perf: Use 4x faster decimal conversion method across all indicators

benchmark_sample.json

@@ -0,0 +1,261 @@
+{
+ "machine_info": {
+ "node": "pkrvmccyg1gnepe",
+ "processor": "x86_64",
+ "machine": "x86_64",
+ "python_compiler": "GCC 13.3.0",
+ "python_implementation": "CPython",
+ "python_implementation_version": "3.12.3",
+ "python_version": "3.12.3",
+ "python_build": [
+ "main",
+ "Aug 14 2025 17:47:21"
+ ],
+ "release": "6.11.0-1018-azure",
+ "system": "Linux",
+ "cpu": {
+ "python_version": "3.12.3.final.0 (64 bit)",
+ "cpuinfo_version": [
+ 9,
+ 0,
+ 0
+ ],
+ "cpuinfo_version_string": "9.0.0",
+ "arch": "X86_64",
+ "bits": 64,
+ "count": 4,
+ "arch_string_raw": "x86_64",
+ "vendor_id_raw": "AuthenticAMD",
+ "brand_raw": "AMD EPYC 7763 64-Core Processor",
+ "hz_advertised_friendly": "3.2409 GHz",
+ "hz_actual_friendly": "3.2409 GHz",
+ "hz_advertised": [
+ 3240884000,
+ 0
+ ],
+ "hz_actual": [
+ 3240884000,
+ 0
+ ],
+ "stepping": 1,
+ "model": 1,
+ "family": 25,
+ "flags": [
+ "3dnowext",
+ "3dnowprefetch",
+ "abm",
+ "adx",
+ "aes",
+ "aperfmperf",
+ "apic",
+ "arat",
+ "avx",
+ "avx2",
+ "bmi1",
+ "bmi2",
+ "clflush",
+ "clflushopt",
+ "clwb",
+ "clzero",
+ "cmov",
+ "cmp_legacy",
+ "constant_tsc",
+ "cpuid",
+ "cr8_legacy",
+ "cx16",
+ "cx8",
+ "de",
+ "decodeassists",
+ "erms",
+ "extd_apicid",
+ "f16c",
+ "flushbyasid",
+ "fma",
+ "fpu",
+ "fsgsbase",
+ "fsrm",
+ "fxsr",
+ "fxsr_opt",
+ "ht",
+ "hypervisor",
+ "invpcid",
+ "lahf_lm",
+ "lm",
+ "mca",
+ "mce",
+ "misalignsse",
+ "mmx",
+ "mmxext",
+ "movbe",
+ "msr",
+ "mtrr",
+ "nonstop_tsc",
+ "nopl",
+ "npt",
+ "nrip_save",
+ "nx",
+ "osvw",
+ "osxsave",
+ "pae",
+ "pat",
+ "pausefilter",
+ "pcid",
+ "pclmulqdq",
+ "pdpe1gb",
+ "pfthreshold",
+ "pge",
+ "pni",
+ "popcnt",
+ "pse",
+ "pse36",
+ "rdpid",
+ "rdpru",
+ "rdrand",
+ "rdrnd",
+ "rdseed",
+ "rdtscp",
+ "rep_good",
+ "sep",
+ "sha",
+ "sha_ni",
+ "smap",
+ "smep",
+ "sse",
+ "sse2",
+ "sse4_1",
+ "sse4_2",
+ "sse4a",
+ "ssse3",
+ "svm",
+ "syscall",
+ "topoext",
+ "tsc",
+ "tsc_known_freq",
+ "tsc_reliable",
+ "tsc_scale",
+ "umip",
+ "user_shstk",
+ "v_vmsave_vmload",
+ "vaes",
+ "vmcb_clean",
+ "vme",
+ "vmmcall",
+ "vpclmulqdq",
+ "xgetbv1",
+ "xsave",
+ "xsavec",
+ "xsaveerptr",
+ "xsaveopt",
+ "xsaves"
+ ],
+ "l3_cache_size": 524288,
+ "l2_cache_size": 1048576,
+ "l1_data_cache_size": 65536,
+ "l1_instruction_cache_size": 65536,
+ "l2_cache_line_size": 512,
+ "l2_cache_associativity": 6
+ }
+ },
+ "commit_info": {
+ "id": "afd8571f32ddd9c4046f35d44be5d0f2f1c9463a",
+ "time": "2025-08-22T23:55:10Z",
+ "author_time": "2025-08-22T23:55:10Z",
+ "dirty": true,
+ "project": "stock-indicators-python",
+ "branch": "copilot/fix-392"
+ },
+ "benchmarks": [
+ {
+ "group": null,
+ "name": "test_benchmark_rsi",
+ "fullname": "benchmarks/test_benchmark_indicators.py::TestPerformance::test_benchmark_rsi",
+ "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": 0.004117928999960441,
+ "max": 0.013795678999997563,
+ "mean": 0.0047654412641540975,
+ "stddev": 0.0016516754227122268,
+ "rounds": 53,
+ "median": 0.00432412300000351,
+ "iqr": 0.0003096714999770711,
+ "q1": 0.004292769750009029,
+ "q3": 0.0046024412499861,
+ "iqr_outliers": 5,
+ "stddev_outliers": 2,
+ "outliers": "2;5",
+ "ld15iqr": 0.004117928999960441,
+ "hd15iqr": 0.0050827120000462855,
+ "ops": 209.84415599076905,
+ "total": 0.25256838700016715,
+ "data": [
+ 0.004580091999969227,
+ 0.004422116999990067,
+ 0.013795678999997563,
+ 0.004366281999978128,
+ 0.004225759999997081,
+ 0.004235747999985051,
+ 0.004223445000036463,
+ 0.004206474000000071,
+ 0.011844792999966103,
+ 0.0045093900000097165,
+ 0.004313873999990392,
+ 0.004351004000000103,
+ 0.004283387000043604,
+ 0.0042905910000285985,
+ 0.004296802999988358,
+ 0.004322300000012547,
+ 0.004293496000002506,
+ 0.004338329999995949,
+ 0.004311410000013893,
+ 0.004333369999983461,
+ 0.004312050000010004,
+ 0.004320806999999149,
+ 0.0043358959999864055,
+ 0.0043086640000069565,
+ 0.004302082000037899,
+ 0.00432412300000351,
+ 0.004307621999998901,
+ 0.004326467999987926,
+ 0.004293646000007811,
+ 0.004287755999996534,
+ 0.0042568480000113595,
+ 0.005241398000009667,
+ 0.00427911999997832,
+ 0.004302371999983734,
+ 0.006015285000046333,
+ 0.004374427000016112,
+ 0.004283697999994729,
+ 0.0050827120000462855,
+ 0.004117928999960441,
+ 0.0041739819999975225,
+ 0.004212424999991526,
+ 0.004307622999988325,
+ 0.004365762000020368,
+ 0.004572388000042338,
+ 0.004558742000028815,
+ 0.0046694890000367195,
+ 0.004747805000022254,
+ 0.004778744000020652,
+ 0.004727838000007978,
+ 0.004700225999954455,
+ 0.0047361030000274695,
+ 0.004718600999979117,
+ 0.004681410999978652
+ ],
+ "iterations": 1
+ }
+ }
+ ],
+ "datetime": "2025-08-29T00:19:11.069407+00:00",
+ "version": "5.1.0"
+}

benchmarks/test_decimal_conversion_performance.py

@@ -0,0 +1,70 @@
+"""Benchmarks comparing performance of different decimal conversion methods."""
+
+import pytest
+from stock_indicators._cstypes import Decimal as CsDecimal
+from stock_indicators._cstypes.decimal import to_pydecimal, to_pydecimal_via_double
+
+
+@pytest.mark.performance
+class TestDecimalConversionPerformance:
+ """Benchmark performance of different decimal conversion methods."""
+
+ def test_benchmark_string_conversion(self, benchmark, raw_data):
+ """Benchmark the current string-based conversion method."""
+ from stock_indicators._cstypes.decimal import to_pydecimal
+
+ raw_data = raw_data * 100 # Use subset for faster testing
+
+ # Pre-convert to CsDecimal to isolate the conversion performance
+ cs_decimals = [CsDecimal(row[2]) for row in raw_data]
+
+ def convert_via_string(cs_decimals):
+ for cs_decimal in cs_decimals:
+ to_pydecimal(cs_decimal)
+
+ benchmark(convert_via_string, cs_decimals)
+
+ def test_benchmark_double_conversion(self, benchmark, raw_data):
+ """Benchmark the new double-based conversion method."""
+ from stock_indicators._cstypes.decimal import to_pydecimal_via_double
+
+ raw_data = raw_data * 100 # Use subset for faster testing
+
+ # Pre-convert to CsDecimal to isolate the conversion performance
+ cs_decimals = [CsDecimal(row[2]) for row in raw_data]
+
+ def convert_via_double(cs_decimals):
+ for cs_decimal in cs_decimals:
+ to_pydecimal_via_double(cs_decimal)
+
+ benchmark(convert_via_double, cs_decimals)
+
+ def test_benchmark_small_dataset_string_conversion(self, benchmark):
+ """Benchmark string conversion with a controlled small dataset."""
+ test_values = [
+ 1996.1012, 123.456789, 0.123456789, 999999.999999,
+ 0.000001, 1000000.0, 1.8e-05, 1.234e10
+ ] * 1000 # Repeat to get meaningful measurements
+
+ cs_decimals = [CsDecimal(val) for val in test_values]
+
+ def convert_via_string(cs_decimals):
+ for cs_decimal in cs_decimals:
+ to_pydecimal(cs_decimal)
+
+ benchmark(convert_via_string, cs_decimals)
+
+ def test_benchmark_small_dataset_double_conversion(self, benchmark):
+ """Benchmark double conversion with a controlled small dataset."""
+ test_values = [
+ 1996.1012, 123.456789, 0.123456789, 999999.999999,
+ 0.000001, 1000000.0, 1.8e-05, 1.234e10
+ ] * 1000 # Repeat to get meaningful measurements
+
+ cs_decimals = [CsDecimal(val) for val in test_values]
+
+ def convert_via_double(cs_decimals):
+ for cs_decimal in cs_decimals:
+ to_pydecimal_via_double(cs_decimal)
+
+ benchmark(convert_via_double, cs_decimals)