Added some documentation.

doc/source/api_reference.rst

@@ -0,0 +1,12 @@
+API reference 
+-------------------------------
+
+.. toctree::
+   :maxdepth: 2
+
+   highlevel
+   methods
+   wavefunction
+   accumulators
+   func3d
+   testwf

doc/source/developer.md → doc/source/benchmarking.md

@@ -1,7 +1,7 @@
-Developer information 
+Benchmarking 
 -------------------------------
 
-### Performance profiling
+### Performance profiling using asv
 Benchmarks are present in the /benchmark directory in the repository. These benchmarks can be run directly, or using [asv](https://asv.readthedocs.io).
 
 

doc/source/conf.py

@@ -48,10 +48,15 @@
 # The theme to use for HTML and HTML Help pages.  See the documentation for
 # a list of builtin themes.
 #
+#extensions.append("faculty_sphinx_theme")
+#html_theme = "faculty-sphinx-theme"
+
 #html_theme = 'alabaster'
 html_theme = 'sphinx_rtd_theme'
 
 # Add any paths that contain custom static files (such as style sheets) here,
 # relative to this directory. They are copied after the builtin static files,
 # so a file named "default.css" will overwrite the builtin "default.css".
 html_static_path = ['_static']
+
+

doc/source/developer.rst

@@ -0,0 +1,8 @@
+Developer information 
+-------------------------------
+
+.. toctree::
+   :maxdepth: 2
+
+   benchmarking
+   api_reference

doc/source/index.rst

@@ -7,25 +7,14 @@ Welcome to pyQMC's documentation!
 =================================
 
 .. toctree::
-   :maxdepth: 2
+   :maxdepth: 1
    :caption: Contents:
 
    install
-
-   highlevel
-
+   tutorial
+   specific_instructions
    developer
 
-   methods
-
-   wavefunction
-
-
-   accumulators
-
-   func3d
-   testwf
-
 Indices and tables
 ==================
 

doc/source/install.rst

@@ -32,6 +32,7 @@ GitHub
     export PYTHONPATH:$PYTHONPATH:/path/to/pyqmc
 
 
+`specific_instructions`
 
 
 

doc/source/specific_instructions.md

@@ -0,0 +1,127 @@
+Specific instructions for HPC machines
+--------------------------------------------------------------
+
+
+### Illinois campus cluster
+#### Setting up your conda installation
+
+Make sure your `.bashrc` is not loading another conda environment and that `PYTHONPATH` is not set. If it is, clear those things, log out, and log back in.
+
+```
+module load anaconda/3
+conda create --name fast-mpi4py python=3.8
+module load openmpi/3.1.1-gcc-7.2.0
+export MPICC=$(which mpicc)
+.  /usr/local/anaconda/5.2.0/python3/etc/profile.d/conda.sh
+conda activate fast-mpi4py
+pip install -v --no-binary mpi4py mpi4py
+pip install -v --no-binary numpy numpy
+pip install pyscf
+module load git
+pip install git+git://github.com/WagnerGroup/pyqmc --upgrade
+pip install snakemake
+```
+
+#### Submission script
+
+```
+#!/bin/bash
+#SBATCH --time=24:00:00                  # Job run time (hh:mm:ss)
+#SBATCH --nodes=1                        # Number of nodes
+#SBATCH --ntasks-per-node=40             # Number of task (cores/ppn) per node
+#SBATCH --job-name=a120
+#SBATCH --partition=wagner
+#SBATCH --output=myjob.o%j              # Name of batch job output file
+
+module load anaconda/3
+module load openmpi/3.1.1-gcc-7.2.0
+.  /usr/local/anaconda/5.2.0/python3/etc/profile.d/conda.sh
+conda activate fast-mpi4py
+
+srun python test.py 
+```
+
+
+
+#### snakemake setup
+
+I would suggest starting with the setup here: https://github.com/WagnerGroup/Energy-Entropy
+
+```
+module load anaconda/3
+module load openmpi/3.1.1-gcc-7.2.0
+.  /usr/local/anaconda/5.2.0/python3/etc/profile.d/conda.sh
+conda activate fast-mpi4py
+pip install cookiecutter
+cookiecutter https://github.com/Snakemake-Profiles/slurm.git
+
+```
+
+The profile will be installed in `~/.config/snakemake/slurm`. I have changed my version of `slurm-submit.py` to include the partition:
+```
+RESOURCE_MAPPING = {
+    "time": ("time", "runtime", "walltime"),
+    "mem": ("mem", "mem_mb", "ram", "memory"),
+    "mem-per-cpu": ("mem-per-cpu", "mem_per_cpu", "mem_per_thread"),
+    "nodes": ("nodes", "nnodes"),
+    "partition": ("partition",)
+}
+```
+
+Then your snakemake can look like the following, if you'd like to run on `qmc_threads` cores on one node and the secondary partition. Note the resources section in the rule.
+```
+rule VMC_MF:
+    input: hf="{dir1}/{dir2}/mf.chk", wffile="{dir1}/{dir2}/vmc_mf_{tol}_{orbs}_{nconfig}.chk"
+    output: "{dir1}/{dir2}/eval_mf_{tol}_{orbs}.chk"
+    threads: qmc_threads
+    resources:
+        walltime="4:00:00", partition="secondary"
+    run:
+        with concurrent.futures.ProcessPoolExecutor(max_workers=qmc_threads) as client:
+            functions.evaluate_vmc(input.hf, None, input.wffile, output[0], 
+                  slater_kws=None, nblocks=1000, client=client, npartitions=qmc_threads)
+```
+TODO: change this to work with MPI.
+
+You can run by doing something like the following:
+```
+module load anaconda/3
+module load openmpi/3.1.1-gcc-7.2.0
+.  /usr/local/anaconda/5.2.0/python3/etc/profile.d/conda.sh
+conda activate fast-mpi4py
+hostname >> run.out
+nohup nice snakemake --jobs=10  --profile=slurm [filenames] > vmc.out & 
+```
+You can run this on a login node and it will execute your jobs.
+
+
+
+### Summit
+
+Some information is available here: https://www.olcf.ornl.gov/wp-content/uploads/2019/02/STW_Feb_20190211_summit_workshop_python.pdf
+
+Set up your environment:
+```
+module load python/3.7.0-anaconda3-5.3.0
+conda create -n pyqmc3.8 python=3.8
+conda init
+. .bashrc 
+conda install numpy pandas h5py
+module load gcc/4.8.5
+CC=gcc MPICC=mpicc pip install --no-binary mpi4py install mpi4py
+```
+
+Install pyscf and pyqmc.
+```
+module load gcc/4.8.5
+module load cmake
+module load openblas/0.3.9-omp
+
+git clone https://github.com/pyscf/pyscf
+cd build
+cmake ..
+make 
+cd [pyscf root directory]
+pip install .
+pip install pyqmc
+```

doc/source/tutorial.md

@@ -0,0 +1,58 @@
+Tutorial
+----------------
+
+Each of these snippets can be run in a separate 
+
+### Mean-field
+
+```
+from pyscf import gto, scf
+
+mol = gto.M(
+    atom="""
+    O 0.0000000, 0.000000, 0.00000000
+    H 0.761561 , 0.478993, 0.00000000
+    H -0.761561, 0.478993, 0.00000000""",
+    basis="ccecp-ccpvtz",
+    ecp={"O": "ccecp"},
+)
+mf = scf.RHF(mol)
+mf.chkfile = "mf.chk"
+mf.run()
+```
+
+
+### Optimize wave function
+
+```
+import pyqmc.recipes
+
+pyqmc.recipes.OPTIMIZE("mf.chk", "opt.chk", nconfig=1000,
+    linemin_kws={"max_iterations": 30 },
+    slater_kws={"optimize_orbitals": False},
+)
+```
+
+### VMC evaluation
+
+```
+import pyqmc.recipes
+
+pyqmc.recipes.VMC("scf.chk", "vmc.chk", 
+    start_from="opt.chk", 
+    accumulators={"energy": True, "rdm1": True},
+    vmc_kws={"nblocks": 100},
+)
+```
+
+### DMC evaluation
+
+```
+import pyqmc.recipes
+
+pyqmc.recipes.DMC("scf.chk", "dmc.chk", 
+    start_from="opt.chk", 
+    accumulators={"energy": True, "rdm1": True},
+    vmc_kws={"nblocks": 100},
+)
+```