GSAS-II is a powerful software for analyzing the crystallographic structure from diffraction data. It is written in Python, with some underlying routines in Fortran and C++. Conveniently, it provides the scriptable interface so that we can perform the refinement in a programming way. This is pretty handy when some batch processing is in need.
The scriptable way of running GSAS-II refinement is basically about writing a Python script, which involves setting up the fitting recipe and conduct the refinement. Detailed documentation can be found in Ref. [1]. To run the scriptable refinement, we need to first install GSAS-II on the machine. Typical installation instructions can be found in Ref. [2]. For Windows and MacOS, there shouldn’t be that much issue in the installation. For Linux, due to the wild variations of Linux flavors (Ubuntu, Centos, OpenSUSE, Arch, Manjaro, Debian, Fedora, you name it…), it is very difficult to come up with a uniform installation solution suitable for all platforms. Though, it is possible to build the codes from the source which is probably the most generic solution. Detailed instructions can be found in Ref. [3].
Once GSAS-II is successfully installed, we want to find out where the codes are installed. For example, on my Linux machine, GSAS-II was installed following Ref. [3] and the installation location is /home/y8z/Dev/gsasii/GSAS-II. Inside the directory, I could see the following file tree,
.
├── backcompat
├── docs
├── GSASII
├── GSASII-bin
├── LICENSE
├── meson.build
├── noxfile.py
├── pixi
├── pyproject.toml
├── README.md
├── sources
└── tests
Now, we can create a conda environment (mamba can be used as well, to replace all conda instances in the following commands) and install some modules (not all of them will be needed but it is safe to install them all),
conda create -n gsasii_dev
conda activate gsasii_dev
conda install python numpy matplotlib wxpython pyopengl scipy git gitpython PyCifRW pillow conda requests hdf5 h5py imageio zarr xmltodict pybaselines seekpath pywin32 -c conda-forge -y
Next, we prepare the Python script for running the GSAS-II refinement, and with the conda environment active from previous step, we should be able to run the script. In the script, we need to provide the full path to where the GSAS-II scriptable file is located – see the example below.
N.B. In my case, my path inserting line would be
sys.path.insert(0, '/home/y8z/Dev/gsasii/GSAS-II/GSASII').
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 | import os import sys import numpy as np sys.path.insert(0, '/full/path/to/GSAS-II/GSASII') import GSASIIscriptable as G2sc # noqa: E402 # +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ # Input parameters # +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ gpx_loc = "/full/path/to/where/gsasii/project/file/will/be/saved/" structure_fn = "/full/path/to/structure.cif" gsa_fn = "/full/path/to/gsa/data/file" prm_fn = "/full/path/to/instrument/parameter/file" output_stem_fn = "output_stem" stype = "N" bank = 5 xmin = 300 xmax = 16667 num_banks = 4 xmin_all = [300, 1500, 2000, 3500] xmax_all = [9000, 10000, 12000, 16667] # +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ def run_gsas2_fit( structure_fn, gsa_fn, prm_fn, output_stem_fn, stype, bank, xmin, xmax): ''' Parameters ---------- structure_fn: str input structure cif filename. gsa_fn: str input gsa filename. instprm_fn: str input instrument profile filename. output_stem_fn: str output stem filename. banks: str bank 1-6. Returns ------- gsas2_poj : str gsas2 .gpx project file ''' def HistStats(gpx): '''prints profile rfactors for all histograms''' print(u"*** profile Rwp, " + os.path.split(gpx.filename)[1]) for hist in gpx.histograms(): print("\t{:20s}: {:.2f}".format(hist.name, hist.get_wR())) print("") print("INFO: Build GSAS-II Project File.") print("******************************") # start GSAS-II refinement # create a project file init_gpx = os.path.join( gpx_loc, output_stem_fn + "_initial.gpx" ) if os.path.exists(init_gpx): os.remove(init_gpx) gpx = G2sc.G2Project(newgpx=init_gpx) # add a bank histogram to the project hists = [] if stype == "N": hist1 = gpx.add_powder_histogram( gsa_fn, prm_fn, fmthint="GSAS powder", databank=bank, instbank=bank ) hist1.set_refinements({'Limits': [xmin, xmax]}) if stype == "X": hist1 = gpx.add_powder_histogram( gsa_fn, prm_fn, fmthint="GSAS powder" ) hist1.set_refinements({'Limits': [xmin, xmax]}) hists.append(hist1) # step 2: add a phase and link it to the previous histograms _ = gpx.add_phase( structure_fn, phasename='structure', fmthint='CIF', histograms=hists ) cell_i = gpx.phase('structure').get_cell() # step 3: increase # of cycles to improve convergence gpx.data['Controls']['data']['max cyc'] = 5 # step 4: start refinement # refinement step 1: turn on Histogram Scale factor refdict1 = { "set": { "Sample Parameters": ["Scale"] }, "call": HistStats, } # refinement step 2: turn on background refinement (Hist) refdict2 = { "set": { "Background": { "type": "chebyschev", "no. coeffs": 6, "refine": True } }, "call": HistStats, } # refinement step 3: refine lattice parameter and Uiso refinement (Phase) refdict3 = { "set": { "Atoms": {"all": "U"}, "Cell": True }, "call": HistStats, } dictList = [refdict1, refdict2, refdict3] # before fit, save project file first. Then in the future, # the refined project file will update this one. ref_gpx = os.path.join( gpx_loc, output_stem_fn + "_refined.gpx" ) gpx.save(ref_gpx) gpx.do_refinements(dictList) print("================") # save results data rw = gpx.histogram(0).get_wR() * 0.01 x = np.array(gpx.histogram(0).getdata('X')) y = np.array(gpx.histogram(0).getdata('Yobs')) ycalc = np.array(gpx.histogram(0).getdata('Ycalc')) dy = np.array(gpx.histogram(0).getdata('Residual')) bkg = np.array(gpx.histogram(0).getdata('Background')) output_cif_fn = os.path.join( gpx_loc, output_stem_fn + "_refined.cif") gpx.phase('structure').export_CIF(output_cif_fn) cell_r = gpx.phase('structure').get_cell() return rw, x, y, ycalc, dy, bkg, cell_i, cell_r def run_gsas2_fit_all( structure_fn, gsa_fn, prm_fn, output_stem_fn, stype, num_banks, xmin_all, xmax_all): ''' Parameters ---------- structure_fn: str input structure cif filename. gsa_fn: str input gsa filename. instprm_fn: str input instrument profile filename. output_stem_fn: str output stem filename. banks: str bank 1-6. Returns ------- gsas2_poj : str gsas2 .gpx project file ''' def HistStats(gpx): '''prints profile rfactors for all histograms''' print(u"*** profile Rwp, " + os.path.split(gpx.filename)[1]) for hist in gpx.histograms(): print("\t{:20s}: {:.2f}".format(hist.name, hist.get_wR())) print("") print("INFO: Build GSAS-II Project File.") print("******************************") # start GSAS-II refinement # create a project file init_gpx = os.path.join( gpx_loc, output_stem_fn + "_initial.gpx" ) if os.path.exists(init_gpx): os.remove(init_gpx) gpx = G2sc.G2Project(newgpx=init_gpx) hists = [] if stype == "N": for bank in range(num_banks): hist_tmp = gpx.add_powder_histogram( gsa_fn, prm_fn, fmthint="GSAS powder", databank=bank + 1, instbank=bank + 1 ) hist_tmp.set_refinements({'Limits': [xmin_all[bank + 1], xmax_all[bank + 1]]}) hists.append(hist_tmp) # step 2: add a phase and link it to the previous histograms _ = gpx.add_phase( structure_fn, phasename='structure', fmthint='CIF', histograms=hists ) cell_i = gpx.phase('structure').get_cell() # step 3: increase # of cycles to improve convergence gpx.data['Controls']['data']['max cyc'] = 5 # step 4: start refinement # refinement step 1: turn on Histogram Scale factor refdict1 = { "set": { "Sample Parameters": ["Scale"] }, "call": HistStats, "histograms": hists } # refinement step 2: turn on background refinement (Hist) refdict2 = { "set": { "Background": { "type": "chebyschev", "no. coeffs": 6, "refine": True } }, "call": HistStats, "histograms": hists } # refinement step 3: refine lattice parameter and Uiso refinement (Phase) refdict3 = { "set": { "Atoms": { "all": "U" }, "Cell": True }, "call": HistStats, "histograms": hists } dictList = [refdict1, refdict2, refdict3] # before fit, save project file first. Then in the future, # the refined project file will update this one. ref_gpx = os.path.join( gpx_loc, output_stem_fn + "_refined.gpx" ) gpx.save(ref_gpx) gpx.do_refinements(dictList) print("================") # save results data rw = list() x = list() y = list() ycalc = list() dy = list() bkg = list() for bank in range(num_banks): rw.append(gpx.histogram(bank).get_wR() * 0.01) x.append(np.array(gpx.histogram(bank).getdata('X'))) y.append(np.array(gpx.histogram(bank).getdata('Yobs'))) ycalc.append(np.array(gpx.histogram(bank).getdata('Ycalc'))) dy.append(np.array(gpx.histogram(bank).getdata('Residual'))) bkg.append(np.array(gpx.histogram(bank).getdata('Background'))) output_cif_fn = os.path.join( gpx_loc, output_stem_fn + "_refined.cif" ) gpx.phase('structure').export_CIF(output_cif_fn) cell_r = gpx.phase('structure').get_cell() return rw, x, y, ycalc, dy, bkg, cell_i, cell_r if __name__ == "__main__": run_gsas2_fit( structure_fn, gsa_fn, prm_fn, output_stem_fn, stype, bank, xmin, xmax ) run_gsas2_fit_all( structure_fn, gsa_fn, prm_fn, output_stem_fn, stype, num_banks, xmin_all, xmax_all ) |
References
[1] https://gsas-ii.readthedocs.io/en/latest/GSASIIscriptable.html
[2] https://advancedphotonsource.github.io/GSAS-II-tutorials/install.html