdataset = None
+				self.colnames = None
+			else:
+				self.rawdataset = dataset
+
+	def recalc(self):
+		if self.indiv_var.get()=='1':
+			counts = dict(zip(self.colnames, [0 for _ in range(len(self.colnames))]))
+			for i,col in enumerate(self.colnames):
+				counts[col] = len(set([x for x in self.rawdataset[i] if x is not None and str(x) != '']))
+			self.dataset = list(counts.items())
+			self.data_labels = ["Variable", "Count"]
+		else:
+			counts = {}
+			ds = columns_to_rows(self.rawdataset)
+			for row in ds:
+				trow = tuple(row)
+				if trow not in counts.keys():
+					counts[trow] = 1
+				else:
+					counts[trow] += 1
+			self.dataset = [list(k) for k in counts.keys()]
+			vals = list(counts.values())
+			for i in range(len(counts)):
+				self.dataset[i].append(vals[i])
+			self.data_labels = self.colnames + ["Count"]
 		tframe, tdata = treeview_table(self.values_frame, self.dataset, self.data_labels)
 		tframe.grid(row=0, column=0, sticky=tk.NSEW)
 		self.dlg.bind("<Control-s>", self.save_data)
@@ -13656,10 +13729,10 @@ class NMFUnmixingDialog(Dialog):
 		frame_panes.grid(row=0, column=0, sticky=tk.NSEW)
 
 		# Variable frame for list of quantitative columns/variables
-		var_frame = tk.Frame(frame_panes, borderwidth=2, relief=tk.RIDGE)
+		var_frame = tk.Frame(frame_panes, width=250, borderwidth=2, relief=tk.RIDGE)
 		var_frame.grid(row=0, column=0, sticky=tk.NSEW)
 		var_frame.rowconfigure(0, weight=1)
-		var_frame.columnconfigure(0, weight=1, minsize=200)
+		var_frame.columnconfigure(0, weight=1)
 		frame_panes.add(var_frame, weight=1)
 
 		# Maybe add a dropdown to select the row ID variable if there are multiple.
@@ -13863,21 +13936,12 @@ class NMFUnmixingDialog(Dialog):
 					# Diagnostics: Total residual, RSS, AIC, RMSE, Frobenius norm of matrix difference, K-L divergence
 					diagnostics.append([em_no, fp_display(arraysum(fit_ds)-arraysum(ds), 4), fp_display(fit_rss,5),
 						fp_display(fit_aic,4), fp_display(fit_rmse,4), fp_display(frobnorm,4), fp_display(kl_div,3)])
-				if len(rmses) > 1:
-					rmse_diffs = [rmses[i] - rmses[i+1] for i in range(len(rmses)-1)]
-					sel_em = rmse_diffs.index(max(rmse_diffs))+2
-					if len(rmse_diffs) > 2:
-						rmse_diffs2 = [rmse_diffs[i] - rmse_diffs[i+1] for i in range(len(rmse_diffs)-1)]
-						sel_em2 = rmse_diffs2.index(max(rmse_diffs2))+2
-						if sel_em != max_ems-1:
-							sel_em = sel_em2
-					else:
-						sel_em = rmse_diffs.index(max(rmse_diffs))+2
-					self.em_var.set(sel_em)
-					sel_em_index = sel_em - 1
+				if len(rmses) > 2:
+					sel_em_index = elbow(rmses, frobs) - 1
 				else:
-					self.em_var.set(1)
-					sel_em_index = 0
+					sel_em_index = length(test_ems) - 1
+				sel_em = test_ems[sel_em_index]
+				self.em_var.set(sel_em)
 				self.em_sel["values"] = test_ems
 				self.em_sel["state"] = tk.NORMAL
 				W, H = unmixes[sel_em_index]

{mapdata-3.20.0 → mapdata-3.22.0/mapdata.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: mapdata
-Version: 3.20.0
+Version: 3.22.0
 Summary: An interactive map and table explorer for geographic coordinates in a spreadsheet, CSV file, or database that includes data plotting and statistical summaries
 Home-page: https:/hg.sr.ht/~rdnielsen/mapdata
 Author: Dreas Nielsen
@@ -24,21 +24,6 @@ Classifier: Topic :: Scientific/Engineering
 Requires-Python: >=3.8
 Description-Content-Type: text/markdown
 License-File: LICENSE.txt
-Requires-Dist: tkintermapview
-Requires-Dist: pyproj
-Requires-Dist: jenkspy
-Requires-Dist: odfpy
-Requires-Dist: openpyxl
-Requires-Dist: xlrd
-Requires-Dist: matplotlib
-Requires-Dist: seaborn
-Requires-Dist: loess
-Requires-Dist: statsmodels
-Requires-Dist: scipy
-Requires-Dist: scikit-learn
-Requires-Dist: pymannkendall
-Requires-Dist: umap-learn
-Requires-Dist: pynndescent
 
 
 *mapdata.py* is a data explorer for data sets containing geographic coordinates.  Data can be read from a CSV file, spreadsheet, or database.  Both a map and a data

{mapdata-3.20.0 → mapdata-3.22.0}/mapdata.egg-info/requires.txt

@@ -1,15 +1,16 @@
-tkintermapview
-pyproj
 jenkspy
+kneeliverse
+loess
+matplotlib
 odfpy
 openpyxl
-xlrd
-matplotlib
+pymannkendall
+pynndescent
+pyproj
+scikit-learn
+scipy
 seaborn
-loess
 statsmodels
-scipy
-scikit-learn
-pymannkendall
+tkintermapview
 umap-learn
-pynndescent
+xlrd

{mapdata-3.20.0 → mapdata-3.22.0}/setup.py

@@ -9,7 +9,7 @@ symbol_files = glob.glob("mapdata/symbols/16x16/*.xbm") + glob.glob("mapdata/sym
 			glob.glob("mapdata/symbols/24x24/*.xbm") + glob.glob("mapdata/symbols/28x28/*.xbm")
 
 setuptools.setup(name='mapdata',
-	version='3.20.0',
+	version='3.22.0',
 	description="An interactive map and table explorer for geographic coordinates in a spreadsheet, CSV file, or database that includes data plotting and statistical summaries",
 	author='Dreas Nielsen',
 	author_email='cortice@tutanota.com',
@@ -18,7 +18,7 @@ setuptools.setup(name='mapdata',
 	scripts=['mapdata/mapdata.py'],
 	data_files=[('symbols', symbol_files), ('config', ['mapdata/configfile/mapdata.conf'])],
     license='GPL',
-	install_requires=['tkintermapview', 'pyproj', 'jenkspy', 'odfpy', 'openpyxl', 'xlrd', 'matplotlib', 'seaborn', 'loess', 'statsmodels', 'scipy', 'scikit-learn', 'pymannkendall', 'umap-learn', 'pynndescent'],
+	install_requires=['tkintermapview', 'pyproj', 'jenkspy', 'odfpy', 'openpyxl', 'xlrd', 'matplotlib', 'seaborn', 'loess', 'statsmodels', 'scipy', 'scikit-learn', 'pymannkendall', 'umap-learn', 'pynndescent', 'kneeliverse'],
 	python_requires = '>=3.8',
 	classifiers=[
 		'Development Status :: 5 - Production/Stable',