area coordinate in ExpressionMatrix from AssignTargets.Label is nan

[kevinyamauchi]

Description

When I make an ExpressionMatrix via AssignTargets.Label, the area coordinate has all nan values.

Steps/Code to Reproduce

# Find spots and decode
p = DetectSpots.BlobDetector(
    min_sigma=1,
    max_sigma=10,
    num_sigma=10,
    threshold=0.005,
    measurement_type='mean',
)
intensities = p.run(low_passed)
decoded = codebook.decode_per_round_max(intensities)

# Get the mask image
label_image = io.imread('perfect_segmentation_image.tif')

# Get the physical ticks
yc = im.xarray.yc.values
xc = im.xarray.xc.values
physical_ticks = {Coordinates.Y: yc, Coordinates.X:xc}

# Make the mask collection
masks = SegmentationMaskCollection.from_label_image(label_image, physical_ticks=physical_ticks)

# Assign the spots to cells
al = AssignTargets.Label()
labeled = al.run(masks, decoded)
cg = labeled.to_expression_matrix()

Expected Results

I think the area coordinate should contain the area of the corresponding mask.

Actual Results

All of the ExpressionMatrix (cg) area values are nan.

Versions

Darwin-16.7.0-x86_64-i386-64bit
Python 3.6.4 (default, Dec 21 2017, 20:33:17)
NumPy 1.16.4
SciPy 1.3.0
scikit-image 0.15.0
sklearn 0.21.2
pandas 0.25.0
xarray 0.12.3
sympy 1.4
starfish 0.1.4