Linux and UNIX Man Pages

Linux & Unix Commands - Search Man Pages

h5totxt(1) [debian man page]

H5TOTXT(1)							      h5utils								H5TOTXT(1)

NAME
h5totxt - generate comma-delimited text from 2d slices of HDF5 files SYNOPSIS
h5totxt [OPTION]... [HDF5FILE]... DESCRIPTION
h5totxt is a utility to generate comma-delimited text (and similar formats) from one-, two-, or more-dimensional slices of numeric datasets in HDF5 files. This way, the data can easily be imported into spreadsheets and similar programs for analysis and visualization. HDF5 is a free, portable binary format and supporting library developed by the National Center for Supercomputing Applications at the Uni- versity of Illinois in Urbana-Champaign. A single h5 file can contain multiple data sets; by default, h5totxt takes the first dataset, but this can be changed via the -d option, or by using the syntax HDF5FILE:DATASET. By default, the entire dataset is dumped to the output. in row-major order. For 3d datasets, this corresponds to a sequence of yz slices, in order of increasing x, separated by blank lines. If -T is specified, outputs in the transposed (column-major) order instead Often, however, you want only a one- or two-dimensional slice of multi-dimensional data. To do this, you specify coordinates in one or more slice dimensions, via the -xyzt options. The most basic usage is something like 'h5totxt foo.h5', which will output comma-delimited text to stdout from the data in foo.h5. OPTIONS
-h Display help on the command-line options and usage. -V Print the version number and copyright info for h5totxt. -v Verbose output. -o file Send text output to file rather than to stdout (the default). -s sep Use the string sep to separate columns of the output rather than a comma (the default). -x ix, -y iy, -z iz, -t it This tells h5totxt to use a particular slice of a multi-dimensional dataset. e.g. -x causes a yz plane (of a 3d dataset) to be used, at an x index of ix (where the indices run from zero to one less than the maximum index in that direction). Here, x/y/z cor- respond to the first/second/third dimensions of the HDF5 dataset. The -t option specifies a slice in the last dimension, whichever that might be. See also the -0 option to shift the origin of the x/y/z slice coordinates to the dataset center. -0 Shift the origin of the x/y/z slice coordinates to the dataset center, so that e.g. -0 -x 0 (or more compactly -0x0) returns the central x plane of the dataset instead of the edge x plane. (-t coordinates are not affected.) -T Transpose the data (interchange the dimension ordering). By default, no transposition is done. -. numdigits Output numdigits digits after the decimal point (defaults to 16). -d name Use dataset name from the input files; otherwise, the first dataset from each file is used. Alternatively, use the syntax HDF5FILE:DATASET, which allows you to specify a different dataset for each file. You can use the h5ls command (included with hdf5) to find the names of datasets within a file. BUGS
Send bug reports to S. G. Johnson, stevenj@alum.mit.edu. AUTHORS
Written by Steven G. Johnson. Copyright (c) 2005 by the Massachusetts Institute of Technology. h5utils March 9, 2002 H5TOTXT(1)

Check Out this Related Man Page

H5TOPNG(1)							      h5utils								H5TOPNG(1)

NAME
h5topng - generate PNG images from 2d slices of HDF5 files SYNOPSIS
h5topng [OPTION]... [HDF5FILE]... DESCRIPTION
h5topng is a utility to generate images in PNG (Portable Network Graphics) format from two-dimensional slices of datasets in HDF5 files. It is designed for quick-and-dirty visualization of scientific data, and for batch processing thereof via shell scripts. HDF5 is a free, portable binary format and supporting library developed by the National Center for Supercomputing Applications at the Uni- versity of Illinois in Urbana-Champaign. A single h5 file can contain multiple data sets; by default, h5topng takes the first dataset, but this can be changed via the -d option, or by using the syntax HDF5FILE:DATASET. For a three- or four-dimensional dataset you must specify coordinates in one or two slice dimensions, respectively, to get a two-dimen- sional slice, via the -xyzt options. Yet more options control things like the colormap and magnification. Still, the most basic usage is something like 'h5topng foo.h5', which will output a file foo.png containing an image from the two-dimensional data in foo.h5. OPTIONS
-h Display help on the command-line options and usage. -V Print the version number and copyright info for h5topng. -v Verbose output. This output includes the minimum and maximum values encountered in the data, which is useful to know for the -mM options. -o file Send PNG output to file rather than to the filename with .h5 replaced with .png (the default). -x ix, -y iy, -z iz, -t it This tells h5topng to use a particular slice of a multi-dimensional dataset. e.g. -x causes a yz plane (of a 3d dataset) to be used, at an x index of ix (where the indices run from zero to one less than the maximum index in that direction). Here, x/y/z cor- respond to the first/second/third dimensions of the HDF5 dataset. The -t option specifies a slice in the last dimension, whichever that might be. See also the -0 option to shift the origin of the x/y/z slice coordinates to the dataset center. Instead of specifying a single index as an argument to these options, you can also specify a range of indices in a Matlab-like nota- tion: start:step:end or start:end (step defaults to 1). This loops over that slice index, from start to end in steps of step, pro- ducing a sequence of output PNG files (with the slice index appended to the filename, before the ".png"). -0 Shift the origin of the x/y/z slice coordinates to the dataset center, so that e.g. -0 -x 0 (or more compactly -0x0) returns the central x plane of the dataset instead of the edge x plane. (-t coordinates are not affected.) -X scalex, -Y scaley, -S scale Scale the x and y dimensions of the image by scalex and scaley respectively. The -S option scales both x and y. The default is to use scale factors of 1.0; i.e. the image has the same dimensions (in pixels) as the data. Linear interpolation is used to fill in the pixels when the scale factors are not 1.0. -s skewangle Skew the image by skewangle (in degrees) to the left or right. The result is a parallelogram, with the leftover space in the (square) image filled with either black or white pixels, depending upon the color map. -T Transpose the data (interchange the image axes). By default, the first (x) coordinate of the data corresponds to the columns, and the second (y) coordinate corresponds to the rows; transposition reverses this convention. -c colormap Use a color map colormap rather than the default gray color map (a grayscale ramp from white to black). colormap is normally the name of one of the color maps provided with h5topng (in the /usr/share/h5utils/colormaps directory), or can instead be the name of a color-map file. Three useful included color maps are hot (black-red-yellow-white, useful for intensity data), bluered (blue-white-red, useful for signed data), and hsv (a multi-color "rainbow"). If you use the bluered color map for signed data, you may also want to use the -Z option so that the center of the color scale (white) corresponds to zero. A color-map file is a sequence of whitespace-separated R G B A quadruples, where each value is in the range 0.0 to 1.0 and indicates the fraction of red/green/blue/alpha. (An alpha of 0 is transparent and of 1 is opaque; this is only used for the -a option, below.) The colors in the color map are linearly interpolated as necessary to provide a continuous color ramp. -r Reverse the ordering of the color map. You can also accomplish this by putting a "-" before the colormap name in the -c or -a option. -Z Center the color scale on the value zero in the data. -m min, -M max Normally, the bottom and top of the color map correspond to the minimum and maximum values in the data. Using these options, you can make the bottom and top of the color map correspond to min and max instead. Data values below or above this range will be treated as if they were min or max respectively. See also the -Z and -R options. -R When multiple files are specified, set the bottom and top of the color maps according to the minimum and maximum over all the data. This is useful to process many files using a consistent color scale, since otherwise the scale is set for each file individually. -C file, -b val Superimpose contour outlines from the first dataset in the file HDF5 file on all of the output images. (If the contour dataset does not have the same dimensions as the output data, it is peridically "tiled" over the output.) You can use the syntax file:dataset to specify a particular dataset within the file. The contour outlines are around a value of val (defaults to middle of value range in file). -A file, -a colormap:opacity Translucently overlay the data from the first dataset in the file HDF5 file, which should have the same dimensions as the input dataset, on all of the output images, using the colormap colormap with opacity (from 0 for completely transparent to 1 for com- pletely opaque) opacity multiplied by the opacity (alpha) values in the colormap. (If the overlay dataset does not have the same dimensions as the output data, it is peridically "tiled" over the output.) You can use the syntax file:dataset to specify a partic- ular dataset within the file. Some predefined colormaps that work particularly well for this feature are yellow (transparent white to opaque yellow) gray (trans- parent white to opaque black), yarg (transparent black to opaque white), green (transparent white to opaque green), and bluered (opaque blue to transparent white to opaque red). You can prepend "-" to the colormap name to reverse the colormap order. (See also -c, above.) The default for -a is yellow:0.3 (yellow colormap multiplied by 30% opacity). -d name Use dataset name from the input files; otherwise, the first dataset from each file is used. Alternatively, use the syntax HDF5FILE:DATASET, which allows you to specify a different dataset for each file. You can use the h5ls command (included with hdf5) to find the names of datasets within a file. -8 Use 8-bit (indexed) color for the PNG output, instead of 24-bit (direct) color (the default). (This shrinks the image size slightly, with some degradation in quality.) Not supported in conjunction with the -A (translucent overlay) option. BUGS
Send bug reports to S. G. Johnson, stevenj@alum.mit.edu. AUTHORS
Written by Steven G. Johnson. Copyright (c) 2004 by the Massachusetts Institute of Technology. h5utils March 9, 2002 H5TOPNG(1)
Man Page