updated readme with infocalc info

.gitignore

@@ -1,2 +1,5 @@
 *~
 *.log
+flycheck_*.py
+\#*#
+test.pdf

README.md

@@ -1,3 +1,5 @@
 # ad-calc
 
 Tools to help calculating values for Axiomatic Design analysis
+
+`infocalc.py` calculates information content based upon a csv file or statistical parameters and upper/lower limits

infocalc.py

@@ -9,8 +9,10 @@ import logging
 import argparse
 from pathlib import PurePath##https://docs.python.org/3/library/pathlib.html#module-pathlib
 import numpy as np
+import matplotlib
 import matplotlib.pyplot as plt
-from scipy.stats import norm
+from scipy.stats import norm,t
+import scipy.stats
 import pandas as pd
 
 #Main program loop
@@ -18,20 +20,43 @@ print("""Axiomatic Design Information Calculator by Joseph. T. Foley<foley AT ru
 From https://gitea.cs.ru.is/AxiomaticDesign/adcalc/""")
 parser = argparse.ArgumentParser(
     description="Axiomatic Design Information Calculator.")
-parser.add_argument('csvfile',
+subparsers = parser.add_subparsers(dest='mode')
+subparsers.required = True
+
+### MODE DATA
+parser_data = subparsers.add_parser("DATA")
+parser_data.add_argument('csvfile',
                     help="CSV file with data and headers")
-parser.add_argument('column',
+parser_data.add_argument('column',
                     help='Which column header to take data from')
-parser.add_argument('minvalue', type=float,
+## MODE SIM
+parser_sim = subparsers.add_parser("SIM")
+parser_sim.add_argument('samplesize', type=int,
+                    help="sample size")
+parser_sim.add_argument('mean', type=float,
+                    help="mean(average) value")
+parser_sim.add_argument('stddev', type=float,
+                    help="sample standard deviation")
+## General Arguments
+parser.add_argument('--lowerbound', type=float,
                     help='Tolerance low limit')
-parser.add_argument('maxvalue', type=float,
+parser.add_argument('--upperbound', type=float,
                     help='Tolerance high limit')
 parser.add_argument('--normalizey', action="store_true",
                     help='Set y-axis to normalized probability density')
 parser.add_argument('--log', default="INFO",
                     help='Console log level:  Number or DEBUG, INFO, WARNING, ERROR')
-parser.add_argument('--graphinfo',
+parser.add_argument('--legend', action="store_true",
+                    help='Put legend on the PDF graph')
+parser.add_argument('--graphinfo', action="store_true",
                     help='Put information on the PDF graph')
+parser.add_argument('--xlabel',
+                    help='X-axis label, if needed')
+parser.add_argument('--outfile',
+                    help="output graph to PDF file")
+parser.add_argument('--fontsize', default=14, type=int,
+                    help="Adjust font size")
+
 args = parser.parse_args()
 
 ## Set up logging
@@ -42,7 +67,7 @@ if not isinstance(numeric_level, int):
 logger = logging.getLogger("app")
 logger.setLevel(numeric_level)
 # log everything to file
-logpath = os.path.splitext(args.csvfile)[0]+".log"
+logpath = "infocalc.log"
 fh = logging.FileHandler(logpath)
 fh.setLevel(logging.DEBUG)
 # log to console
@@ -57,57 +82,99 @@ fh.setFormatter(spamformatter)
 logger.addHandler(ch)
 logger.addHandler(fh)
 
-logger.info("Creating infocalc log file %s", logpath)
+logger.debug("Creating infocalc log file %s", logpath)
 
-# filename pre-processing for output
-inpath = PurePath(args.csvfile)
-print(f"Input: {inpath}")
+# seed values for variable scoping
+mean = 0
+stddev = 1
+samplesize =1
 
-# grab the data and process
-data = np.array(pd.read_csv(inpath)[args.column])
-lowerbound = args.minvalue
-upperbound = args.maxvalue
-logger.debug(f"data:{data}, lower:{lowerbound}, upper:{upperbound}")
+if args.mode == "DATA":
+    # filename pre-processing for output
+    inpath = PurePath(args.csvfile)
+    print(f"Input: {inpath}")    
+    # grab the data and process
+    data = np.array(pd.read_csv(inpath)[args.column])
+    mean = data.mean()
+    stddev = data.std(ddof=1)
+    # Delta Degrees of Freedom: ddof=0 for population, ddof=1 for sample std dev
+    samplesize = len(data)
+elif args.mode == "SIM":
+    mean = args.mean
+    stddev = args.stddev
+    samplesize = args.samplesize
+df = samplesize - 1 
     
-mean = data.mean()
-stddev = data.std(ddof=1)
-# Delta Degrees of Freedom: ddof=0 for population, ddof=1 for sample std dev
-prob = norm.cdf(upperbound, mean, stddev) - norm.cdf(lowerbound, mean, stddev)
+prob = 0
+if args.upperbound and args.lowerbound:
+    prob = t.cdf(df,args.upperbound, mean, stddev) - t.cdf(df,args.lowerbound, mean, stddev)
+elif args.upperbound:
+    prob = t.cdf(df,args.upperbound, mean, stddev)
+elif args.lowerbound:
+    prob = 1 - t.cdf(df,args.lowerbound, mean, stddev)
+else:
+    prob = 1# no bounds set!
 #print("probability: %f", prob)
 info = -np.emath.log2(prob)
 #print("information content: %f bits", info)
+## set default fontsize
+matplotlib.rcParams['font.size']=args.fontsize
 ## place text on plot:  https://matplotlib.org/3.3.4/gallery/recipes/placing_text_boxes.html
 fig, ax = plt.subplots()
-textstr = '\n'.join((
-    r'$n=%d$' % (len(data)),
+if args.graphinfo:#put info on corner of graph
+    textstr = '\n'.join((
+        r'$n=%d$' % (samplesize),
         r'$\mu=%.2f$' % (mean, ),
         r'$\sigma=%.2f$' % (stddev, ),
         r'$P=%.2f$' % (prob, ),
         r'$I=%.2f$ bits' % (info, )))
-# these are matplotlib.patch.Patch properties
-props = dict(boxstyle='round', facecolor='wheat', alpha=0.5)
+    # these are matplotlib.patch.Patch properties
+    props = dict(boxstyle='round', facecolor='wheat', alpha=0.5)
 
-# place a text box in upper left in axes coords
-ax.text(0.05, 0.95, textstr, transform=ax.transAxes, fontsize=14,
+    # place a text box in upper left in axes coords
+    ax.text(0.05, 0.95, textstr, transform=ax.transAxes, fontsize=args.fontsize,
             verticalalignment='top', bbox=props)
+xgraphlimits = {"min": mean-3*stddev, "max": mean+3*stddev}
+if args.lowerbound and xgraphlimits["min"] > args.lowerbound:
+    xgraphlimits["min"] = args.lowerbound
+if args.upperbound and xgraphlimits["max"] < args.upperbound:
+    xgraphlimits["max"] = args.upperbound
 
-x = np.linspace(mean-3*stddev, mean+3*stddev, 500)
+x = np.linspace(xgraphlimits["min"], xgraphlimits["max"], 500)
 y = norm.pdf(x, loc=mean, scale=stddev)
 if args.normalizey:
     y = y * stddev#rescale back to unity area
 plt.axvline(x=mean, color="green", linestyle="dashed", label="mean")
-plt.axvline(lowerbound, color="red")
-plt.axvline(upperbound, color="red")
+if args.lowerbound:
+    plt.axvline(args.lowerbound, color="red")
+if args.upperbound:
+    plt.axvline(args.upperbound, color="red")
 
 plt.plot(x, y, 'b-', label='Normal distribution')
 #yt = scipy.stats.t.pdf(x, len(data)-1, mean, stddev)
 #plt.plot(x, yt, 'g-', label='T Distribution')
-coloredregion = (x >= lowerbound) & ( x <= upperbound ) #select x values
-plt.fill_between(x, 0, y, where=coloredregion, color="grey", alpha=0.5, label="Design range")
-plt.xlabel('X')
-plt.ylabel('Probability density')
-plt.legend()
-plt.grid(True)
 
+# Filter for which region to fill
+coloredregion = x#default fill all
+if args.lowerbound and args.upperbound:
+    coloredregion = (x >= args.lowerbound) & ( x <= args.upperbound )
+elif args.upperbound:
+    coloredregion =  x <= args.upperbound 
+elif args.lowerbound:
+    coloredregion = x >= args.lowerbound
+    
+plt.fill_between(x, 0, y, where=coloredregion, color="grey", alpha=0.5, label="Design range",)
+    
+if args.xlabel:
+    plt.xlabel(args.xlabel)
+plt.ylabel('Probability density')
+if args.legend:
+    plt.legend()
+#plt.grid(True)
 top = plt.ylim()[1]
-plt.show()
+
+if args.outfile:
+    logger.info(f"Graph output to {args.outfile}")
+    plt.savefig(args.outfile,bbox_inches='tight')
+else:
+    plt.show()

normdist.py

@@ -1,57 +0,0 @@
-#!/usr/bin/env python
-import numpy as np
-import matplotlib.pyplot as plt
-from scipy.stats import norm
-
-## Data goes here for now --foley
-data = np.array([1, 1.1, 0.9, 1, 1, 0.9, 0.9])
-lowerbound = 0.9
-upperbound = 1.0
-
-mean = data.mean()
-stddev = data.std(ddof=1)
-# Delta Degrees of Freedom: ddof=0 for population, ddof=1 for sample std dev
-prob = norm.cdf(upperbound, mean, stddev) - norm.cdf(lowerbound, mean, stddev)
-#print("probability: %f", prob)
-info = -np.emath.log2(prob)
-#print("information content: %f bits", info)
-## place text on plot:  https://matplotlib.org/3.3.4/gallery/recipes/placing_text_boxes.html
-fig, ax = plt.subplots()
-textstr = '\n'.join((
-    r'$n=%d$' % (len(data)),
-    r'$\mu=%.2f$' % (mean, ),
-    r'$\sigma=%.2f$' % (stddev, ),
-    r'$P=%.2f$' % (prob, ),
-    r'$I=%.2f$ bits' % (info, )))
-# these are matplotlib.patch.Patch properties
-props = dict(boxstyle='round', facecolor='wheat', alpha=0.5)
-
-# place a text box in upper left in axes coords
-ax.text(0.05, 0.95, textstr, transform=ax.transAxes, fontsize=14,
-        verticalalignment='top', bbox=props)
-
-x = np.linspace(mean-3*stddev, mean+3*stddev, 500)
-y = norm.pdf(x, loc=mean, scale=stddev) * stddev#rescale back to unity area
-plt.axvline(x=mean, color="green", linestyle="dashed", label="mean")
-plt.axvline(lowerbound, color="red")
-plt.axvline(upperbound, color="red")
-
-plt.plot(x, y, 'b-', label='Normal distribution')
-#yt = scipy.stats.t.pdf(x, len(data)-1, mean, stddev)
-#plt.plot(x, yt, 'g-', label='T Distribution')
-coloredregion = (x >= lowerbound) & ( x <= upperbound ) #select x values
-plt.fill_between(x, 0, y, where=coloredregion, color="grey", alpha=0.5, label="Design range")
-plt.xlabel('X')
-plt.ylabel('Probability density')
-plt.legend()
-plt.grid(True)
-
-top = plt.ylim()[1]
-
-
-plt.show()
-# annotate values on X after drawing the graphs
-
-
-
-

pandas-test.py

@@ -1,23 +0,0 @@
-#!/usr/bin/env python
-import pandas as pd
-
-df = pd.DataFrame(
-    {
-        "Name": [
-            "Braund, Mr. Owen Harris",
-            "Allen, Mr. William Hentry",
-            "Bonnell, Miss. Elizabeth",
-            ],
-        "Age": [22, 35, 58],
-        "Sex": ["male", "male", "female"],
-    }       
-)
-
-print(df.describe())
-
-titanic = pd.read_csv("titanic.csv")
-print(titanic.head(8))
-titanic.to_excel("titanic.xlsx", sheet_name="passengers", index=False)
-titanic_xltest = pd.read_excel("titanic.xlsx", sheet_name="passengers")
-print("INFO")
-print(titanic.info())

tests.sh

@@ -0,0 +1,6 @@
+#!/bin/bash
+# Get infocalc.py from https://gitea.cs.ru.is/AxiomaticDesign/adcalc
+echo "Loading data from file"
+./infocalc.py --lowerbound 0.9 --upperbound 1.1 --graphinfo DATA testdata.csv data1
+echo "Creating simulated curve from parameters"
+./infocalc.py --lowerbound 0.9 --upperbound 1.1 SIM 8 1.0 0.5