def printRange(row1, row2, col1, col2, HERCMATRIX):
if row1 < 0:
raise IndexError("row1 may not be less than zero")
if row1 > HERCMATRIX.height:
raise IndexError("row1 is out of bounds")
if row2 < 0:
raise IndexError("row2 may not be less than zero")
if row2 > HERCMATRIX.height:
raise IndexError("row2 is out of bounds")
if col1 < 0:
raise IndexError("col1 may not be less than zero")
if col1 > HERCMATRIX.width:
raise IndexError("col1 is out of bounds")
if col2 < 0:
raise IndexError("col2 may not be less than zero")
if col2 > HERCMATRIX.width:
raise IndexError("col2 is out of bounds")
if row1 > row2:
raise ValueError("row1 larger than row2")
if col1 > col2:
raise ValueError("col1 larger than col2")
TMPMATRIX = libHercMatrix.hercMatrix()
TMPMATRIX.height = abs(row2 - row1 + 1)
TMPMATRIX.width = abs(col2 - col1 + 1)
width = HERCMATRIX.width
height = HERCMATRIX.height
for element in HERCMATRIX.elements:
row = element[0]
col = element[1]
val = element[2]
if (row >= row1) and (row <= row2):
if (col >= col1) and (col <= col2):
try:
TMPMATRIX.setValue(row - row1,
col - col1,
val)
except IndexError as e:
logging.warning("encountered error {0}".format(e))
logging.warning("writing to row {0} col {1} of "
.format(row - row1, col - col1))
logging.warning("row {0} col {1}"
.format(TMPMATRIX.height, TMPMATRIX.width))
TMPMATRIX.makeRowMajor()
displayMatrix(TMPMATRIX)
def read(path):
# hercMatrix instance we will return later
MATRIX = libHercMatrix.hercMatrix()
# CSR matrix contents
row_ptr = numpy.array([]) # row pointer
col_idx = numpy.array([]) # column index
val = numpy.array([]) # values
# read the file
FILE = open(path, 'r')
lines = FILE.readlines()
FILE.close()
# read in the header, split it, and save the contents
header = lines[0].split()
height = int(header[0])
width = int(header[0])
nzentries = int(header[1])
lines.pop(0) # get rid of the header
for line in lines:
if len(line.split()) == 2:
# if the length is 2, we are in the column index + val section
val = numpy.append(val, float(line.split()[0]))
# col_idx is 1-indexed in valcol
col_idx = numpy.append(col_idx, int(line.split()[1]) - 1)
elif len(line.split()) == 1:
# if line length is 1, we are in the row pointer section
row_ptr = numpy.append(row_ptr, int(line))
else:
print("WARNING: malformed line for valcol file: {0}".format(line))
# generate a scipy.sparse.csr_matrix instance of
# libHercMatrix.hercMatrix.replaceContents()
CSRMATRIX = scipy.sparse.csr_matrix((val, col_idx, row_ptr),
shape=(height, width))
# move data into the hercMatrix instance
MATRIX.nzentries = nzentries
MATRIX.height = height
MATRIX.width = width
MATRIX.replaceContents(CSRMATRIX)
# check if the matrix is symmetric
if MATRIX.checkLowerTriangle():
print("Lower triangle is empty, assuming symmetric matrix...")
MATRIX.symmetry = "SYM"
return MATRIX
# utility for exploring and editing the contents of hercm matrix files
import readline
import libHercMatrix
import libHercmIO
import matplotlib
import matplotlib.pyplot
import logging
import os
import pprint
from yapsy.PluginManager import PluginManager
import traceback
import sys
WORKINGMATRIX = libHercMatrix.hercMatrix()
pp = pprint.PrettyPrinter()
pluginManager = PluginManager()
menuItems = []
currentTraceBack = None
def loadPlugins():
global pluginManager
global menuItems
pluginManager.setPluginPlaces(["menuPlugins"])
pluginManager.collectPlugins()
for plugin in pluginManager.getAllPlugins():
menuItems.append(plugin.plugin_object)
def readMatrix(filename, form, showProgress=False):
HERCMATRIX = libHercMatrix.hercMatrix()
logging.info("reading matrix {0} in format {1}".format(filename, form))
if (form == 'hercm') or (form == 'bxf'):
# TODO: exception handling
HERCMATRIX = libBXF.read(filename)
elif form == 'mtx':
from scipy import io
from scipy.sparse import csr_matrix
from numpy import array
# reads in an MTX file and converts it to hercm
try:
if showProgress:
print("reading data from file...")
rawMatrix = scipy.sparse.coo_matrix(scipy.io.mmread(filename))
if 'symmetric' in io.mminfo(filename):
HERCMATRIX.symmetry = "SYM"
else:
HERCMATRIX.symmetry = "ASYM"
hercm = {} # needed to generate verification
if showProgress:
print("generating header data..")
hercm['val'] = rawMatrix.data
hercm['col'] = rawMatrix.col.tolist()
hercm['row'] = rawMatrix.row.tolist()
(matrixWidth, matrixHeight) = rawMatrix.shape
HERCMATRIX.height = int(matrixHeight)
HERCMATRIX.width = int(matrixWidth)
vs = libBXF.generateVerificationSum(hercm)
HERCMATRIX.verification = vs
HERCMATRIX.remarks = []
# I'm not sure why has to be hard...
# http://stackoverflow.com/questions/26018781/numpy-is-it-possible-to-preserve-the-dtype-of-columns-when-using-column-stack
if showProgress:
print("preparing matrix data...")
val = numpy.asarray(hercm['val'], dtype='float64')
col = numpy.asarray(hercm['col'], dtype='int32')
row = numpy.asarray(hercm['row'], dtype='int32')
val = numpy.rec.array(val, dtype=[(('val'), numpy.float64)])
col = numpy.rec.array(col, dtype=[(('col'), numpy.int32)])
row = numpy.rec.array(row, dtype=[(('row'), numpy.int32)])
HERCMATRIX.elements = append_fields(row,
'col',
col,
usemask=False,
dtypes=[numpy.int32])
HERCMATRIX.elements = append_fields(HERCMATRIX.elements,
'val',
val,
usemask=False,
dtypes=[numpy.float64])
HERCMATRIX.nzentries = len(HERCMATRIX.elements['val'])
HERCMATRIX.verification = libBXF.generateVerificationSum(
HERCMATRIX)
if showProgress:
print("finished reading matrix")
except IOError as e: # make sure the file exists and is readable
logging.warning("(lsc-480) could not open matrix file")
raise IOError("could not open matrix file for writing...",
str(e))
elif form == 'mat': # matlab matrices
from scipy import io
from scipy import sparse
from numpy import array
try:
rawMatrix = scipy.sparse.coo_matrix(
scipy.io.loadmat(filename)['matrix'])
hercm = {} # needed to generate verification
hercm['val'] = rawMatrix.data
hercm['col'] = rawMatrix.col.tolist()
hercm['row'] = rawMatrix.row.tolist()
(matrixWidth, matrixHeight) = rawMatrix.shape
HERCMATRIX.height = int(matrixHeight)
HERCMATRIX.width = int(matrixWidth)
vs = libBXF.generateVerificationSum(hercm)
HERCMATRIX.verification = vs
HERCMATRIX.remarks = []
# I'm not sure why has to be hard...
# http://stackoverflow.com/questions/26018781/numpy-is-it-possible-to-preserve-the-dtype-of-columns-when-using-column-stack
val = numpy.asarray(hercm['val'], dtype='float64')
col = numpy.asarray(hercm['col'], dtype='int32')
row = numpy.asarray(hercm['row'], dtype='int32')
val = numpy.rec.array(val, dtype=[(('val'), numpy.float64)])
col = numpy.rec.array(col, dtype=[(('col'), numpy.int32)])
row = numpy.rec.array(row, dtype=[(('row'), numpy.int32)])
HERCMATRIX.elements = append_fields(row,
'col',
col,
usemask=False,
dtypes=[numpy.int32])
HERCMATRIX.elements = append_fields(HERCMATRIX.elements,
'val', val, usemask=False, dtypes=[numpy.float64])
HERCMATRIX.nzentries = len(HERCMATRIX.elements['val'])
if HERCMATRIX.checkSymmetry():
HERCMATRIX.symmetry = 'SYM'
HERCMATRIX.verification = libBXF.generateVerificationSum(
HERCMATRIX)
except IOError as e: # make sure the file exists and is readable
logging.warning("(lsc-536)could not open matrix file")
raise IOError("could not open matrix file for writing...",
str(e))
elif form == 'valcol':
HERCMATRIX = libValcolIO.read(filename)
else:
logging.warning("(lsc-545) format {0} is not valid".format(form))
if showProgress:
print("converting matrix to row-major...")
logging.info("converting matrix to row-major")
HERCMATRIX.makeRowMajor()
if showProgress:
print("matrix is now row major")
if HERCMATRIX.symmetry == 'SYM':
logging.info("matrix is symmetric, truncating upper triangle")
if showProgress:
print("matrix is symmetric, truncating upper triangle...")
HERCMATRIX.makeSymmetrical('truncate')
if showProgress:
print("upper triangle truncated")
return HERCMATRIX
def read(filename):
# reads in the HeRCM file specified by filename
# returns it as an instance of libhsm.hsm
# matrix object we will return later
HERCMATRIX = libHercMatrix.hercMatrix()
# row, col, and val lists we will read the matrix data into later
row = []
col = []
val = []
logging.info("Reading BXF file {0}".format(filename))
fileObject = open(filename, 'r')
# this may raise OSError, which the caller should catch
# Get a list of lines from the file. Might be a good point for future
# optimization - lines could be processed one at a time
lines = fileObject.readlines()
fileObject.close()
# read in the header
header = lines[0]
lines.pop(0)
splitHeader = header.split()
logging.info("read BXF header: " + header)
# stuff we are going to read in from the header
version = splitHeader[0]
width = None
height = None
nzentries = None
symmetry = None
# version specific header parsing logic
# deprecated HERCM (BXF 1.0) and BXF 2.0
if version == "HERCM" or version == "BXF":
if len(splitHeader) > 6:
logging.warning("possibly mangled header - too many " +
"fields for BXF 1.0 header. Attempting to read anyway...")
elif len(splitHeader) < 6:
logging.warning("possibly mangled header - too few fields for " +
"BXF 1.0 header. Attempting to read anyway...")
elif len(splitHeader) < 5:
raise ValueError("Header has too few fields for any known BXF " +
"version - unable to read header")
# read fields from the header, see doc-extra/bxf-spec.md for more
# details
width = int(splitHeader[1])
height = int(splitHeader[2])
nzentries = int(splitHeader[3])
symmetry = splitHeader[4].upper()
elif (version == "BXF21") or (version == "BXF22"): # bxf 2.1 and 2.2 use
# the same header format
if len(splitHeader) != 5:
raise ValueError("Header has incorrect number of fields for BXF " +
" 2.1")
width = int(splitHeader[1])
height = int(splitHeader[2])
nzentries = int(splitHeader[3])
symmetry = splitHeader[4].upper()
else:
raise ValueError("Header did not contain valid BXF version " +
"identifier")
# verify symmetry
if symmetry not in ["SYM", "ASYM"]:
logging.warning("Symmetry {0} is not valid, assuming asymmetric"
.format(symmetry))
symmetry = "ASYM"
logging.info("finished reading header")
HERCMATRIX.width = width
HERCMATRIX.height = height
HERCMATRIX.nzentries = nzentries
HERCMATRIX.symmetry = symmetry
inField = False
currentHeader = ''
fieldname = ''
vtype = ''
currentContents = []
for line in lines:
# we are starting a new field
if not inField:
currentHeader = line.rstrip()
splitHeader = currentHeader.split()
if (version == "BXF21") or (version == "BXF22"):
fieldname = splitHeader[0]
vtype = splitHeader[1]
else:
# if you review previous BXF specifications, field headers
# had three fields, the middle of which was either `LIST` or
# `SINGLE`. These can both be safely treated as lists, as BXF2.1
# does
#
fieldname = splitHeader[0]
vtype = splitHeader[2]
inField = True
# this is the end of a field
elif 'ENDFIELD' in line:
# save the values we read from this field to lists for later use
if fieldname.lower() == "val":
val = currentContents
elif fieldname.lower() == "row":
row = currentContents
elif fieldname.lower() == "col":
col = currentContents
elif fieldname.lower() == "remarks":
pass
else:
logging.warning("Ignoring field with unrecognized name: "
+ fieldname)
# discard the contents of this field
currentContents = []
inField = False
# we are currently reading data from a field
else:
for value in line.split():
# typecast this element according to the vtype
# these may throw TypeError , which the caller should handle
if vtype == 'INT':
currentContents.append(int(value))
elif vtype == 'FLOAT':
currentContents.append(float(value))
else:
currentContents.append(value)
# do some basic validation
if (len(row) != len(col)) or \
(len(row) != len(val)) or \
(len(val) != len(col)):
raise ValueError("one or more vectors have non-matching lengths" +
", not a valid COO matrix")
elif (len(val) != nzentries):
# maybe this should throw an exception?
logging.warning("nzentries does not match number of nonzero entries " +
"read from file - matrix may be mangled")
else:
logging.info("matrix seems sane, it is probably not corrupt")
# copy matrix data into the matrix object
for i in range(0, HERCMATRIX.nzentries):
# this could probably be optimized by generating a scipy.sparse
# matrix then using hercMatrix.replaceContents()
if (version == "HERCM") or (version == "BXF") or (version == "BXF21"):
if HERCMATRIX.symmetry == "SYM":
# perform an inline transpose
HERCMATRIX.addElement([col[i], row[i], val[i]])
else:
HERCMATRIX.addElement([row[i], col[i], val[i]])
else:
HERCMATRIX.addElement([row[i], col[i], val[i]])
HERCMATRIX.removeZeros()
HERCMATRIX.makeRowMajor()
return HERCMATRIX
