Adds radiography requirements 1. and 2.a

radiography/__init__.py

@@ -0,0 +1 @@
+from .operations import fits_to_variable

radiography/operations.py

@@ -0,0 +1,46 @@
+import glob
+import os
+import numpy as np
+import scipp as sc
+from astropy.io import fits
+
+
+def _load_fits(fit_dir):
+    if not os.path.isdir(fit_dir):
+        raise RuntimeError(fit_dir + " is not directory")
+    stack = []
+    path_length = len(fit_dir) + 1
+    filenames = sorted(glob.glob(fit_dir + "/*.fits"))
+    nfiles = len(filenames)
+    count = 0
+    print(f"Loading {nfiles} files from '{fit_dir}'")
+    for filename in filenames:
+        count += 1
+        print('\r{0}: Image {1}, of {2}'.format(filename[path_length:], count, nfiles), end="")
+        with fits.open(os.path.join(fit_dir, filename)) as hdu:
+            data = hdu[0].data
+            stack.append(np.flipud(data))
+
+    if len(stack) == 1:
+        # Fold away a dim
+        stack = stack[0]
+
+    return np.array(stack, dtype=np.float64)
+
+
+def fits_to_variable(fits_dir, average_data=False):
+    """
+    Loads all fits images from the directory into a scipp Variable.
+    """
+    stack = _load_fits(fits_dir)
+    if average_data:
+        stack = stack.mean(axis=0)
+
+    if len(stack.shape) == 3:
+        return sc.Variable(["slice", "x", "y"], values=stack, variances=stack)
+
+    elif len(stack.shape) == 2:
+        return sc.Variable(["x", "y"], values=stack, variances=stack)
+    else:
+        raise IndexError("Expected 2 or 3 dimensions,"
+                         f" found {len(stack.shape)}")

scipp-radiography.ipynb

@@ -0,0 +1,175 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "metadata": {
+    "collapsed": true,
+    "pycharm": {
+     "name": "#%% md\n"
+    }
+   },
+   "source": [
+    "# How to start\n",
+    "\n",
+    "Before starting you must:\n",
+    "- Ensure that `scipp` and `mantid` are on your `PYTHONPATH`.\n",
+    "- Generate the `config.py` file using `make_config.py`. Refer to the `README.md` or `python make_config.py --help` for information.\n",
+    "\n",
+    "For `scipp` and `mantid` follow instructions at: https://scipp.readthedocs.io/en/latest/getting-started/installation.html."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "outputs": [],
+   "source": [
+    "# Some PYTHONPATH sanity checks\n",
+    "try:\n",
+    "    import scipp\n",
+    "    print(\"scipp found\")\n",
+    "except ImportError as e:\n",
+    "    print(\"scipp is not available in the PYTHONPATH\")\n",
+    "    raise e\n",
+    "\n",
+    "try:\n",
+    "    import mantid\n",
+    "    print(\"mantid found\")\n",
+    "except ImportError as e:\n",
+    "    print(\"mantid is not available in the PYTHONPATH\")\n",
+    "    raise e\n",
+    "\n",
+    "try:\n",
+    "    import scippconfig\n",
+    "    print(\"config found\")\n",
+    "except ImportError as e:\n",
+    "    print(\"config is not available. Make sure you have generated it with `make_config.py`.\")\n",
+    "    raise e"
+   ],
+   "metadata": {
+    "collapsed": false,
+    "pycharm": {
+     "name": "#%%\n"
+    }
+   }
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "outputs": [],
+   "source": [
+    "import os\n",
+    "import radiography\n",
+    "import scipp as sc\n",
+    "import numpy as np"
+   ],
+   "metadata": {
+    "collapsed": false,
+    "pycharm": {
+     "name": "#%%\n"
+    }
+   }
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "outputs": [],
+   "source": [
+    "experiment_dir = scippconfig.script_root\n",
+    "\n",
+    "data_dir = os.path.join(experiment_dir, 'radiography', 'IMAT_sample')\n",
+    "\n",
+    "# Since we don't have reference data pretend this folder contains our ref\n",
+    "dark_field_dir = os.path.join(data_dir, \"angle0\")\n",
+    "flat_field_dir = os.path.join(data_dir, \"Flats\", \"RB1730044\", \"Tomo_test\", \"OpenBeam_aft1\")\n",
+    "sample_dir = os.path.join(data_dir, \"angle1\")"
+   ],
+   "metadata": {
+    "collapsed": false,
+    "pycharm": {
+     "name": "#%%\n"
+    }
+   }
+  },
+  {
+   "cell_type": "code",
+   "source": [
+    "raw_data = sc.Dataset()\n",
+    "raw_data[\"sample\"] = radiography.fits_to_variable(sample_dir)\n",
+    "\n",
+    "# Set coords based on sample dims, as reference needs to match\n",
+    "raw_data.coords[\"projection\"] = sc.Variable([\"projection\"], unit=sc.units.dimensionless, values=np.arange(0, raw_data[\"sample\"].shape[0]))\n",
+    "raw_data.coords[\"x\"] = sc.Variable([\"x\"], unit=sc.units.dimensionless, values=np.arange(0, raw_data[\"sample\"].shape[1]))\n",
+    "raw_data.coords[\"y\"] = sc.Variable([\"y\"], unit=sc.units.dimensionless, values=np.arange(0, raw_data[\"sample\"].shape[2]))"
+   ],
+   "metadata": {
+    "collapsed": false,
+    "pycharm": {
+     "name": "#%%\n"
+    }
+   },
+   "execution_count": null,
+   "outputs": []
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "outputs": [],
+   "source": [
+    "dark_field = radiography.fits_to_variable(dark_field_dir, average_data=True)\n",
+    "flat_field = radiography.fits_to_variable(flat_field_dir, average_data=True)\n",
+    "\n",
+    "# This blows through 32GB of memory as raw_data[\"sample\"] is already ~17 GB\n",
+    "raw_data[\"transmission\"] = (raw_data[\"sample\"] - dark_field) / (flat_field - dark_field)"
+   ],
+   "metadata": {
+    "collapsed": false,
+    "pycharm": {
+     "name": "#%%\n",
+     "is_executing": true
+    }
+   }
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "outputs": [],
+   "source": [
+    "# TODO WIP below\n",
+    "roi_x = (100, 200)\n",
+    "roi_y = (100, 200)\n",
+    "\n",
+    "raw_data[\"roi_sample\"] = raw_data[\"sample\"]\n",
+    "\n",
+    "raw_data.masks[\"sample\"] = sc.Variable(['x', 'y'], shape=(raw_data.coords[\"x\"].shape[0], raw_data.coords[\"y\"].shape[0]),\n",
+    "                                       dtype=sc.dtype.bool)"
+   ],
+   "metadata": {
+    "collapsed": false,
+    "pycharm": {
+     "name": "#%%\n"
+    }
+   }
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 2
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython2",
+   "version": "2.7.6"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 0
+}