Updated readme.md, annotation function

Author: Anastomose

Students/ebuer/termp/createplot.py

@@ -1,4 +1,11 @@
+"""createplot.py
+   file creates class object then uses makeSampdict and 
+   plotSampdict to generate the visualization. 
 
+   This file can be further refined in the future to initialized with a 
+   datafile create a full dictionary of samples and then plot each
+   sample/method pair in sequence. 
+   """
 
 import sys
 sys.path.append(r"D:\GitHub\IntroToPython\Students\ebuer\termp")
@@ -13,7 +20,8 @@
 import matplotlib.pyplot as plt
 
 
-# create the instance from the data file
+# create the class instance from the data file
+# script could be enhaced with sys.argv argument for initializing object
 
 
 def callclass():
@@ -22,7 +30,6 @@ def callclass():
 
 # use the class object and methods to select and prepare data for plotting
 # this could easily be turned into an fx w/dict arg for efficient looping
-# with either sys.argv or an additional script to feed values
 
 
 def makeSampdict(**kwargs):
@@ -64,20 +71,28 @@ def plotSampdict(**kwargs):
     ax1.set_ylim(plot_dict['y_lims'])
     ax1.set_ylabel('Concentration in {u_units}'.format(**kwargs), rotation='vertical')
 
-    # create annotation
+    # create title and annotation
     anno_text = 'Results for Sample: {u_sample}  Analytical Method: {u_method}'.format(**kwargs)
-    # ax1.annotate(anno_text, xy=(kwargs['x_ticks'][0]-0.5, kwargs['y_lims'][1]))
     ax1.set_title(anno_text)
 
+    anno_list = zip(kwargs['flag'], kwargs['x_ticks'], kwargs['y_value'])
+    for a in anno_list:
+        l_text = ''
+        if type(a[0]) is not float:
+            l_text = a[0]
+        x = a[1] + 0.15
+        y = a[2] + 1
+        ax1.annotate(l_text, xy = (x, y))
+
 
     # in principle visualization is expensive, so we save it for the end
     d = {'pad': 2, 'w_pad': 1, 'h_pad': 1}  # keyword dict for padding
     fig.set_tight_layout(d)
 
     fig.canvas.draw()
-    # plt.savefig('test_figure.pdf')
-    fig.show() # after many trial runs and tests screen vis not needed
+    plt.savefig('test_figure.pdf')
+    # fig.show() # after many trial runs screen vis not needed
     return (fig, ax1)
 
-# call function with dict
+# call function with dict generated by makeSampledict
 fig1, ax1 = plotSampdict(**plot_dict)

Students/ebuer/termp/readme.md

@@ -1,18 +1,52 @@
-###Term Project Readme  
-Basic project construct: use newfound python knowledge combined with pandas and matplotlib modules to create a simple set of data visualization tools for environmental chemistry and monitoring data.
-
-**Desired Key Elements**  
-* Read in environmental data from format delivered
-* Visualize detected results on a single image
-* Apply estimate flag to qualified results
-* Add axis and constituent labeling as appropriate
-* Plot multiple sets of data on shared axis graphs or a single image for monitoring/environmental data
-
-####Part 1: Read in data
-Environmental chemistry data is delivered in csv or xlsx format.  After validation it is almost always delivered as an xlsx file.  Use pandas module along with other techniques to create a data set for plotting.
-
-Several points to work out:  
-* multiple sample names per sheet -- need unique list of samples
-* create list (or tuple?) of parameters analyzed in each sample
-* take advantage of index feature to pull out flags
-* make dict of plotting parameters to pass to chart
+##Term Project Readme  
+Basic project construct: use newfound python knowledge combined with pandas and matplotlib modules to create a simple data visualization tool for environmental chemistry data.
+
+###plotprep Class Attributes  
+All attributes at this time are mutable to allow for tinkering.
+
+_raw_data_  
+Raw import of the data used in class initialization
+
+_selected_data_  
+Columns selected from raw_data that are processed as part of the plotprep. Original raw_data is left untouched in the event it needs to be examined later.
+
+_samples_  
+A list of unique sample identifiers pulled from selected_data.
+
+_qc_samples_  
+A llist of unique qc sample identifiers pulled from selected_data
+
+_sampleData_  
+Analytical results selected based on the sample id handed to the selectSample method.
+
+_plotData_  
+Analytical results for the sample identified using selectSample and the method provided to selectForPlot. Results do not include spiked sample recoveries.
+
+_plot_dict_  
+Plotting dictionary created by makePlotobj method, requires that both plotData already have been selected and assigned as an attribute of the class instance
+
+
+####Class Methods  
+_showOptions_  
+Automatically called by other methods to display a unique list of options for the user such as sample identifiers or methods used on a specific sample.
+
+_selectSample_  
+Returns sampleData attribute to the class instance.  sampleData are the columns of selected_data reduced down to only the selected sample.
+
+Syntax: selectSample(sample_id)  
+
+Note: type(sample_id) must match what is in the raw_data.
+
+_selectForPlot_  
+Returns plotData attribute to the class instance.  plotData is the set of analytical results from the selected_data reduced down by sample in _selectSample_ and by the method entered.
+
+Syntax: selectForPlot(analytical_method)
+
+_makePlotobj_  
+Returns plotting dictionary attribute to the class instance.  Method takes no arguments, but requires that selectSample and selectForPlot have already been run to be fully populated
+
+###Createplot.py  
+Script initializes plotprep object and creates plotting dictionary for sample-method pair passed in as test_samp dictonary using _makeSampdict_.  
+
+_plotSampdict(**kwargs)_  
+Takes plotting dictionary created in the plotprep instance of _makeSampdict_ and renders a bar chart of results with labeling and laboratory flags applied to the top of each column.  Returns figure and subplot objects for further tinkering.

Students/ebuer/termp/termp_funcs.py

@@ -56,7 +56,7 @@ def __init__(self, dfile):
 
         # samples and qc samples each in their own class attribute
         self.samples = temp_array[mask]
-        self.qc_samples = temp_array[i_mask]
+        self.qc_samples = temp_array[i_mask] # might be useful later
 
         # create a unique list of sample names and print list of samples
         s_dict = {'name': 'samples', 'o_list': self.samples.unique()}
@@ -113,7 +113,7 @@ def makePlotobj(self):
 
         # get values set up for plotting
         x_label = [x for x in self.plotData['chemical_name']]
-        y_value = [float(y.replace(',', '')) 
+        y_value = [float(y.replace(',', ''))
                    for y in self.plotData['result_value']
                    ]
         flag =    [f for f in self.plotData['lab_flag']]
@@ -128,16 +128,11 @@ def makePlotobj(self):
                           'y_lims': (0, max(y_value)+5),
                           'flag': flag,
                           'u_sample': self.plotData['sample_name'].unique()[0],
-                          'u_method': self.plotData['std_anl_method_name'].unique()[0],
+                          'u_method':
+                           self.plotData['std_anl_method_name'].unique()[0],
+
                           'u_units': self.plotData['result_unit'].unique()[0]
                           # add dict entry for results units, call in labeling
                           }
 
         return self.plot_dict
-
-
-
-
-
-
-

Students/ebuer/termp/test_figure.pdf

@@ -72,3 +72,17 @@ def test_makePlotobj():
     assert sample_dict['y_value'][0] is not None
     assert type(sample_dict['y_value'][0]) is float
     assert sample_dict['u_units'] == 'ug/kg'
+    
+
+def test_annoStr():
+    # created to get at flag data types for logic in termp_funcs
+    p_test = callclass()
+    p_test.selectSample('14051305')
+    p_test.selectForPlot('SW8082A')
+    # pdb.set_trace()
+    sample_dict = p_test.makePlotobj()
+
+    f = sample_dict['flag']
+    for fl in f:
+        print type(fl)
+    assert True