Benchmarking via pytest (#316)

Added benchmark function and results to doc

.benchmarks/Linux-CPython-3.8-64bit/4db3fce6f8c51aa6bd15d627b6ec6693725b806d.json

@@ -0,0 +1,230 @@
+{
+    "machine_info": {
+        "node": "bhd0038",
+        "processor": "x86_64",
+        "machine": "x86_64",
+        "python_compiler": "GCC 7.3.0",
+        "python_implementation": "CPython",
+        "python_implementation_version": "3.8.8",
+        "python_version": "3.8.8",
+        "python_build": [
+            "default",
+            "Feb 24 2021 21:46:12"
+        ],
+        "release": "3.10.0-1160.31.1.el7.x86_64",
+        "system": "Linux",
+        "cpu": {
+            "python_version": "3.8.8.final.0 (64 bit)",
+            "cpuinfo_version": [
+                8,
+                0,
+                0
+            ],
+            "cpuinfo_version_string": "8.0.0",
+            "arch": "X86_64",
+            "bits": 64,
+            "count": 36,
+            "arch_string_raw": "x86_64",
+            "vendor_id_raw": "GenuineIntel",
+            "brand_raw": "Intel(R) Xeon(R) Gold 6140 CPU @ 2.30GHz",
+            "hz_advertised_friendly": "2.3000 GHz",
+            "hz_actual_friendly": "1.7326 GHz",
+            "hz_advertised": [
+                2300000000,
+                0
+            ],
+            "hz_actual": [
+                1732580000,
+                0
+            ],
+            "stepping": 4,
+            "model": 85,
+            "family": 6,
+            "flags": [
+                "3dnowprefetch",
+                "abm",
+                "acpi",
+                "adx",
+                "aes",
+                "aperfmperf",
+                "apic",
+                "arat",
+                "arch_perfmon",
+                "art",
+                "avx",
+                "avx2",
+                "avx512bw",
+                "avx512cd",
+                "avx512dq",
+                "avx512f",
+                "avx512vl",
+                "bmi1",
+                "bmi2",
+                "bts",
+                "cat_l3",
+                "cdp_l3",
+                "clflush",
+                "clflushopt",
+                "clwb",
+                "cmov",
+                "constant_tsc",
+                "cqm",
+                "cqm_llc",
+                "cqm_mbm_local",
+                "cqm_mbm_total",
+                "cqm_occup_llc",
+                "cx16",
+                "cx8",
+                "dca",
+                "de",
+                "ds_cpl",
+                "dtes64",
+                "dtherm",
+                "dts",
+                "eagerfpu",
+                "epb",
+                "ept",
+                "erms",
+                "est",
+                "f16c",
+                "flexpriority",
+                "flush_l1d",
+                "fma",
+                "fpu",
+                "fsgsbase",
+                "fxsr",
+                "hle",
+                "ht",
+                "ibpb",
+                "ibrs",
+                "ida",
+                "intel_ppin",
+                "intel_pt",
+                "intel_stibp",
+                "invpcid",
+                "invpcid_single",
+                "lahf_lm",
+                "lm",
+                "mba",
+                "mca",
+                "mce",
+                "md_clear",
+                "mmx",
+                "monitor",
+                "movbe",
+                "mpx",
+                "msr",
+                "mtrr",
+                "nonstop_tsc",
+                "nopl",
+                "nx",
+                "ospke",
+                "pae",
+                "pat",
+                "pbe",
+                "pcid",
+                "pclmulqdq",
+                "pdcm",
+                "pdpe1gb",
+                "pebs",
+                "pge",
+                "pku",
+                "pln",
+                "pni",
+                "popcnt",
+                "pse",
+                "pse36",
+                "pts",
+                "rdrand",
+                "rdseed",
+                "rdt_a",
+                "rdtscp",
+                "rep_good",
+                "rtm",
+                "sdbg",
+                "sep",
+                "smap",
+                "smep",
+                "smx",
+                "spec_ctrl",
+                "ss",
+                "ssbd",
+                "sse",
+                "sse2",
+                "sse4_1",
+                "sse4_2",
+                "ssse3",
+                "stibp",
+                "syscall",
+                "tm",
+                "tm2",
+                "tpr_shadow",
+                "tsc",
+                "tsc_adjust",
+                "tsc_deadline_timer",
+                "vme",
+                "vmx",
+                "vnmi",
+                "vpid",
+                "x2apic",
+                "xgetbv1",
+                "xsave",
+                "xsavec",
+                "xsaveopt",
+                "xtopology",
+                "xtpr"
+            ],
+            "l3_cache_size": 25952256,
+            "l2_cache_size": 1048576,
+            "l1_data_cache_size": 32768,
+            "l1_instruction_cache_size": 32768
+        }
+    },
+    "commit_info": {
+        "id": "4db3fce6f8c51aa6bd15d627b6ec6693725b806d",
+        "time": "2021-07-09T15:19:26-04:00",
+        "author_time": "2021-07-09T15:19:26-04:00",
+        "dirty": true,
+        "project": "hoomd-tf",
+        "branch": "issue-279"
+    },
+    "benchmarks": [
+        {
+            "group": null,
+            "name": "test_lj_benchmark",
+            "fullname": "htf/test-py/benchmark.py::test_lj_benchmark",
+            "params": null,
+            "param": null,
+            "extra_info": {},
+            "options": {
+                "disable_gc": false,
+                "timer": "perf_counter",
+                "min_rounds": 5,
+                "max_time": 1.0,
+                "min_time": 5e-06,
+                "warmup": false
+            },
+            "stats": {
+                "min": 1.9830403740052134,
+                "max": 3.0145940768998116,
+                "mean": 2.2169361331500115,
+                "stddev": 0.448255204004423,
+                "rounds": 5,
+                "median": 2.0034955348819494,
+                "iqr": 0.33898555068299174,
+                "q1": 1.986543036007788,
+                "q3": 2.3255285866907798,
+                "iqr_outliers": 1,
+                "stddev_outliers": 1,
+                "outliers": "1;1",
+                "ld15iqr": 1.9830403740052134,
+                "hd15iqr": 3.0145940768998116,
+                "ops": 0.45107298539047896,
+                "total": 11.084680665750057,
+                "iterations": 1
+            }
+        }
+    ],
+    "datetime": "2021-07-10T02:21:23.047649",
+    "version": "3.4.1"
+}

.github/workflows/test.yml

@@ -44,7 +44,7 @@ jobs:
     - name: Install dependencies
       run: |
         # Install testing dependencies
-        pip install pytest-cov pytest-xdist jupyter notebook matplotlib gsd scipy
+        pip install pytest-cov pytest-xdist jupyter notebook matplotlib gsd scipy pytest-benchmark
 
         # Install software dependencies
         pip install tensorflow==${{ matrix.tensorflow }}
@@ -61,7 +61,7 @@ jobs:
         cd ..
     - name: Run tests
       run: |
-        pytest -v --forked
+        pytest
     - name: Run notebooks with MDAnalysis
       if: matrix.md-analysis == 'true'
       run: |

Release.md

@@ -3,4 +3,5 @@ Release Process
 
 1. Update changelog
 2. Make sure changelog is ISO dates
-3. Modify `version.py` file
+3. Modify `version.py` file
+4. Update benchmarks

htf/test-py/benchmark.py

@@ -0,0 +1,48 @@
+def test_lj_benchmark(benchmark):
+
+    import hoomd.htf as htf
+    from hoomd.htf import tfcompute
+    import hoomd
+    import hoomd.md
+    import hoomd.dump
+    import hoomd.group
+    import numpy as np
+    import tensorflow as tf
+
+    class LJModel(htf.SimModel):
+        def compute(self, nlist, positions, box):
+            # get r
+            rinv = htf.nlist_rinv(nlist)
+            inv_r6 = rinv**6
+            # pairwise energy. Double count -> divide by 2
+            p_energy = 4.0 / 2.0 * (inv_r6 * inv_r6 - inv_r6)
+            # sum over pairwise energy
+            energy = tf.reduce_sum(input_tensor=p_energy, axis=1)
+            forces = htf.compute_nlist_forces(nlist, energy)
+
+            return forces
+
+    N = 256
+    NN = 64
+    model = LJModel(NN)
+    tfcompute = htf.tfcompute(model)
+    hoomd.context.initialize('')
+    rcut = 3.0
+    sqrt_N = int(np.sqrt(N))
+
+    system = hoomd.init.create_lattice(unitcell=hoomd.lattice.sq(a=2.0),
+                                       n=[sqrt_N, sqrt_N])
+    nlist = hoomd.md.nlist.cell(check_period=1)
+    # basic LJ forces from Hoomd
+    lj = hoomd.md.pair.lj(rcut, nlist)
+    lj.pair_coeff.set('A', 'A', epsilon=1.0, sigma=1.0)
+    hoomd.md.integrate.mode_standard(dt=0.005)
+    hoomd.md.integrate.langevin(group=hoomd.group.all(), kT=1.0, seed=42)
+    # equilibrate for 4k steps first
+    hoomd.run(4000)
+    # now attach the LJ force model
+    tfcompute.attach(nlist,
+                     r_cut=rcut)
+
+    # train on 1k timesteps
+    benchmark.pedantic(hoomd.run, (1000,), rounds=5, iterations=1)

pytest-benchmark.ini

@@ -0,0 +1,6 @@
+# pytest-benchmark.ini
+[pytest]
+minversion = 6.0
+addopts = -v --durations=0 --benchmark-autosave
+python_files = htf/test-py/benchmark.py
+required_plugins = pytest-cov pytest-xdist pytest-benchmark

pytest.ini

@@ -0,0 +1,4 @@
+# pytest.ini
+[pytest]
+minversion = 6.0
+addopts = -v --durations=0 --forked