effspm 0.1.5__cp313-cp313-win_amd64.whl → 0.1.7__cp313-cp313-win_amd64.whl

effspm/__init__.py

@@ -1,4 +1,3 @@
-from ._core import mine
+from ._core import PrefixProjection
 
-
-__all__ = ['mine']
+__all__ = ['PrefixProjection']

effspm/_core.cpp

@@ -1,3 +1,7 @@
+#include <ctime>       // std::clock
+#include <cmath>       // std::ceil, std::abs
+#include <algorithm>   // std::max
+#include <iostream>    // optional echo
 #include <pybind11/pybind11.h>
 #include <pybind11/stl.h>
 
@@ -5,24 +9,98 @@
 #include "freq_miner.hpp"
 #include "utility.hpp"
 
-
 namespace py = pybind11;
 
 PYBIND11_MODULE(_core, m) {
     m.doc() = "Efficient Sequential Pattern Mining via Prefix-Projection";
 
-    m.def("mine", [](std::string data_file, double minsup) {
-        ClearCollected();
+    m.def("PrefixProjection",
+        [](py::object data,
+           double minsup,
+           unsigned int time_limit_arg,
+           bool preproc_arg,
+           bool use_dic_arg,
+           bool verbose_arg,
+           const std::string &out_file_arg)
+        {
+            // 1) configure C++ globals
+            time_limit = time_limit_arg;
+            pre_pro    = preproc_arg;
+            use_dic    = use_dic_arg;
+            use_list   = false;
+            b_disp     = verbose_arg;
+            b_write    = !out_file_arg.empty();
+            out_file   = out_file_arg;
+
+            // 2) clear collector & start timer
+            ClearCollected();
+            start_time = std::clock();
+
+            // 3) load either file or in‐memory sequences
+            if (py::isinstance<py::str>(data)) {
+                auto path = data.cast<std::string>();
+                if (!Load_instance(path, minsup))
+                    throw std::runtime_error("Failed to load database from " + path);
+            }
+            else {
+                // convert Python List[List[int]] → C++ items
+                auto seqs = data.cast<std::vector<std::vector<int>>>();
+                items     = std::move(seqs);
+                N         = items.size();
+
+                // a) compute max item ID → L
+                int max_id = 0;
+                for (auto &seq : items)
+                    for (int x : seq)
+                        max_id = std::max(max_id, std::abs(x));
+                L = static_cast<unsigned int>(max_id);
+
+                // b) support threshold θ
+                if (minsup < 1.0)
+                    theta = static_cast<unsigned long long>(std::ceil(minsup * N));
+                else
+                    theta = static_cast<unsigned long long>(minsup);
+
+                // c) initialize DFS stack
+                DFS.clear();
+                DFS.reserve(L);
+                for (unsigned int i = 0; i < L; ++i)
+                    DFS.emplace_back(-static_cast<int>(i) - 1);
+
+                // d) gather dataset stats: max length M, total entries E
+                M = 0;
+                E = 0;
+                for (auto &seq : items) {
+                    M = std::max<unsigned int>(M, static_cast<unsigned int>(seq.size()));
+                    E += seq.size();
+                }
+
+                if (b_disp) {
+                    std::cout << "\nIn-memory dataset: "
+                              << N << " sequences, max len " << M
+                              << ", " << E << " entries, " << L << " items\n";
+                }
+            }
+
+            // 4) run the C++ miner
+            Freq_miner();
 
-        // Note: Pass by value (safe copy)
-        if (!Load_instance(data_file, minsup)) {
-            throw std::runtime_error("Failed to load database from " + data_file);
-        }
+            // 5) collect patterns & timing
+            auto patterns   = GetCollected();
+            double wall_time = give_time(std::clock() - start_time);
 
-        Freq_miner();
-        return GetCollected();
-    },
-    py::arg("data_file"),
-    py::arg("minsup") = 0.01,
-    "Mine sequential patterns from the given data file with minimum support.");
+            // 6) return Python dict
+            py::dict out;
+            out["patterns"] = patterns;
+            out["time"]     = wall_time;
+            return out;
+        },
+        py::arg("data"),
+        py::arg("minsup")      = 0.01,
+        py::arg("time_limit")  = 10 * 3600,
+        py::arg("preproc")     = false,
+        py::arg("use_dic")     = false,
+        py::arg("verbose")     = false,
+        py::arg("out_file")    = ""
+    );
 }

effspm/freq_miner.hpp

@@ -1,7 +1,9 @@
 #pragma once
-
+#include <vector>
 #include "load_inst.hpp"
-
+#include <cstdlib> 
+#include <cmath> 
+using namespace std;
 void Freq_miner();
 void Out_patt(std::vector<int>& seq, unsigned int freq);
 

effspm/load_inst.hpp

@@ -1,25 +1,30 @@
+// effspm/load_inst.hpp
 #pragma once
 
-#include<vector>
-#include<string>
+#include <vector>
+#include <string>
 #include <fstream>
 #include <map>
- // Should work because "effspm" is in include_dirs
-
+#include <ctime>    // for clock_t
 
 using namespace std;
 
+// ------------------------------------------------------------
+// forward declare Pattern (defined in freq_miner.hpp)
+struct Pattern;
+
+// Main entrypoint: load your file on disk into 'items', build DFS, theta, etc.
 bool Load_instance(string &items_file, double thresh);
 
+// storage & globals shared between the C++-CLI & Python bindings
 extern vector<vector<int>> items;
+extern vector<Pattern>     DFS;         // now Pattern is known
+extern vector<int>         item_dic;
 
-extern string out_file;
-
-extern bool b_disp, b_write, use_dic, use_list, pre_pro;
-
-extern unsigned int M, L, time_limit;
-
-extern unsigned long long int N, theta;
+extern string              out_file;
+extern bool                b_disp, b_write, use_dic, use_list, pre_pro;
 
-extern clock_t start_time;
+extern unsigned int        M, L, time_limit;
+extern unsigned long long  N, E, theta; // E = total number of entries
 
+extern clock_t             start_time;

{effspm-0.1.5.dist-info → effspm-0.1.7.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: effspm
-Version: 0.1.5
+Version: 0.1.7
 Summary: Prefix‑Projection sequential pattern mining
 Home-page: https://github.com/yeshu999/effspm
 Author: yeshu999

effspm-0.1.7.dist-info/RECORD

@@ -0,0 +1,14 @@
+effspm/__init__.py,sha256=9edygUUVqPU4ubTsSGcch5NST8moZaYYvOgXPWv5L-w,71
+effspm/_core.cp313-win_amd64.pyd,sha256=b6x8TewWC_-9aXUJCcGqmrLnmS4eQlwRkUGIOcCqzF8,179712
+effspm/_core.cpp,sha256=JzUCIVmmDMVfRIjIsmXtVkSMkRMprE3wpOrx-jYhAQU,3781
+effspm/freq_miner.cpp,sha256=qQDFPoPKY3ICaH2brm1CUKwNBhCy0-0dUWEoV_3FwME,4785
+effspm/freq_miner.hpp,sha256=0Phi-H6Wto9mVRqmR_APFbmLUBDdZOEX8UU0KiHVJz4,789
+effspm/load_inst.cpp,sha256=kTEucQ5YU7xPdRjcM9ixAPk49cLJ8H8YN9gJsTrm7mM,4769
+effspm/load_inst.hpp,sha256=yQKGhYiPWoKC-t6dLPjpVi5qaOBkCUoQMwlWhYzDUj4,936
+effspm/utility.cpp,sha256=OD5K0K0jQKgqVGJm91pSofImXOEVkDnqQvFh1qytvpA,1458
+effspm/utility.hpp,sha256=hECSm-ORd20QJMundbOLkZvo9nj-I2pXd5AokagyGqQ,773
+effspm-0.1.7.dist-info/licenses/LICENSE,sha256=HrhfyXIkWY2tGFK11kg7vPCqhgh5DcxleloqdhrpyMY,11558
+effspm-0.1.7.dist-info/METADATA,sha256=iwRl88P5FDFbOYeY6Y9Xd_Lk6DK_E018VnZJW2d8a0A,1349
+effspm-0.1.7.dist-info/WHEEL,sha256=28q3oNn-jDh_H155N4TqNP6BD4TtmXmAi_15pKLiI-M,101
+effspm-0.1.7.dist-info/top_level.txt,sha256=2O-AuI0nw0pDmJMo2jzM1wvV2rj48AmkjskkAnsuuQk,7
+effspm-0.1.7.dist-info/RECORD,,

{effspm-0.1.5.dist-info → effspm-0.1.7.dist-info}/WHEEL

@@ -1,5 +1,5 @@
 Wheel-Version: 1.0
-Generator: setuptools (78.1.0)
+Generator: setuptools (79.0.1)
 Root-Is-Purelib: false
 Tag: cp313-cp313-win_amd64
 

effspm-0.1.5.dist-info/RECORD

@@ -1,14 +0,0 @@
-effspm/__init__.py,sha256=Nqm5qpeFBGhn0dGdEIhIc9H-u6vpzBdlP5KEAEb-LIM,49
-effspm/_core.cp313-win_amd64.pyd,sha256=wOGnPln3qO0aHSqCTpgbcxn__OuVOOT1Sq4PYtg-1n0,172032
-effspm/_core.cpp,sha256=3uqAf3NrZW6Ok-aPKK8lhWpdpwAu3M2dtPglZWMiJ3g,764
-effspm/freq_miner.cpp,sha256=qQDFPoPKY3ICaH2brm1CUKwNBhCy0-0dUWEoV_3FwME,4785
-effspm/freq_miner.hpp,sha256=Exp_4_7yIV0Y9U2I8kyImPoY8hWUFMIE--Zjf-G2TlI,712
-effspm/load_inst.cpp,sha256=kTEucQ5YU7xPdRjcM9ixAPk49cLJ8H8YN9gJsTrm7mM,4769
-effspm/load_inst.hpp,sha256=VxiVbYWe1I8tQvVZRrvwQ097xPxZM5CUn5CjyXM3c7U,465
-effspm/utility.cpp,sha256=OD5K0K0jQKgqVGJm91pSofImXOEVkDnqQvFh1qytvpA,1458
-effspm/utility.hpp,sha256=hECSm-ORd20QJMundbOLkZvo9nj-I2pXd5AokagyGqQ,773
-effspm-0.1.5.dist-info/licenses/LICENSE,sha256=HrhfyXIkWY2tGFK11kg7vPCqhgh5DcxleloqdhrpyMY,11558
-effspm-0.1.5.dist-info/METADATA,sha256=fVoNUBxIuCDGMc3SzOTtKr1X1WeiA3y6e44EdqUBTns,1349
-effspm-0.1.5.dist-info/WHEEL,sha256=6WoW_bHwIgUwfRGomr3tsp2x5B1WboZF1vodYe1_93Y,101
-effspm-0.1.5.dist-info/top_level.txt,sha256=2O-AuI0nw0pDmJMo2jzM1wvV2rj48AmkjskkAnsuuQk,7
-effspm-0.1.5.dist-info/RECORD,,