DAMASK_EICMD/python/damask/table.py

196 lines
5.8 KiB
Python
Raw Normal View History

2019-10-31 15:15:34 +05:30
import re
import pandas as pd
import numpy as np
class Table():
"""Store spreadsheet-like data."""
2019-12-05 09:30:26 +05:30
def __init__(self,data,shapes,comments=None):
"""
2019-12-05 09:30:26 +05:30
New spreadsheet.
Parameters
----------
2019-12-05 09:30:26 +05:30
data : numpy.ndarray
Data.
2019-12-05 09:30:26 +05:30
shapes : dict with str:tuple pairs
Shapes of the columns. Example 'F':(3,3) for a deformation gradient.
comments : iterable of str, optional
2019-12-05 09:30:26 +05:30
Additional, human-readable information.
"""
2019-12-05 09:30:26 +05:30
self.data = pd.DataFrame(data=data)
2019-12-05 09:30:26 +05:30
labels = {}
i = 0
2019-12-05 09:30:26 +05:30
for label in shapes.keys():
for components in range(np.prod(shapes[label])):
labels[i] = label
i+=1
if i != self.data.shape[1]:
2019-12-05 09:30:26 +05:30
raise IndexError('Shape mismatch between shapes and data')
2019-12-05 09:30:26 +05:30
self.data.rename(columns=labels,inplace=True)
if comments is None:
self.comments = []
else:
self.comments = [c for c in comments]
2019-12-05 09:30:26 +05:30
self.shapes = shapes
@staticmethod
def from_ASCII(fname):
"""
Create table from ASCII file.
The first line needs to indicate the number of subsequent header lines as 'n header'.
2019-12-05 09:30:26 +05:30
Vector data column labels are indicated by '1_v, 2_v, ..., n_v'.
Tensor data column labels are indicated by '3x3:1_T, 3x3:2_T, ..., 3x3:9_T'.
Parameters
----------
fname : file, str, or pathlib.Path
Filename or file for reading.
"""
try:
f = open(fname)
except TypeError:
f = fname
header,keyword = f.readline().split()
if keyword == 'header':
header = int(header)
else:
raise Exception
comments = [f.readline()[:-1] for i in range(header-1)]
labels = f.readline().split()
2019-12-05 09:30:26 +05:30
shapes = {}
for label in labels:
tensor_column = re.search(r'[0-9,x]*?:[0-9]*?_',label)
if tensor_column:
my_shape = tensor_column.group().split(':',1)[0].split('x')
2019-12-05 09:30:26 +05:30
shapes[label.split('_',1)[1]] = tuple([int(d) for d in my_shape])
2019-10-31 15:15:34 +05:30
else:
vector_column = re.match(r'[0-9]*?_',label)
if vector_column:
2019-12-05 09:30:26 +05:30
shapes[label.split('_',1)[1]] = (int(label.split('_',1)[0]),)
2019-10-31 15:15:34 +05:30
else:
2019-12-05 09:30:26 +05:30
shapes[label]=(1,)
2019-12-05 10:15:27 +05:30
data = pd.read_csv(f,names=[i for i in range(len(labels))],sep='\s+').to_numpy()
return Table(data,shapes,comments)
2019-10-31 15:15:34 +05:30
2019-12-05 10:40:27 +05:30
def labels(self):
"""Return the labels of all columns."""
return list(self.shapes.keys())
def get(self,label):
"""
2019-12-05 10:40:27 +05:30
Get column data.
Parameters
----------
label : str
2019-12-05 10:40:27 +05:30
Column label.
"""
if re.match(r'[0-9]*?_',label):
idx,key = label.split('_',1)
return self.data[key].to_numpy()[:,int(idx)-1]
else:
2019-12-05 09:30:26 +05:30
return self.data[label].to_numpy().reshape((-1,)+self.shapes[label])
2019-10-31 15:15:34 +05:30
2019-12-05 10:40:27 +05:30
def set(self,label,data,info=None):
"""
2019-12-05 10:40:27 +05:30
Set column data.
Parameters
----------
label : str
2019-12-05 10:40:27 +05:30
Column label.
data : np.ndarray
New data.
2019-12-05 10:40:27 +05:30
info : str, optional
Human-readable information about the new data.
"""
2019-12-05 10:40:27 +05:30
if info is not None:
if np.prod(data.shape[1:],dtype=int) == 1:
self.comments.append('{}: {}'.format(label,info))
else:
self.comments.append('{} {}: {}'.format(label,data.shape[1:],info))
if re.match(r'[0-9]*?_',label):
idx,key = label.split('_',1)
iloc = self.data.columns.get_loc(key).tolist().index(True) + int(idx) -1
2019-12-05 10:40:27 +05:30
self.data.iloc[:,iloc] = data
else:
2019-12-05 10:40:27 +05:30
self.data[label] = data.reshape(self.data[label].shape)
2019-10-31 15:15:34 +05:30
2019-12-05 10:40:27 +05:30
def add(self,label,data,info=None):
"""
2019-12-05 10:40:27 +05:30
Add column data.
Parameters
----------
label : str
2019-12-05 10:40:27 +05:30
Column label.
data : np.ndarray
Modified data.
info : str, optional
Human-readable information about the modified data.
"""
2019-12-05 10:40:27 +05:30
if info is not None:
if np.prod(data.shape[1:],dtype=int) == 1:
self.comments.append('{}: {}'.format(label,info))
else:
self.comments.append('{} {}: {}'.format(label,data.shape[1:],info))
2019-10-31 15:15:34 +05:30
2019-12-05 10:40:27 +05:30
self.shapes[label] = data.shape[1:] if len(data.shape) > 1 else (1,)
size = np.prod(data.shape[1:],dtype=int)
new_data = pd.DataFrame(data=data.reshape(-1,size),
2019-10-31 15:15:34 +05:30
columns=[label for l in range(size)])
self.data = pd.concat([self.data,new_data],axis=1)
2019-12-05 10:40:27 +05:30
def to_ASCII(self,fname):
"""
Store as plain text file.
Parameters
----------
fname : file, str, or pathlib.Path
Filename or file for reading.
"""
2019-10-31 15:15:34 +05:30
labels = []
2019-12-05 09:30:26 +05:30
for l in self.shapes:
if(self.shapes[l] == (1,)):
2019-10-31 15:15:34 +05:30
labels.append('{}'.format(l))
2019-12-05 09:30:26 +05:30
elif(len(self.shapes[l]) == 1):
2019-10-31 15:15:34 +05:30
labels+=['{}_{}'.format(i+1,l)\
2019-12-05 09:30:26 +05:30
for i in range(self.shapes[l][0])]
2019-10-31 15:15:34 +05:30
else:
2019-12-05 09:30:26 +05:30
labels+=['{}:{}_{}'.format('x'.join([str(d) for d in self.shapes[l]]),i+1,l)\
for i in range(np.prod(self.shapes[l],dtype=int))]
2019-10-31 15:15:34 +05:30
header = ['{} header'.format(len(self.comments)+1)]\
+ self.comments\
+ [' '.join(labels)]
try:
f = open(fname,'w')
except TypeError:
f = fname
for line in header: f.write(line+'\n')
self.data.to_csv(f,sep=' ',index=False,header=False)