ZTWHHH commited on
Commit
6e92363
·
verified ·
1 Parent(s): aa7d6a2

Add files using upload-large-folder tool

Browse files
This view is limited to 50 files because it contains too many changes.   See raw diff
Files changed (50) hide show
  1. evalkit_tf437/lib/python3.10/site-packages/numpy/core/__pycache__/__init__.cpython-310.pyc +0 -0
  2. evalkit_tf437/lib/python3.10/site-packages/numpy/core/__pycache__/_add_newdocs_scalars.cpython-310.pyc +0 -0
  3. evalkit_tf437/lib/python3.10/site-packages/numpy/core/__pycache__/_asarray.cpython-310.pyc +0 -0
  4. evalkit_tf437/lib/python3.10/site-packages/numpy/core/__pycache__/_dtype.cpython-310.pyc +0 -0
  5. evalkit_tf437/lib/python3.10/site-packages/numpy/core/__pycache__/_dtype_ctypes.cpython-310.pyc +0 -0
  6. evalkit_tf437/lib/python3.10/site-packages/numpy/core/__pycache__/_exceptions.cpython-310.pyc +0 -0
  7. evalkit_tf437/lib/python3.10/site-packages/numpy/core/__pycache__/_internal.cpython-310.pyc +0 -0
  8. evalkit_tf437/lib/python3.10/site-packages/numpy/core/__pycache__/_string_helpers.cpython-310.pyc +0 -0
  9. evalkit_tf437/lib/python3.10/site-packages/numpy/core/__pycache__/_type_aliases.cpython-310.pyc +0 -0
  10. evalkit_tf437/lib/python3.10/site-packages/numpy/core/__pycache__/cversions.cpython-310.pyc +0 -0
  11. evalkit_tf437/lib/python3.10/site-packages/numpy/core/__pycache__/defchararray.cpython-310.pyc +0 -0
  12. evalkit_tf437/lib/python3.10/site-packages/numpy/core/__pycache__/function_base.cpython-310.pyc +0 -0
  13. evalkit_tf437/lib/python3.10/site-packages/numpy/core/__pycache__/getlimits.cpython-310.pyc +0 -0
  14. evalkit_tf437/lib/python3.10/site-packages/numpy/core/__pycache__/memmap.cpython-310.pyc +0 -0
  15. evalkit_tf437/lib/python3.10/site-packages/numpy/core/__pycache__/multiarray.cpython-310.pyc +0 -0
  16. evalkit_tf437/lib/python3.10/site-packages/numpy/core/__pycache__/numeric.cpython-310.pyc +0 -0
  17. evalkit_tf437/lib/python3.10/site-packages/numpy/core/__pycache__/umath_tests.cpython-310.pyc +0 -0
  18. evalkit_tf437/lib/python3.10/site-packages/numpy/core/include/numpy/__ufunc_api.c +50 -0
  19. evalkit_tf437/lib/python3.10/site-packages/numpy/core/include/numpy/__ufunc_api.h +314 -0
  20. evalkit_tf437/lib/python3.10/site-packages/numpy/core/include/numpy/_neighborhood_iterator_imp.h +90 -0
  21. evalkit_tf437/lib/python3.10/site-packages/numpy/core/include/numpy/arrayobject.h +12 -0
  22. evalkit_tf437/lib/python3.10/site-packages/numpy/core/include/numpy/experimental_dtype_api.h +365 -0
  23. evalkit_tf437/lib/python3.10/site-packages/numpy/core/include/numpy/npy_3kcompat.h +595 -0
  24. evalkit_tf437/lib/python3.10/site-packages/numpy/core/include/numpy/npy_cpu.h +129 -0
  25. evalkit_tf437/lib/python3.10/site-packages/numpy/core/include/numpy/npy_interrupt.h +56 -0
  26. evalkit_tf437/lib/python3.10/site-packages/numpy/core/include/numpy/npy_no_deprecated_api.h +20 -0
  27. evalkit_tf437/lib/python3.10/site-packages/numpy/core/include/numpy/old_defines.h +187 -0
  28. evalkit_tf437/lib/python3.10/site-packages/numpy/core/include/numpy/random/libdivide.h +2079 -0
  29. evalkit_tf437/lib/python3.10/site-packages/numpy/core/lib/npy-pkg-config/npymath.ini +20 -0
  30. evalkit_tf437/lib/python3.10/site-packages/numpy/core/tests/__pycache__/_locales.cpython-310.pyc +0 -0
  31. evalkit_tf437/lib/python3.10/site-packages/numpy/core/tests/__pycache__/test_argparse.cpython-310.pyc +0 -0
  32. evalkit_tf437/lib/python3.10/site-packages/numpy/core/tests/__pycache__/test_function_base.cpython-310.pyc +0 -0
  33. evalkit_tf437/lib/python3.10/site-packages/numpy/core/tests/__pycache__/test_indexerrors.cpython-310.pyc +0 -0
  34. evalkit_tf437/lib/python3.10/site-packages/numpy/core/tests/__pycache__/test_memmap.cpython-310.pyc +0 -0
  35. evalkit_tf437/lib/python3.10/site-packages/numpy/core/tests/__pycache__/test_print.cpython-310.pyc +0 -0
  36. evalkit_tf437/lib/python3.10/site-packages/numpy/core/tests/__pycache__/test_records.cpython-310.pyc +0 -0
  37. evalkit_tf437/lib/python3.10/site-packages/numpy/core/tests/__pycache__/test_simd.cpython-310.pyc +0 -0
  38. evalkit_tf437/lib/python3.10/site-packages/numpy/core/tests/__pycache__/test_umath_complex.cpython-310.pyc +0 -0
  39. evalkit_tf437/lib/python3.10/site-packages/numpy/core/tests/data/astype_copy.pkl +3 -0
  40. evalkit_tf437/lib/python3.10/site-packages/numpy/core/tests/data/umath-validation-set-log1p.csv +1429 -0
  41. evalkit_tf437/lib/python3.10/site-packages/numpy/core/tests/data/umath-validation-set-log2.csv +1629 -0
  42. evalkit_tf446/lib/python3.10/site-packages/networkx/__pycache__/__init__.cpython-310.pyc +0 -0
  43. evalkit_tf446/lib/python3.10/site-packages/networkx/__pycache__/conftest.cpython-310.pyc +0 -0
  44. evalkit_tf446/lib/python3.10/site-packages/networkx/__pycache__/convert.cpython-310.pyc +0 -0
  45. evalkit_tf446/lib/python3.10/site-packages/networkx/__pycache__/convert_matrix.cpython-310.pyc +0 -0
  46. evalkit_tf446/lib/python3.10/site-packages/networkx/__pycache__/exception.cpython-310.pyc +0 -0
  47. evalkit_tf446/lib/python3.10/site-packages/networkx/__pycache__/lazy_imports.cpython-310.pyc +0 -0
  48. evalkit_tf446/lib/python3.10/site-packages/networkx/__pycache__/relabel.cpython-310.pyc +0 -0
  49. evalkit_tf446/lib/python3.10/site-packages/networkx/algorithms/components/attracting.py +115 -0
  50. evalkit_tf446/lib/python3.10/site-packages/networkx/algorithms/components/strongly_connected.py +351 -0
evalkit_tf437/lib/python3.10/site-packages/numpy/core/__pycache__/__init__.cpython-310.pyc ADDED
Binary file (4 kB). View file
 
evalkit_tf437/lib/python3.10/site-packages/numpy/core/__pycache__/_add_newdocs_scalars.cpython-310.pyc ADDED
Binary file (11 kB). View file
 
evalkit_tf437/lib/python3.10/site-packages/numpy/core/__pycache__/_asarray.cpython-310.pyc ADDED
Binary file (3.76 kB). View file
 
evalkit_tf437/lib/python3.10/site-packages/numpy/core/__pycache__/_dtype.cpython-310.pyc ADDED
Binary file (8.2 kB). View file
 
evalkit_tf437/lib/python3.10/site-packages/numpy/core/__pycache__/_dtype_ctypes.cpython-310.pyc ADDED
Binary file (3.05 kB). View file
 
evalkit_tf437/lib/python3.10/site-packages/numpy/core/__pycache__/_exceptions.cpython-310.pyc ADDED
Binary file (5.82 kB). View file
 
evalkit_tf437/lib/python3.10/site-packages/numpy/core/__pycache__/_internal.cpython-310.pyc ADDED
Binary file (23.2 kB). View file
 
evalkit_tf437/lib/python3.10/site-packages/numpy/core/__pycache__/_string_helpers.cpython-310.pyc ADDED
Binary file (3.03 kB). View file
 
evalkit_tf437/lib/python3.10/site-packages/numpy/core/__pycache__/_type_aliases.cpython-310.pyc ADDED
Binary file (5.35 kB). View file
 
evalkit_tf437/lib/python3.10/site-packages/numpy/core/__pycache__/cversions.cpython-310.pyc ADDED
Binary file (551 Bytes). View file
 
evalkit_tf437/lib/python3.10/site-packages/numpy/core/__pycache__/defchararray.cpython-310.pyc ADDED
Binary file (71 kB). View file
 
evalkit_tf437/lib/python3.10/site-packages/numpy/core/__pycache__/function_base.cpython-310.pyc ADDED
Binary file (17.7 kB). View file
 
evalkit_tf437/lib/python3.10/site-packages/numpy/core/__pycache__/getlimits.cpython-310.pyc ADDED
Binary file (19.1 kB). View file
 
evalkit_tf437/lib/python3.10/site-packages/numpy/core/__pycache__/memmap.cpython-310.pyc ADDED
Binary file (10.5 kB). View file
 
evalkit_tf437/lib/python3.10/site-packages/numpy/core/__pycache__/multiarray.cpython-310.pyc ADDED
Binary file (54.4 kB). View file
 
evalkit_tf437/lib/python3.10/site-packages/numpy/core/__pycache__/numeric.cpython-310.pyc ADDED
Binary file (73 kB). View file
 
evalkit_tf437/lib/python3.10/site-packages/numpy/core/__pycache__/umath_tests.cpython-310.pyc ADDED
Binary file (518 Bytes). View file
 
evalkit_tf437/lib/python3.10/site-packages/numpy/core/include/numpy/__ufunc_api.c ADDED
@@ -0,0 +1,50 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+
2
+ /* These pointers will be stored in the C-object for use in other
3
+ extension modules
4
+ */
5
+
6
+ void *PyUFunc_API[] = {
7
+ (void *) &PyUFunc_Type,
8
+ (void *) PyUFunc_FromFuncAndData,
9
+ (void *) PyUFunc_RegisterLoopForType,
10
+ (void *) PyUFunc_GenericFunction,
11
+ (void *) PyUFunc_f_f_As_d_d,
12
+ (void *) PyUFunc_d_d,
13
+ (void *) PyUFunc_f_f,
14
+ (void *) PyUFunc_g_g,
15
+ (void *) PyUFunc_F_F_As_D_D,
16
+ (void *) PyUFunc_F_F,
17
+ (void *) PyUFunc_D_D,
18
+ (void *) PyUFunc_G_G,
19
+ (void *) PyUFunc_O_O,
20
+ (void *) PyUFunc_ff_f_As_dd_d,
21
+ (void *) PyUFunc_ff_f,
22
+ (void *) PyUFunc_dd_d,
23
+ (void *) PyUFunc_gg_g,
24
+ (void *) PyUFunc_FF_F_As_DD_D,
25
+ (void *) PyUFunc_DD_D,
26
+ (void *) PyUFunc_FF_F,
27
+ (void *) PyUFunc_GG_G,
28
+ (void *) PyUFunc_OO_O,
29
+ (void *) PyUFunc_O_O_method,
30
+ (void *) PyUFunc_OO_O_method,
31
+ (void *) PyUFunc_On_Om,
32
+ (void *) PyUFunc_GetPyValues,
33
+ (void *) PyUFunc_checkfperr,
34
+ (void *) PyUFunc_clearfperr,
35
+ (void *) PyUFunc_getfperr,
36
+ (void *) PyUFunc_handlefperr,
37
+ (void *) PyUFunc_ReplaceLoopBySignature,
38
+ (void *) PyUFunc_FromFuncAndDataAndSignature,
39
+ (void *) PyUFunc_SetUsesArraysAsData,
40
+ (void *) PyUFunc_e_e,
41
+ (void *) PyUFunc_e_e_As_f_f,
42
+ (void *) PyUFunc_e_e_As_d_d,
43
+ (void *) PyUFunc_ee_e,
44
+ (void *) PyUFunc_ee_e_As_ff_f,
45
+ (void *) PyUFunc_ee_e_As_dd_d,
46
+ (void *) PyUFunc_DefaultTypeResolver,
47
+ (void *) PyUFunc_ValidateCasting,
48
+ (void *) PyUFunc_RegisterLoopForDescr,
49
+ (void *) PyUFunc_FromFuncAndDataAndSignatureAndIdentity
50
+ };
evalkit_tf437/lib/python3.10/site-packages/numpy/core/include/numpy/__ufunc_api.h ADDED
@@ -0,0 +1,314 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+
2
+ #ifdef _UMATHMODULE
3
+
4
+ extern NPY_NO_EXPORT PyTypeObject PyUFunc_Type;
5
+
6
+ extern NPY_NO_EXPORT PyTypeObject PyUFunc_Type;
7
+
8
+ NPY_NO_EXPORT PyObject * PyUFunc_FromFuncAndData \
9
+ (PyUFuncGenericFunction *, void **, char *, int, int, int, int, const char *, const char *, int);
10
+ NPY_NO_EXPORT int PyUFunc_RegisterLoopForType \
11
+ (PyUFuncObject *, int, PyUFuncGenericFunction, const int *, void *);
12
+ NPY_NO_EXPORT int PyUFunc_GenericFunction \
13
+ (PyUFuncObject *NPY_UNUSED(ufunc), PyObject *NPY_UNUSED(args), PyObject *NPY_UNUSED(kwds), PyArrayObject **NPY_UNUSED(op));
14
+ NPY_NO_EXPORT void PyUFunc_f_f_As_d_d \
15
+ (char **, npy_intp const *, npy_intp const *, void *);
16
+ NPY_NO_EXPORT void PyUFunc_d_d \
17
+ (char **, npy_intp const *, npy_intp const *, void *);
18
+ NPY_NO_EXPORT void PyUFunc_f_f \
19
+ (char **, npy_intp const *, npy_intp const *, void *);
20
+ NPY_NO_EXPORT void PyUFunc_g_g \
21
+ (char **, npy_intp const *, npy_intp const *, void *);
22
+ NPY_NO_EXPORT void PyUFunc_F_F_As_D_D \
23
+ (char **, npy_intp const *, npy_intp const *, void *);
24
+ NPY_NO_EXPORT void PyUFunc_F_F \
25
+ (char **, npy_intp const *, npy_intp const *, void *);
26
+ NPY_NO_EXPORT void PyUFunc_D_D \
27
+ (char **, npy_intp const *, npy_intp const *, void *);
28
+ NPY_NO_EXPORT void PyUFunc_G_G \
29
+ (char **, npy_intp const *, npy_intp const *, void *);
30
+ NPY_NO_EXPORT void PyUFunc_O_O \
31
+ (char **, npy_intp const *, npy_intp const *, void *);
32
+ NPY_NO_EXPORT void PyUFunc_ff_f_As_dd_d \
33
+ (char **, npy_intp const *, npy_intp const *, void *);
34
+ NPY_NO_EXPORT void PyUFunc_ff_f \
35
+ (char **, npy_intp const *, npy_intp const *, void *);
36
+ NPY_NO_EXPORT void PyUFunc_dd_d \
37
+ (char **, npy_intp const *, npy_intp const *, void *);
38
+ NPY_NO_EXPORT void PyUFunc_gg_g \
39
+ (char **, npy_intp const *, npy_intp const *, void *);
40
+ NPY_NO_EXPORT void PyUFunc_FF_F_As_DD_D \
41
+ (char **, npy_intp const *, npy_intp const *, void *);
42
+ NPY_NO_EXPORT void PyUFunc_DD_D \
43
+ (char **, npy_intp const *, npy_intp const *, void *);
44
+ NPY_NO_EXPORT void PyUFunc_FF_F \
45
+ (char **, npy_intp const *, npy_intp const *, void *);
46
+ NPY_NO_EXPORT void PyUFunc_GG_G \
47
+ (char **, npy_intp const *, npy_intp const *, void *);
48
+ NPY_NO_EXPORT void PyUFunc_OO_O \
49
+ (char **, npy_intp const *, npy_intp const *, void *);
50
+ NPY_NO_EXPORT void PyUFunc_O_O_method \
51
+ (char **, npy_intp const *, npy_intp const *, void *);
52
+ NPY_NO_EXPORT void PyUFunc_OO_O_method \
53
+ (char **, npy_intp const *, npy_intp const *, void *);
54
+ NPY_NO_EXPORT void PyUFunc_On_Om \
55
+ (char **, npy_intp const *, npy_intp const *, void *);
56
+ NPY_NO_EXPORT int PyUFunc_GetPyValues \
57
+ (char *, int *, int *, PyObject **);
58
+ NPY_NO_EXPORT int PyUFunc_checkfperr \
59
+ (int, PyObject *, int *);
60
+ NPY_NO_EXPORT void PyUFunc_clearfperr \
61
+ (void);
62
+ NPY_NO_EXPORT int PyUFunc_getfperr \
63
+ (void);
64
+ NPY_NO_EXPORT int PyUFunc_handlefperr \
65
+ (int, PyObject *, int, int *);
66
+ NPY_NO_EXPORT int PyUFunc_ReplaceLoopBySignature \
67
+ (PyUFuncObject *, PyUFuncGenericFunction, const int *, PyUFuncGenericFunction *);
68
+ NPY_NO_EXPORT PyObject * PyUFunc_FromFuncAndDataAndSignature \
69
+ (PyUFuncGenericFunction *, void **, char *, int, int, int, int, const char *, const char *, int, const char *);
70
+ NPY_NO_EXPORT int PyUFunc_SetUsesArraysAsData \
71
+ (void **NPY_UNUSED(data), size_t NPY_UNUSED(i));
72
+ NPY_NO_EXPORT void PyUFunc_e_e \
73
+ (char **, npy_intp const *, npy_intp const *, void *);
74
+ NPY_NO_EXPORT void PyUFunc_e_e_As_f_f \
75
+ (char **, npy_intp const *, npy_intp const *, void *);
76
+ NPY_NO_EXPORT void PyUFunc_e_e_As_d_d \
77
+ (char **, npy_intp const *, npy_intp const *, void *);
78
+ NPY_NO_EXPORT void PyUFunc_ee_e \
79
+ (char **, npy_intp const *, npy_intp const *, void *);
80
+ NPY_NO_EXPORT void PyUFunc_ee_e_As_ff_f \
81
+ (char **, npy_intp const *, npy_intp const *, void *);
82
+ NPY_NO_EXPORT void PyUFunc_ee_e_As_dd_d \
83
+ (char **, npy_intp const *, npy_intp const *, void *);
84
+ NPY_NO_EXPORT int PyUFunc_DefaultTypeResolver \
85
+ (PyUFuncObject *, NPY_CASTING, PyArrayObject **, PyObject *, PyArray_Descr **);
86
+ NPY_NO_EXPORT int PyUFunc_ValidateCasting \
87
+ (PyUFuncObject *, NPY_CASTING, PyArrayObject **, PyArray_Descr **);
88
+ NPY_NO_EXPORT int PyUFunc_RegisterLoopForDescr \
89
+ (PyUFuncObject *, PyArray_Descr *, PyUFuncGenericFunction, PyArray_Descr **, void *);
90
+ NPY_NO_EXPORT PyObject * PyUFunc_FromFuncAndDataAndSignatureAndIdentity \
91
+ (PyUFuncGenericFunction *, void **, char *, int, int, int, int, const char *, const char *, const int, const char *, PyObject *);
92
+
93
+ #else
94
+
95
+ #if defined(PY_UFUNC_UNIQUE_SYMBOL)
96
+ #define PyUFunc_API PY_UFUNC_UNIQUE_SYMBOL
97
+ #endif
98
+
99
+ #if defined(NO_IMPORT) || defined(NO_IMPORT_UFUNC)
100
+ extern void **PyUFunc_API;
101
+ #else
102
+ #if defined(PY_UFUNC_UNIQUE_SYMBOL)
103
+ void **PyUFunc_API;
104
+ #else
105
+ static void **PyUFunc_API=NULL;
106
+ #endif
107
+ #endif
108
+
109
+ #define PyUFunc_Type (*(PyTypeObject *)PyUFunc_API[0])
110
+ #define PyUFunc_FromFuncAndData \
111
+ (*(PyObject * (*)(PyUFuncGenericFunction *, void **, char *, int, int, int, int, const char *, const char *, int)) \
112
+ PyUFunc_API[1])
113
+ #define PyUFunc_RegisterLoopForType \
114
+ (*(int (*)(PyUFuncObject *, int, PyUFuncGenericFunction, const int *, void *)) \
115
+ PyUFunc_API[2])
116
+ #define PyUFunc_GenericFunction \
117
+ (*(int (*)(PyUFuncObject *NPY_UNUSED(ufunc), PyObject *NPY_UNUSED(args), PyObject *NPY_UNUSED(kwds), PyArrayObject **NPY_UNUSED(op))) \
118
+ PyUFunc_API[3])
119
+ #define PyUFunc_f_f_As_d_d \
120
+ (*(void (*)(char **, npy_intp const *, npy_intp const *, void *)) \
121
+ PyUFunc_API[4])
122
+ #define PyUFunc_d_d \
123
+ (*(void (*)(char **, npy_intp const *, npy_intp const *, void *)) \
124
+ PyUFunc_API[5])
125
+ #define PyUFunc_f_f \
126
+ (*(void (*)(char **, npy_intp const *, npy_intp const *, void *)) \
127
+ PyUFunc_API[6])
128
+ #define PyUFunc_g_g \
129
+ (*(void (*)(char **, npy_intp const *, npy_intp const *, void *)) \
130
+ PyUFunc_API[7])
131
+ #define PyUFunc_F_F_As_D_D \
132
+ (*(void (*)(char **, npy_intp const *, npy_intp const *, void *)) \
133
+ PyUFunc_API[8])
134
+ #define PyUFunc_F_F \
135
+ (*(void (*)(char **, npy_intp const *, npy_intp const *, void *)) \
136
+ PyUFunc_API[9])
137
+ #define PyUFunc_D_D \
138
+ (*(void (*)(char **, npy_intp const *, npy_intp const *, void *)) \
139
+ PyUFunc_API[10])
140
+ #define PyUFunc_G_G \
141
+ (*(void (*)(char **, npy_intp const *, npy_intp const *, void *)) \
142
+ PyUFunc_API[11])
143
+ #define PyUFunc_O_O \
144
+ (*(void (*)(char **, npy_intp const *, npy_intp const *, void *)) \
145
+ PyUFunc_API[12])
146
+ #define PyUFunc_ff_f_As_dd_d \
147
+ (*(void (*)(char **, npy_intp const *, npy_intp const *, void *)) \
148
+ PyUFunc_API[13])
149
+ #define PyUFunc_ff_f \
150
+ (*(void (*)(char **, npy_intp const *, npy_intp const *, void *)) \
151
+ PyUFunc_API[14])
152
+ #define PyUFunc_dd_d \
153
+ (*(void (*)(char **, npy_intp const *, npy_intp const *, void *)) \
154
+ PyUFunc_API[15])
155
+ #define PyUFunc_gg_g \
156
+ (*(void (*)(char **, npy_intp const *, npy_intp const *, void *)) \
157
+ PyUFunc_API[16])
158
+ #define PyUFunc_FF_F_As_DD_D \
159
+ (*(void (*)(char **, npy_intp const *, npy_intp const *, void *)) \
160
+ PyUFunc_API[17])
161
+ #define PyUFunc_DD_D \
162
+ (*(void (*)(char **, npy_intp const *, npy_intp const *, void *)) \
163
+ PyUFunc_API[18])
164
+ #define PyUFunc_FF_F \
165
+ (*(void (*)(char **, npy_intp const *, npy_intp const *, void *)) \
166
+ PyUFunc_API[19])
167
+ #define PyUFunc_GG_G \
168
+ (*(void (*)(char **, npy_intp const *, npy_intp const *, void *)) \
169
+ PyUFunc_API[20])
170
+ #define PyUFunc_OO_O \
171
+ (*(void (*)(char **, npy_intp const *, npy_intp const *, void *)) \
172
+ PyUFunc_API[21])
173
+ #define PyUFunc_O_O_method \
174
+ (*(void (*)(char **, npy_intp const *, npy_intp const *, void *)) \
175
+ PyUFunc_API[22])
176
+ #define PyUFunc_OO_O_method \
177
+ (*(void (*)(char **, npy_intp const *, npy_intp const *, void *)) \
178
+ PyUFunc_API[23])
179
+ #define PyUFunc_On_Om \
180
+ (*(void (*)(char **, npy_intp const *, npy_intp const *, void *)) \
181
+ PyUFunc_API[24])
182
+ #define PyUFunc_GetPyValues \
183
+ (*(int (*)(char *, int *, int *, PyObject **)) \
184
+ PyUFunc_API[25])
185
+ #define PyUFunc_checkfperr \
186
+ (*(int (*)(int, PyObject *, int *)) \
187
+ PyUFunc_API[26])
188
+ #define PyUFunc_clearfperr \
189
+ (*(void (*)(void)) \
190
+ PyUFunc_API[27])
191
+ #define PyUFunc_getfperr \
192
+ (*(int (*)(void)) \
193
+ PyUFunc_API[28])
194
+ #define PyUFunc_handlefperr \
195
+ (*(int (*)(int, PyObject *, int, int *)) \
196
+ PyUFunc_API[29])
197
+ #define PyUFunc_ReplaceLoopBySignature \
198
+ (*(int (*)(PyUFuncObject *, PyUFuncGenericFunction, const int *, PyUFuncGenericFunction *)) \
199
+ PyUFunc_API[30])
200
+ #define PyUFunc_FromFuncAndDataAndSignature \
201
+ (*(PyObject * (*)(PyUFuncGenericFunction *, void **, char *, int, int, int, int, const char *, const char *, int, const char *)) \
202
+ PyUFunc_API[31])
203
+ #define PyUFunc_SetUsesArraysAsData \
204
+ (*(int (*)(void **NPY_UNUSED(data), size_t NPY_UNUSED(i))) \
205
+ PyUFunc_API[32])
206
+ #define PyUFunc_e_e \
207
+ (*(void (*)(char **, npy_intp const *, npy_intp const *, void *)) \
208
+ PyUFunc_API[33])
209
+ #define PyUFunc_e_e_As_f_f \
210
+ (*(void (*)(char **, npy_intp const *, npy_intp const *, void *)) \
211
+ PyUFunc_API[34])
212
+ #define PyUFunc_e_e_As_d_d \
213
+ (*(void (*)(char **, npy_intp const *, npy_intp const *, void *)) \
214
+ PyUFunc_API[35])
215
+ #define PyUFunc_ee_e \
216
+ (*(void (*)(char **, npy_intp const *, npy_intp const *, void *)) \
217
+ PyUFunc_API[36])
218
+ #define PyUFunc_ee_e_As_ff_f \
219
+ (*(void (*)(char **, npy_intp const *, npy_intp const *, void *)) \
220
+ PyUFunc_API[37])
221
+ #define PyUFunc_ee_e_As_dd_d \
222
+ (*(void (*)(char **, npy_intp const *, npy_intp const *, void *)) \
223
+ PyUFunc_API[38])
224
+ #define PyUFunc_DefaultTypeResolver \
225
+ (*(int (*)(PyUFuncObject *, NPY_CASTING, PyArrayObject **, PyObject *, PyArray_Descr **)) \
226
+ PyUFunc_API[39])
227
+ #define PyUFunc_ValidateCasting \
228
+ (*(int (*)(PyUFuncObject *, NPY_CASTING, PyArrayObject **, PyArray_Descr **)) \
229
+ PyUFunc_API[40])
230
+ #define PyUFunc_RegisterLoopForDescr \
231
+ (*(int (*)(PyUFuncObject *, PyArray_Descr *, PyUFuncGenericFunction, PyArray_Descr **, void *)) \
232
+ PyUFunc_API[41])
233
+
234
+ #if NPY_FEATURE_VERSION >= NPY_1_16_API_VERSION
235
+ #define PyUFunc_FromFuncAndDataAndSignatureAndIdentity \
236
+ (*(PyObject * (*)(PyUFuncGenericFunction *, void **, char *, int, int, int, int, const char *, const char *, const int, const char *, PyObject *)) \
237
+ PyUFunc_API[42])
238
+ #endif
239
+
240
+ static inline int
241
+ _import_umath(void)
242
+ {
243
+ PyObject *numpy = PyImport_ImportModule("numpy.core._multiarray_umath");
244
+ PyObject *c_api = NULL;
245
+
246
+ if (numpy == NULL) {
247
+ PyErr_SetString(PyExc_ImportError,
248
+ "numpy.core._multiarray_umath failed to import");
249
+ return -1;
250
+ }
251
+ c_api = PyObject_GetAttrString(numpy, "_UFUNC_API");
252
+ Py_DECREF(numpy);
253
+ if (c_api == NULL) {
254
+ PyErr_SetString(PyExc_AttributeError, "_UFUNC_API not found");
255
+ return -1;
256
+ }
257
+
258
+ if (!PyCapsule_CheckExact(c_api)) {
259
+ PyErr_SetString(PyExc_RuntimeError, "_UFUNC_API is not PyCapsule object");
260
+ Py_DECREF(c_api);
261
+ return -1;
262
+ }
263
+ PyUFunc_API = (void **)PyCapsule_GetPointer(c_api, NULL);
264
+ Py_DECREF(c_api);
265
+ if (PyUFunc_API == NULL) {
266
+ PyErr_SetString(PyExc_RuntimeError, "_UFUNC_API is NULL pointer");
267
+ return -1;
268
+ }
269
+ return 0;
270
+ }
271
+
272
+ #define import_umath() \
273
+ do {\
274
+ UFUNC_NOFPE\
275
+ if (_import_umath() < 0) {\
276
+ PyErr_Print();\
277
+ PyErr_SetString(PyExc_ImportError,\
278
+ "numpy.core.umath failed to import");\
279
+ return NULL;\
280
+ }\
281
+ } while(0)
282
+
283
+ #define import_umath1(ret) \
284
+ do {\
285
+ UFUNC_NOFPE\
286
+ if (_import_umath() < 0) {\
287
+ PyErr_Print();\
288
+ PyErr_SetString(PyExc_ImportError,\
289
+ "numpy.core.umath failed to import");\
290
+ return ret;\
291
+ }\
292
+ } while(0)
293
+
294
+ #define import_umath2(ret, msg) \
295
+ do {\
296
+ UFUNC_NOFPE\
297
+ if (_import_umath() < 0) {\
298
+ PyErr_Print();\
299
+ PyErr_SetString(PyExc_ImportError, msg);\
300
+ return ret;\
301
+ }\
302
+ } while(0)
303
+
304
+ #define import_ufunc() \
305
+ do {\
306
+ UFUNC_NOFPE\
307
+ if (_import_umath() < 0) {\
308
+ PyErr_Print();\
309
+ PyErr_SetString(PyExc_ImportError,\
310
+ "numpy.core.umath failed to import");\
311
+ }\
312
+ } while(0)
313
+
314
+ #endif
evalkit_tf437/lib/python3.10/site-packages/numpy/core/include/numpy/_neighborhood_iterator_imp.h ADDED
@@ -0,0 +1,90 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ #ifndef NUMPY_CORE_INCLUDE_NUMPY__NEIGHBORHOOD_IMP_H_
2
+ #error You should not include this header directly
3
+ #endif
4
+ /*
5
+ * Private API (here for inline)
6
+ */
7
+ static inline int
8
+ _PyArrayNeighborhoodIter_IncrCoord(PyArrayNeighborhoodIterObject* iter);
9
+
10
+ /*
11
+ * Update to next item of the iterator
12
+ *
13
+ * Note: this simply increment the coordinates vector, last dimension
14
+ * incremented first , i.e, for dimension 3
15
+ * ...
16
+ * -1, -1, -1
17
+ * -1, -1, 0
18
+ * -1, -1, 1
19
+ * ....
20
+ * -1, 0, -1
21
+ * -1, 0, 0
22
+ * ....
23
+ * 0, -1, -1
24
+ * 0, -1, 0
25
+ * ....
26
+ */
27
+ #define _UPDATE_COORD_ITER(c) \
28
+ wb = iter->coordinates[c] < iter->bounds[c][1]; \
29
+ if (wb) { \
30
+ iter->coordinates[c] += 1; \
31
+ return 0; \
32
+ } \
33
+ else { \
34
+ iter->coordinates[c] = iter->bounds[c][0]; \
35
+ }
36
+
37
+ static inline int
38
+ _PyArrayNeighborhoodIter_IncrCoord(PyArrayNeighborhoodIterObject* iter)
39
+ {
40
+ npy_intp i, wb;
41
+
42
+ for (i = iter->nd - 1; i >= 0; --i) {
43
+ _UPDATE_COORD_ITER(i)
44
+ }
45
+
46
+ return 0;
47
+ }
48
+
49
+ /*
50
+ * Version optimized for 2d arrays, manual loop unrolling
51
+ */
52
+ static inline int
53
+ _PyArrayNeighborhoodIter_IncrCoord2D(PyArrayNeighborhoodIterObject* iter)
54
+ {
55
+ npy_intp wb;
56
+
57
+ _UPDATE_COORD_ITER(1)
58
+ _UPDATE_COORD_ITER(0)
59
+
60
+ return 0;
61
+ }
62
+ #undef _UPDATE_COORD_ITER
63
+
64
+ /*
65
+ * Advance to the next neighbour
66
+ */
67
+ static inline int
68
+ PyArrayNeighborhoodIter_Next(PyArrayNeighborhoodIterObject* iter)
69
+ {
70
+ _PyArrayNeighborhoodIter_IncrCoord (iter);
71
+ iter->dataptr = iter->translate((PyArrayIterObject*)iter, iter->coordinates);
72
+
73
+ return 0;
74
+ }
75
+
76
+ /*
77
+ * Reset functions
78
+ */
79
+ static inline int
80
+ PyArrayNeighborhoodIter_Reset(PyArrayNeighborhoodIterObject* iter)
81
+ {
82
+ npy_intp i;
83
+
84
+ for (i = 0; i < iter->nd; ++i) {
85
+ iter->coordinates[i] = iter->bounds[i][0];
86
+ }
87
+ iter->dataptr = iter->translate((PyArrayIterObject*)iter, iter->coordinates);
88
+
89
+ return 0;
90
+ }
evalkit_tf437/lib/python3.10/site-packages/numpy/core/include/numpy/arrayobject.h ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ #ifndef NUMPY_CORE_INCLUDE_NUMPY_ARRAYOBJECT_H_
2
+ #define NUMPY_CORE_INCLUDE_NUMPY_ARRAYOBJECT_H_
3
+ #define Py_ARRAYOBJECT_H
4
+
5
+ #include "ndarrayobject.h"
6
+ #include "npy_interrupt.h"
7
+
8
+ #ifdef NPY_NO_PREFIX
9
+ #include "noprefix.h"
10
+ #endif
11
+
12
+ #endif /* NUMPY_CORE_INCLUDE_NUMPY_ARRAYOBJECT_H_ */
evalkit_tf437/lib/python3.10/site-packages/numpy/core/include/numpy/experimental_dtype_api.h ADDED
@@ -0,0 +1,365 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ /*
2
+ * This header exports the new experimental DType API as proposed in
3
+ * NEPs 41 to 43. For background, please check these NEPs. Otherwise,
4
+ * this header also serves as documentation for the time being.
5
+ *
6
+ * The header includes `_dtype_api.h` which holds most definition while this
7
+ * header mainly wraps functions for public consumption.
8
+ *
9
+ * Please do not hesitate to contact @seberg with questions. This is
10
+ * developed together with https://github.com/seberg/experimental_user_dtypes
11
+ * and those interested in experimenting are encouraged to contribute there.
12
+ *
13
+ * To use the functions defined in the header, call::
14
+ *
15
+ * if (import_experimental_dtype_api(version) < 0) {
16
+ * return NULL;
17
+ * }
18
+ *
19
+ * in your module init. (A version mismatch will be reported, just update
20
+ * to the correct one, this will alert you of possible changes.)
21
+ *
22
+ * The following lists the main symbols currently exported. Please do not
23
+ * hesitate to ask for help or clarification:
24
+ *
25
+ * - PyUFunc_AddLoopFromSpec:
26
+ *
27
+ * Register a new loop for a ufunc. This uses the `PyArrayMethod_Spec`
28
+ * which must be filled in (see in-line comments).
29
+ *
30
+ * - PyUFunc_AddWrappingLoop:
31
+ *
32
+ * Register a new loop which reuses an existing one, but modifies the
33
+ * result dtypes. Please search the internal NumPy docs for more info
34
+ * at this point. (Used for physical units dtype.)
35
+ *
36
+ * - PyUFunc_AddPromoter:
37
+ *
38
+ * Register a new promoter for a ufunc. A promoter is a function stored
39
+ * in a PyCapsule (see in-line comments). It is passed the operation and
40
+ * requested DType signatures and can mutate it to attempt a new search
41
+ * for a matching loop/promoter.
42
+ * I.e. for Numba a promoter could even add the desired loop.
43
+ *
44
+ * - PyArrayInitDTypeMeta_FromSpec:
45
+ *
46
+ * Initialize a new DType. It must currently be a static Python C type
47
+ * that is declared as `PyArray_DTypeMeta` and not `PyTypeObject`.
48
+ * Further, it must subclass `np.dtype` and set its type to
49
+ * `PyArrayDTypeMeta_Type` (before calling `PyType_Read()`).
50
+ *
51
+ * - PyArray_CommonDType:
52
+ *
53
+ * Find the common-dtype ("promotion") for two DType classes. Similar
54
+ * to `np.result_type`, but works on the classes and not instances.
55
+ *
56
+ * - PyArray_PromoteDTypeSequence:
57
+ *
58
+ * Same as CommonDType, but works with an arbitrary number of DTypes.
59
+ * This function is smarter and can often return successful and unambiguous
60
+ * results when `common_dtype(common_dtype(dt1, dt2), dt3)` would
61
+ * depend on the operation order or fail. Nevertheless, DTypes should
62
+ * aim to ensure that their common-dtype implementation is associative
63
+ * and commutative! (Mainly, unsigned and signed integers are not.)
64
+ *
65
+ * For guaranteed consistent results DTypes must implement common-Dtype
66
+ * "transitively". If A promotes B and B promotes C, than A must generally
67
+ * also promote C; where "promotes" means implements the promotion.
68
+ * (There are some exceptions for abstract DTypes)
69
+ *
70
+ * - PyArray_GetDefaultDescr:
71
+ *
72
+ * Given a DType class, returns the default instance (descriptor).
73
+ * This is an inline function checking for `singleton` first and only
74
+ * calls the `default_descr` function if necessary.
75
+ *
76
+ * - PyArray_DoubleDType, etc.:
77
+ *
78
+ * Aliases to the DType classes for the builtin NumPy DTypes.
79
+ *
80
+ * WARNING
81
+ * =======
82
+ *
83
+ * By using this header, you understand that this is a fully experimental
84
+ * exposure. Details are expected to change, and some options may have no
85
+ * effect. (Please contact @seberg if you have questions!)
86
+ * If the exposure stops working, please file a bug report with NumPy.
87
+ * Further, a DType created using this API/header should still be expected
88
+ * to be incompatible with some functionality inside and outside of NumPy.
89
+ * In this case crashes must be expected. Please report any such problems
90
+ * so that they can be fixed before final exposure.
91
+ * Furthermore, expect missing checks for programming errors which the final
92
+ * API is expected to have.
93
+ *
94
+ * Symbols with a leading underscore are likely to not be included in the
95
+ * first public version, if these are central to your use-case, please let
96
+ * us know, so that we can reconsider.
97
+ *
98
+ * "Array-like" consumer API not yet under considerations
99
+ * ======================================================
100
+ *
101
+ * The new DType API is designed in a way to make it potentially useful for
102
+ * alternative "array-like" implementations. This will require careful
103
+ * exposure of details and functions and is not part of this experimental API.
104
+ *
105
+ * Brief (incompatibility) changelog
106
+ * =================================
107
+ *
108
+ * 2. None (only additions).
109
+ * 3. New `npy_intp *view_offset` argument for `resolve_descriptors`.
110
+ * This replaces the `NPY_CAST_IS_VIEW` flag. It can be set to 0 if the
111
+ * operation is a view, and is pre-initialized to `NPY_MIN_INTP` indicating
112
+ * that the operation is not a view.
113
+ */
114
+
115
+ #ifndef NUMPY_CORE_INCLUDE_NUMPY_EXPERIMENTAL_DTYPE_API_H_
116
+ #define NUMPY_CORE_INCLUDE_NUMPY_EXPERIMENTAL_DTYPE_API_H_
117
+
118
+ #include <Python.h>
119
+ #include "ndarraytypes.h"
120
+ #include "_dtype_api.h"
121
+
122
+ /*
123
+ * The contents of PyArrayMethodObject are currently opaque (is there a way
124
+ * good way to make them be `PyObject *`?)
125
+ */
126
+ typedef struct PyArrayMethodObject_tag PyArrayMethodObject;
127
+
128
+ /*
129
+ * There must be a better way?! -- Oh well, this is experimental
130
+ * (my issue with it, is that I cannot undef those helpers).
131
+ */
132
+ #if defined(PY_ARRAY_UNIQUE_SYMBOL)
133
+ #define NPY_EXP_DTYPE_API_CONCAT_HELPER2(x, y) x ## y
134
+ #define NPY_EXP_DTYPE_API_CONCAT_HELPER(arg) NPY_EXP_DTYPE_API_CONCAT_HELPER2(arg, __experimental_dtype_api_table)
135
+ #define __experimental_dtype_api_table NPY_EXP_DTYPE_API_CONCAT_HELPER(PY_ARRAY_UNIQUE_SYMBOL)
136
+ #else
137
+ #define __experimental_dtype_api_table __experimental_dtype_api_table
138
+ #endif
139
+
140
+ /* Support for correct multi-file projects: */
141
+ #if defined(NO_IMPORT) || defined(NO_IMPORT_ARRAY)
142
+ extern void **__experimental_dtype_api_table;
143
+ #else
144
+ /*
145
+ * Just a hack so I don't forget importing as much myself, I spend way too
146
+ * much time noticing it the first time around :).
147
+ */
148
+ static void
149
+ __not_imported(void)
150
+ {
151
+ printf("*****\nCritical error, dtype API not imported\n*****\n");
152
+ }
153
+
154
+ static void *__uninitialized_table[] = {
155
+ &__not_imported, &__not_imported, &__not_imported, &__not_imported,
156
+ &__not_imported, &__not_imported, &__not_imported, &__not_imported};
157
+
158
+ #if defined(PY_ARRAY_UNIQUE_SYMBOL)
159
+ void **__experimental_dtype_api_table = __uninitialized_table;
160
+ #else
161
+ static void **__experimental_dtype_api_table = __uninitialized_table;
162
+ #endif
163
+ #endif
164
+
165
+
166
+ typedef int _ufunc_addloop_fromspec_func(
167
+ PyObject *ufunc, PyArrayMethod_Spec *spec);
168
+ /*
169
+ * The main ufunc registration function. This adds a new implementation/loop
170
+ * to a ufunc. It replaces `PyUFunc_RegisterLoopForType`.
171
+ */
172
+ #define PyUFunc_AddLoopFromSpec \
173
+ (*(_ufunc_addloop_fromspec_func *)(__experimental_dtype_api_table[0]))
174
+
175
+
176
+ /* Please see the NumPy definitions in `array_method.h` for details on these */
177
+ typedef int translate_given_descrs_func(int nin, int nout,
178
+ PyArray_DTypeMeta *wrapped_dtypes[],
179
+ PyArray_Descr *given_descrs[], PyArray_Descr *new_descrs[]);
180
+ typedef int translate_loop_descrs_func(int nin, int nout,
181
+ PyArray_DTypeMeta *new_dtypes[], PyArray_Descr *given_descrs[],
182
+ PyArray_Descr *original_descrs[], PyArray_Descr *loop_descrs[]);
183
+
184
+ typedef int _ufunc_wrapping_loop_func(PyObject *ufunc_obj,
185
+ PyArray_DTypeMeta *new_dtypes[], PyArray_DTypeMeta *wrapped_dtypes[],
186
+ translate_given_descrs_func *translate_given_descrs,
187
+ translate_loop_descrs_func *translate_loop_descrs);
188
+ #define PyUFunc_AddWrappingLoop \
189
+ (*(_ufunc_wrapping_loop_func *)(__experimental_dtype_api_table[7]))
190
+
191
+ /*
192
+ * Type of the C promoter function, which must be wrapped into a
193
+ * PyCapsule with name "numpy._ufunc_promoter".
194
+ *
195
+ * Note that currently the output dtypes are always NULL unless they are
196
+ * also part of the signature. This is an implementation detail and could
197
+ * change in the future. However, in general promoters should not have a
198
+ * need for output dtypes.
199
+ * (There are potential use-cases, these are currently unsupported.)
200
+ */
201
+ typedef int promoter_function(PyObject *ufunc,
202
+ PyArray_DTypeMeta *op_dtypes[], PyArray_DTypeMeta *signature[],
203
+ PyArray_DTypeMeta *new_op_dtypes[]);
204
+
205
+ /*
206
+ * Function to register a promoter.
207
+ *
208
+ * @param ufunc The ufunc object to register the promoter with.
209
+ * @param DType_tuple A Python tuple containing DTypes or None matching the
210
+ * number of inputs and outputs of the ufunc.
211
+ * @param promoter A PyCapsule with name "numpy._ufunc_promoter" containing
212
+ * a pointer to a `promoter_function`.
213
+ */
214
+ typedef int _ufunc_addpromoter_func(
215
+ PyObject *ufunc, PyObject *DType_tuple, PyObject *promoter);
216
+ #define PyUFunc_AddPromoter \
217
+ (*(_ufunc_addpromoter_func *)(__experimental_dtype_api_table[1]))
218
+
219
+ #define PyArrayDTypeMeta_Type \
220
+ (*(PyTypeObject *)__experimental_dtype_api_table[2])
221
+ typedef int __dtypemeta_fromspec(
222
+ PyArray_DTypeMeta *DType, PyArrayDTypeMeta_Spec *dtype_spec);
223
+ /*
224
+ * Finalize creation of a DTypeMeta. You must ensure that the DTypeMeta is
225
+ * a proper subclass. The DTypeMeta object has additional fields compared to
226
+ * a normal PyTypeObject!
227
+ * The only (easy) creation of a new DType is to create a static Type which
228
+ * inherits `PyArray_DescrType`, sets its type to `PyArrayDTypeMeta_Type` and
229
+ * uses `PyArray_DTypeMeta` defined above as the C-structure.
230
+ */
231
+ #define PyArrayInitDTypeMeta_FromSpec \
232
+ ((__dtypemeta_fromspec *)(__experimental_dtype_api_table[3]))
233
+
234
+
235
+ /*
236
+ * *************************************
237
+ * WORKING WITH DTYPES
238
+ * *************************************
239
+ */
240
+
241
+ typedef PyArray_DTypeMeta *__common_dtype(
242
+ PyArray_DTypeMeta *DType1, PyArray_DTypeMeta *DType2);
243
+ #define PyArray_CommonDType \
244
+ ((__common_dtype *)(__experimental_dtype_api_table[4]))
245
+
246
+
247
+ typedef PyArray_DTypeMeta *__promote_dtype_sequence(
248
+ npy_intp num, PyArray_DTypeMeta *DTypes[]);
249
+ #define PyArray_PromoteDTypeSequence \
250
+ ((__promote_dtype_sequence *)(__experimental_dtype_api_table[5]))
251
+
252
+
253
+ typedef PyArray_Descr *__get_default_descr(
254
+ PyArray_DTypeMeta *DType);
255
+ #define _PyArray_GetDefaultDescr \
256
+ ((__get_default_descr *)(__experimental_dtype_api_table[6]))
257
+
258
+ static inline PyArray_Descr *
259
+ PyArray_GetDefaultDescr(PyArray_DTypeMeta *DType)
260
+ {
261
+ if (DType->singleton != NULL) {
262
+ Py_INCREF(DType->singleton);
263
+ return DType->singleton;
264
+ }
265
+ return _PyArray_GetDefaultDescr(DType);
266
+ }
267
+
268
+
269
+ /*
270
+ * NumPy's builtin DTypes:
271
+ */
272
+ #define PyArray_BoolDType (*(PyArray_DTypeMeta *)__experimental_dtype_api_table[10])
273
+ /* Integers */
274
+ #define PyArray_ByteDType (*(PyArray_DTypeMeta *)__experimental_dtype_api_table[11])
275
+ #define PyArray_UByteDType (*(PyArray_DTypeMeta *)__experimental_dtype_api_table[12])
276
+ #define PyArray_ShortDType (*(PyArray_DTypeMeta *)__experimental_dtype_api_table[13])
277
+ #define PyArray_UShortDType (*(PyArray_DTypeMeta *)__experimental_dtype_api_table[14])
278
+ #define PyArray_IntDType (*(PyArray_DTypeMeta *)__experimental_dtype_api_table[15])
279
+ #define PyArray_UIntDType (*(PyArray_DTypeMeta *)__experimental_dtype_api_table[16])
280
+ #define PyArray_LongDType (*(PyArray_DTypeMeta *)__experimental_dtype_api_table[17])
281
+ #define PyArray_ULongDType (*(PyArray_DTypeMeta *)__experimental_dtype_api_table[18])
282
+ #define PyArray_LongLongDType (*(PyArray_DTypeMeta *)__experimental_dtype_api_table[19])
283
+ #define PyArray_ULongLongDType (*(PyArray_DTypeMeta *)__experimental_dtype_api_table[20])
284
+ /* Integer aliases */
285
+ #define PyArray_Int8Type (*(PyArray_DTypeMeta *)__experimental_dtype_api_table[21])
286
+ #define PyArray_UInt8DType (*(PyArray_DTypeMeta *)__experimental_dtype_api_table[22])
287
+ #define PyArray_Int16DType (*(PyArray_DTypeMeta *)__experimental_dtype_api_table[23])
288
+ #define PyArray_UInt16DType (*(PyArray_DTypeMeta *)__experimental_dtype_api_table[24])
289
+ #define PyArray_Int32DType (*(PyArray_DTypeMeta *)__experimental_dtype_api_table[25])
290
+ #define PyArray_UInt32DType (*(PyArray_DTypeMeta *)__experimental_dtype_api_table[26])
291
+ #define PyArray_Int64DType (*(PyArray_DTypeMeta *)__experimental_dtype_api_table[27])
292
+ #define PyArray_UInt64DType (*(PyArray_DTypeMeta *)__experimental_dtype_api_table[28])
293
+ #define PyArray_IntpDType (*(PyArray_DTypeMeta *)__experimental_dtype_api_table[29])
294
+ #define PyArray_UIntpDType (*(PyArray_DTypeMeta *)__experimental_dtype_api_table[30])
295
+ /* Floats */
296
+ #define PyArray_HalfType (*(PyArray_DTypeMeta *)__experimental_dtype_api_table[31])
297
+ #define PyArray_FloatDType (*(PyArray_DTypeMeta *)__experimental_dtype_api_table[32])
298
+ #define PyArray_DoubleDType (*(PyArray_DTypeMeta *)__experimental_dtype_api_table[33])
299
+ #define PyArray_LongDoubleDType (*(PyArray_DTypeMeta *)__experimental_dtype_api_table[34])
300
+ /* Complex */
301
+ #define PyArray_CFloatDType (*(PyArray_DTypeMeta *)__experimental_dtype_api_table[35])
302
+ #define PyArray_CDoubleDType (*(PyArray_DTypeMeta *)__experimental_dtype_api_table[36])
303
+ #define PyArray_CLongDoubleDType (*(PyArray_DTypeMeta *)__experimental_dtype_api_table[37])
304
+ /* String/Bytes */
305
+ #define PyArray_StringDType (*(PyArray_DTypeMeta *)__experimental_dtype_api_table[38])
306
+ #define PyArray_UnicodeDType (*(PyArray_DTypeMeta *)__experimental_dtype_api_table[39])
307
+ /* Datetime/Timedelta */
308
+ #define PyArray_DatetimeDType (*(PyArray_DTypeMeta *)__experimental_dtype_api_table[40])
309
+ #define PyArray_TimedeltaDType (*(PyArray_DTypeMeta *)__experimental_dtype_api_table[41])
310
+ /* Object/Void */
311
+ #define PyArray_ObjectDType (*(PyArray_DTypeMeta *)__experimental_dtype_api_table[42])
312
+ #define PyArray_VoidDType (*(PyArray_DTypeMeta *)__experimental_dtype_api_table[43])
313
+
314
+ /*
315
+ * ********************************
316
+ * Initialization
317
+ * ********************************
318
+ *
319
+ * Import the experimental API, the version must match the one defined in
320
+ * the header to ensure changes are taken into account. NumPy will further
321
+ * runtime-check this.
322
+ * You must call this function to use the symbols defined in this file.
323
+ */
324
+ #if !defined(NO_IMPORT) && !defined(NO_IMPORT_ARRAY)
325
+
326
+ static int
327
+ import_experimental_dtype_api(int version)
328
+ {
329
+ if (version != __EXPERIMENTAL_DTYPE_API_VERSION) {
330
+ PyErr_Format(PyExc_RuntimeError,
331
+ "DType API version %d did not match header version %d. Please "
332
+ "update the import statement and check for API changes.",
333
+ version, __EXPERIMENTAL_DTYPE_API_VERSION);
334
+ return -1;
335
+ }
336
+ if (__experimental_dtype_api_table != __uninitialized_table) {
337
+ /* already imported. */
338
+ return 0;
339
+ }
340
+
341
+ PyObject *multiarray = PyImport_ImportModule("numpy.core._multiarray_umath");
342
+ if (multiarray == NULL) {
343
+ return -1;
344
+ }
345
+
346
+ PyObject *api = PyObject_CallMethod(multiarray,
347
+ "_get_experimental_dtype_api", "i", version);
348
+ Py_DECREF(multiarray);
349
+ if (api == NULL) {
350
+ return -1;
351
+ }
352
+ __experimental_dtype_api_table = (void **)PyCapsule_GetPointer(api,
353
+ "experimental_dtype_api_table");
354
+ Py_DECREF(api);
355
+
356
+ if (__experimental_dtype_api_table == NULL) {
357
+ __experimental_dtype_api_table = __uninitialized_table;
358
+ return -1;
359
+ }
360
+ return 0;
361
+ }
362
+
363
+ #endif /* !defined(NO_IMPORT) && !defined(NO_IMPORT_ARRAY) */
364
+
365
+ #endif /* NUMPY_CORE_INCLUDE_NUMPY_EXPERIMENTAL_DTYPE_API_H_ */
evalkit_tf437/lib/python3.10/site-packages/numpy/core/include/numpy/npy_3kcompat.h ADDED
@@ -0,0 +1,595 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ /*
2
+ * This is a convenience header file providing compatibility utilities
3
+ * for supporting different minor versions of Python 3.
4
+ * It was originally used to support the transition from Python 2,
5
+ * hence the "3k" naming.
6
+ *
7
+ * If you want to use this for your own projects, it's recommended to make a
8
+ * copy of it. Although the stuff below is unlikely to change, we don't provide
9
+ * strong backwards compatibility guarantees at the moment.
10
+ */
11
+
12
+ #ifndef NUMPY_CORE_INCLUDE_NUMPY_NPY_3KCOMPAT_H_
13
+ #define NUMPY_CORE_INCLUDE_NUMPY_NPY_3KCOMPAT_H_
14
+
15
+ #include <Python.h>
16
+ #include <stdio.h>
17
+
18
+ #ifndef NPY_PY3K
19
+ #define NPY_PY3K 1
20
+ #endif
21
+
22
+ #include "numpy/npy_common.h"
23
+ #include "numpy/ndarrayobject.h"
24
+
25
+ #ifdef __cplusplus
26
+ extern "C" {
27
+ #endif
28
+
29
+ /*
30
+ * PyInt -> PyLong
31
+ */
32
+
33
+
34
+ /*
35
+ * This is a renamed copy of the Python non-limited API function _PyLong_AsInt. It is
36
+ * included here because it is missing from the PyPy API. It completes the PyLong_As*
37
+ * group of functions and can be useful in replacing PyInt_Check.
38
+ */
39
+ static inline int
40
+ Npy__PyLong_AsInt(PyObject *obj)
41
+ {
42
+ int overflow;
43
+ long result = PyLong_AsLongAndOverflow(obj, &overflow);
44
+
45
+ /* INT_MAX and INT_MIN are defined in Python.h */
46
+ if (overflow || result > INT_MAX || result < INT_MIN) {
47
+ /* XXX: could be cute and give a different
48
+ message for overflow == -1 */
49
+ PyErr_SetString(PyExc_OverflowError,
50
+ "Python int too large to convert to C int");
51
+ return -1;
52
+ }
53
+ return (int)result;
54
+ }
55
+
56
+
57
+ #if defined(NPY_PY3K)
58
+ /* Return True only if the long fits in a C long */
59
+ static inline int PyInt_Check(PyObject *op) {
60
+ int overflow = 0;
61
+ if (!PyLong_Check(op)) {
62
+ return 0;
63
+ }
64
+ PyLong_AsLongAndOverflow(op, &overflow);
65
+ return (overflow == 0);
66
+ }
67
+
68
+
69
+ #define PyInt_FromLong PyLong_FromLong
70
+ #define PyInt_AsLong PyLong_AsLong
71
+ #define PyInt_AS_LONG PyLong_AsLong
72
+ #define PyInt_AsSsize_t PyLong_AsSsize_t
73
+ #define PyNumber_Int PyNumber_Long
74
+
75
+ /* NOTE:
76
+ *
77
+ * Since the PyLong type is very different from the fixed-range PyInt,
78
+ * we don't define PyInt_Type -> PyLong_Type.
79
+ */
80
+ #endif /* NPY_PY3K */
81
+
82
+ /* Py3 changes PySlice_GetIndicesEx' first argument's type to PyObject* */
83
+ #ifdef NPY_PY3K
84
+ # define NpySlice_GetIndicesEx PySlice_GetIndicesEx
85
+ #else
86
+ # define NpySlice_GetIndicesEx(op, nop, start, end, step, slicelength) \
87
+ PySlice_GetIndicesEx((PySliceObject *)op, nop, start, end, step, slicelength)
88
+ #endif
89
+
90
+ #if PY_VERSION_HEX < 0x030900a4
91
+ /* Introduced in https://github.com/python/cpython/commit/d2ec81a8c99796b51fb8c49b77a7fe369863226f */
92
+ #define Py_SET_TYPE(obj, type) ((Py_TYPE(obj) = (type)), (void)0)
93
+ /* Introduced in https://github.com/python/cpython/commit/b10dc3e7a11fcdb97e285882eba6da92594f90f9 */
94
+ #define Py_SET_SIZE(obj, size) ((Py_SIZE(obj) = (size)), (void)0)
95
+ /* Introduced in https://github.com/python/cpython/commit/c86a11221df7e37da389f9c6ce6e47ea22dc44ff */
96
+ #define Py_SET_REFCNT(obj, refcnt) ((Py_REFCNT(obj) = (refcnt)), (void)0)
97
+ #endif
98
+
99
+
100
+ #define Npy_EnterRecursiveCall(x) Py_EnterRecursiveCall(x)
101
+
102
+ /*
103
+ * PyString -> PyBytes
104
+ */
105
+
106
+ #if defined(NPY_PY3K)
107
+
108
+ #define PyString_Type PyBytes_Type
109
+ #define PyString_Check PyBytes_Check
110
+ #define PyStringObject PyBytesObject
111
+ #define PyString_FromString PyBytes_FromString
112
+ #define PyString_FromStringAndSize PyBytes_FromStringAndSize
113
+ #define PyString_AS_STRING PyBytes_AS_STRING
114
+ #define PyString_AsStringAndSize PyBytes_AsStringAndSize
115
+ #define PyString_FromFormat PyBytes_FromFormat
116
+ #define PyString_Concat PyBytes_Concat
117
+ #define PyString_ConcatAndDel PyBytes_ConcatAndDel
118
+ #define PyString_AsString PyBytes_AsString
119
+ #define PyString_GET_SIZE PyBytes_GET_SIZE
120
+ #define PyString_Size PyBytes_Size
121
+
122
+ #define PyUString_Type PyUnicode_Type
123
+ #define PyUString_Check PyUnicode_Check
124
+ #define PyUStringObject PyUnicodeObject
125
+ #define PyUString_FromString PyUnicode_FromString
126
+ #define PyUString_FromStringAndSize PyUnicode_FromStringAndSize
127
+ #define PyUString_FromFormat PyUnicode_FromFormat
128
+ #define PyUString_Concat PyUnicode_Concat2
129
+ #define PyUString_ConcatAndDel PyUnicode_ConcatAndDel
130
+ #define PyUString_GET_SIZE PyUnicode_GET_SIZE
131
+ #define PyUString_Size PyUnicode_Size
132
+ #define PyUString_InternFromString PyUnicode_InternFromString
133
+ #define PyUString_Format PyUnicode_Format
134
+
135
+ #define PyBaseString_Check(obj) (PyUnicode_Check(obj))
136
+
137
+ #else
138
+
139
+ #define PyBytes_Type PyString_Type
140
+ #define PyBytes_Check PyString_Check
141
+ #define PyBytesObject PyStringObject
142
+ #define PyBytes_FromString PyString_FromString
143
+ #define PyBytes_FromStringAndSize PyString_FromStringAndSize
144
+ #define PyBytes_AS_STRING PyString_AS_STRING
145
+ #define PyBytes_AsStringAndSize PyString_AsStringAndSize
146
+ #define PyBytes_FromFormat PyString_FromFormat
147
+ #define PyBytes_Concat PyString_Concat
148
+ #define PyBytes_ConcatAndDel PyString_ConcatAndDel
149
+ #define PyBytes_AsString PyString_AsString
150
+ #define PyBytes_GET_SIZE PyString_GET_SIZE
151
+ #define PyBytes_Size PyString_Size
152
+
153
+ #define PyUString_Type PyString_Type
154
+ #define PyUString_Check PyString_Check
155
+ #define PyUStringObject PyStringObject
156
+ #define PyUString_FromString PyString_FromString
157
+ #define PyUString_FromStringAndSize PyString_FromStringAndSize
158
+ #define PyUString_FromFormat PyString_FromFormat
159
+ #define PyUString_Concat PyString_Concat
160
+ #define PyUString_ConcatAndDel PyString_ConcatAndDel
161
+ #define PyUString_GET_SIZE PyString_GET_SIZE
162
+ #define PyUString_Size PyString_Size
163
+ #define PyUString_InternFromString PyString_InternFromString
164
+ #define PyUString_Format PyString_Format
165
+
166
+ #define PyBaseString_Check(obj) (PyBytes_Check(obj) || PyUnicode_Check(obj))
167
+
168
+ #endif /* NPY_PY3K */
169
+
170
+ /*
171
+ * Macros to protect CRT calls against instant termination when passed an
172
+ * invalid parameter (https://bugs.python.org/issue23524).
173
+ */
174
+ #if defined _MSC_VER && _MSC_VER >= 1900
175
+
176
+ #include <stdlib.h>
177
+
178
+ extern _invalid_parameter_handler _Py_silent_invalid_parameter_handler;
179
+ #define NPY_BEGIN_SUPPRESS_IPH { _invalid_parameter_handler _Py_old_handler = \
180
+ _set_thread_local_invalid_parameter_handler(_Py_silent_invalid_parameter_handler);
181
+ #define NPY_END_SUPPRESS_IPH _set_thread_local_invalid_parameter_handler(_Py_old_handler); }
182
+
183
+ #else
184
+
185
+ #define NPY_BEGIN_SUPPRESS_IPH
186
+ #define NPY_END_SUPPRESS_IPH
187
+
188
+ #endif /* _MSC_VER >= 1900 */
189
+
190
+
191
+ static inline void
192
+ PyUnicode_ConcatAndDel(PyObject **left, PyObject *right)
193
+ {
194
+ Py_SETREF(*left, PyUnicode_Concat(*left, right));
195
+ Py_DECREF(right);
196
+ }
197
+
198
+ static inline void
199
+ PyUnicode_Concat2(PyObject **left, PyObject *right)
200
+ {
201
+ Py_SETREF(*left, PyUnicode_Concat(*left, right));
202
+ }
203
+
204
+ /*
205
+ * PyFile_* compatibility
206
+ */
207
+
208
+ /*
209
+ * Get a FILE* handle to the file represented by the Python object
210
+ */
211
+ static inline FILE*
212
+ npy_PyFile_Dup2(PyObject *file, char *mode, npy_off_t *orig_pos)
213
+ {
214
+ int fd, fd2, unbuf;
215
+ Py_ssize_t fd2_tmp;
216
+ PyObject *ret, *os, *io, *io_raw;
217
+ npy_off_t pos;
218
+ FILE *handle;
219
+
220
+ /* For Python 2 PyFileObject, use PyFile_AsFile */
221
+ #if !defined(NPY_PY3K)
222
+ if (PyFile_Check(file)) {
223
+ return PyFile_AsFile(file);
224
+ }
225
+ #endif
226
+
227
+ /* Flush first to ensure things end up in the file in the correct order */
228
+ ret = PyObject_CallMethod(file, "flush", "");
229
+ if (ret == NULL) {
230
+ return NULL;
231
+ }
232
+ Py_DECREF(ret);
233
+ fd = PyObject_AsFileDescriptor(file);
234
+ if (fd == -1) {
235
+ return NULL;
236
+ }
237
+
238
+ /*
239
+ * The handle needs to be dup'd because we have to call fclose
240
+ * at the end
241
+ */
242
+ os = PyImport_ImportModule("os");
243
+ if (os == NULL) {
244
+ return NULL;
245
+ }
246
+ ret = PyObject_CallMethod(os, "dup", "i", fd);
247
+ Py_DECREF(os);
248
+ if (ret == NULL) {
249
+ return NULL;
250
+ }
251
+ fd2_tmp = PyNumber_AsSsize_t(ret, PyExc_IOError);
252
+ Py_DECREF(ret);
253
+ if (fd2_tmp == -1 && PyErr_Occurred()) {
254
+ return NULL;
255
+ }
256
+ if (fd2_tmp < INT_MIN || fd2_tmp > INT_MAX) {
257
+ PyErr_SetString(PyExc_IOError,
258
+ "Getting an 'int' from os.dup() failed");
259
+ return NULL;
260
+ }
261
+ fd2 = (int)fd2_tmp;
262
+
263
+ /* Convert to FILE* handle */
264
+ #ifdef _WIN32
265
+ NPY_BEGIN_SUPPRESS_IPH
266
+ handle = _fdopen(fd2, mode);
267
+ NPY_END_SUPPRESS_IPH
268
+ #else
269
+ handle = fdopen(fd2, mode);
270
+ #endif
271
+ if (handle == NULL) {
272
+ PyErr_SetString(PyExc_IOError,
273
+ "Getting a FILE* from a Python file object via "
274
+ "_fdopen failed. If you built NumPy, you probably "
275
+ "linked with the wrong debug/release runtime");
276
+ return NULL;
277
+ }
278
+
279
+ /* Record the original raw file handle position */
280
+ *orig_pos = npy_ftell(handle);
281
+ if (*orig_pos == -1) {
282
+ /* The io module is needed to determine if buffering is used */
283
+ io = PyImport_ImportModule("io");
284
+ if (io == NULL) {
285
+ fclose(handle);
286
+ return NULL;
287
+ }
288
+ /* File object instances of RawIOBase are unbuffered */
289
+ io_raw = PyObject_GetAttrString(io, "RawIOBase");
290
+ Py_DECREF(io);
291
+ if (io_raw == NULL) {
292
+ fclose(handle);
293
+ return NULL;
294
+ }
295
+ unbuf = PyObject_IsInstance(file, io_raw);
296
+ Py_DECREF(io_raw);
297
+ if (unbuf == 1) {
298
+ /* Succeed if the IO is unbuffered */
299
+ return handle;
300
+ }
301
+ else {
302
+ PyErr_SetString(PyExc_IOError, "obtaining file position failed");
303
+ fclose(handle);
304
+ return NULL;
305
+ }
306
+ }
307
+
308
+ /* Seek raw handle to the Python-side position */
309
+ ret = PyObject_CallMethod(file, "tell", "");
310
+ if (ret == NULL) {
311
+ fclose(handle);
312
+ return NULL;
313
+ }
314
+ pos = PyLong_AsLongLong(ret);
315
+ Py_DECREF(ret);
316
+ if (PyErr_Occurred()) {
317
+ fclose(handle);
318
+ return NULL;
319
+ }
320
+ if (npy_fseek(handle, pos, SEEK_SET) == -1) {
321
+ PyErr_SetString(PyExc_IOError, "seeking file failed");
322
+ fclose(handle);
323
+ return NULL;
324
+ }
325
+ return handle;
326
+ }
327
+
328
+ /*
329
+ * Close the dup-ed file handle, and seek the Python one to the current position
330
+ */
331
+ static inline int
332
+ npy_PyFile_DupClose2(PyObject *file, FILE* handle, npy_off_t orig_pos)
333
+ {
334
+ int fd, unbuf;
335
+ PyObject *ret, *io, *io_raw;
336
+ npy_off_t position;
337
+
338
+ /* For Python 2 PyFileObject, do nothing */
339
+ #if !defined(NPY_PY3K)
340
+ if (PyFile_Check(file)) {
341
+ return 0;
342
+ }
343
+ #endif
344
+
345
+ position = npy_ftell(handle);
346
+
347
+ /* Close the FILE* handle */
348
+ fclose(handle);
349
+
350
+ /*
351
+ * Restore original file handle position, in order to not confuse
352
+ * Python-side data structures
353
+ */
354
+ fd = PyObject_AsFileDescriptor(file);
355
+ if (fd == -1) {
356
+ return -1;
357
+ }
358
+
359
+ if (npy_lseek(fd, orig_pos, SEEK_SET) == -1) {
360
+
361
+ /* The io module is needed to determine if buffering is used */
362
+ io = PyImport_ImportModule("io");
363
+ if (io == NULL) {
364
+ return -1;
365
+ }
366
+ /* File object instances of RawIOBase are unbuffered */
367
+ io_raw = PyObject_GetAttrString(io, "RawIOBase");
368
+ Py_DECREF(io);
369
+ if (io_raw == NULL) {
370
+ return -1;
371
+ }
372
+ unbuf = PyObject_IsInstance(file, io_raw);
373
+ Py_DECREF(io_raw);
374
+ if (unbuf == 1) {
375
+ /* Succeed if the IO is unbuffered */
376
+ return 0;
377
+ }
378
+ else {
379
+ PyErr_SetString(PyExc_IOError, "seeking file failed");
380
+ return -1;
381
+ }
382
+ }
383
+
384
+ if (position == -1) {
385
+ PyErr_SetString(PyExc_IOError, "obtaining file position failed");
386
+ return -1;
387
+ }
388
+
389
+ /* Seek Python-side handle to the FILE* handle position */
390
+ ret = PyObject_CallMethod(file, "seek", NPY_OFF_T_PYFMT "i", position, 0);
391
+ if (ret == NULL) {
392
+ return -1;
393
+ }
394
+ Py_DECREF(ret);
395
+ return 0;
396
+ }
397
+
398
+ static inline int
399
+ npy_PyFile_Check(PyObject *file)
400
+ {
401
+ int fd;
402
+ /* For Python 2, check if it is a PyFileObject */
403
+ #if !defined(NPY_PY3K)
404
+ if (PyFile_Check(file)) {
405
+ return 1;
406
+ }
407
+ #endif
408
+ fd = PyObject_AsFileDescriptor(file);
409
+ if (fd == -1) {
410
+ PyErr_Clear();
411
+ return 0;
412
+ }
413
+ return 1;
414
+ }
415
+
416
+ static inline PyObject*
417
+ npy_PyFile_OpenFile(PyObject *filename, const char *mode)
418
+ {
419
+ PyObject *open;
420
+ open = PyDict_GetItemString(PyEval_GetBuiltins(), "open");
421
+ if (open == NULL) {
422
+ return NULL;
423
+ }
424
+ return PyObject_CallFunction(open, "Os", filename, mode);
425
+ }
426
+
427
+ static inline int
428
+ npy_PyFile_CloseFile(PyObject *file)
429
+ {
430
+ PyObject *ret;
431
+
432
+ ret = PyObject_CallMethod(file, "close", NULL);
433
+ if (ret == NULL) {
434
+ return -1;
435
+ }
436
+ Py_DECREF(ret);
437
+ return 0;
438
+ }
439
+
440
+
441
+ /* This is a copy of _PyErr_ChainExceptions
442
+ */
443
+ static inline void
444
+ npy_PyErr_ChainExceptions(PyObject *exc, PyObject *val, PyObject *tb)
445
+ {
446
+ if (exc == NULL)
447
+ return;
448
+
449
+ if (PyErr_Occurred()) {
450
+ /* only py3 supports this anyway */
451
+ #ifdef NPY_PY3K
452
+ PyObject *exc2, *val2, *tb2;
453
+ PyErr_Fetch(&exc2, &val2, &tb2);
454
+ PyErr_NormalizeException(&exc, &val, &tb);
455
+ if (tb != NULL) {
456
+ PyException_SetTraceback(val, tb);
457
+ Py_DECREF(tb);
458
+ }
459
+ Py_DECREF(exc);
460
+ PyErr_NormalizeException(&exc2, &val2, &tb2);
461
+ PyException_SetContext(val2, val);
462
+ PyErr_Restore(exc2, val2, tb2);
463
+ #endif
464
+ }
465
+ else {
466
+ PyErr_Restore(exc, val, tb);
467
+ }
468
+ }
469
+
470
+
471
+ /* This is a copy of _PyErr_ChainExceptions, with:
472
+ * - a minimal implementation for python 2
473
+ * - __cause__ used instead of __context__
474
+ */
475
+ static inline void
476
+ npy_PyErr_ChainExceptionsCause(PyObject *exc, PyObject *val, PyObject *tb)
477
+ {
478
+ if (exc == NULL)
479
+ return;
480
+
481
+ if (PyErr_Occurred()) {
482
+ /* only py3 supports this anyway */
483
+ #ifdef NPY_PY3K
484
+ PyObject *exc2, *val2, *tb2;
485
+ PyErr_Fetch(&exc2, &val2, &tb2);
486
+ PyErr_NormalizeException(&exc, &val, &tb);
487
+ if (tb != NULL) {
488
+ PyException_SetTraceback(val, tb);
489
+ Py_DECREF(tb);
490
+ }
491
+ Py_DECREF(exc);
492
+ PyErr_NormalizeException(&exc2, &val2, &tb2);
493
+ PyException_SetCause(val2, val);
494
+ PyErr_Restore(exc2, val2, tb2);
495
+ #endif
496
+ }
497
+ else {
498
+ PyErr_Restore(exc, val, tb);
499
+ }
500
+ }
501
+
502
+ /*
503
+ * PyObject_Cmp
504
+ */
505
+ #if defined(NPY_PY3K)
506
+ static inline int
507
+ PyObject_Cmp(PyObject *i1, PyObject *i2, int *cmp)
508
+ {
509
+ int v;
510
+ v = PyObject_RichCompareBool(i1, i2, Py_LT);
511
+ if (v == 1) {
512
+ *cmp = -1;
513
+ return 1;
514
+ }
515
+ else if (v == -1) {
516
+ return -1;
517
+ }
518
+
519
+ v = PyObject_RichCompareBool(i1, i2, Py_GT);
520
+ if (v == 1) {
521
+ *cmp = 1;
522
+ return 1;
523
+ }
524
+ else if (v == -1) {
525
+ return -1;
526
+ }
527
+
528
+ v = PyObject_RichCompareBool(i1, i2, Py_EQ);
529
+ if (v == 1) {
530
+ *cmp = 0;
531
+ return 1;
532
+ }
533
+ else {
534
+ *cmp = 0;
535
+ return -1;
536
+ }
537
+ }
538
+ #endif
539
+
540
+ /*
541
+ * PyCObject functions adapted to PyCapsules.
542
+ *
543
+ * The main job here is to get rid of the improved error handling
544
+ * of PyCapsules. It's a shame...
545
+ */
546
+ static inline PyObject *
547
+ NpyCapsule_FromVoidPtr(void *ptr, void (*dtor)(PyObject *))
548
+ {
549
+ PyObject *ret = PyCapsule_New(ptr, NULL, dtor);
550
+ if (ret == NULL) {
551
+ PyErr_Clear();
552
+ }
553
+ return ret;
554
+ }
555
+
556
+ static inline PyObject *
557
+ NpyCapsule_FromVoidPtrAndDesc(void *ptr, void* context, void (*dtor)(PyObject *))
558
+ {
559
+ PyObject *ret = NpyCapsule_FromVoidPtr(ptr, dtor);
560
+ if (ret != NULL && PyCapsule_SetContext(ret, context) != 0) {
561
+ PyErr_Clear();
562
+ Py_DECREF(ret);
563
+ ret = NULL;
564
+ }
565
+ return ret;
566
+ }
567
+
568
+ static inline void *
569
+ NpyCapsule_AsVoidPtr(PyObject *obj)
570
+ {
571
+ void *ret = PyCapsule_GetPointer(obj, NULL);
572
+ if (ret == NULL) {
573
+ PyErr_Clear();
574
+ }
575
+ return ret;
576
+ }
577
+
578
+ static inline void *
579
+ NpyCapsule_GetDesc(PyObject *obj)
580
+ {
581
+ return PyCapsule_GetContext(obj);
582
+ }
583
+
584
+ static inline int
585
+ NpyCapsule_Check(PyObject *ptr)
586
+ {
587
+ return PyCapsule_CheckExact(ptr);
588
+ }
589
+
590
+ #ifdef __cplusplus
591
+ }
592
+ #endif
593
+
594
+
595
+ #endif /* NUMPY_CORE_INCLUDE_NUMPY_NPY_3KCOMPAT_H_ */
evalkit_tf437/lib/python3.10/site-packages/numpy/core/include/numpy/npy_cpu.h ADDED
@@ -0,0 +1,129 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ /*
2
+ * This set (target) cpu specific macros:
3
+ * - Possible values:
4
+ * NPY_CPU_X86
5
+ * NPY_CPU_AMD64
6
+ * NPY_CPU_PPC
7
+ * NPY_CPU_PPC64
8
+ * NPY_CPU_PPC64LE
9
+ * NPY_CPU_SPARC
10
+ * NPY_CPU_S390
11
+ * NPY_CPU_IA64
12
+ * NPY_CPU_HPPA
13
+ * NPY_CPU_ALPHA
14
+ * NPY_CPU_ARMEL
15
+ * NPY_CPU_ARMEB
16
+ * NPY_CPU_SH_LE
17
+ * NPY_CPU_SH_BE
18
+ * NPY_CPU_ARCEL
19
+ * NPY_CPU_ARCEB
20
+ * NPY_CPU_RISCV64
21
+ * NPY_CPU_LOONGARCH
22
+ * NPY_CPU_WASM
23
+ */
24
+ #ifndef NUMPY_CORE_INCLUDE_NUMPY_NPY_CPU_H_
25
+ #define NUMPY_CORE_INCLUDE_NUMPY_NPY_CPU_H_
26
+
27
+ #include "numpyconfig.h"
28
+
29
+ #if defined( __i386__ ) || defined(i386) || defined(_M_IX86)
30
+ /*
31
+ * __i386__ is defined by gcc and Intel compiler on Linux,
32
+ * _M_IX86 by VS compiler,
33
+ * i386 by Sun compilers on opensolaris at least
34
+ */
35
+ #define NPY_CPU_X86
36
+ #elif defined(__x86_64__) || defined(__amd64__) || defined(__x86_64) || defined(_M_AMD64)
37
+ /*
38
+ * both __x86_64__ and __amd64__ are defined by gcc
39
+ * __x86_64 defined by sun compiler on opensolaris at least
40
+ * _M_AMD64 defined by MS compiler
41
+ */
42
+ #define NPY_CPU_AMD64
43
+ #elif defined(__powerpc64__) && defined(__LITTLE_ENDIAN__)
44
+ #define NPY_CPU_PPC64LE
45
+ #elif defined(__powerpc64__) && defined(__BIG_ENDIAN__)
46
+ #define NPY_CPU_PPC64
47
+ #elif defined(__ppc__) || defined(__powerpc__) || defined(_ARCH_PPC)
48
+ /*
49
+ * __ppc__ is defined by gcc, I remember having seen __powerpc__ once,
50
+ * but can't find it ATM
51
+ * _ARCH_PPC is used by at least gcc on AIX
52
+ * As __powerpc__ and _ARCH_PPC are also defined by PPC64 check
53
+ * for those specifically first before defaulting to ppc
54
+ */
55
+ #define NPY_CPU_PPC
56
+ #elif defined(__sparc__) || defined(__sparc)
57
+ /* __sparc__ is defined by gcc and Forte (e.g. Sun) compilers */
58
+ #define NPY_CPU_SPARC
59
+ #elif defined(__s390__)
60
+ #define NPY_CPU_S390
61
+ #elif defined(__ia64)
62
+ #define NPY_CPU_IA64
63
+ #elif defined(__hppa)
64
+ #define NPY_CPU_HPPA
65
+ #elif defined(__alpha__)
66
+ #define NPY_CPU_ALPHA
67
+ #elif defined(__arm__) || defined(__aarch64__) || defined(_M_ARM64)
68
+ /* _M_ARM64 is defined in MSVC for ARM64 compilation on Windows */
69
+ #if defined(__ARMEB__) || defined(__AARCH64EB__)
70
+ #if defined(__ARM_32BIT_STATE)
71
+ #define NPY_CPU_ARMEB_AARCH32
72
+ #elif defined(__ARM_64BIT_STATE)
73
+ #define NPY_CPU_ARMEB_AARCH64
74
+ #else
75
+ #define NPY_CPU_ARMEB
76
+ #endif
77
+ #elif defined(__ARMEL__) || defined(__AARCH64EL__) || defined(_M_ARM64)
78
+ #if defined(__ARM_32BIT_STATE)
79
+ #define NPY_CPU_ARMEL_AARCH32
80
+ #elif defined(__ARM_64BIT_STATE) || defined(_M_ARM64) || defined(__AARCH64EL__)
81
+ #define NPY_CPU_ARMEL_AARCH64
82
+ #else
83
+ #define NPY_CPU_ARMEL
84
+ #endif
85
+ #else
86
+ # error Unknown ARM CPU, please report this to numpy maintainers with \
87
+ information about your platform (OS, CPU and compiler)
88
+ #endif
89
+ #elif defined(__sh__) && defined(__LITTLE_ENDIAN__)
90
+ #define NPY_CPU_SH_LE
91
+ #elif defined(__sh__) && defined(__BIG_ENDIAN__)
92
+ #define NPY_CPU_SH_BE
93
+ #elif defined(__MIPSEL__)
94
+ #define NPY_CPU_MIPSEL
95
+ #elif defined(__MIPSEB__)
96
+ #define NPY_CPU_MIPSEB
97
+ #elif defined(__or1k__)
98
+ #define NPY_CPU_OR1K
99
+ #elif defined(__mc68000__)
100
+ #define NPY_CPU_M68K
101
+ #elif defined(__arc__) && defined(__LITTLE_ENDIAN__)
102
+ #define NPY_CPU_ARCEL
103
+ #elif defined(__arc__) && defined(__BIG_ENDIAN__)
104
+ #define NPY_CPU_ARCEB
105
+ #elif defined(__riscv) && defined(__riscv_xlen) && __riscv_xlen == 64
106
+ #define NPY_CPU_RISCV64
107
+ #elif defined(__loongarch__)
108
+ #define NPY_CPU_LOONGARCH
109
+ #elif defined(__EMSCRIPTEN__)
110
+ /* __EMSCRIPTEN__ is defined by emscripten: an LLVM-to-Web compiler */
111
+ #define NPY_CPU_WASM
112
+ #else
113
+ #error Unknown CPU, please report this to numpy maintainers with \
114
+ information about your platform (OS, CPU and compiler)
115
+ #endif
116
+
117
+ /*
118
+ * Except for the following architectures, memory access is limited to the natural
119
+ * alignment of data types otherwise it may lead to bus error or performance regression.
120
+ * For more details about unaligned access, see https://www.kernel.org/doc/Documentation/unaligned-memory-access.txt.
121
+ */
122
+ #if defined(NPY_CPU_X86) || defined(NPY_CPU_AMD64) || defined(__aarch64__) || defined(__powerpc64__)
123
+ #define NPY_ALIGNMENT_REQUIRED 0
124
+ #endif
125
+ #ifndef NPY_ALIGNMENT_REQUIRED
126
+ #define NPY_ALIGNMENT_REQUIRED 1
127
+ #endif
128
+
129
+ #endif /* NUMPY_CORE_INCLUDE_NUMPY_NPY_CPU_H_ */
evalkit_tf437/lib/python3.10/site-packages/numpy/core/include/numpy/npy_interrupt.h ADDED
@@ -0,0 +1,56 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ /*
2
+ * This API is only provided because it is part of publicly exported
3
+ * headers. Its use is considered DEPRECATED, and it will be removed
4
+ * eventually.
5
+ * (This includes the _PyArray_SigintHandler and _PyArray_GetSigintBuf
6
+ * functions which are however, public API, and not headers.)
7
+ *
8
+ * Instead of using these non-threadsafe macros consider periodically
9
+ * querying `PyErr_CheckSignals()` or `PyOS_InterruptOccurred()` will work.
10
+ * Both of these require holding the GIL, although cpython could add a
11
+ * version of `PyOS_InterruptOccurred()` which does not. Such a version
12
+ * actually exists as private API in Python 3.10, and backported to 3.9 and 3.8,
13
+ * see also https://bugs.python.org/issue41037 and
14
+ * https://github.com/python/cpython/pull/20599).
15
+ */
16
+
17
+ #ifndef NUMPY_CORE_INCLUDE_NUMPY_NPY_INTERRUPT_H_
18
+ #define NUMPY_CORE_INCLUDE_NUMPY_NPY_INTERRUPT_H_
19
+
20
+ #ifndef NPY_NO_SIGNAL
21
+
22
+ #include <setjmp.h>
23
+ #include <signal.h>
24
+
25
+ #ifndef sigsetjmp
26
+
27
+ #define NPY_SIGSETJMP(arg1, arg2) setjmp(arg1)
28
+ #define NPY_SIGLONGJMP(arg1, arg2) longjmp(arg1, arg2)
29
+ #define NPY_SIGJMP_BUF jmp_buf
30
+
31
+ #else
32
+
33
+ #define NPY_SIGSETJMP(arg1, arg2) sigsetjmp(arg1, arg2)
34
+ #define NPY_SIGLONGJMP(arg1, arg2) siglongjmp(arg1, arg2)
35
+ #define NPY_SIGJMP_BUF sigjmp_buf
36
+
37
+ #endif
38
+
39
+ # define NPY_SIGINT_ON { \
40
+ PyOS_sighandler_t _npy_sig_save; \
41
+ _npy_sig_save = PyOS_setsig(SIGINT, _PyArray_SigintHandler); \
42
+ if (NPY_SIGSETJMP(*((NPY_SIGJMP_BUF *)_PyArray_GetSigintBuf()), \
43
+ 1) == 0) { \
44
+
45
+ # define NPY_SIGINT_OFF } \
46
+ PyOS_setsig(SIGINT, _npy_sig_save); \
47
+ }
48
+
49
+ #else /* NPY_NO_SIGNAL */
50
+
51
+ #define NPY_SIGINT_ON
52
+ #define NPY_SIGINT_OFF
53
+
54
+ #endif /* HAVE_SIGSETJMP */
55
+
56
+ #endif /* NUMPY_CORE_INCLUDE_NUMPY_NPY_INTERRUPT_H_ */
evalkit_tf437/lib/python3.10/site-packages/numpy/core/include/numpy/npy_no_deprecated_api.h ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ /*
2
+ * This include file is provided for inclusion in Cython *.pyd files where
3
+ * one would like to define the NPY_NO_DEPRECATED_API macro. It can be
4
+ * included by
5
+ *
6
+ * cdef extern from "npy_no_deprecated_api.h": pass
7
+ *
8
+ */
9
+ #ifndef NPY_NO_DEPRECATED_API
10
+
11
+ /* put this check here since there may be multiple includes in C extensions. */
12
+ #if defined(NUMPY_CORE_INCLUDE_NUMPY_NDARRAYTYPES_H_) || \
13
+ defined(NUMPY_CORE_INCLUDE_NUMPY_NPY_DEPRECATED_API_H) || \
14
+ defined(NUMPY_CORE_INCLUDE_NUMPY_OLD_DEFINES_H_)
15
+ #error "npy_no_deprecated_api.h" must be first among numpy includes.
16
+ #else
17
+ #define NPY_NO_DEPRECATED_API NPY_API_VERSION
18
+ #endif
19
+
20
+ #endif /* NPY_NO_DEPRECATED_API */
evalkit_tf437/lib/python3.10/site-packages/numpy/core/include/numpy/old_defines.h ADDED
@@ -0,0 +1,187 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ /* This header is deprecated as of NumPy 1.7 */
2
+ #ifndef NUMPY_CORE_INCLUDE_NUMPY_OLD_DEFINES_H_
3
+ #define NUMPY_CORE_INCLUDE_NUMPY_OLD_DEFINES_H_
4
+
5
+ #if defined(NPY_NO_DEPRECATED_API) && NPY_NO_DEPRECATED_API >= NPY_1_7_API_VERSION
6
+ #error The header "old_defines.h" is deprecated as of NumPy 1.7.
7
+ #endif
8
+
9
+ #define NDARRAY_VERSION NPY_VERSION
10
+
11
+ #define PyArray_MIN_BUFSIZE NPY_MIN_BUFSIZE
12
+ #define PyArray_MAX_BUFSIZE NPY_MAX_BUFSIZE
13
+ #define PyArray_BUFSIZE NPY_BUFSIZE
14
+
15
+ #define PyArray_PRIORITY NPY_PRIORITY
16
+ #define PyArray_SUBTYPE_PRIORITY NPY_PRIORITY
17
+ #define PyArray_NUM_FLOATTYPE NPY_NUM_FLOATTYPE
18
+
19
+ #define NPY_MAX PyArray_MAX
20
+ #define NPY_MIN PyArray_MIN
21
+
22
+ #define PyArray_TYPES NPY_TYPES
23
+ #define PyArray_BOOL NPY_BOOL
24
+ #define PyArray_BYTE NPY_BYTE
25
+ #define PyArray_UBYTE NPY_UBYTE
26
+ #define PyArray_SHORT NPY_SHORT
27
+ #define PyArray_USHORT NPY_USHORT
28
+ #define PyArray_INT NPY_INT
29
+ #define PyArray_UINT NPY_UINT
30
+ #define PyArray_LONG NPY_LONG
31
+ #define PyArray_ULONG NPY_ULONG
32
+ #define PyArray_LONGLONG NPY_LONGLONG
33
+ #define PyArray_ULONGLONG NPY_ULONGLONG
34
+ #define PyArray_HALF NPY_HALF
35
+ #define PyArray_FLOAT NPY_FLOAT
36
+ #define PyArray_DOUBLE NPY_DOUBLE
37
+ #define PyArray_LONGDOUBLE NPY_LONGDOUBLE
38
+ #define PyArray_CFLOAT NPY_CFLOAT
39
+ #define PyArray_CDOUBLE NPY_CDOUBLE
40
+ #define PyArray_CLONGDOUBLE NPY_CLONGDOUBLE
41
+ #define PyArray_OBJECT NPY_OBJECT
42
+ #define PyArray_STRING NPY_STRING
43
+ #define PyArray_UNICODE NPY_UNICODE
44
+ #define PyArray_VOID NPY_VOID
45
+ #define PyArray_DATETIME NPY_DATETIME
46
+ #define PyArray_TIMEDELTA NPY_TIMEDELTA
47
+ #define PyArray_NTYPES NPY_NTYPES
48
+ #define PyArray_NOTYPE NPY_NOTYPE
49
+ #define PyArray_CHAR NPY_CHAR
50
+ #define PyArray_USERDEF NPY_USERDEF
51
+ #define PyArray_NUMUSERTYPES NPY_NUMUSERTYPES
52
+
53
+ #define PyArray_INTP NPY_INTP
54
+ #define PyArray_UINTP NPY_UINTP
55
+
56
+ #define PyArray_INT8 NPY_INT8
57
+ #define PyArray_UINT8 NPY_UINT8
58
+ #define PyArray_INT16 NPY_INT16
59
+ #define PyArray_UINT16 NPY_UINT16
60
+ #define PyArray_INT32 NPY_INT32
61
+ #define PyArray_UINT32 NPY_UINT32
62
+
63
+ #ifdef NPY_INT64
64
+ #define PyArray_INT64 NPY_INT64
65
+ #define PyArray_UINT64 NPY_UINT64
66
+ #endif
67
+
68
+ #ifdef NPY_INT128
69
+ #define PyArray_INT128 NPY_INT128
70
+ #define PyArray_UINT128 NPY_UINT128
71
+ #endif
72
+
73
+ #ifdef NPY_FLOAT16
74
+ #define PyArray_FLOAT16 NPY_FLOAT16
75
+ #define PyArray_COMPLEX32 NPY_COMPLEX32
76
+ #endif
77
+
78
+ #ifdef NPY_FLOAT80
79
+ #define PyArray_FLOAT80 NPY_FLOAT80
80
+ #define PyArray_COMPLEX160 NPY_COMPLEX160
81
+ #endif
82
+
83
+ #ifdef NPY_FLOAT96
84
+ #define PyArray_FLOAT96 NPY_FLOAT96
85
+ #define PyArray_COMPLEX192 NPY_COMPLEX192
86
+ #endif
87
+
88
+ #ifdef NPY_FLOAT128
89
+ #define PyArray_FLOAT128 NPY_FLOAT128
90
+ #define PyArray_COMPLEX256 NPY_COMPLEX256
91
+ #endif
92
+
93
+ #define PyArray_FLOAT32 NPY_FLOAT32
94
+ #define PyArray_COMPLEX64 NPY_COMPLEX64
95
+ #define PyArray_FLOAT64 NPY_FLOAT64
96
+ #define PyArray_COMPLEX128 NPY_COMPLEX128
97
+
98
+
99
+ #define PyArray_TYPECHAR NPY_TYPECHAR
100
+ #define PyArray_BOOLLTR NPY_BOOLLTR
101
+ #define PyArray_BYTELTR NPY_BYTELTR
102
+ #define PyArray_UBYTELTR NPY_UBYTELTR
103
+ #define PyArray_SHORTLTR NPY_SHORTLTR
104
+ #define PyArray_USHORTLTR NPY_USHORTLTR
105
+ #define PyArray_INTLTR NPY_INTLTR
106
+ #define PyArray_UINTLTR NPY_UINTLTR
107
+ #define PyArray_LONGLTR NPY_LONGLTR
108
+ #define PyArray_ULONGLTR NPY_ULONGLTR
109
+ #define PyArray_LONGLONGLTR NPY_LONGLONGLTR
110
+ #define PyArray_ULONGLONGLTR NPY_ULONGLONGLTR
111
+ #define PyArray_HALFLTR NPY_HALFLTR
112
+ #define PyArray_FLOATLTR NPY_FLOATLTR
113
+ #define PyArray_DOUBLELTR NPY_DOUBLELTR
114
+ #define PyArray_LONGDOUBLELTR NPY_LONGDOUBLELTR
115
+ #define PyArray_CFLOATLTR NPY_CFLOATLTR
116
+ #define PyArray_CDOUBLELTR NPY_CDOUBLELTR
117
+ #define PyArray_CLONGDOUBLELTR NPY_CLONGDOUBLELTR
118
+ #define PyArray_OBJECTLTR NPY_OBJECTLTR
119
+ #define PyArray_STRINGLTR NPY_STRINGLTR
120
+ #define PyArray_STRINGLTR2 NPY_STRINGLTR2
121
+ #define PyArray_UNICODELTR NPY_UNICODELTR
122
+ #define PyArray_VOIDLTR NPY_VOIDLTR
123
+ #define PyArray_DATETIMELTR NPY_DATETIMELTR
124
+ #define PyArray_TIMEDELTALTR NPY_TIMEDELTALTR
125
+ #define PyArray_CHARLTR NPY_CHARLTR
126
+ #define PyArray_INTPLTR NPY_INTPLTR
127
+ #define PyArray_UINTPLTR NPY_UINTPLTR
128
+ #define PyArray_GENBOOLLTR NPY_GENBOOLLTR
129
+ #define PyArray_SIGNEDLTR NPY_SIGNEDLTR
130
+ #define PyArray_UNSIGNEDLTR NPY_UNSIGNEDLTR
131
+ #define PyArray_FLOATINGLTR NPY_FLOATINGLTR
132
+ #define PyArray_COMPLEXLTR NPY_COMPLEXLTR
133
+
134
+ #define PyArray_QUICKSORT NPY_QUICKSORT
135
+ #define PyArray_HEAPSORT NPY_HEAPSORT
136
+ #define PyArray_MERGESORT NPY_MERGESORT
137
+ #define PyArray_SORTKIND NPY_SORTKIND
138
+ #define PyArray_NSORTS NPY_NSORTS
139
+
140
+ #define PyArray_NOSCALAR NPY_NOSCALAR
141
+ #define PyArray_BOOL_SCALAR NPY_BOOL_SCALAR
142
+ #define PyArray_INTPOS_SCALAR NPY_INTPOS_SCALAR
143
+ #define PyArray_INTNEG_SCALAR NPY_INTNEG_SCALAR
144
+ #define PyArray_FLOAT_SCALAR NPY_FLOAT_SCALAR
145
+ #define PyArray_COMPLEX_SCALAR NPY_COMPLEX_SCALAR
146
+ #define PyArray_OBJECT_SCALAR NPY_OBJECT_SCALAR
147
+ #define PyArray_SCALARKIND NPY_SCALARKIND
148
+ #define PyArray_NSCALARKINDS NPY_NSCALARKINDS
149
+
150
+ #define PyArray_ANYORDER NPY_ANYORDER
151
+ #define PyArray_CORDER NPY_CORDER
152
+ #define PyArray_FORTRANORDER NPY_FORTRANORDER
153
+ #define PyArray_ORDER NPY_ORDER
154
+
155
+ #define PyDescr_ISBOOL PyDataType_ISBOOL
156
+ #define PyDescr_ISUNSIGNED PyDataType_ISUNSIGNED
157
+ #define PyDescr_ISSIGNED PyDataType_ISSIGNED
158
+ #define PyDescr_ISINTEGER PyDataType_ISINTEGER
159
+ #define PyDescr_ISFLOAT PyDataType_ISFLOAT
160
+ #define PyDescr_ISNUMBER PyDataType_ISNUMBER
161
+ #define PyDescr_ISSTRING PyDataType_ISSTRING
162
+ #define PyDescr_ISCOMPLEX PyDataType_ISCOMPLEX
163
+ #define PyDescr_ISPYTHON PyDataType_ISPYTHON
164
+ #define PyDescr_ISFLEXIBLE PyDataType_ISFLEXIBLE
165
+ #define PyDescr_ISUSERDEF PyDataType_ISUSERDEF
166
+ #define PyDescr_ISEXTENDED PyDataType_ISEXTENDED
167
+ #define PyDescr_ISOBJECT PyDataType_ISOBJECT
168
+ #define PyDescr_HASFIELDS PyDataType_HASFIELDS
169
+
170
+ #define PyArray_LITTLE NPY_LITTLE
171
+ #define PyArray_BIG NPY_BIG
172
+ #define PyArray_NATIVE NPY_NATIVE
173
+ #define PyArray_SWAP NPY_SWAP
174
+ #define PyArray_IGNORE NPY_IGNORE
175
+
176
+ #define PyArray_NATBYTE NPY_NATBYTE
177
+ #define PyArray_OPPBYTE NPY_OPPBYTE
178
+
179
+ #define PyArray_MAX_ELSIZE NPY_MAX_ELSIZE
180
+
181
+ #define PyArray_USE_PYMEM NPY_USE_PYMEM
182
+
183
+ #define PyArray_RemoveLargest PyArray_RemoveSmallest
184
+
185
+ #define PyArray_UCS4 npy_ucs4
186
+
187
+ #endif /* NUMPY_CORE_INCLUDE_NUMPY_OLD_DEFINES_H_ */
evalkit_tf437/lib/python3.10/site-packages/numpy/core/include/numpy/random/libdivide.h ADDED
@@ -0,0 +1,2079 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ // libdivide.h - Optimized integer division
2
+ // https://libdivide.com
3
+ //
4
+ // Copyright (C) 2010 - 2019 ridiculous_fish, <libdivide@ridiculousfish.com>
5
+ // Copyright (C) 2016 - 2019 Kim Walisch, <kim.walisch@gmail.com>
6
+ //
7
+ // libdivide is dual-licensed under the Boost or zlib licenses.
8
+ // You may use libdivide under the terms of either of these.
9
+ // See LICENSE.txt for more details.
10
+
11
+ #ifndef NUMPY_CORE_INCLUDE_NUMPY_LIBDIVIDE_LIBDIVIDE_H_
12
+ #define NUMPY_CORE_INCLUDE_NUMPY_LIBDIVIDE_LIBDIVIDE_H_
13
+
14
+ #define LIBDIVIDE_VERSION "3.0"
15
+ #define LIBDIVIDE_VERSION_MAJOR 3
16
+ #define LIBDIVIDE_VERSION_MINOR 0
17
+
18
+ #include <stdint.h>
19
+
20
+ #if defined(__cplusplus)
21
+ #include <cstdlib>
22
+ #include <cstdio>
23
+ #include <type_traits>
24
+ #else
25
+ #include <stdlib.h>
26
+ #include <stdio.h>
27
+ #endif
28
+
29
+ #if defined(LIBDIVIDE_AVX512)
30
+ #include <immintrin.h>
31
+ #elif defined(LIBDIVIDE_AVX2)
32
+ #include <immintrin.h>
33
+ #elif defined(LIBDIVIDE_SSE2)
34
+ #include <emmintrin.h>
35
+ #endif
36
+
37
+ #if defined(_MSC_VER)
38
+ #include <intrin.h>
39
+ // disable warning C4146: unary minus operator applied
40
+ // to unsigned type, result still unsigned
41
+ #pragma warning(disable: 4146)
42
+ #define LIBDIVIDE_VC
43
+ #endif
44
+
45
+ #if !defined(__has_builtin)
46
+ #define __has_builtin(x) 0
47
+ #endif
48
+
49
+ #if defined(__SIZEOF_INT128__)
50
+ #define HAS_INT128_T
51
+ // clang-cl on Windows does not yet support 128-bit division
52
+ #if !(defined(__clang__) && defined(LIBDIVIDE_VC))
53
+ #define HAS_INT128_DIV
54
+ #endif
55
+ #endif
56
+
57
+ #if defined(__x86_64__) || defined(_M_X64)
58
+ #define LIBDIVIDE_X86_64
59
+ #endif
60
+
61
+ #if defined(__i386__)
62
+ #define LIBDIVIDE_i386
63
+ #endif
64
+
65
+ #if defined(__GNUC__) || defined(__clang__)
66
+ #define LIBDIVIDE_GCC_STYLE_ASM
67
+ #endif
68
+
69
+ #if defined(__cplusplus) || defined(LIBDIVIDE_VC)
70
+ #define LIBDIVIDE_FUNCTION __FUNCTION__
71
+ #else
72
+ #define LIBDIVIDE_FUNCTION __func__
73
+ #endif
74
+
75
+ #define LIBDIVIDE_ERROR(msg) \
76
+ do { \
77
+ fprintf(stderr, "libdivide.h:%d: %s(): Error: %s\n", \
78
+ __LINE__, LIBDIVIDE_FUNCTION, msg); \
79
+ abort(); \
80
+ } while (0)
81
+
82
+ #if defined(LIBDIVIDE_ASSERTIONS_ON)
83
+ #define LIBDIVIDE_ASSERT(x) \
84
+ do { \
85
+ if (!(x)) { \
86
+ fprintf(stderr, "libdivide.h:%d: %s(): Assertion failed: %s\n", \
87
+ __LINE__, LIBDIVIDE_FUNCTION, #x); \
88
+ abort(); \
89
+ } \
90
+ } while (0)
91
+ #else
92
+ #define LIBDIVIDE_ASSERT(x)
93
+ #endif
94
+
95
+ #ifdef __cplusplus
96
+ namespace libdivide {
97
+ #endif
98
+
99
+ // pack divider structs to prevent compilers from padding.
100
+ // This reduces memory usage by up to 43% when using a large
101
+ // array of libdivide dividers and improves performance
102
+ // by up to 10% because of reduced memory bandwidth.
103
+ #pragma pack(push, 1)
104
+
105
+ struct libdivide_u32_t {
106
+ uint32_t magic;
107
+ uint8_t more;
108
+ };
109
+
110
+ struct libdivide_s32_t {
111
+ int32_t magic;
112
+ uint8_t more;
113
+ };
114
+
115
+ struct libdivide_u64_t {
116
+ uint64_t magic;
117
+ uint8_t more;
118
+ };
119
+
120
+ struct libdivide_s64_t {
121
+ int64_t magic;
122
+ uint8_t more;
123
+ };
124
+
125
+ struct libdivide_u32_branchfree_t {
126
+ uint32_t magic;
127
+ uint8_t more;
128
+ };
129
+
130
+ struct libdivide_s32_branchfree_t {
131
+ int32_t magic;
132
+ uint8_t more;
133
+ };
134
+
135
+ struct libdivide_u64_branchfree_t {
136
+ uint64_t magic;
137
+ uint8_t more;
138
+ };
139
+
140
+ struct libdivide_s64_branchfree_t {
141
+ int64_t magic;
142
+ uint8_t more;
143
+ };
144
+
145
+ #pragma pack(pop)
146
+
147
+ // Explanation of the "more" field:
148
+ //
149
+ // * Bits 0-5 is the shift value (for shift path or mult path).
150
+ // * Bit 6 is the add indicator for mult path.
151
+ // * Bit 7 is set if the divisor is negative. We use bit 7 as the negative
152
+ // divisor indicator so that we can efficiently use sign extension to
153
+ // create a bitmask with all bits set to 1 (if the divisor is negative)
154
+ // or 0 (if the divisor is positive).
155
+ //
156
+ // u32: [0-4] shift value
157
+ // [5] ignored
158
+ // [6] add indicator
159
+ // magic number of 0 indicates shift path
160
+ //
161
+ // s32: [0-4] shift value
162
+ // [5] ignored
163
+ // [6] add indicator
164
+ // [7] indicates negative divisor
165
+ // magic number of 0 indicates shift path
166
+ //
167
+ // u64: [0-5] shift value
168
+ // [6] add indicator
169
+ // magic number of 0 indicates shift path
170
+ //
171
+ // s64: [0-5] shift value
172
+ // [6] add indicator
173
+ // [7] indicates negative divisor
174
+ // magic number of 0 indicates shift path
175
+ //
176
+ // In s32 and s64 branchfree modes, the magic number is negated according to
177
+ // whether the divisor is negated. In branchfree strategy, it is not negated.
178
+
179
+ enum {
180
+ LIBDIVIDE_32_SHIFT_MASK = 0x1F,
181
+ LIBDIVIDE_64_SHIFT_MASK = 0x3F,
182
+ LIBDIVIDE_ADD_MARKER = 0x40,
183
+ LIBDIVIDE_NEGATIVE_DIVISOR = 0x80
184
+ };
185
+
186
+ static inline struct libdivide_s32_t libdivide_s32_gen(int32_t d);
187
+ static inline struct libdivide_u32_t libdivide_u32_gen(uint32_t d);
188
+ static inline struct libdivide_s64_t libdivide_s64_gen(int64_t d);
189
+ static inline struct libdivide_u64_t libdivide_u64_gen(uint64_t d);
190
+
191
+ static inline struct libdivide_s32_branchfree_t libdivide_s32_branchfree_gen(int32_t d);
192
+ static inline struct libdivide_u32_branchfree_t libdivide_u32_branchfree_gen(uint32_t d);
193
+ static inline struct libdivide_s64_branchfree_t libdivide_s64_branchfree_gen(int64_t d);
194
+ static inline struct libdivide_u64_branchfree_t libdivide_u64_branchfree_gen(uint64_t d);
195
+
196
+ static inline int32_t libdivide_s32_do(int32_t numer, const struct libdivide_s32_t *denom);
197
+ static inline uint32_t libdivide_u32_do(uint32_t numer, const struct libdivide_u32_t *denom);
198
+ static inline int64_t libdivide_s64_do(int64_t numer, const struct libdivide_s64_t *denom);
199
+ static inline uint64_t libdivide_u64_do(uint64_t numer, const struct libdivide_u64_t *denom);
200
+
201
+ static inline int32_t libdivide_s32_branchfree_do(int32_t numer, const struct libdivide_s32_branchfree_t *denom);
202
+ static inline uint32_t libdivide_u32_branchfree_do(uint32_t numer, const struct libdivide_u32_branchfree_t *denom);
203
+ static inline int64_t libdivide_s64_branchfree_do(int64_t numer, const struct libdivide_s64_branchfree_t *denom);
204
+ static inline uint64_t libdivide_u64_branchfree_do(uint64_t numer, const struct libdivide_u64_branchfree_t *denom);
205
+
206
+ static inline int32_t libdivide_s32_recover(const struct libdivide_s32_t *denom);
207
+ static inline uint32_t libdivide_u32_recover(const struct libdivide_u32_t *denom);
208
+ static inline int64_t libdivide_s64_recover(const struct libdivide_s64_t *denom);
209
+ static inline uint64_t libdivide_u64_recover(const struct libdivide_u64_t *denom);
210
+
211
+ static inline int32_t libdivide_s32_branchfree_recover(const struct libdivide_s32_branchfree_t *denom);
212
+ static inline uint32_t libdivide_u32_branchfree_recover(const struct libdivide_u32_branchfree_t *denom);
213
+ static inline int64_t libdivide_s64_branchfree_recover(const struct libdivide_s64_branchfree_t *denom);
214
+ static inline uint64_t libdivide_u64_branchfree_recover(const struct libdivide_u64_branchfree_t *denom);
215
+
216
+ //////// Internal Utility Functions
217
+
218
+ static inline uint32_t libdivide_mullhi_u32(uint32_t x, uint32_t y) {
219
+ uint64_t xl = x, yl = y;
220
+ uint64_t rl = xl * yl;
221
+ return (uint32_t)(rl >> 32);
222
+ }
223
+
224
+ static inline int32_t libdivide_mullhi_s32(int32_t x, int32_t y) {
225
+ int64_t xl = x, yl = y;
226
+ int64_t rl = xl * yl;
227
+ // needs to be arithmetic shift
228
+ return (int32_t)(rl >> 32);
229
+ }
230
+
231
+ static inline uint64_t libdivide_mullhi_u64(uint64_t x, uint64_t y) {
232
+ #if defined(LIBDIVIDE_VC) && \
233
+ defined(LIBDIVIDE_X86_64)
234
+ return __umulh(x, y);
235
+ #elif defined(HAS_INT128_T)
236
+ __uint128_t xl = x, yl = y;
237
+ __uint128_t rl = xl * yl;
238
+ return (uint64_t)(rl >> 64);
239
+ #else
240
+ // full 128 bits are x0 * y0 + (x0 * y1 << 32) + (x1 * y0 << 32) + (x1 * y1 << 64)
241
+ uint32_t mask = 0xFFFFFFFF;
242
+ uint32_t x0 = (uint32_t)(x & mask);
243
+ uint32_t x1 = (uint32_t)(x >> 32);
244
+ uint32_t y0 = (uint32_t)(y & mask);
245
+ uint32_t y1 = (uint32_t)(y >> 32);
246
+ uint32_t x0y0_hi = libdivide_mullhi_u32(x0, y0);
247
+ uint64_t x0y1 = x0 * (uint64_t)y1;
248
+ uint64_t x1y0 = x1 * (uint64_t)y0;
249
+ uint64_t x1y1 = x1 * (uint64_t)y1;
250
+ uint64_t temp = x1y0 + x0y0_hi;
251
+ uint64_t temp_lo = temp & mask;
252
+ uint64_t temp_hi = temp >> 32;
253
+
254
+ return x1y1 + temp_hi + ((temp_lo + x0y1) >> 32);
255
+ #endif
256
+ }
257
+
258
+ static inline int64_t libdivide_mullhi_s64(int64_t x, int64_t y) {
259
+ #if defined(LIBDIVIDE_VC) && \
260
+ defined(LIBDIVIDE_X86_64)
261
+ return __mulh(x, y);
262
+ #elif defined(HAS_INT128_T)
263
+ __int128_t xl = x, yl = y;
264
+ __int128_t rl = xl * yl;
265
+ return (int64_t)(rl >> 64);
266
+ #else
267
+ // full 128 bits are x0 * y0 + (x0 * y1 << 32) + (x1 * y0 << 32) + (x1 * y1 << 64)
268
+ uint32_t mask = 0xFFFFFFFF;
269
+ uint32_t x0 = (uint32_t)(x & mask);
270
+ uint32_t y0 = (uint32_t)(y & mask);
271
+ int32_t x1 = (int32_t)(x >> 32);
272
+ int32_t y1 = (int32_t)(y >> 32);
273
+ uint32_t x0y0_hi = libdivide_mullhi_u32(x0, y0);
274
+ int64_t t = x1 * (int64_t)y0 + x0y0_hi;
275
+ int64_t w1 = x0 * (int64_t)y1 + (t & mask);
276
+
277
+ return x1 * (int64_t)y1 + (t >> 32) + (w1 >> 32);
278
+ #endif
279
+ }
280
+
281
+ static inline int32_t libdivide_count_leading_zeros32(uint32_t val) {
282
+ #if defined(__GNUC__) || \
283
+ __has_builtin(__builtin_clz)
284
+ // Fast way to count leading zeros
285
+ return __builtin_clz(val);
286
+ #elif defined(LIBDIVIDE_VC)
287
+ unsigned long result;
288
+ if (_BitScanReverse(&result, val)) {
289
+ return 31 - result;
290
+ }
291
+ return 0;
292
+ #else
293
+ if (val == 0)
294
+ return 32;
295
+ int32_t result = 8;
296
+ uint32_t hi = 0xFFU << 24;
297
+ while ((val & hi) == 0) {
298
+ hi >>= 8;
299
+ result += 8;
300
+ }
301
+ while (val & hi) {
302
+ result -= 1;
303
+ hi <<= 1;
304
+ }
305
+ return result;
306
+ #endif
307
+ }
308
+
309
+ static inline int32_t libdivide_count_leading_zeros64(uint64_t val) {
310
+ #if defined(__GNUC__) || \
311
+ __has_builtin(__builtin_clzll)
312
+ // Fast way to count leading zeros
313
+ return __builtin_clzll(val);
314
+ #elif defined(LIBDIVIDE_VC) && defined(_WIN64)
315
+ unsigned long result;
316
+ if (_BitScanReverse64(&result, val)) {
317
+ return 63 - result;
318
+ }
319
+ return 0;
320
+ #else
321
+ uint32_t hi = val >> 32;
322
+ uint32_t lo = val & 0xFFFFFFFF;
323
+ if (hi != 0) return libdivide_count_leading_zeros32(hi);
324
+ return 32 + libdivide_count_leading_zeros32(lo);
325
+ #endif
326
+ }
327
+
328
+ // libdivide_64_div_32_to_32: divides a 64-bit uint {u1, u0} by a 32-bit
329
+ // uint {v}. The result must fit in 32 bits.
330
+ // Returns the quotient directly and the remainder in *r
331
+ static inline uint32_t libdivide_64_div_32_to_32(uint32_t u1, uint32_t u0, uint32_t v, uint32_t *r) {
332
+ #if (defined(LIBDIVIDE_i386) || defined(LIBDIVIDE_X86_64)) && \
333
+ defined(LIBDIVIDE_GCC_STYLE_ASM)
334
+ uint32_t result;
335
+ __asm__("divl %[v]"
336
+ : "=a"(result), "=d"(*r)
337
+ : [v] "r"(v), "a"(u0), "d"(u1)
338
+ );
339
+ return result;
340
+ #else
341
+ uint64_t n = ((uint64_t)u1 << 32) | u0;
342
+ uint32_t result = (uint32_t)(n / v);
343
+ *r = (uint32_t)(n - result * (uint64_t)v);
344
+ return result;
345
+ #endif
346
+ }
347
+
348
+ // libdivide_128_div_64_to_64: divides a 128-bit uint {u1, u0} by a 64-bit
349
+ // uint {v}. The result must fit in 64 bits.
350
+ // Returns the quotient directly and the remainder in *r
351
+ static uint64_t libdivide_128_div_64_to_64(uint64_t u1, uint64_t u0, uint64_t v, uint64_t *r) {
352
+ #if defined(LIBDIVIDE_X86_64) && \
353
+ defined(LIBDIVIDE_GCC_STYLE_ASM)
354
+ uint64_t result;
355
+ __asm__("divq %[v]"
356
+ : "=a"(result), "=d"(*r)
357
+ : [v] "r"(v), "a"(u0), "d"(u1)
358
+ );
359
+ return result;
360
+ #elif defined(HAS_INT128_T) && \
361
+ defined(HAS_INT128_DIV)
362
+ __uint128_t n = ((__uint128_t)u1 << 64) | u0;
363
+ uint64_t result = (uint64_t)(n / v);
364
+ *r = (uint64_t)(n - result * (__uint128_t)v);
365
+ return result;
366
+ #else
367
+ // Code taken from Hacker's Delight:
368
+ // http://www.hackersdelight.org/HDcode/divlu.c.
369
+ // License permits inclusion here per:
370
+ // http://www.hackersdelight.org/permissions.htm
371
+
372
+ const uint64_t b = (1ULL << 32); // Number base (32 bits)
373
+ uint64_t un1, un0; // Norm. dividend LSD's
374
+ uint64_t vn1, vn0; // Norm. divisor digits
375
+ uint64_t q1, q0; // Quotient digits
376
+ uint64_t un64, un21, un10; // Dividend digit pairs
377
+ uint64_t rhat; // A remainder
378
+ int32_t s; // Shift amount for norm
379
+
380
+ // If overflow, set rem. to an impossible value,
381
+ // and return the largest possible quotient
382
+ if (u1 >= v) {
383
+ *r = (uint64_t) -1;
384
+ return (uint64_t) -1;
385
+ }
386
+
387
+ // count leading zeros
388
+ s = libdivide_count_leading_zeros64(v);
389
+ if (s > 0) {
390
+ // Normalize divisor
391
+ v = v << s;
392
+ un64 = (u1 << s) | (u0 >> (64 - s));
393
+ un10 = u0 << s; // Shift dividend left
394
+ } else {
395
+ // Avoid undefined behavior of (u0 >> 64).
396
+ // The behavior is undefined if the right operand is
397
+ // negative, or greater than or equal to the length
398
+ // in bits of the promoted left operand.
399
+ un64 = u1;
400
+ un10 = u0;
401
+ }
402
+
403
+ // Break divisor up into two 32-bit digits
404
+ vn1 = v >> 32;
405
+ vn0 = v & 0xFFFFFFFF;
406
+
407
+ // Break right half of dividend into two digits
408
+ un1 = un10 >> 32;
409
+ un0 = un10 & 0xFFFFFFFF;
410
+
411
+ // Compute the first quotient digit, q1
412
+ q1 = un64 / vn1;
413
+ rhat = un64 - q1 * vn1;
414
+
415
+ while (q1 >= b || q1 * vn0 > b * rhat + un1) {
416
+ q1 = q1 - 1;
417
+ rhat = rhat + vn1;
418
+ if (rhat >= b)
419
+ break;
420
+ }
421
+
422
+ // Multiply and subtract
423
+ un21 = un64 * b + un1 - q1 * v;
424
+
425
+ // Compute the second quotient digit
426
+ q0 = un21 / vn1;
427
+ rhat = un21 - q0 * vn1;
428
+
429
+ while (q0 >= b || q0 * vn0 > b * rhat + un0) {
430
+ q0 = q0 - 1;
431
+ rhat = rhat + vn1;
432
+ if (rhat >= b)
433
+ break;
434
+ }
435
+
436
+ *r = (un21 * b + un0 - q0 * v) >> s;
437
+ return q1 * b + q0;
438
+ #endif
439
+ }
440
+
441
+ // Bitshift a u128 in place, left (signed_shift > 0) or right (signed_shift < 0)
442
+ static inline void libdivide_u128_shift(uint64_t *u1, uint64_t *u0, int32_t signed_shift) {
443
+ if (signed_shift > 0) {
444
+ uint32_t shift = signed_shift;
445
+ *u1 <<= shift;
446
+ *u1 |= *u0 >> (64 - shift);
447
+ *u0 <<= shift;
448
+ }
449
+ else if (signed_shift < 0) {
450
+ uint32_t shift = -signed_shift;
451
+ *u0 >>= shift;
452
+ *u0 |= *u1 << (64 - shift);
453
+ *u1 >>= shift;
454
+ }
455
+ }
456
+
457
+ // Computes a 128 / 128 -> 64 bit division, with a 128 bit remainder.
458
+ static uint64_t libdivide_128_div_128_to_64(uint64_t u_hi, uint64_t u_lo, uint64_t v_hi, uint64_t v_lo, uint64_t *r_hi, uint64_t *r_lo) {
459
+ #if defined(HAS_INT128_T) && \
460
+ defined(HAS_INT128_DIV)
461
+ __uint128_t ufull = u_hi;
462
+ __uint128_t vfull = v_hi;
463
+ ufull = (ufull << 64) | u_lo;
464
+ vfull = (vfull << 64) | v_lo;
465
+ uint64_t res = (uint64_t)(ufull / vfull);
466
+ __uint128_t remainder = ufull - (vfull * res);
467
+ *r_lo = (uint64_t)remainder;
468
+ *r_hi = (uint64_t)(remainder >> 64);
469
+ return res;
470
+ #else
471
+ // Adapted from "Unsigned Doubleword Division" in Hacker's Delight
472
+ // We want to compute u / v
473
+ typedef struct { uint64_t hi; uint64_t lo; } u128_t;
474
+ u128_t u = {u_hi, u_lo};
475
+ u128_t v = {v_hi, v_lo};
476
+
477
+ if (v.hi == 0) {
478
+ // divisor v is a 64 bit value, so we just need one 128/64 division
479
+ // Note that we are simpler than Hacker's Delight here, because we know
480
+ // the quotient fits in 64 bits whereas Hacker's Delight demands a full
481
+ // 128 bit quotient
482
+ *r_hi = 0;
483
+ return libdivide_128_div_64_to_64(u.hi, u.lo, v.lo, r_lo);
484
+ }
485
+ // Here v >= 2**64
486
+ // We know that v.hi != 0, so count leading zeros is OK
487
+ // We have 0 <= n <= 63
488
+ uint32_t n = libdivide_count_leading_zeros64(v.hi);
489
+
490
+ // Normalize the divisor so its MSB is 1
491
+ u128_t v1t = v;
492
+ libdivide_u128_shift(&v1t.hi, &v1t.lo, n);
493
+ uint64_t v1 = v1t.hi; // i.e. v1 = v1t >> 64
494
+
495
+ // To ensure no overflow
496
+ u128_t u1 = u;
497
+ libdivide_u128_shift(&u1.hi, &u1.lo, -1);
498
+
499
+ // Get quotient from divide unsigned insn.
500
+ uint64_t rem_ignored;
501
+ uint64_t q1 = libdivide_128_div_64_to_64(u1.hi, u1.lo, v1, &rem_ignored);
502
+
503
+ // Undo normalization and division of u by 2.
504
+ u128_t q0 = {0, q1};
505
+ libdivide_u128_shift(&q0.hi, &q0.lo, n);
506
+ libdivide_u128_shift(&q0.hi, &q0.lo, -63);
507
+
508
+ // Make q0 correct or too small by 1
509
+ // Equivalent to `if (q0 != 0) q0 = q0 - 1;`
510
+ if (q0.hi != 0 || q0.lo != 0) {
511
+ q0.hi -= (q0.lo == 0); // borrow
512
+ q0.lo -= 1;
513
+ }
514
+
515
+ // Now q0 is correct.
516
+ // Compute q0 * v as q0v
517
+ // = (q0.hi << 64 + q0.lo) * (v.hi << 64 + v.lo)
518
+ // = (q0.hi * v.hi << 128) + (q0.hi * v.lo << 64) +
519
+ // (q0.lo * v.hi << 64) + q0.lo * v.lo)
520
+ // Each term is 128 bit
521
+ // High half of full product (upper 128 bits!) are dropped
522
+ u128_t q0v = {0, 0};
523
+ q0v.hi = q0.hi*v.lo + q0.lo*v.hi + libdivide_mullhi_u64(q0.lo, v.lo);
524
+ q0v.lo = q0.lo*v.lo;
525
+
526
+ // Compute u - q0v as u_q0v
527
+ // This is the remainder
528
+ u128_t u_q0v = u;
529
+ u_q0v.hi -= q0v.hi + (u.lo < q0v.lo); // second term is borrow
530
+ u_q0v.lo -= q0v.lo;
531
+
532
+ // Check if u_q0v >= v
533
+ // This checks if our remainder is larger than the divisor
534
+ if ((u_q0v.hi > v.hi) ||
535
+ (u_q0v.hi == v.hi && u_q0v.lo >= v.lo)) {
536
+ // Increment q0
537
+ q0.lo += 1;
538
+ q0.hi += (q0.lo == 0); // carry
539
+
540
+ // Subtract v from remainder
541
+ u_q0v.hi -= v.hi + (u_q0v.lo < v.lo);
542
+ u_q0v.lo -= v.lo;
543
+ }
544
+
545
+ *r_hi = u_q0v.hi;
546
+ *r_lo = u_q0v.lo;
547
+
548
+ LIBDIVIDE_ASSERT(q0.hi == 0);
549
+ return q0.lo;
550
+ #endif
551
+ }
552
+
553
+ ////////// UINT32
554
+
555
+ static inline struct libdivide_u32_t libdivide_internal_u32_gen(uint32_t d, int branchfree) {
556
+ if (d == 0) {
557
+ LIBDIVIDE_ERROR("divider must be != 0");
558
+ }
559
+
560
+ struct libdivide_u32_t result;
561
+ uint32_t floor_log_2_d = 31 - libdivide_count_leading_zeros32(d);
562
+
563
+ // Power of 2
564
+ if ((d & (d - 1)) == 0) {
565
+ // We need to subtract 1 from the shift value in case of an unsigned
566
+ // branchfree divider because there is a hardcoded right shift by 1
567
+ // in its division algorithm. Because of this we also need to add back
568
+ // 1 in its recovery algorithm.
569
+ result.magic = 0;
570
+ result.more = (uint8_t)(floor_log_2_d - (branchfree != 0));
571
+ } else {
572
+ uint8_t more;
573
+ uint32_t rem, proposed_m;
574
+ proposed_m = libdivide_64_div_32_to_32(1U << floor_log_2_d, 0, d, &rem);
575
+
576
+ LIBDIVIDE_ASSERT(rem > 0 && rem < d);
577
+ const uint32_t e = d - rem;
578
+
579
+ // This power works if e < 2**floor_log_2_d.
580
+ if (!branchfree && (e < (1U << floor_log_2_d))) {
581
+ // This power works
582
+ more = floor_log_2_d;
583
+ } else {
584
+ // We have to use the general 33-bit algorithm. We need to compute
585
+ // (2**power) / d. However, we already have (2**(power-1))/d and
586
+ // its remainder. By doubling both, and then correcting the
587
+ // remainder, we can compute the larger division.
588
+ // don't care about overflow here - in fact, we expect it
589
+ proposed_m += proposed_m;
590
+ const uint32_t twice_rem = rem + rem;
591
+ if (twice_rem >= d || twice_rem < rem) proposed_m += 1;
592
+ more = floor_log_2_d | LIBDIVIDE_ADD_MARKER;
593
+ }
594
+ result.magic = 1 + proposed_m;
595
+ result.more = more;
596
+ // result.more's shift should in general be ceil_log_2_d. But if we
597
+ // used the smaller power, we subtract one from the shift because we're
598
+ // using the smaller power. If we're using the larger power, we
599
+ // subtract one from the shift because it's taken care of by the add
600
+ // indicator. So floor_log_2_d happens to be correct in both cases.
601
+ }
602
+ return result;
603
+ }
604
+
605
+ struct libdivide_u32_t libdivide_u32_gen(uint32_t d) {
606
+ return libdivide_internal_u32_gen(d, 0);
607
+ }
608
+
609
+ struct libdivide_u32_branchfree_t libdivide_u32_branchfree_gen(uint32_t d) {
610
+ if (d == 1) {
611
+ LIBDIVIDE_ERROR("branchfree divider must be != 1");
612
+ }
613
+ struct libdivide_u32_t tmp = libdivide_internal_u32_gen(d, 1);
614
+ struct libdivide_u32_branchfree_t ret = {tmp.magic, (uint8_t)(tmp.more & LIBDIVIDE_32_SHIFT_MASK)};
615
+ return ret;
616
+ }
617
+
618
+ uint32_t libdivide_u32_do(uint32_t numer, const struct libdivide_u32_t *denom) {
619
+ uint8_t more = denom->more;
620
+ if (!denom->magic) {
621
+ return numer >> more;
622
+ }
623
+ else {
624
+ uint32_t q = libdivide_mullhi_u32(denom->magic, numer);
625
+ if (more & LIBDIVIDE_ADD_MARKER) {
626
+ uint32_t t = ((numer - q) >> 1) + q;
627
+ return t >> (more & LIBDIVIDE_32_SHIFT_MASK);
628
+ }
629
+ else {
630
+ // All upper bits are 0,
631
+ // don't need to mask them off.
632
+ return q >> more;
633
+ }
634
+ }
635
+ }
636
+
637
+ uint32_t libdivide_u32_branchfree_do(uint32_t numer, const struct libdivide_u32_branchfree_t *denom) {
638
+ uint32_t q = libdivide_mullhi_u32(denom->magic, numer);
639
+ uint32_t t = ((numer - q) >> 1) + q;
640
+ return t >> denom->more;
641
+ }
642
+
643
+ uint32_t libdivide_u32_recover(const struct libdivide_u32_t *denom) {
644
+ uint8_t more = denom->more;
645
+ uint8_t shift = more & LIBDIVIDE_32_SHIFT_MASK;
646
+
647
+ if (!denom->magic) {
648
+ return 1U << shift;
649
+ } else if (!(more & LIBDIVIDE_ADD_MARKER)) {
650
+ // We compute q = n/d = n*m / 2^(32 + shift)
651
+ // Therefore we have d = 2^(32 + shift) / m
652
+ // We need to ceil it.
653
+ // We know d is not a power of 2, so m is not a power of 2,
654
+ // so we can just add 1 to the floor
655
+ uint32_t hi_dividend = 1U << shift;
656
+ uint32_t rem_ignored;
657
+ return 1 + libdivide_64_div_32_to_32(hi_dividend, 0, denom->magic, &rem_ignored);
658
+ } else {
659
+ // Here we wish to compute d = 2^(32+shift+1)/(m+2^32).
660
+ // Notice (m + 2^32) is a 33 bit number. Use 64 bit division for now
661
+ // Also note that shift may be as high as 31, so shift + 1 will
662
+ // overflow. So we have to compute it as 2^(32+shift)/(m+2^32), and
663
+ // then double the quotient and remainder.
664
+ uint64_t half_n = 1ULL << (32 + shift);
665
+ uint64_t d = (1ULL << 32) | denom->magic;
666
+ // Note that the quotient is guaranteed <= 32 bits, but the remainder
667
+ // may need 33!
668
+ uint32_t half_q = (uint32_t)(half_n / d);
669
+ uint64_t rem = half_n % d;
670
+ // We computed 2^(32+shift)/(m+2^32)
671
+ // Need to double it, and then add 1 to the quotient if doubling th
672
+ // remainder would increase the quotient.
673
+ // Note that rem<<1 cannot overflow, since rem < d and d is 33 bits
674
+ uint32_t full_q = half_q + half_q + ((rem<<1) >= d);
675
+
676
+ // We rounded down in gen (hence +1)
677
+ return full_q + 1;
678
+ }
679
+ }
680
+
681
+ uint32_t libdivide_u32_branchfree_recover(const struct libdivide_u32_branchfree_t *denom) {
682
+ uint8_t more = denom->more;
683
+ uint8_t shift = more & LIBDIVIDE_32_SHIFT_MASK;
684
+
685
+ if (!denom->magic) {
686
+ return 1U << (shift + 1);
687
+ } else {
688
+ // Here we wish to compute d = 2^(32+shift+1)/(m+2^32).
689
+ // Notice (m + 2^32) is a 33 bit number. Use 64 bit division for now
690
+ // Also note that shift may be as high as 31, so shift + 1 will
691
+ // overflow. So we have to compute it as 2^(32+shift)/(m+2^32), and
692
+ // then double the quotient and remainder.
693
+ uint64_t half_n = 1ULL << (32 + shift);
694
+ uint64_t d = (1ULL << 32) | denom->magic;
695
+ // Note that the quotient is guaranteed <= 32 bits, but the remainder
696
+ // may need 33!
697
+ uint32_t half_q = (uint32_t)(half_n / d);
698
+ uint64_t rem = half_n % d;
699
+ // We computed 2^(32+shift)/(m+2^32)
700
+ // Need to double it, and then add 1 to the quotient if doubling th
701
+ // remainder would increase the quotient.
702
+ // Note that rem<<1 cannot overflow, since rem < d and d is 33 bits
703
+ uint32_t full_q = half_q + half_q + ((rem<<1) >= d);
704
+
705
+ // We rounded down in gen (hence +1)
706
+ return full_q + 1;
707
+ }
708
+ }
709
+
710
+ /////////// UINT64
711
+
712
+ static inline struct libdivide_u64_t libdivide_internal_u64_gen(uint64_t d, int branchfree) {
713
+ if (d == 0) {
714
+ LIBDIVIDE_ERROR("divider must be != 0");
715
+ }
716
+
717
+ struct libdivide_u64_t result;
718
+ uint32_t floor_log_2_d = 63 - libdivide_count_leading_zeros64(d);
719
+
720
+ // Power of 2
721
+ if ((d & (d - 1)) == 0) {
722
+ // We need to subtract 1 from the shift value in case of an unsigned
723
+ // branchfree divider because there is a hardcoded right shift by 1
724
+ // in its division algorithm. Because of this we also need to add back
725
+ // 1 in its recovery algorithm.
726
+ result.magic = 0;
727
+ result.more = (uint8_t)(floor_log_2_d - (branchfree != 0));
728
+ } else {
729
+ uint64_t proposed_m, rem;
730
+ uint8_t more;
731
+ // (1 << (64 + floor_log_2_d)) / d
732
+ proposed_m = libdivide_128_div_64_to_64(1ULL << floor_log_2_d, 0, d, &rem);
733
+
734
+ LIBDIVIDE_ASSERT(rem > 0 && rem < d);
735
+ const uint64_t e = d - rem;
736
+
737
+ // This power works if e < 2**floor_log_2_d.
738
+ if (!branchfree && e < (1ULL << floor_log_2_d)) {
739
+ // This power works
740
+ more = floor_log_2_d;
741
+ } else {
742
+ // We have to use the general 65-bit algorithm. We need to compute
743
+ // (2**power) / d. However, we already have (2**(power-1))/d and
744
+ // its remainder. By doubling both, and then correcting the
745
+ // remainder, we can compute the larger division.
746
+ // don't care about overflow here - in fact, we expect it
747
+ proposed_m += proposed_m;
748
+ const uint64_t twice_rem = rem + rem;
749
+ if (twice_rem >= d || twice_rem < rem) proposed_m += 1;
750
+ more = floor_log_2_d | LIBDIVIDE_ADD_MARKER;
751
+ }
752
+ result.magic = 1 + proposed_m;
753
+ result.more = more;
754
+ // result.more's shift should in general be ceil_log_2_d. But if we
755
+ // used the smaller power, we subtract one from the shift because we're
756
+ // using the smaller power. If we're using the larger power, we
757
+ // subtract one from the shift because it's taken care of by the add
758
+ // indicator. So floor_log_2_d happens to be correct in both cases,
759
+ // which is why we do it outside of the if statement.
760
+ }
761
+ return result;
762
+ }
763
+
764
+ struct libdivide_u64_t libdivide_u64_gen(uint64_t d) {
765
+ return libdivide_internal_u64_gen(d, 0);
766
+ }
767
+
768
+ struct libdivide_u64_branchfree_t libdivide_u64_branchfree_gen(uint64_t d) {
769
+ if (d == 1) {
770
+ LIBDIVIDE_ERROR("branchfree divider must be != 1");
771
+ }
772
+ struct libdivide_u64_t tmp = libdivide_internal_u64_gen(d, 1);
773
+ struct libdivide_u64_branchfree_t ret = {tmp.magic, (uint8_t)(tmp.more & LIBDIVIDE_64_SHIFT_MASK)};
774
+ return ret;
775
+ }
776
+
777
+ uint64_t libdivide_u64_do(uint64_t numer, const struct libdivide_u64_t *denom) {
778
+ uint8_t more = denom->more;
779
+ if (!denom->magic) {
780
+ return numer >> more;
781
+ }
782
+ else {
783
+ uint64_t q = libdivide_mullhi_u64(denom->magic, numer);
784
+ if (more & LIBDIVIDE_ADD_MARKER) {
785
+ uint64_t t = ((numer - q) >> 1) + q;
786
+ return t >> (more & LIBDIVIDE_64_SHIFT_MASK);
787
+ }
788
+ else {
789
+ // All upper bits are 0,
790
+ // don't need to mask them off.
791
+ return q >> more;
792
+ }
793
+ }
794
+ }
795
+
796
+ uint64_t libdivide_u64_branchfree_do(uint64_t numer, const struct libdivide_u64_branchfree_t *denom) {
797
+ uint64_t q = libdivide_mullhi_u64(denom->magic, numer);
798
+ uint64_t t = ((numer - q) >> 1) + q;
799
+ return t >> denom->more;
800
+ }
801
+
802
+ uint64_t libdivide_u64_recover(const struct libdivide_u64_t *denom) {
803
+ uint8_t more = denom->more;
804
+ uint8_t shift = more & LIBDIVIDE_64_SHIFT_MASK;
805
+
806
+ if (!denom->magic) {
807
+ return 1ULL << shift;
808
+ } else if (!(more & LIBDIVIDE_ADD_MARKER)) {
809
+ // We compute q = n/d = n*m / 2^(64 + shift)
810
+ // Therefore we have d = 2^(64 + shift) / m
811
+ // We need to ceil it.
812
+ // We know d is not a power of 2, so m is not a power of 2,
813
+ // so we can just add 1 to the floor
814
+ uint64_t hi_dividend = 1ULL << shift;
815
+ uint64_t rem_ignored;
816
+ return 1 + libdivide_128_div_64_to_64(hi_dividend, 0, denom->magic, &rem_ignored);
817
+ } else {
818
+ // Here we wish to compute d = 2^(64+shift+1)/(m+2^64).
819
+ // Notice (m + 2^64) is a 65 bit number. This gets hairy. See
820
+ // libdivide_u32_recover for more on what we do here.
821
+ // TODO: do something better than 128 bit math
822
+
823
+ // Full n is a (potentially) 129 bit value
824
+ // half_n is a 128 bit value
825
+ // Compute the hi half of half_n. Low half is 0.
826
+ uint64_t half_n_hi = 1ULL << shift, half_n_lo = 0;
827
+ // d is a 65 bit value. The high bit is always set to 1.
828
+ const uint64_t d_hi = 1, d_lo = denom->magic;
829
+ // Note that the quotient is guaranteed <= 64 bits,
830
+ // but the remainder may need 65!
831
+ uint64_t r_hi, r_lo;
832
+ uint64_t half_q = libdivide_128_div_128_to_64(half_n_hi, half_n_lo, d_hi, d_lo, &r_hi, &r_lo);
833
+ // We computed 2^(64+shift)/(m+2^64)
834
+ // Double the remainder ('dr') and check if that is larger than d
835
+ // Note that d is a 65 bit value, so r1 is small and so r1 + r1
836
+ // cannot overflow
837
+ uint64_t dr_lo = r_lo + r_lo;
838
+ uint64_t dr_hi = r_hi + r_hi + (dr_lo < r_lo); // last term is carry
839
+ int dr_exceeds_d = (dr_hi > d_hi) || (dr_hi == d_hi && dr_lo >= d_lo);
840
+ uint64_t full_q = half_q + half_q + (dr_exceeds_d ? 1 : 0);
841
+ return full_q + 1;
842
+ }
843
+ }
844
+
845
+ uint64_t libdivide_u64_branchfree_recover(const struct libdivide_u64_branchfree_t *denom) {
846
+ uint8_t more = denom->more;
847
+ uint8_t shift = more & LIBDIVIDE_64_SHIFT_MASK;
848
+
849
+ if (!denom->magic) {
850
+ return 1ULL << (shift + 1);
851
+ } else {
852
+ // Here we wish to compute d = 2^(64+shift+1)/(m+2^64).
853
+ // Notice (m + 2^64) is a 65 bit number. This gets hairy. See
854
+ // libdivide_u32_recover for more on what we do here.
855
+ // TODO: do something better than 128 bit math
856
+
857
+ // Full n is a (potentially) 129 bit value
858
+ // half_n is a 128 bit value
859
+ // Compute the hi half of half_n. Low half is 0.
860
+ uint64_t half_n_hi = 1ULL << shift, half_n_lo = 0;
861
+ // d is a 65 bit value. The high bit is always set to 1.
862
+ const uint64_t d_hi = 1, d_lo = denom->magic;
863
+ // Note that the quotient is guaranteed <= 64 bits,
864
+ // but the remainder may need 65!
865
+ uint64_t r_hi, r_lo;
866
+ uint64_t half_q = libdivide_128_div_128_to_64(half_n_hi, half_n_lo, d_hi, d_lo, &r_hi, &r_lo);
867
+ // We computed 2^(64+shift)/(m+2^64)
868
+ // Double the remainder ('dr') and check if that is larger than d
869
+ // Note that d is a 65 bit value, so r1 is small and so r1 + r1
870
+ // cannot overflow
871
+ uint64_t dr_lo = r_lo + r_lo;
872
+ uint64_t dr_hi = r_hi + r_hi + (dr_lo < r_lo); // last term is carry
873
+ int dr_exceeds_d = (dr_hi > d_hi) || (dr_hi == d_hi && dr_lo >= d_lo);
874
+ uint64_t full_q = half_q + half_q + (dr_exceeds_d ? 1 : 0);
875
+ return full_q + 1;
876
+ }
877
+ }
878
+
879
+ /////////// SINT32
880
+
881
+ static inline struct libdivide_s32_t libdivide_internal_s32_gen(int32_t d, int branchfree) {
882
+ if (d == 0) {
883
+ LIBDIVIDE_ERROR("divider must be != 0");
884
+ }
885
+
886
+ struct libdivide_s32_t result;
887
+
888
+ // If d is a power of 2, or negative a power of 2, we have to use a shift.
889
+ // This is especially important because the magic algorithm fails for -1.
890
+ // To check if d is a power of 2 or its inverse, it suffices to check
891
+ // whether its absolute value has exactly one bit set. This works even for
892
+ // INT_MIN, because abs(INT_MIN) == INT_MIN, and INT_MIN has one bit set
893
+ // and is a power of 2.
894
+ uint32_t ud = (uint32_t)d;
895
+ uint32_t absD = (d < 0) ? -ud : ud;
896
+ uint32_t floor_log_2_d = 31 - libdivide_count_leading_zeros32(absD);
897
+ // check if exactly one bit is set,
898
+ // don't care if absD is 0 since that's divide by zero
899
+ if ((absD & (absD - 1)) == 0) {
900
+ // Branchfree and normal paths are exactly the same
901
+ result.magic = 0;
902
+ result.more = floor_log_2_d | (d < 0 ? LIBDIVIDE_NEGATIVE_DIVISOR : 0);
903
+ } else {
904
+ LIBDIVIDE_ASSERT(floor_log_2_d >= 1);
905
+
906
+ uint8_t more;
907
+ // the dividend here is 2**(floor_log_2_d + 31), so the low 32 bit word
908
+ // is 0 and the high word is floor_log_2_d - 1
909
+ uint32_t rem, proposed_m;
910
+ proposed_m = libdivide_64_div_32_to_32(1U << (floor_log_2_d - 1), 0, absD, &rem);
911
+ const uint32_t e = absD - rem;
912
+
913
+ // We are going to start with a power of floor_log_2_d - 1.
914
+ // This works if works if e < 2**floor_log_2_d.
915
+ if (!branchfree && e < (1U << floor_log_2_d)) {
916
+ // This power works
917
+ more = floor_log_2_d - 1;
918
+ } else {
919
+ // We need to go one higher. This should not make proposed_m
920
+ // overflow, but it will make it negative when interpreted as an
921
+ // int32_t.
922
+ proposed_m += proposed_m;
923
+ const uint32_t twice_rem = rem + rem;
924
+ if (twice_rem >= absD || twice_rem < rem) proposed_m += 1;
925
+ more = floor_log_2_d | LIBDIVIDE_ADD_MARKER;
926
+ }
927
+
928
+ proposed_m += 1;
929
+ int32_t magic = (int32_t)proposed_m;
930
+
931
+ // Mark if we are negative. Note we only negate the magic number in the
932
+ // branchfull case.
933
+ if (d < 0) {
934
+ more |= LIBDIVIDE_NEGATIVE_DIVISOR;
935
+ if (!branchfree) {
936
+ magic = -magic;
937
+ }
938
+ }
939
+
940
+ result.more = more;
941
+ result.magic = magic;
942
+ }
943
+ return result;
944
+ }
945
+
946
+ struct libdivide_s32_t libdivide_s32_gen(int32_t d) {
947
+ return libdivide_internal_s32_gen(d, 0);
948
+ }
949
+
950
+ struct libdivide_s32_branchfree_t libdivide_s32_branchfree_gen(int32_t d) {
951
+ struct libdivide_s32_t tmp = libdivide_internal_s32_gen(d, 1);
952
+ struct libdivide_s32_branchfree_t result = {tmp.magic, tmp.more};
953
+ return result;
954
+ }
955
+
956
+ int32_t libdivide_s32_do(int32_t numer, const struct libdivide_s32_t *denom) {
957
+ uint8_t more = denom->more;
958
+ uint8_t shift = more & LIBDIVIDE_32_SHIFT_MASK;
959
+
960
+ if (!denom->magic) {
961
+ uint32_t sign = (int8_t)more >> 7;
962
+ uint32_t mask = (1U << shift) - 1;
963
+ uint32_t uq = numer + ((numer >> 31) & mask);
964
+ int32_t q = (int32_t)uq;
965
+ q >>= shift;
966
+ q = (q ^ sign) - sign;
967
+ return q;
968
+ } else {
969
+ uint32_t uq = (uint32_t)libdivide_mullhi_s32(denom->magic, numer);
970
+ if (more & LIBDIVIDE_ADD_MARKER) {
971
+ // must be arithmetic shift and then sign extend
972
+ int32_t sign = (int8_t)more >> 7;
973
+ // q += (more < 0 ? -numer : numer)
974
+ // cast required to avoid UB
975
+ uq += ((uint32_t)numer ^ sign) - sign;
976
+ }
977
+ int32_t q = (int32_t)uq;
978
+ q >>= shift;
979
+ q += (q < 0);
980
+ return q;
981
+ }
982
+ }
983
+
984
+ int32_t libdivide_s32_branchfree_do(int32_t numer, const struct libdivide_s32_branchfree_t *denom) {
985
+ uint8_t more = denom->more;
986
+ uint8_t shift = more & LIBDIVIDE_32_SHIFT_MASK;
987
+ // must be arithmetic shift and then sign extend
988
+ int32_t sign = (int8_t)more >> 7;
989
+ int32_t magic = denom->magic;
990
+ int32_t q = libdivide_mullhi_s32(magic, numer);
991
+ q += numer;
992
+
993
+ // If q is non-negative, we have nothing to do
994
+ // If q is negative, we want to add either (2**shift)-1 if d is a power of
995
+ // 2, or (2**shift) if it is not a power of 2
996
+ uint32_t is_power_of_2 = (magic == 0);
997
+ uint32_t q_sign = (uint32_t)(q >> 31);
998
+ q += q_sign & ((1U << shift) - is_power_of_2);
999
+
1000
+ // Now arithmetic right shift
1001
+ q >>= shift;
1002
+ // Negate if needed
1003
+ q = (q ^ sign) - sign;
1004
+
1005
+ return q;
1006
+ }
1007
+
1008
+ int32_t libdivide_s32_recover(const struct libdivide_s32_t *denom) {
1009
+ uint8_t more = denom->more;
1010
+ uint8_t shift = more & LIBDIVIDE_32_SHIFT_MASK;
1011
+ if (!denom->magic) {
1012
+ uint32_t absD = 1U << shift;
1013
+ if (more & LIBDIVIDE_NEGATIVE_DIVISOR) {
1014
+ absD = -absD;
1015
+ }
1016
+ return (int32_t)absD;
1017
+ } else {
1018
+ // Unsigned math is much easier
1019
+ // We negate the magic number only in the branchfull case, and we don't
1020
+ // know which case we're in. However we have enough information to
1021
+ // determine the correct sign of the magic number. The divisor was
1022
+ // negative if LIBDIVIDE_NEGATIVE_DIVISOR is set. If ADD_MARKER is set,
1023
+ // the magic number's sign is opposite that of the divisor.
1024
+ // We want to compute the positive magic number.
1025
+ int negative_divisor = (more & LIBDIVIDE_NEGATIVE_DIVISOR);
1026
+ int magic_was_negated = (more & LIBDIVIDE_ADD_MARKER)
1027
+ ? denom->magic > 0 : denom->magic < 0;
1028
+
1029
+ // Handle the power of 2 case (including branchfree)
1030
+ if (denom->magic == 0) {
1031
+ int32_t result = 1U << shift;
1032
+ return negative_divisor ? -result : result;
1033
+ }
1034
+
1035
+ uint32_t d = (uint32_t)(magic_was_negated ? -denom->magic : denom->magic);
1036
+ uint64_t n = 1ULL << (32 + shift); // this shift cannot exceed 30
1037
+ uint32_t q = (uint32_t)(n / d);
1038
+ int32_t result = (int32_t)q;
1039
+ result += 1;
1040
+ return negative_divisor ? -result : result;
1041
+ }
1042
+ }
1043
+
1044
+ int32_t libdivide_s32_branchfree_recover(const struct libdivide_s32_branchfree_t *denom) {
1045
+ return libdivide_s32_recover((const struct libdivide_s32_t *)denom);
1046
+ }
1047
+
1048
+ ///////////// SINT64
1049
+
1050
+ static inline struct libdivide_s64_t libdivide_internal_s64_gen(int64_t d, int branchfree) {
1051
+ if (d == 0) {
1052
+ LIBDIVIDE_ERROR("divider must be != 0");
1053
+ }
1054
+
1055
+ struct libdivide_s64_t result;
1056
+
1057
+ // If d is a power of 2, or negative a power of 2, we have to use a shift.
1058
+ // This is especially important because the magic algorithm fails for -1.
1059
+ // To check if d is a power of 2 or its inverse, it suffices to check
1060
+ // whether its absolute value has exactly one bit set. This works even for
1061
+ // INT_MIN, because abs(INT_MIN) == INT_MIN, and INT_MIN has one bit set
1062
+ // and is a power of 2.
1063
+ uint64_t ud = (uint64_t)d;
1064
+ uint64_t absD = (d < 0) ? -ud : ud;
1065
+ uint32_t floor_log_2_d = 63 - libdivide_count_leading_zeros64(absD);
1066
+ // check if exactly one bit is set,
1067
+ // don't care if absD is 0 since that's divide by zero
1068
+ if ((absD & (absD - 1)) == 0) {
1069
+ // Branchfree and non-branchfree cases are the same
1070
+ result.magic = 0;
1071
+ result.more = floor_log_2_d | (d < 0 ? LIBDIVIDE_NEGATIVE_DIVISOR : 0);
1072
+ } else {
1073
+ // the dividend here is 2**(floor_log_2_d + 63), so the low 64 bit word
1074
+ // is 0 and the high word is floor_log_2_d - 1
1075
+ uint8_t more;
1076
+ uint64_t rem, proposed_m;
1077
+ proposed_m = libdivide_128_div_64_to_64(1ULL << (floor_log_2_d - 1), 0, absD, &rem);
1078
+ const uint64_t e = absD - rem;
1079
+
1080
+ // We are going to start with a power of floor_log_2_d - 1.
1081
+ // This works if works if e < 2**floor_log_2_d.
1082
+ if (!branchfree && e < (1ULL << floor_log_2_d)) {
1083
+ // This power works
1084
+ more = floor_log_2_d - 1;
1085
+ } else {
1086
+ // We need to go one higher. This should not make proposed_m
1087
+ // overflow, but it will make it negative when interpreted as an
1088
+ // int32_t.
1089
+ proposed_m += proposed_m;
1090
+ const uint64_t twice_rem = rem + rem;
1091
+ if (twice_rem >= absD || twice_rem < rem) proposed_m += 1;
1092
+ // note that we only set the LIBDIVIDE_NEGATIVE_DIVISOR bit if we
1093
+ // also set ADD_MARKER this is an annoying optimization that
1094
+ // enables algorithm #4 to avoid the mask. However we always set it
1095
+ // in the branchfree case
1096
+ more = floor_log_2_d | LIBDIVIDE_ADD_MARKER;
1097
+ }
1098
+ proposed_m += 1;
1099
+ int64_t magic = (int64_t)proposed_m;
1100
+
1101
+ // Mark if we are negative
1102
+ if (d < 0) {
1103
+ more |= LIBDIVIDE_NEGATIVE_DIVISOR;
1104
+ if (!branchfree) {
1105
+ magic = -magic;
1106
+ }
1107
+ }
1108
+
1109
+ result.more = more;
1110
+ result.magic = magic;
1111
+ }
1112
+ return result;
1113
+ }
1114
+
1115
+ struct libdivide_s64_t libdivide_s64_gen(int64_t d) {
1116
+ return libdivide_internal_s64_gen(d, 0);
1117
+ }
1118
+
1119
+ struct libdivide_s64_branchfree_t libdivide_s64_branchfree_gen(int64_t d) {
1120
+ struct libdivide_s64_t tmp = libdivide_internal_s64_gen(d, 1);
1121
+ struct libdivide_s64_branchfree_t ret = {tmp.magic, tmp.more};
1122
+ return ret;
1123
+ }
1124
+
1125
+ int64_t libdivide_s64_do(int64_t numer, const struct libdivide_s64_t *denom) {
1126
+ uint8_t more = denom->more;
1127
+ uint8_t shift = more & LIBDIVIDE_64_SHIFT_MASK;
1128
+
1129
+ if (!denom->magic) { // shift path
1130
+ uint64_t mask = (1ULL << shift) - 1;
1131
+ uint64_t uq = numer + ((numer >> 63) & mask);
1132
+ int64_t q = (int64_t)uq;
1133
+ q >>= shift;
1134
+ // must be arithmetic shift and then sign-extend
1135
+ int64_t sign = (int8_t)more >> 7;
1136
+ q = (q ^ sign) - sign;
1137
+ return q;
1138
+ } else {
1139
+ uint64_t uq = (uint64_t)libdivide_mullhi_s64(denom->magic, numer);
1140
+ if (more & LIBDIVIDE_ADD_MARKER) {
1141
+ // must be arithmetic shift and then sign extend
1142
+ int64_t sign = (int8_t)more >> 7;
1143
+ // q += (more < 0 ? -numer : numer)
1144
+ // cast required to avoid UB
1145
+ uq += ((uint64_t)numer ^ sign) - sign;
1146
+ }
1147
+ int64_t q = (int64_t)uq;
1148
+ q >>= shift;
1149
+ q += (q < 0);
1150
+ return q;
1151
+ }
1152
+ }
1153
+
1154
+ int64_t libdivide_s64_branchfree_do(int64_t numer, const struct libdivide_s64_branchfree_t *denom) {
1155
+ uint8_t more = denom->more;
1156
+ uint8_t shift = more & LIBDIVIDE_64_SHIFT_MASK;
1157
+ // must be arithmetic shift and then sign extend
1158
+ int64_t sign = (int8_t)more >> 7;
1159
+ int64_t magic = denom->magic;
1160
+ int64_t q = libdivide_mullhi_s64(magic, numer);
1161
+ q += numer;
1162
+
1163
+ // If q is non-negative, we have nothing to do.
1164
+ // If q is negative, we want to add either (2**shift)-1 if d is a power of
1165
+ // 2, or (2**shift) if it is not a power of 2.
1166
+ uint64_t is_power_of_2 = (magic == 0);
1167
+ uint64_t q_sign = (uint64_t)(q >> 63);
1168
+ q += q_sign & ((1ULL << shift) - is_power_of_2);
1169
+
1170
+ // Arithmetic right shift
1171
+ q >>= shift;
1172
+ // Negate if needed
1173
+ q = (q ^ sign) - sign;
1174
+
1175
+ return q;
1176
+ }
1177
+
1178
+ int64_t libdivide_s64_recover(const struct libdivide_s64_t *denom) {
1179
+ uint8_t more = denom->more;
1180
+ uint8_t shift = more & LIBDIVIDE_64_SHIFT_MASK;
1181
+ if (denom->magic == 0) { // shift path
1182
+ uint64_t absD = 1ULL << shift;
1183
+ if (more & LIBDIVIDE_NEGATIVE_DIVISOR) {
1184
+ absD = -absD;
1185
+ }
1186
+ return (int64_t)absD;
1187
+ } else {
1188
+ // Unsigned math is much easier
1189
+ int negative_divisor = (more & LIBDIVIDE_NEGATIVE_DIVISOR);
1190
+ int magic_was_negated = (more & LIBDIVIDE_ADD_MARKER)
1191
+ ? denom->magic > 0 : denom->magic < 0;
1192
+
1193
+ uint64_t d = (uint64_t)(magic_was_negated ? -denom->magic : denom->magic);
1194
+ uint64_t n_hi = 1ULL << shift, n_lo = 0;
1195
+ uint64_t rem_ignored;
1196
+ uint64_t q = libdivide_128_div_64_to_64(n_hi, n_lo, d, &rem_ignored);
1197
+ int64_t result = (int64_t)(q + 1);
1198
+ if (negative_divisor) {
1199
+ result = -result;
1200
+ }
1201
+ return result;
1202
+ }
1203
+ }
1204
+
1205
+ int64_t libdivide_s64_branchfree_recover(const struct libdivide_s64_branchfree_t *denom) {
1206
+ return libdivide_s64_recover((const struct libdivide_s64_t *)denom);
1207
+ }
1208
+
1209
+ #if defined(LIBDIVIDE_AVX512)
1210
+
1211
+ static inline __m512i libdivide_u32_do_vector(__m512i numers, const struct libdivide_u32_t *denom);
1212
+ static inline __m512i libdivide_s32_do_vector(__m512i numers, const struct libdivide_s32_t *denom);
1213
+ static inline __m512i libdivide_u64_do_vector(__m512i numers, const struct libdivide_u64_t *denom);
1214
+ static inline __m512i libdivide_s64_do_vector(__m512i numers, const struct libdivide_s64_t *denom);
1215
+
1216
+ static inline __m512i libdivide_u32_branchfree_do_vector(__m512i numers, const struct libdivide_u32_branchfree_t *denom);
1217
+ static inline __m512i libdivide_s32_branchfree_do_vector(__m512i numers, const struct libdivide_s32_branchfree_t *denom);
1218
+ static inline __m512i libdivide_u64_branchfree_do_vector(__m512i numers, const struct libdivide_u64_branchfree_t *denom);
1219
+ static inline __m512i libdivide_s64_branchfree_do_vector(__m512i numers, const struct libdivide_s64_branchfree_t *denom);
1220
+
1221
+ //////// Internal Utility Functions
1222
+
1223
+ static inline __m512i libdivide_s64_signbits(__m512i v) {;
1224
+ return _mm512_srai_epi64(v, 63);
1225
+ }
1226
+
1227
+ static inline __m512i libdivide_s64_shift_right_vector(__m512i v, int amt) {
1228
+ return _mm512_srai_epi64(v, amt);
1229
+ }
1230
+
1231
+ // Here, b is assumed to contain one 32-bit value repeated.
1232
+ static inline __m512i libdivide_mullhi_u32_vector(__m512i a, __m512i b) {
1233
+ __m512i hi_product_0Z2Z = _mm512_srli_epi64(_mm512_mul_epu32(a, b), 32);
1234
+ __m512i a1X3X = _mm512_srli_epi64(a, 32);
1235
+ __m512i mask = _mm512_set_epi32(-1, 0, -1, 0, -1, 0, -1, 0, -1, 0, -1, 0, -1, 0, -1, 0);
1236
+ __m512i hi_product_Z1Z3 = _mm512_and_si512(_mm512_mul_epu32(a1X3X, b), mask);
1237
+ return _mm512_or_si512(hi_product_0Z2Z, hi_product_Z1Z3);
1238
+ }
1239
+
1240
+ // b is one 32-bit value repeated.
1241
+ static inline __m512i libdivide_mullhi_s32_vector(__m512i a, __m512i b) {
1242
+ __m512i hi_product_0Z2Z = _mm512_srli_epi64(_mm512_mul_epi32(a, b), 32);
1243
+ __m512i a1X3X = _mm512_srli_epi64(a, 32);
1244
+ __m512i mask = _mm512_set_epi32(-1, 0, -1, 0, -1, 0, -1, 0, -1, 0, -1, 0, -1, 0, -1, 0);
1245
+ __m512i hi_product_Z1Z3 = _mm512_and_si512(_mm512_mul_epi32(a1X3X, b), mask);
1246
+ return _mm512_or_si512(hi_product_0Z2Z, hi_product_Z1Z3);
1247
+ }
1248
+
1249
+ // Here, y is assumed to contain one 64-bit value repeated.
1250
+ // https://stackoverflow.com/a/28827013
1251
+ static inline __m512i libdivide_mullhi_u64_vector(__m512i x, __m512i y) {
1252
+ __m512i lomask = _mm512_set1_epi64(0xffffffff);
1253
+ __m512i xh = _mm512_shuffle_epi32(x, (_MM_PERM_ENUM) 0xB1);
1254
+ __m512i yh = _mm512_shuffle_epi32(y, (_MM_PERM_ENUM) 0xB1);
1255
+ __m512i w0 = _mm512_mul_epu32(x, y);
1256
+ __m512i w1 = _mm512_mul_epu32(x, yh);
1257
+ __m512i w2 = _mm512_mul_epu32(xh, y);
1258
+ __m512i w3 = _mm512_mul_epu32(xh, yh);
1259
+ __m512i w0h = _mm512_srli_epi64(w0, 32);
1260
+ __m512i s1 = _mm512_add_epi64(w1, w0h);
1261
+ __m512i s1l = _mm512_and_si512(s1, lomask);
1262
+ __m512i s1h = _mm512_srli_epi64(s1, 32);
1263
+ __m512i s2 = _mm512_add_epi64(w2, s1l);
1264
+ __m512i s2h = _mm512_srli_epi64(s2, 32);
1265
+ __m512i hi = _mm512_add_epi64(w3, s1h);
1266
+ hi = _mm512_add_epi64(hi, s2h);
1267
+
1268
+ return hi;
1269
+ }
1270
+
1271
+ // y is one 64-bit value repeated.
1272
+ static inline __m512i libdivide_mullhi_s64_vector(__m512i x, __m512i y) {
1273
+ __m512i p = libdivide_mullhi_u64_vector(x, y);
1274
+ __m512i t1 = _mm512_and_si512(libdivide_s64_signbits(x), y);
1275
+ __m512i t2 = _mm512_and_si512(libdivide_s64_signbits(y), x);
1276
+ p = _mm512_sub_epi64(p, t1);
1277
+ p = _mm512_sub_epi64(p, t2);
1278
+ return p;
1279
+ }
1280
+
1281
+ ////////// UINT32
1282
+
1283
+ __m512i libdivide_u32_do_vector(__m512i numers, const struct libdivide_u32_t *denom) {
1284
+ uint8_t more = denom->more;
1285
+ if (!denom->magic) {
1286
+ return _mm512_srli_epi32(numers, more);
1287
+ }
1288
+ else {
1289
+ __m512i q = libdivide_mullhi_u32_vector(numers, _mm512_set1_epi32(denom->magic));
1290
+ if (more & LIBDIVIDE_ADD_MARKER) {
1291
+ // uint32_t t = ((numer - q) >> 1) + q;
1292
+ // return t >> denom->shift;
1293
+ uint32_t shift = more & LIBDIVIDE_32_SHIFT_MASK;
1294
+ __m512i t = _mm512_add_epi32(_mm512_srli_epi32(_mm512_sub_epi32(numers, q), 1), q);
1295
+ return _mm512_srli_epi32(t, shift);
1296
+ }
1297
+ else {
1298
+ return _mm512_srli_epi32(q, more);
1299
+ }
1300
+ }
1301
+ }
1302
+
1303
+ __m512i libdivide_u32_branchfree_do_vector(__m512i numers, const struct libdivide_u32_branchfree_t *denom) {
1304
+ __m512i q = libdivide_mullhi_u32_vector(numers, _mm512_set1_epi32(denom->magic));
1305
+ __m512i t = _mm512_add_epi32(_mm512_srli_epi32(_mm512_sub_epi32(numers, q), 1), q);
1306
+ return _mm512_srli_epi32(t, denom->more);
1307
+ }
1308
+
1309
+ ////////// UINT64
1310
+
1311
+ __m512i libdivide_u64_do_vector(__m512i numers, const struct libdivide_u64_t *denom) {
1312
+ uint8_t more = denom->more;
1313
+ if (!denom->magic) {
1314
+ return _mm512_srli_epi64(numers, more);
1315
+ }
1316
+ else {
1317
+ __m512i q = libdivide_mullhi_u64_vector(numers, _mm512_set1_epi64(denom->magic));
1318
+ if (more & LIBDIVIDE_ADD_MARKER) {
1319
+ // uint32_t t = ((numer - q) >> 1) + q;
1320
+ // return t >> denom->shift;
1321
+ uint32_t shift = more & LIBDIVIDE_64_SHIFT_MASK;
1322
+ __m512i t = _mm512_add_epi64(_mm512_srli_epi64(_mm512_sub_epi64(numers, q), 1), q);
1323
+ return _mm512_srli_epi64(t, shift);
1324
+ }
1325
+ else {
1326
+ return _mm512_srli_epi64(q, more);
1327
+ }
1328
+ }
1329
+ }
1330
+
1331
+ __m512i libdivide_u64_branchfree_do_vector(__m512i numers, const struct libdivide_u64_branchfree_t *denom) {
1332
+ __m512i q = libdivide_mullhi_u64_vector(numers, _mm512_set1_epi64(denom->magic));
1333
+ __m512i t = _mm512_add_epi64(_mm512_srli_epi64(_mm512_sub_epi64(numers, q), 1), q);
1334
+ return _mm512_srli_epi64(t, denom->more);
1335
+ }
1336
+
1337
+ ////////// SINT32
1338
+
1339
+ __m512i libdivide_s32_do_vector(__m512i numers, const struct libdivide_s32_t *denom) {
1340
+ uint8_t more = denom->more;
1341
+ if (!denom->magic) {
1342
+ uint32_t shift = more & LIBDIVIDE_32_SHIFT_MASK;
1343
+ uint32_t mask = (1U << shift) - 1;
1344
+ __m512i roundToZeroTweak = _mm512_set1_epi32(mask);
1345
+ // q = numer + ((numer >> 31) & roundToZeroTweak);
1346
+ __m512i q = _mm512_add_epi32(numers, _mm512_and_si512(_mm512_srai_epi32(numers, 31), roundToZeroTweak));
1347
+ q = _mm512_srai_epi32(q, shift);
1348
+ __m512i sign = _mm512_set1_epi32((int8_t)more >> 7);
1349
+ // q = (q ^ sign) - sign;
1350
+ q = _mm512_sub_epi32(_mm512_xor_si512(q, sign), sign);
1351
+ return q;
1352
+ }
1353
+ else {
1354
+ __m512i q = libdivide_mullhi_s32_vector(numers, _mm512_set1_epi32(denom->magic));
1355
+ if (more & LIBDIVIDE_ADD_MARKER) {
1356
+ // must be arithmetic shift
1357
+ __m512i sign = _mm512_set1_epi32((int8_t)more >> 7);
1358
+ // q += ((numer ^ sign) - sign);
1359
+ q = _mm512_add_epi32(q, _mm512_sub_epi32(_mm512_xor_si512(numers, sign), sign));
1360
+ }
1361
+ // q >>= shift
1362
+ q = _mm512_srai_epi32(q, more & LIBDIVIDE_32_SHIFT_MASK);
1363
+ q = _mm512_add_epi32(q, _mm512_srli_epi32(q, 31)); // q += (q < 0)
1364
+ return q;
1365
+ }
1366
+ }
1367
+
1368
+ __m512i libdivide_s32_branchfree_do_vector(__m512i numers, const struct libdivide_s32_branchfree_t *denom) {
1369
+ int32_t magic = denom->magic;
1370
+ uint8_t more = denom->more;
1371
+ uint8_t shift = more & LIBDIVIDE_32_SHIFT_MASK;
1372
+ // must be arithmetic shift
1373
+ __m512i sign = _mm512_set1_epi32((int8_t)more >> 7);
1374
+ __m512i q = libdivide_mullhi_s32_vector(numers, _mm512_set1_epi32(magic));
1375
+ q = _mm512_add_epi32(q, numers); // q += numers
1376
+
1377
+ // If q is non-negative, we have nothing to do
1378
+ // If q is negative, we want to add either (2**shift)-1 if d is
1379
+ // a power of 2, or (2**shift) if it is not a power of 2
1380
+ uint32_t is_power_of_2 = (magic == 0);
1381
+ __m512i q_sign = _mm512_srai_epi32(q, 31); // q_sign = q >> 31
1382
+ __m512i mask = _mm512_set1_epi32((1U << shift) - is_power_of_2);
1383
+ q = _mm512_add_epi32(q, _mm512_and_si512(q_sign, mask)); // q = q + (q_sign & mask)
1384
+ q = _mm512_srai_epi32(q, shift); // q >>= shift
1385
+ q = _mm512_sub_epi32(_mm512_xor_si512(q, sign), sign); // q = (q ^ sign) - sign
1386
+ return q;
1387
+ }
1388
+
1389
+ ////////// SINT64
1390
+
1391
+ __m512i libdivide_s64_do_vector(__m512i numers, const struct libdivide_s64_t *denom) {
1392
+ uint8_t more = denom->more;
1393
+ int64_t magic = denom->magic;
1394
+ if (magic == 0) { // shift path
1395
+ uint32_t shift = more & LIBDIVIDE_64_SHIFT_MASK;
1396
+ uint64_t mask = (1ULL << shift) - 1;
1397
+ __m512i roundToZeroTweak = _mm512_set1_epi64(mask);
1398
+ // q = numer + ((numer >> 63) & roundToZeroTweak);
1399
+ __m512i q = _mm512_add_epi64(numers, _mm512_and_si512(libdivide_s64_signbits(numers), roundToZeroTweak));
1400
+ q = libdivide_s64_shift_right_vector(q, shift);
1401
+ __m512i sign = _mm512_set1_epi32((int8_t)more >> 7);
1402
+ // q = (q ^ sign) - sign;
1403
+ q = _mm512_sub_epi64(_mm512_xor_si512(q, sign), sign);
1404
+ return q;
1405
+ }
1406
+ else {
1407
+ __m512i q = libdivide_mullhi_s64_vector(numers, _mm512_set1_epi64(magic));
1408
+ if (more & LIBDIVIDE_ADD_MARKER) {
1409
+ // must be arithmetic shift
1410
+ __m512i sign = _mm512_set1_epi32((int8_t)more >> 7);
1411
+ // q += ((numer ^ sign) - sign);
1412
+ q = _mm512_add_epi64(q, _mm512_sub_epi64(_mm512_xor_si512(numers, sign), sign));
1413
+ }
1414
+ // q >>= denom->mult_path.shift
1415
+ q = libdivide_s64_shift_right_vector(q, more & LIBDIVIDE_64_SHIFT_MASK);
1416
+ q = _mm512_add_epi64(q, _mm512_srli_epi64(q, 63)); // q += (q < 0)
1417
+ return q;
1418
+ }
1419
+ }
1420
+
1421
+ __m512i libdivide_s64_branchfree_do_vector(__m512i numers, const struct libdivide_s64_branchfree_t *denom) {
1422
+ int64_t magic = denom->magic;
1423
+ uint8_t more = denom->more;
1424
+ uint8_t shift = more & LIBDIVIDE_64_SHIFT_MASK;
1425
+ // must be arithmetic shift
1426
+ __m512i sign = _mm512_set1_epi32((int8_t)more >> 7);
1427
+
1428
+ // libdivide_mullhi_s64(numers, magic);
1429
+ __m512i q = libdivide_mullhi_s64_vector(numers, _mm512_set1_epi64(magic));
1430
+ q = _mm512_add_epi64(q, numers); // q += numers
1431
+
1432
+ // If q is non-negative, we have nothing to do.
1433
+ // If q is negative, we want to add either (2**shift)-1 if d is
1434
+ // a power of 2, or (2**shift) if it is not a power of 2.
1435
+ uint32_t is_power_of_2 = (magic == 0);
1436
+ __m512i q_sign = libdivide_s64_signbits(q); // q_sign = q >> 63
1437
+ __m512i mask = _mm512_set1_epi64((1ULL << shift) - is_power_of_2);
1438
+ q = _mm512_add_epi64(q, _mm512_and_si512(q_sign, mask)); // q = q + (q_sign & mask)
1439
+ q = libdivide_s64_shift_right_vector(q, shift); // q >>= shift
1440
+ q = _mm512_sub_epi64(_mm512_xor_si512(q, sign), sign); // q = (q ^ sign) - sign
1441
+ return q;
1442
+ }
1443
+
1444
+ #elif defined(LIBDIVIDE_AVX2)
1445
+
1446
+ static inline __m256i libdivide_u32_do_vector(__m256i numers, const struct libdivide_u32_t *denom);
1447
+ static inline __m256i libdivide_s32_do_vector(__m256i numers, const struct libdivide_s32_t *denom);
1448
+ static inline __m256i libdivide_u64_do_vector(__m256i numers, const struct libdivide_u64_t *denom);
1449
+ static inline __m256i libdivide_s64_do_vector(__m256i numers, const struct libdivide_s64_t *denom);
1450
+
1451
+ static inline __m256i libdivide_u32_branchfree_do_vector(__m256i numers, const struct libdivide_u32_branchfree_t *denom);
1452
+ static inline __m256i libdivide_s32_branchfree_do_vector(__m256i numers, const struct libdivide_s32_branchfree_t *denom);
1453
+ static inline __m256i libdivide_u64_branchfree_do_vector(__m256i numers, const struct libdivide_u64_branchfree_t *denom);
1454
+ static inline __m256i libdivide_s64_branchfree_do_vector(__m256i numers, const struct libdivide_s64_branchfree_t *denom);
1455
+
1456
+ //////// Internal Utility Functions
1457
+
1458
+ // Implementation of _mm256_srai_epi64(v, 63) (from AVX512).
1459
+ static inline __m256i libdivide_s64_signbits(__m256i v) {
1460
+ __m256i hiBitsDuped = _mm256_shuffle_epi32(v, _MM_SHUFFLE(3, 3, 1, 1));
1461
+ __m256i signBits = _mm256_srai_epi32(hiBitsDuped, 31);
1462
+ return signBits;
1463
+ }
1464
+
1465
+ // Implementation of _mm256_srai_epi64 (from AVX512).
1466
+ static inline __m256i libdivide_s64_shift_right_vector(__m256i v, int amt) {
1467
+ const int b = 64 - amt;
1468
+ __m256i m = _mm256_set1_epi64x(1ULL << (b - 1));
1469
+ __m256i x = _mm256_srli_epi64(v, amt);
1470
+ __m256i result = _mm256_sub_epi64(_mm256_xor_si256(x, m), m);
1471
+ return result;
1472
+ }
1473
+
1474
+ // Here, b is assumed to contain one 32-bit value repeated.
1475
+ static inline __m256i libdivide_mullhi_u32_vector(__m256i a, __m256i b) {
1476
+ __m256i hi_product_0Z2Z = _mm256_srli_epi64(_mm256_mul_epu32(a, b), 32);
1477
+ __m256i a1X3X = _mm256_srli_epi64(a, 32);
1478
+ __m256i mask = _mm256_set_epi32(-1, 0, -1, 0, -1, 0, -1, 0);
1479
+ __m256i hi_product_Z1Z3 = _mm256_and_si256(_mm256_mul_epu32(a1X3X, b), mask);
1480
+ return _mm256_or_si256(hi_product_0Z2Z, hi_product_Z1Z3);
1481
+ }
1482
+
1483
+ // b is one 32-bit value repeated.
1484
+ static inline __m256i libdivide_mullhi_s32_vector(__m256i a, __m256i b) {
1485
+ __m256i hi_product_0Z2Z = _mm256_srli_epi64(_mm256_mul_epi32(a, b), 32);
1486
+ __m256i a1X3X = _mm256_srli_epi64(a, 32);
1487
+ __m256i mask = _mm256_set_epi32(-1, 0, -1, 0, -1, 0, -1, 0);
1488
+ __m256i hi_product_Z1Z3 = _mm256_and_si256(_mm256_mul_epi32(a1X3X, b), mask);
1489
+ return _mm256_or_si256(hi_product_0Z2Z, hi_product_Z1Z3);
1490
+ }
1491
+
1492
+ // Here, y is assumed to contain one 64-bit value repeated.
1493
+ // https://stackoverflow.com/a/28827013
1494
+ static inline __m256i libdivide_mullhi_u64_vector(__m256i x, __m256i y) {
1495
+ __m256i lomask = _mm256_set1_epi64x(0xffffffff);
1496
+ __m256i xh = _mm256_shuffle_epi32(x, 0xB1); // x0l, x0h, x1l, x1h
1497
+ __m256i yh = _mm256_shuffle_epi32(y, 0xB1); // y0l, y0h, y1l, y1h
1498
+ __m256i w0 = _mm256_mul_epu32(x, y); // x0l*y0l, x1l*y1l
1499
+ __m256i w1 = _mm256_mul_epu32(x, yh); // x0l*y0h, x1l*y1h
1500
+ __m256i w2 = _mm256_mul_epu32(xh, y); // x0h*y0l, x1h*y0l
1501
+ __m256i w3 = _mm256_mul_epu32(xh, yh); // x0h*y0h, x1h*y1h
1502
+ __m256i w0h = _mm256_srli_epi64(w0, 32);
1503
+ __m256i s1 = _mm256_add_epi64(w1, w0h);
1504
+ __m256i s1l = _mm256_and_si256(s1, lomask);
1505
+ __m256i s1h = _mm256_srli_epi64(s1, 32);
1506
+ __m256i s2 = _mm256_add_epi64(w2, s1l);
1507
+ __m256i s2h = _mm256_srli_epi64(s2, 32);
1508
+ __m256i hi = _mm256_add_epi64(w3, s1h);
1509
+ hi = _mm256_add_epi64(hi, s2h);
1510
+
1511
+ return hi;
1512
+ }
1513
+
1514
+ // y is one 64-bit value repeated.
1515
+ static inline __m256i libdivide_mullhi_s64_vector(__m256i x, __m256i y) {
1516
+ __m256i p = libdivide_mullhi_u64_vector(x, y);
1517
+ __m256i t1 = _mm256_and_si256(libdivide_s64_signbits(x), y);
1518
+ __m256i t2 = _mm256_and_si256(libdivide_s64_signbits(y), x);
1519
+ p = _mm256_sub_epi64(p, t1);
1520
+ p = _mm256_sub_epi64(p, t2);
1521
+ return p;
1522
+ }
1523
+
1524
+ ////////// UINT32
1525
+
1526
+ __m256i libdivide_u32_do_vector(__m256i numers, const struct libdivide_u32_t *denom) {
1527
+ uint8_t more = denom->more;
1528
+ if (!denom->magic) {
1529
+ return _mm256_srli_epi32(numers, more);
1530
+ }
1531
+ else {
1532
+ __m256i q = libdivide_mullhi_u32_vector(numers, _mm256_set1_epi32(denom->magic));
1533
+ if (more & LIBDIVIDE_ADD_MARKER) {
1534
+ // uint32_t t = ((numer - q) >> 1) + q;
1535
+ // return t >> denom->shift;
1536
+ uint32_t shift = more & LIBDIVIDE_32_SHIFT_MASK;
1537
+ __m256i t = _mm256_add_epi32(_mm256_srli_epi32(_mm256_sub_epi32(numers, q), 1), q);
1538
+ return _mm256_srli_epi32(t, shift);
1539
+ }
1540
+ else {
1541
+ return _mm256_srli_epi32(q, more);
1542
+ }
1543
+ }
1544
+ }
1545
+
1546
+ __m256i libdivide_u32_branchfree_do_vector(__m256i numers, const struct libdivide_u32_branchfree_t *denom) {
1547
+ __m256i q = libdivide_mullhi_u32_vector(numers, _mm256_set1_epi32(denom->magic));
1548
+ __m256i t = _mm256_add_epi32(_mm256_srli_epi32(_mm256_sub_epi32(numers, q), 1), q);
1549
+ return _mm256_srli_epi32(t, denom->more);
1550
+ }
1551
+
1552
+ ////////// UINT64
1553
+
1554
+ __m256i libdivide_u64_do_vector(__m256i numers, const struct libdivide_u64_t *denom) {
1555
+ uint8_t more = denom->more;
1556
+ if (!denom->magic) {
1557
+ return _mm256_srli_epi64(numers, more);
1558
+ }
1559
+ else {
1560
+ __m256i q = libdivide_mullhi_u64_vector(numers, _mm256_set1_epi64x(denom->magic));
1561
+ if (more & LIBDIVIDE_ADD_MARKER) {
1562
+ // uint32_t t = ((numer - q) >> 1) + q;
1563
+ // return t >> denom->shift;
1564
+ uint32_t shift = more & LIBDIVIDE_64_SHIFT_MASK;
1565
+ __m256i t = _mm256_add_epi64(_mm256_srli_epi64(_mm256_sub_epi64(numers, q), 1), q);
1566
+ return _mm256_srli_epi64(t, shift);
1567
+ }
1568
+ else {
1569
+ return _mm256_srli_epi64(q, more);
1570
+ }
1571
+ }
1572
+ }
1573
+
1574
+ __m256i libdivide_u64_branchfree_do_vector(__m256i numers, const struct libdivide_u64_branchfree_t *denom) {
1575
+ __m256i q = libdivide_mullhi_u64_vector(numers, _mm256_set1_epi64x(denom->magic));
1576
+ __m256i t = _mm256_add_epi64(_mm256_srli_epi64(_mm256_sub_epi64(numers, q), 1), q);
1577
+ return _mm256_srli_epi64(t, denom->more);
1578
+ }
1579
+
1580
+ ////////// SINT32
1581
+
1582
+ __m256i libdivide_s32_do_vector(__m256i numers, const struct libdivide_s32_t *denom) {
1583
+ uint8_t more = denom->more;
1584
+ if (!denom->magic) {
1585
+ uint32_t shift = more & LIBDIVIDE_32_SHIFT_MASK;
1586
+ uint32_t mask = (1U << shift) - 1;
1587
+ __m256i roundToZeroTweak = _mm256_set1_epi32(mask);
1588
+ // q = numer + ((numer >> 31) & roundToZeroTweak);
1589
+ __m256i q = _mm256_add_epi32(numers, _mm256_and_si256(_mm256_srai_epi32(numers, 31), roundToZeroTweak));
1590
+ q = _mm256_srai_epi32(q, shift);
1591
+ __m256i sign = _mm256_set1_epi32((int8_t)more >> 7);
1592
+ // q = (q ^ sign) - sign;
1593
+ q = _mm256_sub_epi32(_mm256_xor_si256(q, sign), sign);
1594
+ return q;
1595
+ }
1596
+ else {
1597
+ __m256i q = libdivide_mullhi_s32_vector(numers, _mm256_set1_epi32(denom->magic));
1598
+ if (more & LIBDIVIDE_ADD_MARKER) {
1599
+ // must be arithmetic shift
1600
+ __m256i sign = _mm256_set1_epi32((int8_t)more >> 7);
1601
+ // q += ((numer ^ sign) - sign);
1602
+ q = _mm256_add_epi32(q, _mm256_sub_epi32(_mm256_xor_si256(numers, sign), sign));
1603
+ }
1604
+ // q >>= shift
1605
+ q = _mm256_srai_epi32(q, more & LIBDIVIDE_32_SHIFT_MASK);
1606
+ q = _mm256_add_epi32(q, _mm256_srli_epi32(q, 31)); // q += (q < 0)
1607
+ return q;
1608
+ }
1609
+ }
1610
+
1611
+ __m256i libdivide_s32_branchfree_do_vector(__m256i numers, const struct libdivide_s32_branchfree_t *denom) {
1612
+ int32_t magic = denom->magic;
1613
+ uint8_t more = denom->more;
1614
+ uint8_t shift = more & LIBDIVIDE_32_SHIFT_MASK;
1615
+ // must be arithmetic shift
1616
+ __m256i sign = _mm256_set1_epi32((int8_t)more >> 7);
1617
+ __m256i q = libdivide_mullhi_s32_vector(numers, _mm256_set1_epi32(magic));
1618
+ q = _mm256_add_epi32(q, numers); // q += numers
1619
+
1620
+ // If q is non-negative, we have nothing to do
1621
+ // If q is negative, we want to add either (2**shift)-1 if d is
1622
+ // a power of 2, or (2**shift) if it is not a power of 2
1623
+ uint32_t is_power_of_2 = (magic == 0);
1624
+ __m256i q_sign = _mm256_srai_epi32(q, 31); // q_sign = q >> 31
1625
+ __m256i mask = _mm256_set1_epi32((1U << shift) - is_power_of_2);
1626
+ q = _mm256_add_epi32(q, _mm256_and_si256(q_sign, mask)); // q = q + (q_sign & mask)
1627
+ q = _mm256_srai_epi32(q, shift); // q >>= shift
1628
+ q = _mm256_sub_epi32(_mm256_xor_si256(q, sign), sign); // q = (q ^ sign) - sign
1629
+ return q;
1630
+ }
1631
+
1632
+ ////////// SINT64
1633
+
1634
+ __m256i libdivide_s64_do_vector(__m256i numers, const struct libdivide_s64_t *denom) {
1635
+ uint8_t more = denom->more;
1636
+ int64_t magic = denom->magic;
1637
+ if (magic == 0) { // shift path
1638
+ uint32_t shift = more & LIBDIVIDE_64_SHIFT_MASK;
1639
+ uint64_t mask = (1ULL << shift) - 1;
1640
+ __m256i roundToZeroTweak = _mm256_set1_epi64x(mask);
1641
+ // q = numer + ((numer >> 63) & roundToZeroTweak);
1642
+ __m256i q = _mm256_add_epi64(numers, _mm256_and_si256(libdivide_s64_signbits(numers), roundToZeroTweak));
1643
+ q = libdivide_s64_shift_right_vector(q, shift);
1644
+ __m256i sign = _mm256_set1_epi32((int8_t)more >> 7);
1645
+ // q = (q ^ sign) - sign;
1646
+ q = _mm256_sub_epi64(_mm256_xor_si256(q, sign), sign);
1647
+ return q;
1648
+ }
1649
+ else {
1650
+ __m256i q = libdivide_mullhi_s64_vector(numers, _mm256_set1_epi64x(magic));
1651
+ if (more & LIBDIVIDE_ADD_MARKER) {
1652
+ // must be arithmetic shift
1653
+ __m256i sign = _mm256_set1_epi32((int8_t)more >> 7);
1654
+ // q += ((numer ^ sign) - sign);
1655
+ q = _mm256_add_epi64(q, _mm256_sub_epi64(_mm256_xor_si256(numers, sign), sign));
1656
+ }
1657
+ // q >>= denom->mult_path.shift
1658
+ q = libdivide_s64_shift_right_vector(q, more & LIBDIVIDE_64_SHIFT_MASK);
1659
+ q = _mm256_add_epi64(q, _mm256_srli_epi64(q, 63)); // q += (q < 0)
1660
+ return q;
1661
+ }
1662
+ }
1663
+
1664
+ __m256i libdivide_s64_branchfree_do_vector(__m256i numers, const struct libdivide_s64_branchfree_t *denom) {
1665
+ int64_t magic = denom->magic;
1666
+ uint8_t more = denom->more;
1667
+ uint8_t shift = more & LIBDIVIDE_64_SHIFT_MASK;
1668
+ // must be arithmetic shift
1669
+ __m256i sign = _mm256_set1_epi32((int8_t)more >> 7);
1670
+
1671
+ // libdivide_mullhi_s64(numers, magic);
1672
+ __m256i q = libdivide_mullhi_s64_vector(numers, _mm256_set1_epi64x(magic));
1673
+ q = _mm256_add_epi64(q, numers); // q += numers
1674
+
1675
+ // If q is non-negative, we have nothing to do.
1676
+ // If q is negative, we want to add either (2**shift)-1 if d is
1677
+ // a power of 2, or (2**shift) if it is not a power of 2.
1678
+ uint32_t is_power_of_2 = (magic == 0);
1679
+ __m256i q_sign = libdivide_s64_signbits(q); // q_sign = q >> 63
1680
+ __m256i mask = _mm256_set1_epi64x((1ULL << shift) - is_power_of_2);
1681
+ q = _mm256_add_epi64(q, _mm256_and_si256(q_sign, mask)); // q = q + (q_sign & mask)
1682
+ q = libdivide_s64_shift_right_vector(q, shift); // q >>= shift
1683
+ q = _mm256_sub_epi64(_mm256_xor_si256(q, sign), sign); // q = (q ^ sign) - sign
1684
+ return q;
1685
+ }
1686
+
1687
+ #elif defined(LIBDIVIDE_SSE2)
1688
+
1689
+ static inline __m128i libdivide_u32_do_vector(__m128i numers, const struct libdivide_u32_t *denom);
1690
+ static inline __m128i libdivide_s32_do_vector(__m128i numers, const struct libdivide_s32_t *denom);
1691
+ static inline __m128i libdivide_u64_do_vector(__m128i numers, const struct libdivide_u64_t *denom);
1692
+ static inline __m128i libdivide_s64_do_vector(__m128i numers, const struct libdivide_s64_t *denom);
1693
+
1694
+ static inline __m128i libdivide_u32_branchfree_do_vector(__m128i numers, const struct libdivide_u32_branchfree_t *denom);
1695
+ static inline __m128i libdivide_s32_branchfree_do_vector(__m128i numers, const struct libdivide_s32_branchfree_t *denom);
1696
+ static inline __m128i libdivide_u64_branchfree_do_vector(__m128i numers, const struct libdivide_u64_branchfree_t *denom);
1697
+ static inline __m128i libdivide_s64_branchfree_do_vector(__m128i numers, const struct libdivide_s64_branchfree_t *denom);
1698
+
1699
+ //////// Internal Utility Functions
1700
+
1701
+ // Implementation of _mm_srai_epi64(v, 63) (from AVX512).
1702
+ static inline __m128i libdivide_s64_signbits(__m128i v) {
1703
+ __m128i hiBitsDuped = _mm_shuffle_epi32(v, _MM_SHUFFLE(3, 3, 1, 1));
1704
+ __m128i signBits = _mm_srai_epi32(hiBitsDuped, 31);
1705
+ return signBits;
1706
+ }
1707
+
1708
+ // Implementation of _mm_srai_epi64 (from AVX512).
1709
+ static inline __m128i libdivide_s64_shift_right_vector(__m128i v, int amt) {
1710
+ const int b = 64 - amt;
1711
+ __m128i m = _mm_set1_epi64x(1ULL << (b - 1));
1712
+ __m128i x = _mm_srli_epi64(v, amt);
1713
+ __m128i result = _mm_sub_epi64(_mm_xor_si128(x, m), m);
1714
+ return result;
1715
+ }
1716
+
1717
+ // Here, b is assumed to contain one 32-bit value repeated.
1718
+ static inline __m128i libdivide_mullhi_u32_vector(__m128i a, __m128i b) {
1719
+ __m128i hi_product_0Z2Z = _mm_srli_epi64(_mm_mul_epu32(a, b), 32);
1720
+ __m128i a1X3X = _mm_srli_epi64(a, 32);
1721
+ __m128i mask = _mm_set_epi32(-1, 0, -1, 0);
1722
+ __m128i hi_product_Z1Z3 = _mm_and_si128(_mm_mul_epu32(a1X3X, b), mask);
1723
+ return _mm_or_si128(hi_product_0Z2Z, hi_product_Z1Z3);
1724
+ }
1725
+
1726
+ // SSE2 does not have a signed multiplication instruction, but we can convert
1727
+ // unsigned to signed pretty efficiently. Again, b is just a 32 bit value
1728
+ // repeated four times.
1729
+ static inline __m128i libdivide_mullhi_s32_vector(__m128i a, __m128i b) {
1730
+ __m128i p = libdivide_mullhi_u32_vector(a, b);
1731
+ // t1 = (a >> 31) & y, arithmetic shift
1732
+ __m128i t1 = _mm_and_si128(_mm_srai_epi32(a, 31), b);
1733
+ __m128i t2 = _mm_and_si128(_mm_srai_epi32(b, 31), a);
1734
+ p = _mm_sub_epi32(p, t1);
1735
+ p = _mm_sub_epi32(p, t2);
1736
+ return p;
1737
+ }
1738
+
1739
+ // Here, y is assumed to contain one 64-bit value repeated.
1740
+ // https://stackoverflow.com/a/28827013
1741
+ static inline __m128i libdivide_mullhi_u64_vector(__m128i x, __m128i y) {
1742
+ __m128i lomask = _mm_set1_epi64x(0xffffffff);
1743
+ __m128i xh = _mm_shuffle_epi32(x, 0xB1); // x0l, x0h, x1l, x1h
1744
+ __m128i yh = _mm_shuffle_epi32(y, 0xB1); // y0l, y0h, y1l, y1h
1745
+ __m128i w0 = _mm_mul_epu32(x, y); // x0l*y0l, x1l*y1l
1746
+ __m128i w1 = _mm_mul_epu32(x, yh); // x0l*y0h, x1l*y1h
1747
+ __m128i w2 = _mm_mul_epu32(xh, y); // x0h*y0l, x1h*y0l
1748
+ __m128i w3 = _mm_mul_epu32(xh, yh); // x0h*y0h, x1h*y1h
1749
+ __m128i w0h = _mm_srli_epi64(w0, 32);
1750
+ __m128i s1 = _mm_add_epi64(w1, w0h);
1751
+ __m128i s1l = _mm_and_si128(s1, lomask);
1752
+ __m128i s1h = _mm_srli_epi64(s1, 32);
1753
+ __m128i s2 = _mm_add_epi64(w2, s1l);
1754
+ __m128i s2h = _mm_srli_epi64(s2, 32);
1755
+ __m128i hi = _mm_add_epi64(w3, s1h);
1756
+ hi = _mm_add_epi64(hi, s2h);
1757
+
1758
+ return hi;
1759
+ }
1760
+
1761
+ // y is one 64-bit value repeated.
1762
+ static inline __m128i libdivide_mullhi_s64_vector(__m128i x, __m128i y) {
1763
+ __m128i p = libdivide_mullhi_u64_vector(x, y);
1764
+ __m128i t1 = _mm_and_si128(libdivide_s64_signbits(x), y);
1765
+ __m128i t2 = _mm_and_si128(libdivide_s64_signbits(y), x);
1766
+ p = _mm_sub_epi64(p, t1);
1767
+ p = _mm_sub_epi64(p, t2);
1768
+ return p;
1769
+ }
1770
+
1771
+ ////////// UINT32
1772
+
1773
+ __m128i libdivide_u32_do_vector(__m128i numers, const struct libdivide_u32_t *denom) {
1774
+ uint8_t more = denom->more;
1775
+ if (!denom->magic) {
1776
+ return _mm_srli_epi32(numers, more);
1777
+ }
1778
+ else {
1779
+ __m128i q = libdivide_mullhi_u32_vector(numers, _mm_set1_epi32(denom->magic));
1780
+ if (more & LIBDIVIDE_ADD_MARKER) {
1781
+ // uint32_t t = ((numer - q) >> 1) + q;
1782
+ // return t >> denom->shift;
1783
+ uint32_t shift = more & LIBDIVIDE_32_SHIFT_MASK;
1784
+ __m128i t = _mm_add_epi32(_mm_srli_epi32(_mm_sub_epi32(numers, q), 1), q);
1785
+ return _mm_srli_epi32(t, shift);
1786
+ }
1787
+ else {
1788
+ return _mm_srli_epi32(q, more);
1789
+ }
1790
+ }
1791
+ }
1792
+
1793
+ __m128i libdivide_u32_branchfree_do_vector(__m128i numers, const struct libdivide_u32_branchfree_t *denom) {
1794
+ __m128i q = libdivide_mullhi_u32_vector(numers, _mm_set1_epi32(denom->magic));
1795
+ __m128i t = _mm_add_epi32(_mm_srli_epi32(_mm_sub_epi32(numers, q), 1), q);
1796
+ return _mm_srli_epi32(t, denom->more);
1797
+ }
1798
+
1799
+ ////////// UINT64
1800
+
1801
+ __m128i libdivide_u64_do_vector(__m128i numers, const struct libdivide_u64_t *denom) {
1802
+ uint8_t more = denom->more;
1803
+ if (!denom->magic) {
1804
+ return _mm_srli_epi64(numers, more);
1805
+ }
1806
+ else {
1807
+ __m128i q = libdivide_mullhi_u64_vector(numers, _mm_set1_epi64x(denom->magic));
1808
+ if (more & LIBDIVIDE_ADD_MARKER) {
1809
+ // uint32_t t = ((numer - q) >> 1) + q;
1810
+ // return t >> denom->shift;
1811
+ uint32_t shift = more & LIBDIVIDE_64_SHIFT_MASK;
1812
+ __m128i t = _mm_add_epi64(_mm_srli_epi64(_mm_sub_epi64(numers, q), 1), q);
1813
+ return _mm_srli_epi64(t, shift);
1814
+ }
1815
+ else {
1816
+ return _mm_srli_epi64(q, more);
1817
+ }
1818
+ }
1819
+ }
1820
+
1821
+ __m128i libdivide_u64_branchfree_do_vector(__m128i numers, const struct libdivide_u64_branchfree_t *denom) {
1822
+ __m128i q = libdivide_mullhi_u64_vector(numers, _mm_set1_epi64x(denom->magic));
1823
+ __m128i t = _mm_add_epi64(_mm_srli_epi64(_mm_sub_epi64(numers, q), 1), q);
1824
+ return _mm_srli_epi64(t, denom->more);
1825
+ }
1826
+
1827
+ ////////// SINT32
1828
+
1829
+ __m128i libdivide_s32_do_vector(__m128i numers, const struct libdivide_s32_t *denom) {
1830
+ uint8_t more = denom->more;
1831
+ if (!denom->magic) {
1832
+ uint32_t shift = more & LIBDIVIDE_32_SHIFT_MASK;
1833
+ uint32_t mask = (1U << shift) - 1;
1834
+ __m128i roundToZeroTweak = _mm_set1_epi32(mask);
1835
+ // q = numer + ((numer >> 31) & roundToZeroTweak);
1836
+ __m128i q = _mm_add_epi32(numers, _mm_and_si128(_mm_srai_epi32(numers, 31), roundToZeroTweak));
1837
+ q = _mm_srai_epi32(q, shift);
1838
+ __m128i sign = _mm_set1_epi32((int8_t)more >> 7);
1839
+ // q = (q ^ sign) - sign;
1840
+ q = _mm_sub_epi32(_mm_xor_si128(q, sign), sign);
1841
+ return q;
1842
+ }
1843
+ else {
1844
+ __m128i q = libdivide_mullhi_s32_vector(numers, _mm_set1_epi32(denom->magic));
1845
+ if (more & LIBDIVIDE_ADD_MARKER) {
1846
+ // must be arithmetic shift
1847
+ __m128i sign = _mm_set1_epi32((int8_t)more >> 7);
1848
+ // q += ((numer ^ sign) - sign);
1849
+ q = _mm_add_epi32(q, _mm_sub_epi32(_mm_xor_si128(numers, sign), sign));
1850
+ }
1851
+ // q >>= shift
1852
+ q = _mm_srai_epi32(q, more & LIBDIVIDE_32_SHIFT_MASK);
1853
+ q = _mm_add_epi32(q, _mm_srli_epi32(q, 31)); // q += (q < 0)
1854
+ return q;
1855
+ }
1856
+ }
1857
+
1858
+ __m128i libdivide_s32_branchfree_do_vector(__m128i numers, const struct libdivide_s32_branchfree_t *denom) {
1859
+ int32_t magic = denom->magic;
1860
+ uint8_t more = denom->more;
1861
+ uint8_t shift = more & LIBDIVIDE_32_SHIFT_MASK;
1862
+ // must be arithmetic shift
1863
+ __m128i sign = _mm_set1_epi32((int8_t)more >> 7);
1864
+ __m128i q = libdivide_mullhi_s32_vector(numers, _mm_set1_epi32(magic));
1865
+ q = _mm_add_epi32(q, numers); // q += numers
1866
+
1867
+ // If q is non-negative, we have nothing to do
1868
+ // If q is negative, we want to add either (2**shift)-1 if d is
1869
+ // a power of 2, or (2**shift) if it is not a power of 2
1870
+ uint32_t is_power_of_2 = (magic == 0);
1871
+ __m128i q_sign = _mm_srai_epi32(q, 31); // q_sign = q >> 31
1872
+ __m128i mask = _mm_set1_epi32((1U << shift) - is_power_of_2);
1873
+ q = _mm_add_epi32(q, _mm_and_si128(q_sign, mask)); // q = q + (q_sign & mask)
1874
+ q = _mm_srai_epi32(q, shift); // q >>= shift
1875
+ q = _mm_sub_epi32(_mm_xor_si128(q, sign), sign); // q = (q ^ sign) - sign
1876
+ return q;
1877
+ }
1878
+
1879
+ ////////// SINT64
1880
+
1881
+ __m128i libdivide_s64_do_vector(__m128i numers, const struct libdivide_s64_t *denom) {
1882
+ uint8_t more = denom->more;
1883
+ int64_t magic = denom->magic;
1884
+ if (magic == 0) { // shift path
1885
+ uint32_t shift = more & LIBDIVIDE_64_SHIFT_MASK;
1886
+ uint64_t mask = (1ULL << shift) - 1;
1887
+ __m128i roundToZeroTweak = _mm_set1_epi64x(mask);
1888
+ // q = numer + ((numer >> 63) & roundToZeroTweak);
1889
+ __m128i q = _mm_add_epi64(numers, _mm_and_si128(libdivide_s64_signbits(numers), roundToZeroTweak));
1890
+ q = libdivide_s64_shift_right_vector(q, shift);
1891
+ __m128i sign = _mm_set1_epi32((int8_t)more >> 7);
1892
+ // q = (q ^ sign) - sign;
1893
+ q = _mm_sub_epi64(_mm_xor_si128(q, sign), sign);
1894
+ return q;
1895
+ }
1896
+ else {
1897
+ __m128i q = libdivide_mullhi_s64_vector(numers, _mm_set1_epi64x(magic));
1898
+ if (more & LIBDIVIDE_ADD_MARKER) {
1899
+ // must be arithmetic shift
1900
+ __m128i sign = _mm_set1_epi32((int8_t)more >> 7);
1901
+ // q += ((numer ^ sign) - sign);
1902
+ q = _mm_add_epi64(q, _mm_sub_epi64(_mm_xor_si128(numers, sign), sign));
1903
+ }
1904
+ // q >>= denom->mult_path.shift
1905
+ q = libdivide_s64_shift_right_vector(q, more & LIBDIVIDE_64_SHIFT_MASK);
1906
+ q = _mm_add_epi64(q, _mm_srli_epi64(q, 63)); // q += (q < 0)
1907
+ return q;
1908
+ }
1909
+ }
1910
+
1911
+ __m128i libdivide_s64_branchfree_do_vector(__m128i numers, const struct libdivide_s64_branchfree_t *denom) {
1912
+ int64_t magic = denom->magic;
1913
+ uint8_t more = denom->more;
1914
+ uint8_t shift = more & LIBDIVIDE_64_SHIFT_MASK;
1915
+ // must be arithmetic shift
1916
+ __m128i sign = _mm_set1_epi32((int8_t)more >> 7);
1917
+
1918
+ // libdivide_mullhi_s64(numers, magic);
1919
+ __m128i q = libdivide_mullhi_s64_vector(numers, _mm_set1_epi64x(magic));
1920
+ q = _mm_add_epi64(q, numers); // q += numers
1921
+
1922
+ // If q is non-negative, we have nothing to do.
1923
+ // If q is negative, we want to add either (2**shift)-1 if d is
1924
+ // a power of 2, or (2**shift) if it is not a power of 2.
1925
+ uint32_t is_power_of_2 = (magic == 0);
1926
+ __m128i q_sign = libdivide_s64_signbits(q); // q_sign = q >> 63
1927
+ __m128i mask = _mm_set1_epi64x((1ULL << shift) - is_power_of_2);
1928
+ q = _mm_add_epi64(q, _mm_and_si128(q_sign, mask)); // q = q + (q_sign & mask)
1929
+ q = libdivide_s64_shift_right_vector(q, shift); // q >>= shift
1930
+ q = _mm_sub_epi64(_mm_xor_si128(q, sign), sign); // q = (q ^ sign) - sign
1931
+ return q;
1932
+ }
1933
+
1934
+ #endif
1935
+
1936
+ /////////// C++ stuff
1937
+
1938
+ #ifdef __cplusplus
1939
+
1940
+ // The C++ divider class is templated on both an integer type
1941
+ // (like uint64_t) and an algorithm type.
1942
+ // * BRANCHFULL is the default algorithm type.
1943
+ // * BRANCHFREE is the branchfree algorithm type.
1944
+ enum {
1945
+ BRANCHFULL,
1946
+ BRANCHFREE
1947
+ };
1948
+
1949
+ #if defined(LIBDIVIDE_AVX512)
1950
+ #define LIBDIVIDE_VECTOR_TYPE __m512i
1951
+ #elif defined(LIBDIVIDE_AVX2)
1952
+ #define LIBDIVIDE_VECTOR_TYPE __m256i
1953
+ #elif defined(LIBDIVIDE_SSE2)
1954
+ #define LIBDIVIDE_VECTOR_TYPE __m128i
1955
+ #endif
1956
+
1957
+ #if !defined(LIBDIVIDE_VECTOR_TYPE)
1958
+ #define LIBDIVIDE_DIVIDE_VECTOR(ALGO)
1959
+ #else
1960
+ #define LIBDIVIDE_DIVIDE_VECTOR(ALGO) \
1961
+ LIBDIVIDE_VECTOR_TYPE divide(LIBDIVIDE_VECTOR_TYPE n) const { \
1962
+ return libdivide_##ALGO##_do_vector(n, &denom); \
1963
+ }
1964
+ #endif
1965
+
1966
+ // The DISPATCHER_GEN() macro generates C++ methods (for the given integer
1967
+ // and algorithm types) that redirect to libdivide's C API.
1968
+ #define DISPATCHER_GEN(T, ALGO) \
1969
+ libdivide_##ALGO##_t denom; \
1970
+ dispatcher() { } \
1971
+ dispatcher(T d) \
1972
+ : denom(libdivide_##ALGO##_gen(d)) \
1973
+ { } \
1974
+ T divide(T n) const { \
1975
+ return libdivide_##ALGO##_do(n, &denom); \
1976
+ } \
1977
+ LIBDIVIDE_DIVIDE_VECTOR(ALGO) \
1978
+ T recover() const { \
1979
+ return libdivide_##ALGO##_recover(&denom); \
1980
+ }
1981
+
1982
+ // The dispatcher selects a specific division algorithm for a given
1983
+ // type and ALGO using partial template specialization.
1984
+ template<bool IS_INTEGRAL, bool IS_SIGNED, int SIZEOF, int ALGO> struct dispatcher { };
1985
+
1986
+ template<> struct dispatcher<true, true, sizeof(int32_t), BRANCHFULL> { DISPATCHER_GEN(int32_t, s32) };
1987
+ template<> struct dispatcher<true, true, sizeof(int32_t), BRANCHFREE> { DISPATCHER_GEN(int32_t, s32_branchfree) };
1988
+ template<> struct dispatcher<true, false, sizeof(uint32_t), BRANCHFULL> { DISPATCHER_GEN(uint32_t, u32) };
1989
+ template<> struct dispatcher<true, false, sizeof(uint32_t), BRANCHFREE> { DISPATCHER_GEN(uint32_t, u32_branchfree) };
1990
+ template<> struct dispatcher<true, true, sizeof(int64_t), BRANCHFULL> { DISPATCHER_GEN(int64_t, s64) };
1991
+ template<> struct dispatcher<true, true, sizeof(int64_t), BRANCHFREE> { DISPATCHER_GEN(int64_t, s64_branchfree) };
1992
+ template<> struct dispatcher<true, false, sizeof(uint64_t), BRANCHFULL> { DISPATCHER_GEN(uint64_t, u64) };
1993
+ template<> struct dispatcher<true, false, sizeof(uint64_t), BRANCHFREE> { DISPATCHER_GEN(uint64_t, u64_branchfree) };
1994
+
1995
+ // This is the main divider class for use by the user (C++ API).
1996
+ // The actual division algorithm is selected using the dispatcher struct
1997
+ // based on the integer and algorithm template parameters.
1998
+ template<typename T, int ALGO = BRANCHFULL>
1999
+ class divider {
2000
+ public:
2001
+ // We leave the default constructor empty so that creating
2002
+ // an array of dividers and then initializing them
2003
+ // later doesn't slow us down.
2004
+ divider() { }
2005
+
2006
+ // Constructor that takes the divisor as a parameter
2007
+ divider(T d) : div(d) { }
2008
+
2009
+ // Divides n by the divisor
2010
+ T divide(T n) const {
2011
+ return div.divide(n);
2012
+ }
2013
+
2014
+ // Recovers the divisor, returns the value that was
2015
+ // used to initialize this divider object.
2016
+ T recover() const {
2017
+ return div.recover();
2018
+ }
2019
+
2020
+ bool operator==(const divider<T, ALGO>& other) const {
2021
+ return div.denom.magic == other.denom.magic &&
2022
+ div.denom.more == other.denom.more;
2023
+ }
2024
+
2025
+ bool operator!=(const divider<T, ALGO>& other) const {
2026
+ return !(*this == other);
2027
+ }
2028
+
2029
+ #if defined(LIBDIVIDE_VECTOR_TYPE)
2030
+ // Treats the vector as packed integer values with the same type as
2031
+ // the divider (e.g. s32, u32, s64, u64) and divides each of
2032
+ // them by the divider, returning the packed quotients.
2033
+ LIBDIVIDE_VECTOR_TYPE divide(LIBDIVIDE_VECTOR_TYPE n) const {
2034
+ return div.divide(n);
2035
+ }
2036
+ #endif
2037
+
2038
+ private:
2039
+ // Storage for the actual divisor
2040
+ dispatcher<std::is_integral<T>::value,
2041
+ std::is_signed<T>::value, sizeof(T), ALGO> div;
2042
+ };
2043
+
2044
+ // Overload of operator / for scalar division
2045
+ template<typename T, int ALGO>
2046
+ T operator/(T n, const divider<T, ALGO>& div) {
2047
+ return div.divide(n);
2048
+ }
2049
+
2050
+ // Overload of operator /= for scalar division
2051
+ template<typename T, int ALGO>
2052
+ T& operator/=(T& n, const divider<T, ALGO>& div) {
2053
+ n = div.divide(n);
2054
+ return n;
2055
+ }
2056
+
2057
+ #if defined(LIBDIVIDE_VECTOR_TYPE)
2058
+ // Overload of operator / for vector division
2059
+ template<typename T, int ALGO>
2060
+ LIBDIVIDE_VECTOR_TYPE operator/(LIBDIVIDE_VECTOR_TYPE n, const divider<T, ALGO>& div) {
2061
+ return div.divide(n);
2062
+ }
2063
+ // Overload of operator /= for vector division
2064
+ template<typename T, int ALGO>
2065
+ LIBDIVIDE_VECTOR_TYPE& operator/=(LIBDIVIDE_VECTOR_TYPE& n, const divider<T, ALGO>& div) {
2066
+ n = div.divide(n);
2067
+ return n;
2068
+ }
2069
+ #endif
2070
+
2071
+ // libdivdie::branchfree_divider<T>
2072
+ template <typename T>
2073
+ using branchfree_divider = divider<T, BRANCHFREE>;
2074
+
2075
+ } // namespace libdivide
2076
+
2077
+ #endif // __cplusplus
2078
+
2079
+ #endif // NUMPY_CORE_INCLUDE_NUMPY_LIBDIVIDE_LIBDIVIDE_H_
evalkit_tf437/lib/python3.10/site-packages/numpy/core/lib/npy-pkg-config/npymath.ini ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ [meta]
2
+ Name=npymath
3
+ Description=Portable, core math library implementing C99 standard
4
+ Version=0.1
5
+
6
+ [variables]
7
+ pkgname=numpy.core
8
+ prefix=${pkgdir}
9
+ libdir=${prefix}/lib
10
+ includedir=${prefix}/include
11
+
12
+ [default]
13
+ Libs=-L${libdir} -lnpymath
14
+ Cflags=-I${includedir}
15
+ Requires=mlib
16
+
17
+ [msvc]
18
+ Libs=/LIBPATH:${libdir} npymath.lib
19
+ Cflags=/INCLUDE:${includedir}
20
+ Requires=mlib
evalkit_tf437/lib/python3.10/site-packages/numpy/core/tests/__pycache__/_locales.cpython-310.pyc ADDED
Binary file (2.52 kB). View file
 
evalkit_tf437/lib/python3.10/site-packages/numpy/core/tests/__pycache__/test_argparse.cpython-310.pyc ADDED
Binary file (2.48 kB). View file
 
evalkit_tf437/lib/python3.10/site-packages/numpy/core/tests/__pycache__/test_function_base.cpython-310.pyc ADDED
Binary file (16.5 kB). View file
 
evalkit_tf437/lib/python3.10/site-packages/numpy/core/tests/__pycache__/test_indexerrors.cpython-310.pyc ADDED
Binary file (7.38 kB). View file
 
evalkit_tf437/lib/python3.10/site-packages/numpy/core/tests/__pycache__/test_memmap.cpython-310.pyc ADDED
Binary file (7.73 kB). View file
 
evalkit_tf437/lib/python3.10/site-packages/numpy/core/tests/__pycache__/test_print.cpython-310.pyc ADDED
Binary file (5.95 kB). View file
 
evalkit_tf437/lib/python3.10/site-packages/numpy/core/tests/__pycache__/test_records.cpython-310.pyc ADDED
Binary file (19.7 kB). View file
 
evalkit_tf437/lib/python3.10/site-packages/numpy/core/tests/__pycache__/test_simd.cpython-310.pyc ADDED
Binary file (42.8 kB). View file
 
evalkit_tf437/lib/python3.10/site-packages/numpy/core/tests/__pycache__/test_umath_complex.cpython-310.pyc ADDED
Binary file (17.1 kB). View file
 
evalkit_tf437/lib/python3.10/site-packages/numpy/core/tests/data/astype_copy.pkl ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9564b309cbf3441ff0a6e4468fddaca46230fab34f15c77d87025a455bdf59d9
3
+ size 716
evalkit_tf437/lib/python3.10/site-packages/numpy/core/tests/data/umath-validation-set-log1p.csv ADDED
@@ -0,0 +1,1429 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ dtype,input,output,ulperrortol
2
+ np.float32,0x3e10aca8,0x3e075347,2
3
+ np.float32,0x3f776e66,0x3f2d2003,2
4
+ np.float32,0xbf34e8ce,0xbf9cfd5c,2
5
+ np.float32,0xbf0260ee,0xbf363f69,2
6
+ np.float32,0x3ed285e8,0x3eb05870,2
7
+ np.float32,0x262b88,0x262b88,2
8
+ np.float32,0x3eeffd6c,0x3ec4cfdb,2
9
+ np.float32,0x3ee86808,0x3ebf9f54,2
10
+ np.float32,0x3f36eba8,0x3f0a0524,2
11
+ np.float32,0xbf1c047a,0xbf70afc7,2
12
+ np.float32,0x3ead2916,0x3e952902,2
13
+ np.float32,0x61c9c9,0x61c9c9,2
14
+ np.float32,0xff7fffff,0xffc00000,2
15
+ np.float32,0x7f64ee52,0x42b138e0,2
16
+ np.float32,0x7ed00b1e,0x42afa4ff,2
17
+ np.float32,0x3db53340,0x3dada0b2,2
18
+ np.float32,0x3e6b0a4a,0x3e5397a4,2
19
+ np.float32,0x7ed5d64f,0x42afb310,2
20
+ np.float32,0xbf12bc5f,0xbf59f5ee,2
21
+ np.float32,0xbda12710,0xbda7d8b5,2
22
+ np.float32,0xbe2e89d8,0xbe3f5a9f,2
23
+ np.float32,0x3f5bee75,0x3f1ebea4,2
24
+ np.float32,0x9317a,0x9317a,2
25
+ np.float32,0x7ee00130,0x42afcad8,2
26
+ np.float32,0x7ef0d16d,0x42afefe7,2
27
+ np.float32,0xbec7463a,0xbefc6a44,2
28
+ np.float32,0xbf760ecc,0xc04fe59c,2
29
+ np.float32,0xbecacb3c,0xbf011ae3,2
30
+ np.float32,0x3ead92be,0x3e9577f0,2
31
+ np.float32,0xbf41510d,0xbfb41b3a,2
32
+ np.float32,0x7f71d489,0x42b154f1,2
33
+ np.float32,0x8023bcd5,0x8023bcd5,2
34
+ np.float32,0x801d33d8,0x801d33d8,2
35
+ np.float32,0x3f3f545d,0x3f0ee0d4,2
36
+ np.float32,0xbf700682,0xc0318c25,2
37
+ np.float32,0xbe54e990,0xbe6eb0a3,2
38
+ np.float32,0x7f0289bf,0x42b01941,2
39
+ np.float32,0xbd61ac90,0xbd682113,2
40
+ np.float32,0xbf2ff310,0xbf94cd6f,2
41
+ np.float32,0x7f10064a,0x42b04b98,2
42
+ np.float32,0x804d0d6d,0x804d0d6d,2
43
+ np.float32,0x80317b0a,0x80317b0a,2
44
+ np.float32,0xbddfef18,0xbded2640,2
45
+ np.float32,0x3f00c9ab,0x3ed0a5bd,2
46
+ np.float32,0x7f04b905,0x42b021c1,2
47
+ np.float32,0x7fc00000,0x7fc00000,2
48
+ np.float32,0x6524c4,0x6524c4,2
49
+ np.float32,0x3da08ae0,0x3d9a8f88,2
50
+ np.float32,0x293ea9,0x293ea9,2
51
+ np.float32,0x71499e,0x71499e,2
52
+ np.float32,0xbf14f54d,0xbf5f38a5,2
53
+ np.float32,0x806e60f5,0x806e60f5,2
54
+ np.float32,0x3f5f34bb,0x3f207fff,2
55
+ np.float32,0x80513427,0x80513427,2
56
+ np.float32,0x7f379670,0x42b0c7dc,2
57
+ np.float32,0x3efba888,0x3eccb20b,2
58
+ np.float32,0x3eeadd1b,0x3ec14f4b,2
59
+ np.float32,0x7ec5a27f,0x42af8ab8,2
60
+ np.float32,0x3f2afe4e,0x3f02f7a2,2
61
+ np.float32,0x5591c8,0x5591c8,2
62
+ np.float32,0x3dbb7240,0x3db35bab,2
63
+ np.float32,0x805b911b,0x805b911b,2
64
+ np.float32,0x800000,0x800000,2
65
+ np.float32,0x7e784c04,0x42ae9cab,2
66
+ np.float32,0x7ebaae14,0x42af6d86,2
67
+ np.float32,0xbec84f7a,0xbefe1d42,2
68
+ np.float32,0x7cea8281,0x42aa56bf,2
69
+ np.float32,0xbf542cf6,0xbfe1eb1b,2
70
+ np.float32,0xbf6bfb13,0xc0231a5b,2
71
+ np.float32,0x7d6eeaef,0x42abc32c,2
72
+ np.float32,0xbf062f6b,0xbf3e2000,2
73
+ np.float32,0x8073d8e9,0x8073d8e9,2
74
+ np.float32,0xbea4db14,0xbec6f485,2
75
+ np.float32,0x7d7e8d62,0x42abe3a0,2
76
+ np.float32,0x7e8fc34e,0x42aee7c6,2
77
+ np.float32,0x7dcbb0c3,0x42acd464,2
78
+ np.float32,0x7e123c,0x7e123c,2
79
+ np.float32,0x3d77af62,0x3d707c34,2
80
+ np.float32,0x498cc8,0x498cc8,2
81
+ np.float32,0x7f4e2206,0x42b1032a,2
82
+ np.float32,0x3f734e0a,0x3f2b04a1,2
83
+ np.float32,0x8053a9d0,0x8053a9d0,2
84
+ np.float32,0xbe8a67e0,0xbea15be9,2
85
+ np.float32,0xbf78e0ea,0xc065409e,2
86
+ np.float32,0x352bdd,0x352bdd,2
87
+ np.float32,0x3ee42be7,0x3ebcb38a,2
88
+ np.float32,0x7f482d10,0x42b0f427,2
89
+ np.float32,0xbf23155e,0xbf81b993,2
90
+ np.float32,0x594920,0x594920,2
91
+ np.float32,0x63f53f,0x63f53f,2
92
+ np.float32,0x363592,0x363592,2
93
+ np.float32,0x7dafbb78,0x42ac88cc,2
94
+ np.float32,0x7f69516c,0x42b14298,2
95
+ np.float32,0x3e1d5be2,0x3e126131,2
96
+ np.float32,0x410c23,0x410c23,2
97
+ np.float32,0x7ec9563c,0x42af9439,2
98
+ np.float32,0xbedd3a0e,0xbf10d705,2
99
+ np.float32,0x7f7c4f1f,0x42b16aa8,2
100
+ np.float32,0xbe99b34e,0xbeb6c2d3,2
101
+ np.float32,0x6cdc84,0x6cdc84,2
102
+ np.float32,0x5b3bbe,0x5b3bbe,2
103
+ np.float32,0x252178,0x252178,2
104
+ np.float32,0x7d531865,0x42ab83c8,2
105
+ np.float32,0xbf565b44,0xbfe873bf,2
106
+ np.float32,0x5977ce,0x5977ce,2
107
+ np.float32,0x588a58,0x588a58,2
108
+ np.float32,0x3eae7054,0x3e961d51,2
109
+ np.float32,0x725049,0x725049,2
110
+ np.float32,0x7f2b9386,0x42b0a538,2
111
+ np.float32,0xbe674714,0xbe831245,2
112
+ np.float32,0x8044f0d8,0x8044f0d8,2
113
+ np.float32,0x800a3c21,0x800a3c21,2
114
+ np.float32,0x807b275b,0x807b275b,2
115
+ np.float32,0xbf2463b6,0xbf83896e,2
116
+ np.float32,0x801cca42,0x801cca42,2
117
+ np.float32,0xbf28f2d0,0xbf8a121a,2
118
+ np.float32,0x3f4168c2,0x3f1010ce,2
119
+ np.float32,0x6f91a1,0x6f91a1,2
120
+ np.float32,0xbf2b9eeb,0xbf8e0fc5,2
121
+ np.float32,0xbea4c858,0xbec6d8e4,2
122
+ np.float32,0xbf7abba0,0xc0788e88,2
123
+ np.float32,0x802f18f7,0x802f18f7,2
124
+ np.float32,0xbf7f6c75,0xc0c3145c,2
125
+ np.float32,0xbe988210,0xbeb50f5e,2
126
+ np.float32,0xbf219b7e,0xbf7f6a3b,2
127
+ np.float32,0x7f800000,0x7f800000,2
128
+ np.float32,0x7f7fffff,0x42b17218,2
129
+ np.float32,0xbdca8d90,0xbdd5487e,2
130
+ np.float32,0xbef683b0,0xbf2821b0,2
131
+ np.float32,0x8043e648,0x8043e648,2
132
+ np.float32,0xbf4319a4,0xbfb7cd1b,2
133
+ np.float32,0x62c2b2,0x62c2b2,2
134
+ np.float32,0xbf479ccd,0xbfc1a7b1,2
135
+ np.float32,0x806c8a32,0x806c8a32,2
136
+ np.float32,0x7f004447,0x42b01045,2
137
+ np.float32,0x3f737d36,0x3f2b1ccf,2
138
+ np.float32,0x3ee71f24,0x3ebebced,2
139
+ np.float32,0x3ea0b6b4,0x3e8bc606,2
140
+ np.float32,0x358fd7,0x358fd7,2
141
+ np.float32,0xbe69780c,0xbe847d17,2
142
+ np.float32,0x7f6bed18,0x42b14849,2
143
+ np.float32,0xbf6a5113,0xc01dfe1d,2
144
+ np.float32,0xbf255693,0xbf84de88,2
145
+ np.float32,0x7f34acac,0x42b0bfac,2
146
+ np.float32,0xbe8a3b6a,0xbea11efe,2
147
+ np.float32,0x3f470d84,0x3f1342ab,2
148
+ np.float32,0xbf2cbde3,0xbf8fc602,2
149
+ np.float32,0x47c103,0x47c103,2
150
+ np.float32,0xe3c94,0xe3c94,2
151
+ np.float32,0xbec07afa,0xbef1693a,2
152
+ np.float32,0x6a9cfe,0x6a9cfe,2
153
+ np.float32,0xbe4339e0,0xbe5899da,2
154
+ np.float32,0x7ea9bf1e,0x42af3cd6,2
155
+ np.float32,0x3f6378b4,0x3f22c4c4,2
156
+ np.float32,0xbd989ff0,0xbd9e9c77,2
157
+ np.float32,0xbe6f2f50,0xbe88343d,2
158
+ np.float32,0x3f7f2ac5,0x3f310764,2
159
+ np.float32,0x3f256704,0x3eff2fb2,2
160
+ np.float32,0x80786aca,0x80786aca,2
161
+ np.float32,0x65d02f,0x65d02f,2
162
+ np.float32,0x50d1c3,0x50d1c3,2
163
+ np.float32,0x3f4a9d76,0x3f1541b4,2
164
+ np.float32,0x802cf491,0x802cf491,2
165
+ np.float32,0x3e935cec,0x3e81829b,2
166
+ np.float32,0x3e2ad478,0x3e1dfd81,2
167
+ np.float32,0xbf107cbd,0xbf54bef2,2
168
+ np.float32,0xbf58c02e,0xbff007fe,2
169
+ np.float32,0x80090808,0x80090808,2
170
+ np.float32,0x805d1f66,0x805d1f66,2
171
+ np.float32,0x6aec95,0x6aec95,2
172
+ np.float32,0xbee3fc6e,0xbf16dc73,2
173
+ np.float32,0x7f63314b,0x42b134f9,2
174
+ np.float32,0x550443,0x550443,2
175
+ np.float32,0xbefa8174,0xbf2c026e,2
176
+ np.float32,0x3f7fb380,0x3f314bd5,2
177
+ np.float32,0x80171f2c,0x80171f2c,2
178
+ np.float32,0x3f2f56ae,0x3f058f2d,2
179
+ np.float32,0x3eacaecb,0x3e94cd97,2
180
+ np.float32,0xbe0c4f0c,0xbe16e69d,2
181
+ np.float32,0x3f48e4cb,0x3f144b42,2
182
+ np.float32,0x7f03efe2,0x42b01eb7,2
183
+ np.float32,0xbf1019ac,0xbf53dbe9,2
184
+ np.float32,0x3e958524,0x3e832eb5,2
185
+ np.float32,0xbf1b23c6,0xbf6e72f2,2
186
+ np.float32,0x12c554,0x12c554,2
187
+ np.float32,0x7dee588c,0x42ad24d6,2
188
+ np.float32,0xbe8c216c,0xbea3ba70,2
189
+ np.float32,0x804553cb,0x804553cb,2
190
+ np.float32,0xbe446324,0xbe5a0966,2
191
+ np.float32,0xbef7150a,0xbf28adff,2
192
+ np.float32,0xbf087282,0xbf42ec6e,2
193
+ np.float32,0x3eeef15c,0x3ec41937,2
194
+ np.float32,0x61bbd2,0x61bbd2,2
195
+ np.float32,0x3e51b28d,0x3e3ec538,2
196
+ np.float32,0x57e869,0x57e869,2
197
+ np.float32,0x7e5e7711,0x42ae646c,2
198
+ np.float32,0x8050b173,0x8050b173,2
199
+ np.float32,0xbf63c90c,0xc00d2438,2
200
+ np.float32,0xbeba774c,0xbee7dcf8,2
201
+ np.float32,0x8016faac,0x8016faac,2
202
+ np.float32,0xbe8b448c,0xbea28aaf,2
203
+ np.float32,0x3e8cd448,0x3e78d29e,2
204
+ np.float32,0x80484e02,0x80484e02,2
205
+ np.float32,0x3f63ba68,0x3f22e78c,2
206
+ np.float32,0x2e87bb,0x2e87bb,2
207
+ np.float32,0x230496,0x230496,2
208
+ np.float32,0x1327b2,0x1327b2,2
209
+ np.float32,0xbf046c56,0xbf3a72d2,2
210
+ np.float32,0x3ecefe60,0x3eadd69a,2
211
+ np.float32,0x49c56e,0x49c56e,2
212
+ np.float32,0x3df22d60,0x3de4e550,2
213
+ np.float32,0x3f67c19d,0x3f250707,2
214
+ np.float32,0x3f20eb9c,0x3ef9b624,2
215
+ np.float32,0x3f05ca75,0x3ed742fa,2
216
+ np.float32,0xbe8514f8,0xbe9a1d45,2
217
+ np.float32,0x8070a003,0x8070a003,2
218
+ np.float32,0x7e49650e,0x42ae317a,2
219
+ np.float32,0x3de16ce9,0x3dd5dc3e,2
220
+ np.float32,0xbf4ae952,0xbfc95f1f,2
221
+ np.float32,0xbe44dd84,0xbe5aa0db,2
222
+ np.float32,0x803c3bc0,0x803c3bc0,2
223
+ np.float32,0x3eebb9e8,0x3ec1e692,2
224
+ np.float32,0x80588275,0x80588275,2
225
+ np.float32,0xbea1e69a,0xbec29d86,2
226
+ np.float32,0x3f7b4bf8,0x3f2f154c,2
227
+ np.float32,0x7eb47ecc,0x42af5c46,2
228
+ np.float32,0x3d441e00,0x3d3f911a,2
229
+ np.float32,0x7f54d40e,0x42b11388,2
230
+ np.float32,0xbf47f17e,0xbfc26882,2
231
+ np.float32,0x3ea7da57,0x3e912db4,2
232
+ np.float32,0x3f59cc7b,0x3f1d984e,2
233
+ np.float32,0x570e08,0x570e08,2
234
+ np.float32,0x3e99560c,0x3e8620a2,2
235
+ np.float32,0x3ecfbd14,0x3eae5e55,2
236
+ np.float32,0x7e86be08,0x42aec698,2
237
+ np.float32,0x3f10f28a,0x3ee5b5d3,2
238
+ np.float32,0x7f228722,0x42b0897a,2
239
+ np.float32,0x3f4b979b,0x3f15cd30,2
240
+ np.float32,0xbf134283,0xbf5b30f9,2
241
+ np.float32,0x3f2ae16a,0x3f02e64f,2
242
+ np.float32,0x3e98e158,0x3e85c6cc,2
243
+ np.float32,0x7ec39f27,0x42af857a,2
244
+ np.float32,0x3effedb0,0x3ecf8cea,2
245
+ np.float32,0xbd545620,0xbd5a09c1,2
246
+ np.float32,0x503a28,0x503a28,2
247
+ np.float32,0x3f712744,0x3f29e9a1,2
248
+ np.float32,0x3edc6194,0x3eb748b1,2
249
+ np.float32,0xbf4ec1e5,0xbfd2ff5f,2
250
+ np.float32,0x3f46669e,0x3f12e4b5,2
251
+ np.float32,0xabad3,0xabad3,2
252
+ np.float32,0x80000000,0x80000000,2
253
+ np.float32,0x803f2e6d,0x803f2e6d,2
254
+ np.float32,0xbf431542,0xbfb7c3e6,2
255
+ np.float32,0x3f6f2d53,0x3f28e496,2
256
+ np.float32,0x546bd8,0x546bd8,2
257
+ np.float32,0x25c80a,0x25c80a,2
258
+ np.float32,0x3e50883c,0x3e3dcd7e,2
259
+ np.float32,0xbf5fa2ba,0xc0045c14,2
260
+ np.float32,0x80271c07,0x80271c07,2
261
+ np.float32,0x8043755d,0x8043755d,2
262
+ np.float32,0xbf3c5cea,0xbfaa5ee9,2
263
+ np.float32,0x3f2fea38,0x3f05e6af,2
264
+ np.float32,0x6da3dc,0x6da3dc,2
265
+ np.float32,0xbf095945,0xbf44dc70,2
266
+ np.float32,0xbe33d584,0xbe45c1f5,2
267
+ np.float32,0x7eb41b2e,0x42af5b2b,2
268
+ np.float32,0xbf0feb74,0xbf537242,2
269
+ np.float32,0xbe96225a,0xbeb1b0b1,2
270
+ np.float32,0x3f63b95f,0x3f22e700,2
271
+ np.float32,0x0,0x0,2
272
+ np.float32,0x3e20b0cc,0x3e154374,2
273
+ np.float32,0xbf79880c,0xc06b6801,2
274
+ np.float32,0xbea690b6,0xbec97b93,2
275
+ np.float32,0xbf3e11ca,0xbfada449,2
276
+ np.float32,0x7e7e6292,0x42aea912,2
277
+ np.float32,0x3e793350,0x3e5f0b7b,2
278
+ np.float32,0x802e7183,0x802e7183,2
279
+ np.float32,0x3f1b3695,0x3ef2a788,2
280
+ np.float32,0x801efa20,0x801efa20,2
281
+ np.float32,0x3f1ec43a,0x3ef70f42,2
282
+ np.float32,0xbf12c5ed,0xbf5a0c52,2
283
+ np.float32,0x8005e99c,0x8005e99c,2
284
+ np.float32,0xbf79f5e7,0xc06fcca5,2
285
+ np.float32,0x3ecbaf50,0x3eab7a03,2
286
+ np.float32,0x46b0fd,0x46b0fd,2
287
+ np.float32,0x3edb9023,0x3eb6b631,2
288
+ np.float32,0x7f24bc41,0x42b09063,2
289
+ np.float32,0xbd8d9328,0xbd92b4c6,2
290
+ np.float32,0x3f2c5d7f,0x3f03c9d9,2
291
+ np.float32,0x807bebc9,0x807bebc9,2
292
+ np.float32,0x7f797a99,0x42b164e2,2
293
+ np.float32,0x756e3c,0x756e3c,2
294
+ np.float32,0x80416f8a,0x80416f8a,2
295
+ np.float32,0x3e0d512a,0x3e04611a,2
296
+ np.float32,0x3f7be3e6,0x3f2f61ec,2
297
+ np.float32,0x80075c41,0x80075c41,2
298
+ np.float32,0xbe850294,0xbe9a046c,2
299
+ np.float32,0x684679,0x684679,2
300
+ np.float32,0x3eb393c4,0x3e99eed2,2
301
+ np.float32,0x3f4177c6,0x3f10195b,2
302
+ np.float32,0x3dd1f402,0x3dc7dfe5,2
303
+ np.float32,0x3ef484d4,0x3ec7e2e1,2
304
+ np.float32,0x53eb8f,0x53eb8f,2
305
+ np.float32,0x7f072cb6,0x42b02b20,2
306
+ np.float32,0xbf1b6b55,0xbf6f28d4,2
307
+ np.float32,0xbd8a98d8,0xbd8f827d,2
308
+ np.float32,0x3eafb418,0x3e970e96,2
309
+ np.float32,0x6555af,0x6555af,2
310
+ np.float32,0x7dd5118e,0x42aceb6f,2
311
+ np.float32,0x800a13f7,0x800a13f7,2
312
+ np.float32,0x331a9d,0x331a9d,2
313
+ np.float32,0x8063773f,0x8063773f,2
314
+ np.float32,0x3e95e068,0x3e837553,2
315
+ np.float32,0x80654b32,0x80654b32,2
316
+ np.float32,0x3dabe0e0,0x3da50bb3,2
317
+ np.float32,0xbf6283c3,0xc00a5280,2
318
+ np.float32,0x80751cc5,0x80751cc5,2
319
+ np.float32,0x3f668eb6,0x3f2465c0,2
320
+ np.float32,0x3e13c058,0x3e0a048c,2
321
+ np.float32,0x77780c,0x77780c,2
322
+ np.float32,0x3f7d6e48,0x3f302868,2
323
+ np.float32,0x7e31f9e3,0x42adf22f,2
324
+ np.float32,0x246c7b,0x246c7b,2
325
+ np.float32,0xbe915bf0,0xbeaafa6c,2
326
+ np.float32,0xbf800000,0xff800000,2
327
+ np.float32,0x3f698f42,0x3f25f8e0,2
328
+ np.float32,0x7e698885,0x42ae7d48,2
329
+ np.float32,0x3f5bbd42,0x3f1ea42c,2
330
+ np.float32,0x5b8444,0x5b8444,2
331
+ np.float32,0xbf6065f6,0xc005e2c6,2
332
+ np.float32,0xbeb95036,0xbee60dad,2
333
+ np.float32,0xbf44f846,0xbfbbcade,2
334
+ np.float32,0xc96e5,0xc96e5,2
335
+ np.float32,0xbf213e90,0xbf7e6eae,2
336
+ np.float32,0xbeb309cc,0xbedc4fe6,2
337
+ np.float32,0xbe781cf4,0xbe8e0fe6,2
338
+ np.float32,0x7f0cf0db,0x42b04083,2
339
+ np.float32,0xbf7b6143,0xc08078f9,2
340
+ np.float32,0x80526fc6,0x80526fc6,2
341
+ np.float32,0x3f092bf3,0x3edbaeec,2
342
+ np.float32,0x3ecdf154,0x3ead16df,2
343
+ np.float32,0x2fe85b,0x2fe85b,2
344
+ np.float32,0xbf5100a0,0xbfd8f871,2
345
+ np.float32,0xbec09d40,0xbef1a028,2
346
+ np.float32,0x5e6a85,0x5e6a85,2
347
+ np.float32,0xbec0e2a0,0xbef20f6b,2
348
+ np.float32,0x3f72e788,0x3f2ad00d,2
349
+ np.float32,0x880a6,0x880a6,2
350
+ np.float32,0x3d9e90bf,0x3d98b9fc,2
351
+ np.float32,0x15cf25,0x15cf25,2
352
+ np.float32,0x10171b,0x10171b,2
353
+ np.float32,0x805cf1aa,0x805cf1aa,2
354
+ np.float32,0x3f19bd36,0x3ef0d0d2,2
355
+ np.float32,0x3ebe2bda,0x3ea1b774,2
356
+ np.float32,0xbecd8192,0xbf035c49,2
357
+ np.float32,0x3e2ce508,0x3e1fc21b,2
358
+ np.float32,0x290f,0x290f,2
359
+ np.float32,0x803b679f,0x803b679f,2
360
+ np.float32,0x1,0x1,2
361
+ np.float32,0x807a9c76,0x807a9c76,2
362
+ np.float32,0xbf65fced,0xc01257f8,2
363
+ np.float32,0x3f783414,0x3f2d8475,2
364
+ np.float32,0x3f2d9d92,0x3f0488da,2
365
+ np.float32,0xbddb5798,0xbde80018,2
366
+ np.float32,0x3e91afb8,0x3e8034e7,2
367
+ np.float32,0xbf1b775a,0xbf6f476d,2
368
+ np.float32,0xbf73a32c,0xc041f3ba,2
369
+ np.float32,0xbea39364,0xbec5121b,2
370
+ np.float32,0x80375b94,0x80375b94,2
371
+ np.float32,0x3f331252,0x3f07c3e9,2
372
+ np.float32,0xbf285774,0xbf892e74,2
373
+ np.float32,0x3e699bb8,0x3e526d55,2
374
+ np.float32,0x3f08208a,0x3eda523a,2
375
+ np.float32,0xbf42fb4a,0xbfb78d60,2
376
+ np.float32,0x8029c894,0x8029c894,2
377
+ np.float32,0x3e926c0c,0x3e80c76e,2
378
+ np.float32,0x801e4715,0x801e4715,2
379
+ np.float32,0x3e4b36d8,0x3e395ffd,2
380
+ np.float32,0x8041556b,0x8041556b,2
381
+ np.float32,0xbf2d99ba,0xbf9119bd,2
382
+ np.float32,0x3ed83ea8,0x3eb46250,2
383
+ np.float32,0xbe94a280,0xbeaf92b4,2
384
+ np.float32,0x7f4c7a64,0x42b0ff0a,2
385
+ np.float32,0x806d4022,0x806d4022,2
386
+ np.float32,0xbed382f8,0xbf086d26,2
387
+ np.float32,0x1846fe,0x1846fe,2
388
+ np.float32,0xbe702558,0xbe88d4d8,2
389
+ np.float32,0xbe650ee0,0xbe81a3cc,2
390
+ np.float32,0x3ee9d088,0x3ec0970c,2
391
+ np.float32,0x7f6d4498,0x42b14b30,2
392
+ np.float32,0xbef9f9e6,0xbf2b7ddb,2
393
+ np.float32,0xbf70c384,0xc0349370,2
394
+ np.float32,0xbeff9e9e,0xbf3110c8,2
395
+ np.float32,0xbef06372,0xbf224aa9,2
396
+ np.float32,0xbf15a692,0xbf60e1fa,2
397
+ np.float32,0x8058c117,0x8058c117,2
398
+ np.float32,0xbd9f74b8,0xbda6017b,2
399
+ np.float32,0x801bf130,0x801bf130,2
400
+ np.float32,0x805da84c,0x805da84c,2
401
+ np.float32,0xff800000,0xffc00000,2
402
+ np.float32,0xbeb01de2,0xbed7d6d6,2
403
+ np.float32,0x8077de08,0x8077de08,2
404
+ np.float32,0x3e327668,0x3e2482c1,2
405
+ np.float32,0xbe7add88,0xbe8fe1ab,2
406
+ np.float32,0x805a3c2e,0x805a3c2e,2
407
+ np.float32,0x80326a73,0x80326a73,2
408
+ np.float32,0x800b8a34,0x800b8a34,2
409
+ np.float32,0x8048c83a,0x8048c83a,2
410
+ np.float32,0xbf3799d6,0xbfa1a975,2
411
+ np.float32,0x807649c7,0x807649c7,2
412
+ np.float32,0x3dfdbf90,0x3def3798,2
413
+ np.float32,0xbf1b538a,0xbf6eec4c,2
414
+ np.float32,0xbf1e5989,0xbf76baa0,2
415
+ np.float32,0xc7a80,0xc7a80,2
416
+ np.float32,0x8001be54,0x8001be54,2
417
+ np.float32,0x3f435bbc,0x3f112c6d,2
418
+ np.float32,0xbeabcff8,0xbed151d1,2
419
+ np.float32,0x7de20c78,0x42ad09b7,2
420
+ np.float32,0x3f0e6d2e,0x3ee27b1e,2
421
+ np.float32,0xbf0cb352,0xbf4c3267,2
422
+ np.float32,0x7f6ec06f,0x42b14e61,2
423
+ np.float32,0x7f6fa8ef,0x42b15053,2
424
+ np.float32,0xbf3d2a6a,0xbfabe623,2
425
+ np.float32,0x7f077a4c,0x42b02c46,2
426
+ np.float32,0xbf2a68dc,0xbf8c3cc4,2
427
+ np.float32,0x802a5dbe,0x802a5dbe,2
428
+ np.float32,0x807f631c,0x807f631c,2
429
+ np.float32,0x3dc9b8,0x3dc9b8,2
430
+ np.float32,0x3ebdc1b7,0x3ea16a0a,2
431
+ np.float32,0x7ef29dab,0x42aff3b5,2
432
+ np.float32,0x3e8ab1cc,0x3e757806,2
433
+ np.float32,0x3f27e88e,0x3f011c6d,2
434
+ np.float32,0x3cfd1455,0x3cf93fb5,2
435
+ np.float32,0x7f7eebf5,0x42b16fef,2
436
+ np.float32,0x3c9b2140,0x3c99ade9,2
437
+ np.float32,0x7e928601,0x42aef183,2
438
+ np.float32,0xbd7d2db0,0xbd82abae,2
439
+ np.float32,0x3e6f0df3,0x3e56da20,2
440
+ np.float32,0x7d36a2fc,0x42ab39a3,2
441
+ np.float32,0xbf49d3a2,0xbfc6c859,2
442
+ np.float32,0x7ee541d3,0x42afd6b6,2
443
+ np.float32,0x80753dc0,0x80753dc0,2
444
+ np.float32,0x3f4ce486,0x3f16865d,2
445
+ np.float32,0x39e701,0x39e701,2
446
+ np.float32,0x3f3d9ede,0x3f0de5fa,2
447
+ np.float32,0x7fafb2,0x7fafb2,2
448
+ np.float32,0x3e013fdc,0x3df37090,2
449
+ np.float32,0x807b6a2c,0x807b6a2c,2
450
+ np.float32,0xbe86800a,0xbe9c08c7,2
451
+ np.float32,0x7f40f080,0x42b0e14d,2
452
+ np.float32,0x7eef5afe,0x42afecc8,2
453
+ np.float32,0x7ec30052,0x42af83da,2
454
+ np.float32,0x3eacf768,0x3e9503e1,2
455
+ np.float32,0x7f13ef0e,0x42b0594e,2
456
+ np.float32,0x80419f4a,0x80419f4a,2
457
+ np.float32,0xbf485932,0xbfc3562a,2
458
+ np.float32,0xbe8a24d6,0xbea10011,2
459
+ np.float32,0xbda791c0,0xbdaed2bc,2
460
+ np.float32,0x3e9b5169,0x3e87a67d,2
461
+ np.float32,0x807dd882,0x807dd882,2
462
+ np.float32,0x7f40170e,0x42b0df0a,2
463
+ np.float32,0x7f02f7f9,0x42b01af1,2
464
+ np.float32,0x3ea38bf9,0x3e8decde,2
465
+ np.float32,0x3e2e7ce8,0x3e211ed4,2
466
+ np.float32,0x70a7a6,0x70a7a6,2
467
+ np.float32,0x7d978592,0x42ac3ce7,2
468
+ np.float32,0x804d12d0,0x804d12d0,2
469
+ np.float32,0x80165dc8,0x80165dc8,2
470
+ np.float32,0x80000001,0x80000001,2
471
+ np.float32,0x3e325da0,0x3e246da6,2
472
+ np.float32,0xbe063bb8,0xbe0fe281,2
473
+ np.float32,0x160b8,0x160b8,2
474
+ np.float32,0xbe5687a4,0xbe70bbef,2
475
+ np.float32,0x7f11ab34,0x42b05168,2
476
+ np.float32,0xc955c,0xc955c,2
477
+ np.float32,0xbea0003a,0xbebfd826,2
478
+ np.float32,0x3f7fbdd9,0x3f315102,2
479
+ np.float32,0xbe61aefc,0xbe7ef121,2
480
+ np.float32,0xbf1b9873,0xbf6f9bc3,2
481
+ np.float32,0x3a6d14,0x3a6d14,2
482
+ np.float32,0xbf1ad3b4,0xbf6da808,2
483
+ np.float32,0x3ed2dd24,0x3eb0963d,2
484
+ np.float32,0xbe81a4ca,0xbe957d52,2
485
+ np.float32,0x7f1be3e9,0x42b07421,2
486
+ np.float32,0x7f5ce943,0x42b1269e,2
487
+ np.float32,0x7eebcbdf,0x42afe51d,2
488
+ np.float32,0x807181b5,0x807181b5,2
489
+ np.float32,0xbecb03ba,0xbf0149ad,2
490
+ np.float32,0x42edb8,0x42edb8,2
491
+ np.float32,0xbf3aeec8,0xbfa7b13f,2
492
+ np.float32,0xbd0c4f00,0xbd0ec4a0,2
493
+ np.float32,0x3e48d260,0x3e376070,2
494
+ np.float32,0x1a9731,0x1a9731,2
495
+ np.float32,0x7f323be4,0x42b0b8b5,2
496
+ np.float32,0x1a327f,0x1a327f,2
497
+ np.float32,0x17f1fc,0x17f1fc,2
498
+ np.float32,0xbf2f4f9b,0xbf93c91a,2
499
+ np.float32,0x3ede8934,0x3eb8c9c3,2
500
+ np.float32,0xbf56aaac,0xbfe968bb,2
501
+ np.float32,0x3e22cb5a,0x3e17148c,2
502
+ np.float32,0x7d9def,0x7d9def,2
503
+ np.float32,0x8045b963,0x8045b963,2
504
+ np.float32,0x77404f,0x77404f,2
505
+ np.float32,0x7e2c9efb,0x42ade28b,2
506
+ np.float32,0x8058ad89,0x8058ad89,2
507
+ np.float32,0x7f4139,0x7f4139,2
508
+ np.float32,0x8020e12a,0x8020e12a,2
509
+ np.float32,0x800c9daa,0x800c9daa,2
510
+ np.float32,0x7f2c5ac5,0x42b0a789,2
511
+ np.float32,0x3f04a47b,0x3ed5c043,2
512
+ np.float32,0x804692d5,0x804692d5,2
513
+ np.float32,0xbf6e7fa4,0xc02bb493,2
514
+ np.float32,0x80330756,0x80330756,2
515
+ np.float32,0x7f3e29ad,0x42b0d9e1,2
516
+ np.float32,0xbebf689a,0xbeefb24d,2
517
+ np.float32,0x3f29a86c,0x3f022a56,2
518
+ np.float32,0x3e3bd1c0,0x3e2c72b3,2
519
+ np.float32,0x3f78f2e8,0x3f2de546,2
520
+ np.float32,0x3f3709be,0x3f0a16af,2
521
+ np.float32,0x3e11f150,0x3e086f97,2
522
+ np.float32,0xbf5867ad,0xbfeee8a0,2
523
+ np.float32,0xbebfb328,0xbef0296c,2
524
+ np.float32,0x2f7f15,0x2f7f15,2
525
+ np.float32,0x805cfe84,0x805cfe84,2
526
+ np.float32,0xbf504e01,0xbfd71589,2
527
+ np.float32,0x3ee0903c,0x3eba330c,2
528
+ np.float32,0xbd838990,0xbd87f399,2
529
+ np.float32,0x3f14444e,0x3ee9ee7d,2
530
+ np.float32,0x7e352583,0x42adfb3a,2
531
+ np.float32,0x7e76f824,0x42ae99ec,2
532
+ np.float32,0x3f772d00,0x3f2cfebf,2
533
+ np.float32,0x801f7763,0x801f7763,2
534
+ np.float32,0x3f760bf5,0x3f2c6b87,2
535
+ np.float32,0xbf0bb696,0xbf4a03a5,2
536
+ np.float32,0x3f175d2c,0x3eedd6d2,2
537
+ np.float32,0xbf5723f8,0xbfeae288,2
538
+ np.float32,0x24de0a,0x24de0a,2
539
+ np.float32,0x3cd73f80,0x3cd47801,2
540
+ np.float32,0x7f013305,0x42b013fa,2
541
+ np.float32,0x3e3ad425,0x3e2b9c50,2
542
+ np.float32,0x7d3d16,0x7d3d16,2
543
+ np.float32,0x3ef49738,0x3ec7ef54,2
544
+ np.float32,0x3f5b8612,0x3f1e8678,2
545
+ np.float32,0x7f0eeb5c,0x42b047a7,2
546
+ np.float32,0x7e9d7cb0,0x42af1675,2
547
+ np.float32,0xbdd1cfb0,0xbddd5aa0,2
548
+ np.float32,0xbf645dba,0xc00e78fe,2
549
+ np.float32,0x3f511174,0x3f18d56c,2
550
+ np.float32,0x3d91ad00,0x3d8cba62,2
551
+ np.float32,0x805298da,0x805298da,2
552
+ np.float32,0xbedb6af4,0xbf0f4090,2
553
+ np.float32,0x3d23b1ba,0x3d208205,2
554
+ np.float32,0xbea5783e,0xbec7dc87,2
555
+ np.float32,0x79d191,0x79d191,2
556
+ np.float32,0x3e894413,0x3e7337da,2
557
+ np.float32,0x80800000,0x80800000,2
558
+ np.float32,0xbf34a8d3,0xbf9c907b,2
559
+ np.float32,0x3bae779a,0x3bae011f,2
560
+ np.float32,0x8049284d,0x8049284d,2
561
+ np.float32,0x3eb42cc4,0x3e9a600b,2
562
+ np.float32,0x3da1e2d0,0x3d9bce5f,2
563
+ np.float32,0x3f364b8a,0x3f09a7af,2
564
+ np.float32,0x3d930b10,0x3d8e0118,2
565
+ np.float32,0x8061f8d7,0x8061f8d7,2
566
+ np.float32,0x3f473213,0x3f13573b,2
567
+ np.float32,0x3f1e2a38,0x3ef65102,2
568
+ np.float32,0x8068f7d9,0x8068f7d9,2
569
+ np.float32,0x3f181ef8,0x3eeeca2c,2
570
+ np.float32,0x3eeb6168,0x3ec1a9f5,2
571
+ np.float32,0xc2db6,0xc2db6,2
572
+ np.float32,0x3ef7b578,0x3eca0a69,2
573
+ np.float32,0xbf5b5a84,0xbff8d075,2
574
+ np.float32,0x7f479d5f,0x42b0f2b7,2
575
+ np.float32,0x3e6f3c24,0x3e56ff92,2
576
+ np.float32,0x3f45543a,0x3f1249f0,2
577
+ np.float32,0xbea7c1fa,0xbecb40d2,2
578
+ np.float32,0x7de082,0x7de082,2
579
+ np.float32,0x383729,0x383729,2
580
+ np.float32,0xbd91cb90,0xbd973eb3,2
581
+ np.float32,0x7f320218,0x42b0b80f,2
582
+ np.float32,0x5547f2,0x5547f2,2
583
+ np.float32,0x291fe4,0x291fe4,2
584
+ np.float32,0xbe078ba0,0xbe11655f,2
585
+ np.float32,0x7e0c0658,0x42ad7764,2
586
+ np.float32,0x7e129a2b,0x42ad8ee5,2
587
+ np.float32,0x3f7c96d4,0x3f2fbc0c,2
588
+ np.float32,0x3f800000,0x3f317218,2
589
+ np.float32,0x7f131754,0x42b05662,2
590
+ np.float32,0x15f833,0x15f833,2
591
+ np.float32,0x80392ced,0x80392ced,2
592
+ np.float32,0x3f7c141a,0x3f2f7a36,2
593
+ np.float32,0xbf71c03f,0xc038dcfd,2
594
+ np.float32,0xbe14fb2c,0xbe20fff3,2
595
+ np.float32,0xbee0bac6,0xbf13f14c,2
596
+ np.float32,0x801a32dd,0x801a32dd,2
597
+ np.float32,0x8e12d,0x8e12d,2
598
+ np.float32,0x3f48c606,0x3f143a04,2
599
+ np.float32,0x7f418af5,0x42b0e2e6,2
600
+ np.float32,0x3f1f2918,0x3ef78bb7,2
601
+ np.float32,0x11141b,0x11141b,2
602
+ np.float32,0x3e9fc9e8,0x3e8b11ad,2
603
+ np.float32,0xbea5447a,0xbec79010,2
604
+ np.float32,0xbe31d904,0xbe4359db,2
605
+ np.float32,0x80184667,0x80184667,2
606
+ np.float32,0xbf00503c,0xbf3212c2,2
607
+ np.float32,0x3e0328cf,0x3df6d425,2
608
+ np.float32,0x7ee8e1b7,0x42afdebe,2
609
+ np.float32,0xbef95e24,0xbf2ae5db,2
610
+ np.float32,0x7f3e4eed,0x42b0da45,2
611
+ np.float32,0x3f43ee85,0x3f117fa0,2
612
+ np.float32,0xbcfa2ac0,0xbcfe10fe,2
613
+ np.float32,0x80162774,0x80162774,2
614
+ np.float32,0x372e8b,0x372e8b,2
615
+ np.float32,0x3f263802,0x3f0016b0,2
616
+ np.float32,0x8008725f,0x8008725f,2
617
+ np.float32,0x800beb40,0x800beb40,2
618
+ np.float32,0xbe93308e,0xbead8a77,2
619
+ np.float32,0x3d8a4240,0x3d85cab8,2
620
+ np.float32,0x80179de0,0x80179de0,2
621
+ np.float32,0x7f4a98f2,0x42b0fa4f,2
622
+ np.float32,0x3f0d214e,0x3ee0cff1,2
623
+ np.float32,0x80536c2c,0x80536c2c,2
624
+ np.float32,0x7e7038ed,0x42ae8bbe,2
625
+ np.float32,0x7f345af9,0x42b0bec4,2
626
+ np.float32,0xbf243219,0xbf83442f,2
627
+ np.float32,0x7e0d5555,0x42ad7c27,2
628
+ np.float32,0x762e95,0x762e95,2
629
+ np.float32,0x7ebf4548,0x42af79f6,2
630
+ np.float32,0x8079639e,0x8079639e,2
631
+ np.float32,0x3ef925c0,0x3ecb0260,2
632
+ np.float32,0x3f708695,0x3f2996d6,2
633
+ np.float32,0xfca9f,0xfca9f,2
634
+ np.float32,0x8060dbf4,0x8060dbf4,2
635
+ np.float32,0x4c8840,0x4c8840,2
636
+ np.float32,0xbea922ee,0xbecd4ed5,2
637
+ np.float32,0xbf4f28a9,0xbfd40b98,2
638
+ np.float32,0xbe25ad48,0xbe34ba1b,2
639
+ np.float32,0x3f2fb254,0x3f05c58c,2
640
+ np.float32,0x3f73bcc2,0x3f2b3d5f,2
641
+ np.float32,0xbf479a07,0xbfc1a165,2
642
+ np.float32,0xbeb9a808,0xbee69763,2
643
+ np.float32,0x7eb16a65,0x42af5376,2
644
+ np.float32,0xbeb3e442,0xbedda042,2
645
+ np.float32,0x3d8f439c,0x3d8a79ac,2
646
+ np.float32,0x80347516,0x80347516,2
647
+ np.float32,0x3e8a0c5d,0x3e74738c,2
648
+ np.float32,0xbf0383a4,0xbf389289,2
649
+ np.float32,0x806be8f5,0x806be8f5,2
650
+ np.float32,0x8023f0c5,0x8023f0c5,2
651
+ np.float32,0x2060e9,0x2060e9,2
652
+ np.float32,0xbf759eba,0xc04d239f,2
653
+ np.float32,0x3d84cc5a,0x3d80ab96,2
654
+ np.float32,0xbf57746b,0xbfebdf87,2
655
+ np.float32,0x3e418417,0x3e31401f,2
656
+ np.float32,0xaecce,0xaecce,2
657
+ np.float32,0x3cd1766f,0x3cced45c,2
658
+ np.float32,0x53724a,0x53724a,2
659
+ np.float32,0x3f773710,0x3f2d03de,2
660
+ np.float32,0x8013d040,0x8013d040,2
661
+ np.float32,0x4d0eb2,0x4d0eb2,2
662
+ np.float32,0x8014364a,0x8014364a,2
663
+ np.float32,0x7f3c56c9,0x42b0d4f2,2
664
+ np.float32,0x3eee1e1c,0x3ec3891a,2
665
+ np.float32,0xbdda3eb8,0xbde6c5a0,2
666
+ np.float32,0x26ef4a,0x26ef4a,2
667
+ np.float32,0x7ed3370c,0x42afacbf,2
668
+ np.float32,0xbf06e31b,0xbf3f9ab7,2
669
+ np.float32,0xbe3185f0,0xbe42f556,2
670
+ np.float32,0x3dcf9abe,0x3dc5be41,2
671
+ np.float32,0xbf3696d9,0xbf9fe2bd,2
672
+ np.float32,0x3e68ee50,0x3e51e01a,2
673
+ np.float32,0x3f3d4cc2,0x3f0db6ca,2
674
+ np.float32,0x7fa00000,0x7fe00000,2
675
+ np.float32,0xbf03070c,0xbf3792d0,2
676
+ np.float32,0x3ea79e6c,0x3e910092,2
677
+ np.float32,0xbf1a393a,0xbf6c2251,2
678
+ np.float32,0x3f41eb0e,0x3f105afc,2
679
+ np.float32,0x3ceadb2f,0x3ce78d79,2
680
+ np.float32,0xbf5dc105,0xc000be2c,2
681
+ np.float32,0x7ebb5a0e,0x42af6f5c,2
682
+ np.float32,0xbf7c44eb,0xc0875058,2
683
+ np.float32,0x6aaaf4,0x6aaaf4,2
684
+ np.float32,0x807d8f23,0x807d8f23,2
685
+ np.float32,0xbee6b142,0xbf194fef,2
686
+ np.float32,0xbe83f256,0xbe989526,2
687
+ np.float32,0x7d588e,0x7d588e,2
688
+ np.float32,0x7cc80131,0x42aa0542,2
689
+ np.float32,0x3e0ab198,0x3e02124f,2
690
+ np.float32,0xbf6e64db,0xc02b52eb,2
691
+ np.float32,0x3d238b56,0x3d205d1b,2
692
+ np.float32,0xbeb408e2,0xbeddd8bc,2
693
+ np.float32,0x3f78340d,0x3f2d8471,2
694
+ np.float32,0x806162a3,0x806162a3,2
695
+ np.float32,0x804e484f,0x804e484f,2
696
+ np.float32,0xbeb8c576,0xbee53466,2
697
+ np.float32,0x807aab15,0x807aab15,2
698
+ np.float32,0x3f523e20,0x3f197ab8,2
699
+ np.float32,0xbf009190,0xbf3295de,2
700
+ np.float32,0x3df43da5,0x3de6bd82,2
701
+ np.float32,0x7f639aea,0x42b135e6,2
702
+ np.float32,0x3f1e638a,0x3ef697da,2
703
+ np.float32,0xbf4884de,0xbfc3bac3,2
704
+ np.float32,0xbe9336b6,0xbead931b,2
705
+ np.float32,0x6daf7f,0x6daf7f,2
706
+ np.float32,0xbf1fc152,0xbf7a70b1,2
707
+ np.float32,0x3f103720,0x3ee4c649,2
708
+ np.float32,0x3eeaa227,0x3ec126df,2
709
+ np.float32,0x7f7ea945,0x42b16f69,2
710
+ np.float32,0x3d3cd800,0x3d389ead,2
711
+ np.float32,0x3f3d7268,0x3f0dcc6e,2
712
+ np.float32,0xbf3c1b41,0xbfa9e2e3,2
713
+ np.float32,0x3ecf3818,0x3eadffb2,2
714
+ np.float32,0x3f1af312,0x3ef25372,2
715
+ np.float32,0x48fae4,0x48fae4,2
716
+ np.float64,0x7fedaa1ee4fb543d,0x40862da7ca7c308e,2
717
+ np.float64,0x8007d2d810efa5b1,0x8007d2d810efa5b1,2
718
+ np.float64,0x3fc385e069270bc0,0x3fc22b8884cf2c3b,2
719
+ np.float64,0x68ed4130d1da9,0x68ed4130d1da9,2
720
+ np.float64,0x8008e93e58d1d27d,0x8008e93e58d1d27d,2
721
+ np.float64,0xbfd3d62852a7ac50,0xbfd7be3a7ad1af02,2
722
+ np.float64,0xbfc1fa0ba923f418,0xbfc35f0f19447df7,2
723
+ np.float64,0xbfe01b8cec20371a,0xbfe6658c7e6c8e50,2
724
+ np.float64,0xbfeda81a147b5034,0xc004e9c94f2b91c1,2
725
+ np.float64,0xbfe1c36a97e386d5,0xbfe9ead4d6beaa92,2
726
+ np.float64,0x3fe50be51f2a17ca,0x3fe02c8067d9e5c5,2
727
+ np.float64,0x3febed4d3337da9a,0x3fe413956466134f,2
728
+ np.float64,0x80068ea59ced1d4c,0x80068ea59ced1d4c,2
729
+ np.float64,0x3febe77d5877cefb,0x3fe4107ac088bc71,2
730
+ np.float64,0x800ae77617d5ceed,0x800ae77617d5ceed,2
731
+ np.float64,0x3fd0546b60a0a8d7,0x3fcd16c2e995ab23,2
732
+ np.float64,0xbfe33e1476667c29,0xbfed6d7faec4db2f,2
733
+ np.float64,0x3fe9d2fd51b3a5fb,0x3fe2eef834310219,2
734
+ np.float64,0x8004249878284932,0x8004249878284932,2
735
+ np.float64,0xbfd5b485c72b690c,0xbfda828ccc6a7a5c,2
736
+ np.float64,0x7fcd6e6b6b3adcd6,0x408622807f04768e,2
737
+ np.float64,0x3fd7f9c32caff386,0x3fd45d024514b8da,2
738
+ np.float64,0x7f87eb9d702fd73a,0x40860aa99fcff27f,2
739
+ np.float64,0xbfc5d1f6fb2ba3ec,0xbfc7ec367cb3fecc,2
740
+ np.float64,0x8008316a44d062d5,0x8008316a44d062d5,2
741
+ np.float64,0xbfd54e4358aa9c86,0xbfd9e889d2998a4a,2
742
+ np.float64,0xda65facdb4cc0,0xda65facdb4cc0,2
743
+ np.float64,0x3fc5b4f6f32b69f0,0x3fc40d13aa8e248b,2
744
+ np.float64,0x3fd825a5d5b04b4c,0x3fd47ce73e04d3ff,2
745
+ np.float64,0x7ac9d56ef593b,0x7ac9d56ef593b,2
746
+ np.float64,0xbfd0a51977214a32,0xbfd34702071428be,2
747
+ np.float64,0x3fd21f620b243ec4,0x3fcfea0c02193640,2
748
+ np.float64,0x3fe6fb3f1b2df67e,0x3fe151ffb18c983b,2
749
+ np.float64,0x700de022e01bd,0x700de022e01bd,2
750
+ np.float64,0xbfbb76b81236ed70,0xbfbd0d31deea1ec7,2
751
+ np.float64,0x3fecfc3856f9f870,0x3fe4a2fcadf221e0,2
752
+ np.float64,0x3fede286517bc50c,0x3fe51af2fbd6ef63,2
753
+ np.float64,0x7fdc8da96c391b52,0x408627ce09cfef2b,2
754
+ np.float64,0x8000edfcfb81dbfb,0x8000edfcfb81dbfb,2
755
+ np.float64,0x8009ebc42af3d789,0x8009ebc42af3d789,2
756
+ np.float64,0x7fd658aaf8acb155,0x408625d80cd1ccc9,2
757
+ np.float64,0x3feea584a37d4b09,0x3fe57f29a73729cd,2
758
+ np.float64,0x4cfe494699fca,0x4cfe494699fca,2
759
+ np.float64,0xbfe9d96460b3b2c9,0xbffa62ecfa026c77,2
760
+ np.float64,0x7fdb3852c3b670a5,0x4086276c191dc9b1,2
761
+ np.float64,0xbfe4d1fc9ee9a3f9,0xbff0d37ce37cf479,2
762
+ np.float64,0xffefffffffffffff,0xfff8000000000000,2
763
+ np.float64,0xbfd1c43d7fa3887a,0xbfd4cfbefb5f2c43,2
764
+ np.float64,0x3fec4a8e0d78951c,0x3fe4453a82ca2570,2
765
+ np.float64,0x7fafed74583fdae8,0x4086181017b8dac9,2
766
+ np.float64,0x80076c4ebcced89e,0x80076c4ebcced89e,2
767
+ np.float64,0x8001a9aa7b235356,0x8001a9aa7b235356,2
768
+ np.float64,0x121260fe2424d,0x121260fe2424d,2
769
+ np.float64,0x3fddd028e3bba052,0x3fd87998c4c43c5b,2
770
+ np.float64,0x800ed1cf4a9da39f,0x800ed1cf4a9da39f,2
771
+ np.float64,0xbfef2e63d7fe5cc8,0xc00d53480b16971b,2
772
+ np.float64,0xbfedde3309fbbc66,0xc005ab55b7a7c127,2
773
+ np.float64,0x3fda3e1e85b47c3d,0x3fd5fddafd8d6729,2
774
+ np.float64,0x8007c6443c6f8c89,0x8007c6443c6f8c89,2
775
+ np.float64,0xbfe101705f2202e0,0xbfe8420817665121,2
776
+ np.float64,0x7fe0bff3c1e17fe7,0x4086291539c56d80,2
777
+ np.float64,0x7fe6001dab6c003a,0x40862b43aa7cb060,2
778
+ np.float64,0x7fbdecf7de3bd9ef,0x40861d170b1c51a5,2
779
+ np.float64,0xbfc0fd508c21faa0,0xbfc23a5876e99fa3,2
780
+ np.float64,0xbfcf6eb14f3edd64,0xbfd208cbf742c8ea,2
781
+ np.float64,0x3f6d40ea403a81d5,0x3f6d33934ab8e799,2
782
+ np.float64,0x7fc32600b6264c00,0x40861f10302357e0,2
783
+ np.float64,0x3fd05870baa0b0e0,0x3fcd1d2af420fac7,2
784
+ np.float64,0x80051d5120aa3aa3,0x80051d5120aa3aa3,2
785
+ np.float64,0x3fdb783fcfb6f080,0x3fd6db229658c083,2
786
+ np.float64,0x3fe0b61199e16c24,0x3fdae41e277be2eb,2
787
+ np.float64,0x3daf62167b5ed,0x3daf62167b5ed,2
788
+ np.float64,0xbfec3c53b6f878a7,0xc0011f0ce7a78a2a,2
789
+ np.float64,0x800fc905161f920a,0x800fc905161f920a,2
790
+ np.float64,0x3fdc7b9cc138f73a,0x3fd78f9c2360e661,2
791
+ np.float64,0x7fe4079e97a80f3c,0x40862a83795f2443,2
792
+ np.float64,0x8010000000000000,0x8010000000000000,2
793
+ np.float64,0x7fe6da5345adb4a6,0x40862b9183c1e4b0,2
794
+ np.float64,0xbfd0a76667214ecc,0xbfd34a1e0c1f6186,2
795
+ np.float64,0x37fb0b906ff62,0x37fb0b906ff62,2
796
+ np.float64,0x7fe170e59fa2e1ca,0x408629680a55e5c5,2
797
+ np.float64,0x3fea900c77752019,0x3fe356eec75aa345,2
798
+ np.float64,0x3fc575c63a2aeb8c,0x3fc3d701167d76b5,2
799
+ np.float64,0x3fe8b45da87168bc,0x3fe24ecbb778fd44,2
800
+ np.float64,0xbfcb990ab5373214,0xbfcf1596c076813c,2
801
+ np.float64,0xf146fdfbe28e0,0xf146fdfbe28e0,2
802
+ np.float64,0x8001fcd474c3f9aa,0x8001fcd474c3f9aa,2
803
+ np.float64,0xbfe9b555eeb36aac,0xbffa0630c3bb485b,2
804
+ np.float64,0x800f950be83f2a18,0x800f950be83f2a18,2
805
+ np.float64,0x7feb0e03ab761c06,0x40862ceb30e36887,2
806
+ np.float64,0x7fca51bd4a34a37a,0x4086219b9dfd35c9,2
807
+ np.float64,0xbfdc27c34cb84f86,0xbfe28ccde8d6bc08,2
808
+ np.float64,0x80009ce1714139c4,0x80009ce1714139c4,2
809
+ np.float64,0x8005290fb1ea5220,0x8005290fb1ea5220,2
810
+ np.float64,0xbfee81e6473d03cd,0xc00885972ca1699b,2
811
+ np.float64,0x7fcfb11a373f6233,0x408623180b8f75d9,2
812
+ np.float64,0xbfcb9c4bfd373898,0xbfcf19bd25881928,2
813
+ np.float64,0x7feaec5885f5d8b0,0x40862ce136050e6c,2
814
+ np.float64,0x8009e17a4a53c2f5,0x8009e17a4a53c2f5,2
815
+ np.float64,0xbfe1cceb9e6399d7,0xbfea0038bd3def20,2
816
+ np.float64,0x8009170bd7122e18,0x8009170bd7122e18,2
817
+ np.float64,0xb2b6f7f1656df,0xb2b6f7f1656df,2
818
+ np.float64,0x3fc75bfd1f2eb7f8,0x3fc574c858332265,2
819
+ np.float64,0x3fa24c06ec249800,0x3fa1fa462ffcb8ec,2
820
+ np.float64,0xaa9a4d2d5534a,0xaa9a4d2d5534a,2
821
+ np.float64,0xbfd7b76208af6ec4,0xbfdda0c3200dcc9f,2
822
+ np.float64,0x7f8cbab73039756d,0x40860c20cba57a94,2
823
+ np.float64,0x3fdbcf9f48b79f3f,0x3fd71827a60e8b6d,2
824
+ np.float64,0xbfdd60f71a3ac1ee,0xbfe3a94bc8cf134d,2
825
+ np.float64,0xb9253589724a7,0xb9253589724a7,2
826
+ np.float64,0xbfcf28e37e3e51c8,0xbfd1da9977b741e3,2
827
+ np.float64,0x80011457f7e228b1,0x80011457f7e228b1,2
828
+ np.float64,0x7fec33df737867be,0x40862d404a897122,2
829
+ np.float64,0xae55f8f95cabf,0xae55f8f95cabf,2
830
+ np.float64,0xbfc1ab9397235728,0xbfc303e5533d4a5f,2
831
+ np.float64,0x7fef0f84b3be1f08,0x40862e05f9ba7118,2
832
+ np.float64,0x7fdc94f328b929e5,0x408627d01449d825,2
833
+ np.float64,0x3fee1b598c7c36b3,0x3fe53847be166834,2
834
+ np.float64,0x3fee8326f37d064e,0x3fe56d96f3fbcf43,2
835
+ np.float64,0x3fe7b18a83ef6316,0x3fe1bb6a6d48c675,2
836
+ np.float64,0x3fe5db969c6bb72e,0x3fe0a8d7d151996c,2
837
+ np.float64,0x3e3391d27c673,0x3e3391d27c673,2
838
+ np.float64,0x3fe79a46d76f348e,0x3fe1ae09a96ea628,2
839
+ np.float64,0x7ff4000000000000,0x7ffc000000000000,2
840
+ np.float64,0x7fe57d6505aafac9,0x40862b13925547f1,2
841
+ np.float64,0x3fc433371d28666e,0x3fc2c196a764c47b,2
842
+ np.float64,0x8008dbf69cd1b7ee,0x8008dbf69cd1b7ee,2
843
+ np.float64,0xbfe744f459ee89e8,0xbff4c847ad3ee152,2
844
+ np.float64,0x80098aa245331545,0x80098aa245331545,2
845
+ np.float64,0x6747112ece8e3,0x6747112ece8e3,2
846
+ np.float64,0x5d342a40ba69,0x5d342a40ba69,2
847
+ np.float64,0xf7a17739ef42f,0xf7a17739ef42f,2
848
+ np.float64,0x3fe1b34a9d236695,0x3fdc2d7c4e2c347a,2
849
+ np.float64,0x7fb53bf5ec2a77eb,0x40861a585ec8f7ff,2
850
+ np.float64,0xbfe6256f1cec4ade,0xbff2d89a36be65ae,2
851
+ np.float64,0xb783bc9b6f078,0xb783bc9b6f078,2
852
+ np.float64,0xbfedf74a3bfbee94,0xc0060bb6f2bc11ef,2
853
+ np.float64,0x3fda2a5eccb454be,0x3fd5efd7f18b8e81,2
854
+ np.float64,0xbfb3838ab2270718,0xbfb44c337fbca3c3,2
855
+ np.float64,0x3fb4ac6dc22958e0,0x3fb3e194ca01a502,2
856
+ np.float64,0x76c11aaaed824,0x76c11aaaed824,2
857
+ np.float64,0x80025bb1af04b764,0x80025bb1af04b764,2
858
+ np.float64,0x3fdc02740ab804e8,0x3fd73b8cd6f95f19,2
859
+ np.float64,0x3fe71856f5ee30ae,0x3fe162e9fafb4428,2
860
+ np.float64,0x800236f332646de7,0x800236f332646de7,2
861
+ np.float64,0x7fe13fd9d2e27fb3,0x408629516b42a317,2
862
+ np.float64,0x7fdf6bbd34bed779,0x40862892069d805c,2
863
+ np.float64,0x3fd4727beba8e4f8,0x3fd1be5b48d9e282,2
864
+ np.float64,0x800e0fac9e5c1f59,0x800e0fac9e5c1f59,2
865
+ np.float64,0xfb54423ff6a89,0xfb54423ff6a89,2
866
+ np.float64,0x800fbf7ed47f7efe,0x800fbf7ed47f7efe,2
867
+ np.float64,0x3fe9d41fa2f3a840,0x3fe2ef98dc1fd463,2
868
+ np.float64,0x800d733e805ae67d,0x800d733e805ae67d,2
869
+ np.float64,0x3feebe4c46fd7c98,0x3fe58bcf7f47264e,2
870
+ np.float64,0x7fe1ab77b5e356ee,0x40862982bb3dce34,2
871
+ np.float64,0xbfdddac05abbb580,0xbfe41aa45f72d5a2,2
872
+ np.float64,0x3fe14219dee28434,0x3fdb9b137d1f1220,2
873
+ np.float64,0x3fe25d3d5a24ba7b,0x3fdd06e1cf32d35a,2
874
+ np.float64,0x8000fa4fbe81f4a0,0x8000fa4fbe81f4a0,2
875
+ np.float64,0x3fe303e23e6607c4,0x3fddd94982efa9f1,2
876
+ np.float64,0x3fe89cf5d83139ec,0x3fe24193a2e12f75,2
877
+ np.float64,0x3fe9b36ef87366de,0x3fe2dd7cdc25a4a5,2
878
+ np.float64,0xbfdb8b38f8371672,0xbfe2023ba7e002bb,2
879
+ np.float64,0xafc354955f86b,0xafc354955f86b,2
880
+ np.float64,0xbfe2f3d49e65e7a9,0xbfecb557a94123d3,2
881
+ np.float64,0x800496617c092cc4,0x800496617c092cc4,2
882
+ np.float64,0x32db0cfa65b62,0x32db0cfa65b62,2
883
+ np.float64,0xbfd893bfa2b12780,0xbfdf02a8c1e545aa,2
884
+ np.float64,0x7fd5ac927d2b5924,0x408625997e7c1f9b,2
885
+ np.float64,0x3fde9defb8bd3be0,0x3fd9056190986349,2
886
+ np.float64,0x80030cfeb54619fe,0x80030cfeb54619fe,2
887
+ np.float64,0x3fcba85b273750b8,0x3fc90a5ca976594f,2
888
+ np.float64,0x3fe98f6f5cf31edf,0x3fe2c97fcb4eca25,2
889
+ np.float64,0x3fe33dbf90667b80,0x3fde21b83321b993,2
890
+ np.float64,0x3fe4686636e8d0cc,0x3fdf928cdca751b3,2
891
+ np.float64,0x80018ade6ce315be,0x80018ade6ce315be,2
892
+ np.float64,0x7fa9af70c8335ee1,0x408616528cd5a906,2
893
+ np.float64,0x3fbeb460aa3d68c0,0x3fbcff96b00a2193,2
894
+ np.float64,0x7fa82c869830590c,0x408615d6598d9368,2
895
+ np.float64,0xd08c0e6fa1182,0xd08c0e6fa1182,2
896
+ np.float64,0x3fef4eb750fe9d6f,0x3fe5d522fd4e7f64,2
897
+ np.float64,0xbfc586f5492b0dec,0xbfc791eaae92aad1,2
898
+ np.float64,0x7fede64ac7bbcc95,0x40862db7f444fa7b,2
899
+ np.float64,0x3fe540003d6a8000,0x3fe04bdfc2916a0b,2
900
+ np.float64,0x8009417fe6f28300,0x8009417fe6f28300,2
901
+ np.float64,0x3fe6959cf16d2b3a,0x3fe116a1ce01887b,2
902
+ np.float64,0x3fb0a40036214800,0x3fb01f447778219a,2
903
+ np.float64,0x3feff26e91ffe4dd,0x3fe627798fc859a7,2
904
+ np.float64,0x7fed8e46cd7b1c8d,0x40862da044a1d102,2
905
+ np.float64,0x7fec4eb774f89d6e,0x40862d47e43edb53,2
906
+ np.float64,0x3fe800e5e07001cc,0x3fe1e8e2b9105fc2,2
907
+ np.float64,0x800f4eb2f9be9d66,0x800f4eb2f9be9d66,2
908
+ np.float64,0x800611659bcc22cc,0x800611659bcc22cc,2
909
+ np.float64,0x3fd66e65d2acdccc,0x3fd33ad63a5e1000,2
910
+ np.float64,0x800a9085b7f5210c,0x800a9085b7f5210c,2
911
+ np.float64,0x7fdf933a3fbf2673,0x4086289c0e292f2b,2
912
+ np.float64,0x1cd1ba7a39a38,0x1cd1ba7a39a38,2
913
+ np.float64,0xbfefd0b10fffa162,0xc0149ded900ed851,2
914
+ np.float64,0xbfe8c63485b18c69,0xbff7cf3078b1574f,2
915
+ np.float64,0x3fecde56ca79bcae,0x3fe4934afbd7dda9,2
916
+ np.float64,0x8006cd6888cd9ad2,0x8006cd6888cd9ad2,2
917
+ np.float64,0x3fd7a391c2af4724,0x3fd41e2f74df2329,2
918
+ np.float64,0x3fe6a8ad58ed515a,0x3fe121ccfb28e6f5,2
919
+ np.float64,0x7fe18a80dd631501,0x40862973c09086b9,2
920
+ np.float64,0xbf74fd6d8029fb00,0xbf750b3e368ebe6b,2
921
+ np.float64,0x3fdd35e93dba6bd4,0x3fd810071faaffad,2
922
+ np.float64,0x3feb0d8f57361b1f,0x3fe39b3abdef8b7a,2
923
+ np.float64,0xbfd5ec7288abd8e6,0xbfdad764df0d2ca1,2
924
+ np.float64,0x7fdc848272b90904,0x408627cb78f3fb9e,2
925
+ np.float64,0x800ed3eda91da7db,0x800ed3eda91da7db,2
926
+ np.float64,0x3fefac64857f58c9,0x3fe60459dbaad1ba,2
927
+ np.float64,0x3fd1df7a5ba3bef4,0x3fcf864a39b926ff,2
928
+ np.float64,0xfe26ca4bfc4da,0xfe26ca4bfc4da,2
929
+ np.float64,0xbfd1099f8da21340,0xbfd3cf6e6efe934b,2
930
+ np.float64,0xbfe15de9a7a2bbd4,0xbfe909cc895f8795,2
931
+ np.float64,0x3fe89714ed712e2a,0x3fe23e40d31242a4,2
932
+ np.float64,0x800387113e470e23,0x800387113e470e23,2
933
+ np.float64,0x3fe4f80730e9f00e,0x3fe0208219314cf1,2
934
+ np.float64,0x2f95a97c5f2b6,0x2f95a97c5f2b6,2
935
+ np.float64,0x800ea7cdd87d4f9c,0x800ea7cdd87d4f9c,2
936
+ np.float64,0xbf64b967c0297300,0xbf64c020a145b7a5,2
937
+ np.float64,0xbfc5a91a342b5234,0xbfc7bafd77a61d81,2
938
+ np.float64,0xbfe2226fe76444e0,0xbfeac33eb1d1b398,2
939
+ np.float64,0x3fc6aaa8d42d5552,0x3fc4de79f5c68cd4,2
940
+ np.float64,0x3fe54fd4c1ea9faa,0x3fe05561a9a5922b,2
941
+ np.float64,0x80029c1f75653840,0x80029c1f75653840,2
942
+ np.float64,0xbfcb4a84a2369508,0xbfceb1a23bac3995,2
943
+ np.float64,0x80010abeff02157f,0x80010abeff02157f,2
944
+ np.float64,0x7f92d12cf825a259,0x40860e49bde3a5b6,2
945
+ np.float64,0x800933e7027267ce,0x800933e7027267ce,2
946
+ np.float64,0x3fc022b12e204562,0x3fbe64acc53ed887,2
947
+ np.float64,0xbfe35f938de6bf27,0xbfedc1f3e443c016,2
948
+ np.float64,0x1f8d9bae3f1b4,0x1f8d9bae3f1b4,2
949
+ np.float64,0x3fe552f22ceaa5e4,0x3fe057404072350f,2
950
+ np.float64,0xbfa73753442e6ea0,0xbfa7c24a100190f1,2
951
+ np.float64,0x7fb3e2982827c52f,0x408619d1efa676b6,2
952
+ np.float64,0xbfd80cb7a5301970,0xbfde28e65f344f33,2
953
+ np.float64,0xbfcde835973bd06c,0xbfd10806fba46c8f,2
954
+ np.float64,0xbfd4e3c749a9c78e,0xbfd949aff65de39c,2
955
+ np.float64,0x3fcb4b9d6f36973b,0x3fc8be02ad6dc0d3,2
956
+ np.float64,0x1a63000034c7,0x1a63000034c7,2
957
+ np.float64,0x7fdc9c751e3938e9,0x408627d22df71959,2
958
+ np.float64,0x3fd74f3f712e9e7f,0x3fd3e07df0c37ec1,2
959
+ np.float64,0xbfceab74d33d56e8,0xbfd187e99bf82903,2
960
+ np.float64,0x7ff0000000000000,0x7ff0000000000000,2
961
+ np.float64,0xbfb2cca466259948,0xbfb3868208e8de30,2
962
+ np.float64,0x800204688b8408d2,0x800204688b8408d2,2
963
+ np.float64,0x3e4547407c8aa,0x3e4547407c8aa,2
964
+ np.float64,0xbfe4668846e8cd10,0xbff03c85189f3818,2
965
+ np.float64,0x800dd350245ba6a0,0x800dd350245ba6a0,2
966
+ np.float64,0xbfbc13c160382780,0xbfbdbd56ce996d16,2
967
+ np.float64,0x7fe25a628a24b4c4,0x408629d06eb2d64d,2
968
+ np.float64,0x3fd19dabbc233b57,0x3fcf1f3ed1d34c8c,2
969
+ np.float64,0x547e20faa8fc5,0x547e20faa8fc5,2
970
+ np.float64,0xbfe19392c6232726,0xbfe97ffe4f303335,2
971
+ np.float64,0x3f87f9f6702ff400,0x3f87d64fb471bb04,2
972
+ np.float64,0x9dfc52db3bf8b,0x9dfc52db3bf8b,2
973
+ np.float64,0x800e1f5a9adc3eb5,0x800e1f5a9adc3eb5,2
974
+ np.float64,0xbfddbd09c8bb7a14,0xbfe3fed7d7cffc70,2
975
+ np.float64,0xbfeda71af87b4e36,0xc004e6631c514544,2
976
+ np.float64,0xbfdbfcfe1bb7f9fc,0xbfe266b5d4a56265,2
977
+ np.float64,0x3fe4ee78cd69dcf2,0x3fe01abba4e81fc9,2
978
+ np.float64,0x800f13b820de2770,0x800f13b820de2770,2
979
+ np.float64,0x3f861e09702c3c00,0x3f85ffae83b02c4f,2
980
+ np.float64,0xbfc0972479212e48,0xbfc1c4bf70b30cbc,2
981
+ np.float64,0x7fef057ef57e0afd,0x40862e036479f6a9,2
982
+ np.float64,0x8bdbabe517b76,0x8bdbabe517b76,2
983
+ np.float64,0xbfec495417f892a8,0xc0013ade88746d18,2
984
+ np.float64,0x3fec680ab3f8d015,0x3fe454dd304b560d,2
985
+ np.float64,0xbfae7ce60c3cf9d0,0xbfaf6eef15bbe56b,2
986
+ np.float64,0x3fec314124786282,0x3fe437ca06294f5a,2
987
+ np.float64,0x7fd5ed05b82bda0a,0x408625b125518e58,2
988
+ np.float64,0x3feac9f02f3593e0,0x3fe3768104dd5cb7,2
989
+ np.float64,0x0,0x0,2
990
+ np.float64,0xbfddd2abd5bba558,0xbfe41312b8ea20de,2
991
+ np.float64,0xbfedf9558c7bf2ab,0xc00613c53e0bb33a,2
992
+ np.float64,0x3fef245ffefe48c0,0x3fe5bfb4dfe3b7a5,2
993
+ np.float64,0x7fe178604922f0c0,0x4086296b77d5eaef,2
994
+ np.float64,0x10000000000000,0x10000000000000,2
995
+ np.float64,0x7fed026766ba04ce,0x40862d7a0dc45643,2
996
+ np.float64,0xbfde27d8c3bc4fb2,0xbfe46336b6447697,2
997
+ np.float64,0x3fe9485d9cb290bb,0x3fe2a1e4b6419423,2
998
+ np.float64,0xbfe27b8a7464f715,0xbfeb9382f5b16f65,2
999
+ np.float64,0x5c34d274b869b,0x5c34d274b869b,2
1000
+ np.float64,0xbfeee0b7453dc16f,0xc00acdb46459b6e6,2
1001
+ np.float64,0x7fe3dfb4d4e7bf69,0x40862a73785fdf12,2
1002
+ np.float64,0xb4635eef68c6c,0xb4635eef68c6c,2
1003
+ np.float64,0xbfe522a2c82a4546,0xbff148912a59a1d6,2
1004
+ np.float64,0x8009ba38a9737472,0x8009ba38a9737472,2
1005
+ np.float64,0xbfc056ff3820ae00,0xbfc17b2205fa180d,2
1006
+ np.float64,0x7fe1c8b8a0239170,0x4086298feeee6133,2
1007
+ np.float64,0x3fe2d2c6b9e5a58e,0x3fdd9b907471031b,2
1008
+ np.float64,0x3fa0a161bc2142c0,0x3fa05db36f6a073b,2
1009
+ np.float64,0x3fdef4268ebde84c,0x3fd93f980794d1e7,2
1010
+ np.float64,0x800ecd9fe2fd9b40,0x800ecd9fe2fd9b40,2
1011
+ np.float64,0xbfc9fbd45e33f7a8,0xbfcd0afc47c340f6,2
1012
+ np.float64,0x3fe8c3035b718606,0x3fe2570eb65551a1,2
1013
+ np.float64,0xbfe78c4ad2ef1896,0xbff54d25b3328742,2
1014
+ np.float64,0x8006f5dcf8adebbb,0x8006f5dcf8adebbb,2
1015
+ np.float64,0x800301dca2a603ba,0x800301dca2a603ba,2
1016
+ np.float64,0xad4289e55a851,0xad4289e55a851,2
1017
+ np.float64,0x80037764f9e6eecb,0x80037764f9e6eecb,2
1018
+ np.float64,0xbfe73575b26e6aec,0xbff4abfb5e985c62,2
1019
+ np.float64,0xbfc6cb91652d9724,0xbfc91a8001b33ec2,2
1020
+ np.float64,0xbfe3a918ffe75232,0xbfee7e6e4fd34c53,2
1021
+ np.float64,0x9bc84e2b3790a,0x9bc84e2b3790a,2
1022
+ np.float64,0x7fdeec303cbdd85f,0x408628714a49d996,2
1023
+ np.float64,0x3fe1d1dcb763a3ba,0x3fdc54ce060dc7f4,2
1024
+ np.float64,0x8008ae6432b15cc9,0x8008ae6432b15cc9,2
1025
+ np.float64,0x3fd8022fa2b00460,0x3fd46322bf02a609,2
1026
+ np.float64,0xbfc55b64472ab6c8,0xbfc75d9568f462e0,2
1027
+ np.float64,0xbfe8b165437162ca,0xbff7a15e2ead645f,2
1028
+ np.float64,0x7f759330feeb3,0x7f759330feeb3,2
1029
+ np.float64,0xbfd504f68eaa09ee,0xbfd97b06c01d7473,2
1030
+ np.float64,0x54702d5aa8e06,0x54702d5aa8e06,2
1031
+ np.float64,0xbfed1779337a2ef2,0xc0032f7109ef5a51,2
1032
+ np.float64,0xe248bd4dc4918,0xe248bd4dc4918,2
1033
+ np.float64,0xbfd8c59150318b22,0xbfdf53bca6ca8b1e,2
1034
+ np.float64,0xbfe3b9d942e773b2,0xbfeea9fcad277ba7,2
1035
+ np.float64,0x800934ec127269d9,0x800934ec127269d9,2
1036
+ np.float64,0xbfbb7f535a36fea8,0xbfbd16d61b6c52b8,2
1037
+ np.float64,0xccb185a199631,0xccb185a199631,2
1038
+ np.float64,0x3fe3dda76fe7bb4e,0x3fdee83bc6094301,2
1039
+ np.float64,0xbfe0c902f5e19206,0xbfe7ca7c0e888006,2
1040
+ np.float64,0xbfefeed08cbfdda1,0xc018aadc483c8724,2
1041
+ np.float64,0x7fd0c05c52a180b8,0x40862389daf64aac,2
1042
+ np.float64,0xbfd28e3323a51c66,0xbfd5e9ba278fb685,2
1043
+ np.float64,0xbef4103b7de82,0xbef4103b7de82,2
1044
+ np.float64,0x3fe7661fd12ecc40,0x3fe18ff7dfb696e2,2
1045
+ np.float64,0x3fddd5f2f0bbabe4,0x3fd87d8bb6719c3b,2
1046
+ np.float64,0x800b3914cfd6722a,0x800b3914cfd6722a,2
1047
+ np.float64,0xf3f09a97e7e14,0xf3f09a97e7e14,2
1048
+ np.float64,0x7f97092b502e1256,0x40860fe8054cf54e,2
1049
+ np.float64,0xbfdbec7917b7d8f2,0xbfe2580b4b792c79,2
1050
+ np.float64,0x7fe7ff215aaffe42,0x40862bf5887fa062,2
1051
+ np.float64,0x80080186e570030e,0x80080186e570030e,2
1052
+ np.float64,0xbfc27f05e624fe0c,0xbfc3fa214be4adc4,2
1053
+ np.float64,0x3fe4481be1689038,0x3fdf6b11e9c4ca72,2
1054
+ np.float64,0x3fd642cc9cac8598,0x3fd31a857fe70227,2
1055
+ np.float64,0xbef8782d7df0f,0xbef8782d7df0f,2
1056
+ np.float64,0x8003077dc2e60efc,0x8003077dc2e60efc,2
1057
+ np.float64,0x80083eb5a2507d6c,0x80083eb5a2507d6c,2
1058
+ np.float64,0x800e8d1eb77d1a3e,0x800e8d1eb77d1a3e,2
1059
+ np.float64,0xbfc7737cd22ee6f8,0xbfc9e7716f03f1fc,2
1060
+ np.float64,0xbfe9a2b4ddf3456a,0xbff9d71664a8fc78,2
1061
+ np.float64,0x7fe67c7d322cf8f9,0x40862b7066465194,2
1062
+ np.float64,0x3fec080ce2b8101a,0x3fe421dac225be46,2
1063
+ np.float64,0xbfe6d27beb6da4f8,0xbff3fbb1add521f7,2
1064
+ np.float64,0x3fdd4f96ceba9f2e,0x3fd821a638986dbe,2
1065
+ np.float64,0x3fbd89f1303b13e2,0x3fbbf49223a9d002,2
1066
+ np.float64,0xbfe94e2b9d329c57,0xbff907e549c534f5,2
1067
+ np.float64,0x3fe2f2cc51e5e599,0x3fddc3d6b4a834a1,2
1068
+ np.float64,0xfdcb5b49fb96c,0xfdcb5b49fb96c,2
1069
+ np.float64,0xbfea7108fa74e212,0xbffc01b392f4897b,2
1070
+ np.float64,0x3fd38baef7a7175c,0x3fd10e7fd3b958dd,2
1071
+ np.float64,0x3fa75bf9cc2eb800,0x3fa6d792ecdedb8e,2
1072
+ np.float64,0x7fd19fd20aa33fa3,0x408623f1e2cd04c3,2
1073
+ np.float64,0x3fd62c708dac58e0,0x3fd309ec7818d16e,2
1074
+ np.float64,0x3fdf489047be9120,0x3fd978640617c758,2
1075
+ np.float64,0x1,0x1,2
1076
+ np.float64,0xbfe21e7c3ea43cf8,0xbfeaba21320697d3,2
1077
+ np.float64,0xbfd3649047a6c920,0xbfd71a6f14223744,2
1078
+ np.float64,0xbfd68ca68c2d194e,0xbfdbcce6784e5d44,2
1079
+ np.float64,0x3fdb26b0ea364d62,0x3fd6a1f86f64ff74,2
1080
+ np.float64,0xbfd843821cb08704,0xbfde80e90805ab3f,2
1081
+ np.float64,0x3fd508a27aaa1144,0x3fd22fc203a7b9d8,2
1082
+ np.float64,0xbfdb951c7eb72a38,0xbfe20aeaec13699b,2
1083
+ np.float64,0x3fef556ba57eaad7,0x3fe5d8865cce0a6d,2
1084
+ np.float64,0x3fd0d224b3a1a448,0x3fcdde7be5d7e21e,2
1085
+ np.float64,0x8007ff272baffe4f,0x8007ff272baffe4f,2
1086
+ np.float64,0x3fe1c7bddf638f7c,0x3fdc47cc6cf2f5cd,2
1087
+ np.float64,0x7ff8000000000000,0x7ff8000000000000,2
1088
+ np.float64,0x2016d560402f,0x2016d560402f,2
1089
+ np.float64,0xbfcca10be9394218,0xbfd033f36b94fc54,2
1090
+ np.float64,0xbfdb833628b7066c,0xbfe1fb344b840c70,2
1091
+ np.float64,0x3fd8529cb3b0a539,0x3fd49d847fe77218,2
1092
+ np.float64,0xbfc0b0ebab2161d8,0xbfc1e260c60ffd1b,2
1093
+ np.float64,0xbfea8b9a79f51735,0xbffc4ee6be8a0fa2,2
1094
+ np.float64,0x7feca8fab7f951f4,0x40862d613e454646,2
1095
+ np.float64,0x7fd8c52d82318a5a,0x408626aaf37423a3,2
1096
+ np.float64,0xbfe364ad4526c95a,0xbfedcee39bc93ff5,2
1097
+ np.float64,0x800b78161256f02d,0x800b78161256f02d,2
1098
+ np.float64,0xbfd55f0153aabe02,0xbfda01a78f72d494,2
1099
+ np.float64,0x800315a5f0662b4d,0x800315a5f0662b4d,2
1100
+ np.float64,0x7fe4c0dca02981b8,0x40862acc27e4819f,2
1101
+ np.float64,0x8009825c703304b9,0x8009825c703304b9,2
1102
+ np.float64,0x3fe6e94e1cadd29c,0x3fe1478ccc634f49,2
1103
+ np.float64,0x7fe622d8586c45b0,0x40862b504177827e,2
1104
+ np.float64,0x3fe4458600688b0c,0x3fdf67e79a84b953,2
1105
+ np.float64,0xbfdd75d8a1baebb2,0xbfe3bc9e6ca1bbb5,2
1106
+ np.float64,0x3fde789c6bbcf138,0x3fd8ec1d435531b3,2
1107
+ np.float64,0x3fe7052b94ee0a58,0x3fe157c5c4418dc1,2
1108
+ np.float64,0x7fef31652abe62c9,0x40862e0eaeabcfc0,2
1109
+ np.float64,0x3fe279691ee4f2d2,0x3fdd2aa41eb43cd4,2
1110
+ np.float64,0xbfd533fa95aa67f6,0xbfd9c12f516d29d7,2
1111
+ np.float64,0x3fe6d057f96da0b0,0x3fe138fd96693a6a,2
1112
+ np.float64,0x800bad984f775b31,0x800bad984f775b31,2
1113
+ np.float64,0x7fdd6fdba4badfb6,0x4086280c73d8ef97,2
1114
+ np.float64,0x7fe9b5c0eef36b81,0x40862c82c6f57a53,2
1115
+ np.float64,0x8000bc02ece17807,0x8000bc02ece17807,2
1116
+ np.float64,0xbff0000000000000,0xfff0000000000000,2
1117
+ np.float64,0xbfed430be3fa8618,0xc003aaf338c75b3c,2
1118
+ np.float64,0x3fee17b759fc2f6f,0x3fe53668696bf48b,2
1119
+ np.float64,0x3f8d4cf9d03a9a00,0x3f8d17d2f532afdc,2
1120
+ np.float64,0x8005d6257b8bac4c,0x8005d6257b8bac4c,2
1121
+ np.float64,0xbfd17a6df9a2f4dc,0xbfd469e3848adc6e,2
1122
+ np.float64,0xb28a293965145,0xb28a293965145,2
1123
+ np.float64,0xbfe7d011e42fa024,0xbff5cf818998c8ec,2
1124
+ np.float64,0xbfe74f0f136e9e1e,0xbff4dad6ebb0443c,2
1125
+ np.float64,0x800f249fc9be4940,0x800f249fc9be4940,2
1126
+ np.float64,0x2542f8fe4a860,0x2542f8fe4a860,2
1127
+ np.float64,0xc48d40cd891a8,0xc48d40cd891a8,2
1128
+ np.float64,0x3fe4e64bc8e9cc98,0x3fe015c9eb3caa53,2
1129
+ np.float64,0x3fd33881eca67104,0x3fd0cea886be2457,2
1130
+ np.float64,0xbfd01748fba02e92,0xbfd28875959e6901,2
1131
+ np.float64,0x7fb7ab01f22f5603,0x40861b369927bf53,2
1132
+ np.float64,0xbfe340274ce6804e,0xbfed72b39f0ebb24,2
1133
+ np.float64,0x7fc16c0c3422d817,0x40861e4eaf1a286c,2
1134
+ np.float64,0x3fc26944a324d288,0x3fc133a77b356ac4,2
1135
+ np.float64,0xa149d7134293b,0xa149d7134293b,2
1136
+ np.float64,0x800837382d106e71,0x800837382d106e71,2
1137
+ np.float64,0x797d1740f2fa4,0x797d1740f2fa4,2
1138
+ np.float64,0xc3f15b7787e2c,0xc3f15b7787e2c,2
1139
+ np.float64,0x80cad1b90195a,0x80cad1b90195a,2
1140
+ np.float64,0x3fdd8f1142bb1e23,0x3fd84d21490d1ce6,2
1141
+ np.float64,0xbfbde6c9123bcd90,0xbfbfcc030a86836a,2
1142
+ np.float64,0x8007f77e032feefd,0x8007f77e032feefd,2
1143
+ np.float64,0x3fe74fed1c6e9fda,0x3fe18322cf19cb61,2
1144
+ np.float64,0xbfd8a40bbcb14818,0xbfdf1d23520ba74b,2
1145
+ np.float64,0xbfeb7a0e6076f41d,0xbfff4ddfb926efa5,2
1146
+ np.float64,0xbfcb8c5f663718c0,0xbfcf0570f702bda9,2
1147
+ np.float64,0xf668cd97ecd1a,0xf668cd97ecd1a,2
1148
+ np.float64,0xbfe92accf572559a,0xbff8b4393878ffdb,2
1149
+ np.float64,0xbfeaa955567552ab,0xbffca70c7d73eee5,2
1150
+ np.float64,0xbfe083a14f610742,0xbfe739d84bc35077,2
1151
+ np.float64,0x78290568f0521,0x78290568f0521,2
1152
+ np.float64,0x3fe94bae2372975c,0x3fe2a3beac5c9858,2
1153
+ np.float64,0x3fca4fbab9349f78,0x3fc7edbca2492acb,2
1154
+ np.float64,0x8000000000000000,0x8000000000000000,2
1155
+ np.float64,0x7fb9eb505433d6a0,0x40861bf0adedb74d,2
1156
+ np.float64,0x7fdc66f72a38cded,0x408627c32aeecf0f,2
1157
+ np.float64,0x2e8e6f445d1cf,0x2e8e6f445d1cf,2
1158
+ np.float64,0xbfec43195af88633,0xc0012d7e3f91b7e8,2
1159
+ np.float64,0x7fcdb971e93b72e3,0x40862294c9e3a7bc,2
1160
+ np.float64,0x800cabc461195789,0x800cabc461195789,2
1161
+ np.float64,0x2c79709c58f2f,0x2c79709c58f2f,2
1162
+ np.float64,0x8005d772d3cbaee6,0x8005d772d3cbaee6,2
1163
+ np.float64,0x3fe84d8c03709b18,0x3fe21490ce3673dd,2
1164
+ np.float64,0x7fe5578adc2aaf15,0x40862b056e8437d4,2
1165
+ np.float64,0xbf91298c58225320,0xbf914ec86c32d11f,2
1166
+ np.float64,0xc7ed2b6d8fda6,0xc7ed2b6d8fda6,2
1167
+ np.float64,0x2761404c4ec29,0x2761404c4ec29,2
1168
+ np.float64,0x3fbad3c48835a789,0x3fb9833c02385305,2
1169
+ np.float64,0x3fa46fee5428dfe0,0x3fa40a357fb24c23,2
1170
+ np.float64,0xbfe3900c6fe72019,0xbfee3dba29dd9d43,2
1171
+ np.float64,0x3fe7a9e41a6f53c8,0x3fe1b704dfb9884b,2
1172
+ np.float64,0xbfe74a7a1eee94f4,0xbff4d269cacb1f29,2
1173
+ np.float64,0xbfee609c72fcc139,0xc007da8499d34123,2
1174
+ np.float64,0x3fef2d5fc23e5ac0,0x3fe5c44414e59cb4,2
1175
+ np.float64,0xbfd7bdc0402f7b80,0xbfddaae1e7bb78fb,2
1176
+ np.float64,0xd71ee01dae3dc,0xd71ee01dae3dc,2
1177
+ np.float64,0x3fe98cbcdef3197a,0x3fe2c7ffe33c4541,2
1178
+ np.float64,0x8000f8dbb3a1f1b8,0x8000f8dbb3a1f1b8,2
1179
+ np.float64,0x3fe3e98ad567d316,0x3fdef6e58058313f,2
1180
+ np.float64,0x41ad0bfc835a2,0x41ad0bfc835a2,2
1181
+ np.float64,0x7fdcc2dc0d3985b7,0x408627dce39f77af,2
1182
+ np.float64,0xbfe47b980de8f730,0xbff059acdccd6e2b,2
1183
+ np.float64,0xbfef49b6577e936d,0xc00e714f46b2ccc1,2
1184
+ np.float64,0x3fac31816c386300,0x3fab71cb92b0db8f,2
1185
+ np.float64,0x3fe59097e76b2130,0x3fe07c299fd1127c,2
1186
+ np.float64,0xbfecf0df5cf9e1bf,0xc002c7ebdd65039c,2
1187
+ np.float64,0x3fd2b7d0b6a56fa1,0x3fd06b638990ae02,2
1188
+ np.float64,0xbfeb68deecf6d1be,0xbfff1187e042d3e4,2
1189
+ np.float64,0x3fd44a9771a8952f,0x3fd1a01867c5e302,2
1190
+ np.float64,0xf79a9dedef354,0xf79a9dedef354,2
1191
+ np.float64,0x800c25a170d84b43,0x800c25a170d84b43,2
1192
+ np.float64,0x3ff0000000000000,0x3fe62e42fefa39ef,2
1193
+ np.float64,0x3fbff4f7623fe9f0,0x3fbe1d3878f4c417,2
1194
+ np.float64,0xd284c845a5099,0xd284c845a5099,2
1195
+ np.float64,0xbfe3c7815f678f02,0xbfeecdab5ca2e651,2
1196
+ np.float64,0x3fc19c934e233927,0x3fc08036104b1f23,2
1197
+ np.float64,0x800b6096de16c12e,0x800b6096de16c12e,2
1198
+ np.float64,0xbfe962a67e32c54d,0xbff9392313a112a1,2
1199
+ np.float64,0x2b9d0116573a1,0x2b9d0116573a1,2
1200
+ np.float64,0x3fcab269ed3564d4,0x3fc83f7e1c3095b7,2
1201
+ np.float64,0x3fc8c78d86318f1b,0x3fc6a6cde5696f99,2
1202
+ np.float64,0xd5b1e9b5ab63d,0xd5b1e9b5ab63d,2
1203
+ np.float64,0xbfed802a47fb0054,0xc00465cad3b5b0ef,2
1204
+ np.float64,0xbfd73aaf08ae755e,0xbfdcdbd62b8af271,2
1205
+ np.float64,0xbfd4f13c0229e278,0xbfd95dacff79e570,2
1206
+ np.float64,0xbfe9622808f2c450,0xbff937f13c397e8d,2
1207
+ np.float64,0xbfeddfa62efbbf4c,0xc005b0c835eed829,2
1208
+ np.float64,0x3fd65663d4acacc8,0x3fd3290cd0e675dc,2
1209
+ np.float64,0x8005e890f1abd123,0x8005e890f1abd123,2
1210
+ np.float64,0xbfe924919fb24923,0xbff8a5a827a28756,2
1211
+ np.float64,0x3fe8cdf490719be9,0x3fe25d39535e8366,2
1212
+ np.float64,0x7fc229e6ff2453cd,0x40861ea40ef87a5a,2
1213
+ np.float64,0x3fe5cf53ceeb9ea8,0x3fe0a18e0b65f27e,2
1214
+ np.float64,0xa79cf6fb4f39f,0xa79cf6fb4f39f,2
1215
+ np.float64,0x7fddbb3c0f3b7677,0x40862820d5edf310,2
1216
+ np.float64,0x3e1011de7c203,0x3e1011de7c203,2
1217
+ np.float64,0x3fc0b59a83216b38,0x3fbf6916510ff411,2
1218
+ np.float64,0x8647f98d0c8ff,0x8647f98d0c8ff,2
1219
+ np.float64,0x8005dad33ecbb5a7,0x8005dad33ecbb5a7,2
1220
+ np.float64,0x8a80d0631501a,0x8a80d0631501a,2
1221
+ np.float64,0xbfe18f7d6ee31efb,0xbfe976f06713afc1,2
1222
+ np.float64,0xbfe06eaed560dd5e,0xbfe70eac696933e6,2
1223
+ np.float64,0xbfed8ef93c7b1df2,0xc00495bfa3195b53,2
1224
+ np.float64,0x3febe9c24677d385,0x3fe411b10db16c42,2
1225
+ np.float64,0x7fd5d80c1fabb017,0x408625a97a7787ba,2
1226
+ np.float64,0x3fca79b59334f368,0x3fc8108a521341dc,2
1227
+ np.float64,0xbfccf8db4339f1b8,0xbfd06c9a5424aadb,2
1228
+ np.float64,0xbfea5ac5a574b58b,0xbffbc21d1405d840,2
1229
+ np.float64,0x800ce2bf4b19c57f,0x800ce2bf4b19c57f,2
1230
+ np.float64,0xbfe8df896d31bf13,0xbff807ab38ac41ab,2
1231
+ np.float64,0x3feab83da9f5707c,0x3fe36cdd827c0eff,2
1232
+ np.float64,0x3fee717683bce2ed,0x3fe564879171719b,2
1233
+ np.float64,0x80025e5577c4bcac,0x80025e5577c4bcac,2
1234
+ np.float64,0x3fe3e5378e67ca70,0x3fdef1902c5d1efd,2
1235
+ np.float64,0x3fa014bb7c202980,0x3f9faacf9238d499,2
1236
+ np.float64,0x3fddbf5e16bb7ebc,0x3fd86e2311cb0f6d,2
1237
+ np.float64,0x3fd24e50e6a49ca0,0x3fd0198f04f82186,2
1238
+ np.float64,0x656b5214cad6b,0x656b5214cad6b,2
1239
+ np.float64,0x8b0a4bfd1614a,0x8b0a4bfd1614a,2
1240
+ np.float64,0xbfeeb6bd9e7d6d7b,0xc009b669285e319e,2
1241
+ np.float64,0x8000000000000001,0x8000000000000001,2
1242
+ np.float64,0xbfe719feceee33fe,0xbff47a4c8cbf0cca,2
1243
+ np.float64,0xbfd14fa8c8a29f52,0xbfd42f27b1aced39,2
1244
+ np.float64,0x7fec9dcb80f93b96,0x40862d5e1e70bbb9,2
1245
+ np.float64,0x7fecacb826f9596f,0x40862d6249746915,2
1246
+ np.float64,0x973459f52e68b,0x973459f52e68b,2
1247
+ np.float64,0x7f40a59e00214b3b,0x4085f194f45f82b1,2
1248
+ np.float64,0x7fc5dbaec32bb75d,0x4086201f3e7065d9,2
1249
+ np.float64,0x82d0801305a10,0x82d0801305a10,2
1250
+ np.float64,0x7fec81c0f4790381,0x40862d5643c0fc85,2
1251
+ np.float64,0xbfe2d81e9ee5b03d,0xbfec71a8e864ea40,2
1252
+ np.float64,0x6c545c9ad8a8c,0x6c545c9ad8a8c,2
1253
+ np.float64,0x3f9be95a5037d2b5,0x3f9b89b48ac8f5d8,2
1254
+ np.float64,0x8000cae9702195d4,0x8000cae9702195d4,2
1255
+ np.float64,0xbfd375f45126ebe8,0xbfd733677e54a80d,2
1256
+ np.float64,0x3fd29a5b81a534b7,0x3fd05494bf200278,2
1257
+ np.float64,0xfff0000000000000,0xfff8000000000000,2
1258
+ np.float64,0x7fca8fc195351f82,0x408621ae61aa6c13,2
1259
+ np.float64,0x1b28e2ae3651d,0x1b28e2ae3651d,2
1260
+ np.float64,0x3fe7fdbd14effb7a,0x3fe1e714884b46a8,2
1261
+ np.float64,0x3fdf1ce068be39c0,0x3fd95b054e0fad3d,2
1262
+ np.float64,0x3fe79f9a636f3f34,0x3fe1b11a40c00b3e,2
1263
+ np.float64,0x3fe60eb7036c1d6e,0x3fe0c72a02176874,2
1264
+ np.float64,0x229da17e453b5,0x229da17e453b5,2
1265
+ np.float64,0x3fc1a921b5235240,0x3fc08b3f35e47fb1,2
1266
+ np.float64,0xbb92d2af7725b,0xbb92d2af7725b,2
1267
+ np.float64,0x3fe4110cb1e8221a,0x3fdf2787de6c73f7,2
1268
+ np.float64,0xbfbc87771a390ef0,0xbfbe3f6e95622363,2
1269
+ np.float64,0xbfe74025dfee804c,0xbff4bf7b1895e697,2
1270
+ np.float64,0x964eb6592c9d7,0x964eb6592c9d7,2
1271
+ np.float64,0x3f951689b82a2d00,0x3f94dfb38d746fdf,2
1272
+ np.float64,0x800356271be6ac4f,0x800356271be6ac4f,2
1273
+ np.float64,0x7fefffffffffffff,0x40862e42fefa39ef,2
1274
+ np.float64,0xbfed5ce250fab9c5,0xc003f7ddfeb94345,2
1275
+ np.float64,0x3fec3d5dc1387abc,0x3fe43e39c02d86f4,2
1276
+ np.float64,0x3999897e73332,0x3999897e73332,2
1277
+ np.float64,0xbfdcb57744b96aee,0xbfe30c4b98f3d088,2
1278
+ np.float64,0x7f961fb0b82c3f60,0x40860f9549c3a380,2
1279
+ np.float64,0x67d6efcacfadf,0x67d6efcacfadf,2
1280
+ np.float64,0x8002c9498f859294,0x8002c9498f859294,2
1281
+ np.float64,0xbfa3033800260670,0xbfa35fe3bf43e188,2
1282
+ np.float64,0xbfeab2fc157565f8,0xbffcc413c486b4eb,2
1283
+ np.float64,0x3fe25e62f364bcc6,0x3fdd0856e19e3430,2
1284
+ np.float64,0x7fb2f42dda25e85b,0x4086196fb34a65fd,2
1285
+ np.float64,0x3fe0f1a5af61e34c,0x3fdb3235a1786efb,2
1286
+ np.float64,0x800a340ca1f4681a,0x800a340ca1f4681a,2
1287
+ np.float64,0x7c20b9def8418,0x7c20b9def8418,2
1288
+ np.float64,0xdf0842a1be109,0xdf0842a1be109,2
1289
+ np.float64,0x3fe9f22cc2f3e45a,0x3fe300359b842bf0,2
1290
+ np.float64,0x3fe389ed73e713da,0x3fde809780fe4432,2
1291
+ np.float64,0x9500fb932a020,0x9500fb932a020,2
1292
+ np.float64,0x3fd8a21ffdb14440,0x3fd4d70862345d86,2
1293
+ np.float64,0x800d99c15cbb3383,0x800d99c15cbb3383,2
1294
+ np.float64,0x3fd96c98c932d932,0x3fd568959c9b028f,2
1295
+ np.float64,0x7fc228483a24508f,0x40861ea358420976,2
1296
+ np.float64,0x7fc6737bef2ce6f7,0x408620560ffc6a98,2
1297
+ np.float64,0xbfb2c27cee2584f8,0xbfb37b8cc7774b5f,2
1298
+ np.float64,0xbfd18409f9230814,0xbfd4771d1a9a24fb,2
1299
+ np.float64,0x3fb53cb3f42a7968,0x3fb466f06f88044b,2
1300
+ np.float64,0x3fef61d0187ec3a0,0x3fe5dec8a9d13dd9,2
1301
+ np.float64,0x3fe59a6ffd2b34e0,0x3fe0820a99c6143d,2
1302
+ np.float64,0x3fce18aff43c3160,0x3fcb07c7b523f0d1,2
1303
+ np.float64,0xbfb1319a62226338,0xbfb1cc62f31b2b40,2
1304
+ np.float64,0xa00cce6d4019a,0xa00cce6d4019a,2
1305
+ np.float64,0x80068ae8e0ed15d3,0x80068ae8e0ed15d3,2
1306
+ np.float64,0x3fecef353239de6a,0x3fe49c280adc607b,2
1307
+ np.float64,0x3fdf1a7fb0be34ff,0x3fd9596bafe2d766,2
1308
+ np.float64,0x3feb5e12eeb6bc26,0x3fe3c6be3ede8d07,2
1309
+ np.float64,0x3fdeff5cd43dfeba,0x3fd947262ec96b05,2
1310
+ np.float64,0x3f995e75e832bd00,0x3f990f511f4c7f1c,2
1311
+ np.float64,0xbfeb5b3ed0b6b67e,0xbffee24fc0fc2881,2
1312
+ np.float64,0x7fb82aad0a305559,0x40861b614d901182,2
1313
+ np.float64,0xbfe5c3a4926b8749,0xbff23cd0ad144fe6,2
1314
+ np.float64,0x3fef47da373e8fb4,0x3fe5d1aaa4031993,2
1315
+ np.float64,0x7fc6a8c3872d5186,0x40862068f5ca84be,2
1316
+ np.float64,0x7fc0c2276221844e,0x40861dff2566d001,2
1317
+ np.float64,0x7fc9ce7d28339cf9,0x40862173541f84d1,2
1318
+ np.float64,0x3fce2c34933c5869,0x3fcb179428ad241d,2
1319
+ np.float64,0xbfcf864c293f0c98,0xbfd21872c4821cfc,2
1320
+ np.float64,0x3fc51fd1f82a3fa4,0x3fc38d4f1685c166,2
1321
+ np.float64,0xbfe2707b70a4e0f7,0xbfeb795fbd5bb444,2
1322
+ np.float64,0x46629b568cc54,0x46629b568cc54,2
1323
+ np.float64,0x7fe5f821f32bf043,0x40862b40c2cdea3f,2
1324
+ np.float64,0x3fedd2c9457ba592,0x3fe512ce92394526,2
1325
+ np.float64,0x7fe6dcb8ceadb971,0x40862b925a7dc05d,2
1326
+ np.float64,0x3fd1b983b4a37307,0x3fcf4ae2545cf64e,2
1327
+ np.float64,0xbfe1c93104639262,0xbfe9f7d28e4c0c82,2
1328
+ np.float64,0x995ebc2932bd8,0x995ebc2932bd8,2
1329
+ np.float64,0x800a4c3ee614987e,0x800a4c3ee614987e,2
1330
+ np.float64,0x3fbb58766e36b0f0,0x3fb9fb3b9810ec16,2
1331
+ np.float64,0xbfe36d636666dac7,0xbfede5080f69053c,2
1332
+ np.float64,0x3f4feee1003fddc2,0x3f4feae5f05443d1,2
1333
+ np.float64,0x3fed0b772ffa16ee,0x3fe4aafb924903c6,2
1334
+ np.float64,0x800bb3faef3767f6,0x800bb3faef3767f6,2
1335
+ np.float64,0x3fe285cda5e50b9c,0x3fdd3a58df06c427,2
1336
+ np.float64,0x7feb9d560bb73aab,0x40862d152362bb94,2
1337
+ np.float64,0x3fecd1f447f9a3e9,0x3fe48cc78288cb3f,2
1338
+ np.float64,0x3fca927b0c3524f6,0x3fc8250f49ba28df,2
1339
+ np.float64,0x7fcc19944e383328,0x40862221b02fcf43,2
1340
+ np.float64,0xbfd8ddf41db1bbe8,0xbfdf7b92073ff2fd,2
1341
+ np.float64,0x80006fe736e0dfcf,0x80006fe736e0dfcf,2
1342
+ np.float64,0x800bbeb66d577d6d,0x800bbeb66d577d6d,2
1343
+ np.float64,0xbfe4329353e86526,0xbfefeaf19ab92b42,2
1344
+ np.float64,0x2fad72805f5af,0x2fad72805f5af,2
1345
+ np.float64,0x3fe1b827aa637050,0x3fdc33bf46012c0d,2
1346
+ np.float64,0x3fc3f3f8e227e7f2,0x3fc28aeb86d65278,2
1347
+ np.float64,0x3fec018933780312,0x3fe41e619aa4285c,2
1348
+ np.float64,0xbfd92428e0b24852,0xbfdfeecb08d154df,2
1349
+ np.float64,0x2d7046845ae0a,0x2d7046845ae0a,2
1350
+ np.float64,0x7fde7fd2233cffa3,0x408628550f8a948f,2
1351
+ np.float64,0x8000a32cd241465a,0x8000a32cd241465a,2
1352
+ np.float64,0x8004267a45084cf5,0x8004267a45084cf5,2
1353
+ np.float64,0xbfe6b422556d6844,0xbff3c71f67661e6e,2
1354
+ np.float64,0x3fe3a37d922746fb,0x3fdea04e04d6195c,2
1355
+ np.float64,0xbfddcc54b53b98aa,0xbfe40d2389cdb848,2
1356
+ np.float64,0x3fe18b4b92a31697,0x3fdbf9e68cbf5794,2
1357
+ np.float64,0x7fc9c5b2ee338b65,0x408621709a17a47a,2
1358
+ np.float64,0x1ebd1ce03d7b,0x1ebd1ce03d7b,2
1359
+ np.float64,0x8008a6fc39d14df9,0x8008a6fc39d14df9,2
1360
+ np.float64,0x3fec11384c782270,0x3fe426bdaedd2965,2
1361
+ np.float64,0x3fefc28344ff8507,0x3fe60f75d34fc3d2,2
1362
+ np.float64,0xc35f379786be7,0xc35f379786be7,2
1363
+ np.float64,0x3feef51f4a7dea3e,0x3fe5a7b95d7786b5,2
1364
+ np.float64,0x3fec9b9f0379373e,0x3fe4702477abbb63,2
1365
+ np.float64,0x3fde94f8cdbd29f0,0x3fd8ff50f7df0a6f,2
1366
+ np.float64,0xbfed32d1cdfa65a4,0xc0037c1470f6f979,2
1367
+ np.float64,0x800d3ba44f5a7749,0x800d3ba44f5a7749,2
1368
+ np.float64,0x3fe3c56c8fe78ad9,0x3fdeca4eb9bb8918,2
1369
+ np.float64,0xbfe7c97242ef92e4,0xbff5c2950dfd6f69,2
1370
+ np.float64,0xbd9440057b288,0xbd9440057b288,2
1371
+ np.float64,0x7feb2fc111f65f81,0x40862cf524bd2001,2
1372
+ np.float64,0x800a431e2df4863d,0x800a431e2df4863d,2
1373
+ np.float64,0x80038a3b79e71478,0x80038a3b79e71478,2
1374
+ np.float64,0x80000c93d4601928,0x80000c93d4601928,2
1375
+ np.float64,0x7fe9fec022f3fd7f,0x40862c995db8ada0,2
1376
+ np.float64,0x3fead0129c35a025,0x3fe379d7a92c8f79,2
1377
+ np.float64,0x3fdd8cbaf7bb1974,0x3fd84b87ff0c26c7,2
1378
+ np.float64,0x3fe8fb7c60b1f6f9,0x3fe276d5339e7135,2
1379
+ np.float64,0x85a255e10b44b,0x85a255e10b44b,2
1380
+ np.float64,0xbfe507c23fea0f84,0xbff1212d2260022a,2
1381
+ np.float64,0x3fc5487c7b2a90f9,0x3fc3b03222d3d148,2
1382
+ np.float64,0x7fec0bdcb8f817b8,0x40862d34e8fd11e7,2
1383
+ np.float64,0xbfc5f34b4f2be698,0xbfc8146a899c7a0c,2
1384
+ np.float64,0xbfa2a49c14254940,0xbfa2fdab2eae3826,2
1385
+ np.float64,0x800ec52f15dd8a5e,0x800ec52f15dd8a5e,2
1386
+ np.float64,0xbfe3ba4b12a77496,0xbfeeab256b3e9422,2
1387
+ np.float64,0x80034d6c7ba69ada,0x80034d6c7ba69ada,2
1388
+ np.float64,0x7fd394d4202729a7,0x408624c98a216742,2
1389
+ np.float64,0xbfd4493a38289274,0xbfd865d67af2de91,2
1390
+ np.float64,0xe47d6203c8fad,0xe47d6203c8fad,2
1391
+ np.float64,0x98eb4e4b31d6a,0x98eb4e4b31d6a,2
1392
+ np.float64,0x4507fb128a100,0x4507fb128a100,2
1393
+ np.float64,0xbfc77032e42ee064,0xbfc9e36ab747a14d,2
1394
+ np.float64,0xa1f8a03b43f14,0xa1f8a03b43f14,2
1395
+ np.float64,0xbfc3d4da8527a9b4,0xbfc58c27af2476b0,2
1396
+ np.float64,0x3fc0eb7d6921d6fb,0x3fbfc858a077ed61,2
1397
+ np.float64,0x7fddb2e9403b65d2,0x4086281e98443709,2
1398
+ np.float64,0xbfa7ea62942fd4c0,0xbfa87dfd06b05d2a,2
1399
+ np.float64,0xbfe7d5c5426fab8a,0xbff5daa969c6d9e5,2
1400
+ np.float64,0x3fbf7cba0c3ef974,0x3fbdb23cd8fe875b,2
1401
+ np.float64,0x7fe92021eb324043,0x40862c53aee8b154,2
1402
+ np.float64,0x7fefbaa1827f7542,0x40862e3194737072,2
1403
+ np.float64,0x3fc6f82c402df059,0x3fc520432cbc533f,2
1404
+ np.float64,0x7fb37679a826ecf2,0x408619a5f857e27f,2
1405
+ np.float64,0x79ec1528f3d83,0x79ec1528f3d83,2
1406
+ np.float64,0x3fbefe1d0c3dfc3a,0x3fbd41650ba2c893,2
1407
+ np.float64,0x3fc3e5e11827cbc2,0x3fc27eb9b47c9c42,2
1408
+ np.float64,0x16aed1922d5db,0x16aed1922d5db,2
1409
+ np.float64,0x800124f7e58249f1,0x800124f7e58249f1,2
1410
+ np.float64,0x8004f7d12489efa3,0x8004f7d12489efa3,2
1411
+ np.float64,0x3fef80b8e27f0172,0x3fe5ee5fd43322c6,2
1412
+ np.float64,0xbfe7740c88eee819,0xbff51f823c8da14d,2
1413
+ np.float64,0xbfe6e1f1f6edc3e4,0xbff416bcb1302e7c,2
1414
+ np.float64,0x8001a2c4a7e3458a,0x8001a2c4a7e3458a,2
1415
+ np.float64,0x3fe861e155f0c3c2,0x3fe2201d3000c329,2
1416
+ np.float64,0x3fd00a101a201420,0x3fcca01087dbd728,2
1417
+ np.float64,0x7fdf0eb1133e1d61,0x4086287a327839b8,2
1418
+ np.float64,0x95e3ffdb2bc80,0x95e3ffdb2bc80,2
1419
+ np.float64,0x3fd87a1e8230f43d,0x3fd4ba1eb9be1270,2
1420
+ np.float64,0x3fedc4792afb88f2,0x3fe50b6529080f73,2
1421
+ np.float64,0x7fc9e81fa833d03e,0x4086217b428cc6ff,2
1422
+ np.float64,0xbfd21f1ba5a43e38,0xbfd54e048b988e09,2
1423
+ np.float64,0xbfbf52af5a3ea560,0xbfc0b4ab3b81fafc,2
1424
+ np.float64,0x7fe475f8e268ebf1,0x40862aaf14fee029,2
1425
+ np.float64,0x3fcf56899f3ead10,0x3fcc081de28ae9cf,2
1426
+ np.float64,0x917d407122fa8,0x917d407122fa8,2
1427
+ np.float64,0x22e23e3245c49,0x22e23e3245c49,2
1428
+ np.float64,0xbfeec2814f3d8503,0xc00a00ecca27b426,2
1429
+ np.float64,0xbfd97fee1c32ffdc,0xbfe04351dfe306ec,2
evalkit_tf437/lib/python3.10/site-packages/numpy/core/tests/data/umath-validation-set-log2.csv ADDED
@@ -0,0 +1,1629 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ dtype,input,output,ulperrortol
2
+ np.float32,0x80000000,0xff800000,3
3
+ np.float32,0x7f12870a,0x42fe63db,3
4
+ np.float32,0x3ef29cf5,0xbf89eb12,3
5
+ np.float32,0x3d6ba8fb,0xc083d26c,3
6
+ np.float32,0x3d9907e8,0xc06f8230,3
7
+ np.float32,0x4ee592,0xc2fd656e,3
8
+ np.float32,0x58d8b1,0xc2fd0db3,3
9
+ np.float32,0x7ba103,0xc2fc19aa,3
10
+ np.float32,0x7f52e90e,0x42ff70e4,3
11
+ np.float32,0x7fcb15,0xc2fc0132,3
12
+ np.float32,0x7cb7129f,0x42f50855,3
13
+ np.float32,0x9faba,0xc301ae59,3
14
+ np.float32,0x7f300a,0xc2fc04b4,3
15
+ np.float32,0x3f0bf047,0xbf5f10cb,3
16
+ np.float32,0x2fb1fb,0xc2fed934,3
17
+ np.float32,0x3eedb0d1,0xbf8db417,3
18
+ np.float32,0x3d7a0b40,0xc0811638,3
19
+ np.float32,0x2e0bac,0xc2fef334,3
20
+ np.float32,0x6278c1,0xc2fcc1b9,3
21
+ np.float32,0x7f61ab2e,0x42ffa2d9,3
22
+ np.float32,0x8fe7c,0xc301d4be,3
23
+ np.float32,0x3f25e6ee,0xbf203536,3
24
+ np.float32,0x7efc78f0,0x42fdf5c0,3
25
+ np.float32,0x6d7304,0xc2fc73a7,3
26
+ np.float32,0x7f1a472a,0x42fe89ed,3
27
+ np.float32,0x7dd029a6,0x42f96734,3
28
+ np.float32,0x3e9b9327,0xbfdbf8f7,3
29
+ np.float32,0x3f4eefc1,0xbe9d2942,3
30
+ np.float32,0x7f5b9b64,0x42ff8ebc,3
31
+ np.float32,0x3e458ee1,0xc017ed6e,3
32
+ np.float32,0x3f7b766b,0xbcd35acf,3
33
+ np.float32,0x3e616070,0xc00bc378,3
34
+ np.float32,0x7f20e633,0x42fea8f8,3
35
+ np.float32,0x3ee3b461,0xbf95a126,3
36
+ np.float32,0x7e7722ba,0x42fbe5f8,3
37
+ np.float32,0x3f0873d7,0xbf6861fa,3
38
+ np.float32,0x7b4cb2,0xc2fc1ba3,3
39
+ np.float32,0x3f0b6b02,0xbf60712e,3
40
+ np.float32,0x9bff4,0xc301b6f2,3
41
+ np.float32,0x3f07be25,0xbf6a4f0c,3
42
+ np.float32,0x3ef10e57,0xbf8b1b75,3
43
+ np.float32,0x46ad75,0xc2fdb6b1,3
44
+ np.float32,0x3f7bc542,0xbcc4e3a9,3
45
+ np.float32,0x3f6673d4,0xbe1b509c,3
46
+ np.float32,0x7f19fe59,0x42fe8890,3
47
+ np.float32,0x7f800000,0x7f800000,3
48
+ np.float32,0x7f2fe696,0x42feead0,3
49
+ np.float32,0x3dc9432d,0xc0563655,3
50
+ np.float32,0x3ee47623,0xbf950446,3
51
+ np.float32,0x3f1f8817,0xbf2eab51,3
52
+ np.float32,0x7f220ec5,0x42feae44,3
53
+ np.float32,0x2325e3,0xc2ffbab1,3
54
+ np.float32,0x29dfc8,0xc2ff395a,3
55
+ np.float32,0x7f524950,0x42ff6eb3,3
56
+ np.float32,0x3e2234e0,0xc02a21c8,3
57
+ np.float32,0x7f1c6f5a,0x42fe942f,3
58
+ np.float32,0x3b6a61,0xc2fe36e7,3
59
+ np.float32,0x3f1df90e,0xbf324ba9,3
60
+ np.float32,0xb57f0,0xc3017f07,3
61
+ np.float32,0x7d0eba,0xc2fc112e,3
62
+ np.float32,0x403aa9,0xc2fdfd5c,3
63
+ np.float32,0x3e74ecc7,0xc004155f,3
64
+ np.float32,0x17509c,0xc30074f2,3
65
+ np.float32,0x7f62196b,0x42ffa442,3
66
+ np.float32,0x3ecef9a9,0xbfa7417a,3
67
+ np.float32,0x7f14b158,0x42fe6eb1,3
68
+ np.float32,0x3ede12be,0xbf9a40fe,3
69
+ np.float32,0x42cfaa,0xc2fde03f,3
70
+ np.float32,0x3f407b0f,0xbed2a6f5,3
71
+ np.float32,0x7f7fffff,0x43000000,3
72
+ np.float32,0x5467c6,0xc2fd3394,3
73
+ np.float32,0x7ea6b80f,0x42fcc336,3
74
+ np.float32,0x3f21e7b2,0xbf293704,3
75
+ np.float32,0x3dc7e9eb,0xc056d542,3
76
+ np.float32,0x7f3e6e67,0x42ff2571,3
77
+ np.float32,0x3e3e809d,0xc01b4911,3
78
+ np.float32,0x3f800000,0x0,3
79
+ np.float32,0x3d8fd238,0xc0753d52,3
80
+ np.float32,0x3f74aa65,0xbd85cd0e,3
81
+ np.float32,0x7ec30305,0x42fd36ff,3
82
+ np.float32,0x3e97bb93,0xbfe0971d,3
83
+ np.float32,0x3e109d9c,0xc034bb1b,3
84
+ np.float32,0x3f4a0b67,0xbeaed537,3
85
+ np.float32,0x3f25a7aa,0xbf20c228,3
86
+ np.float32,0x3ebc05eb,0xbfb8fd6b,3
87
+ np.float32,0x3eebe749,0xbf8f18e5,3
88
+ np.float32,0x3e9dc479,0xbfd96356,3
89
+ np.float32,0x7f245200,0x42feb882,3
90
+ np.float32,0x1573a8,0xc30093b5,3
91
+ np.float32,0x3e66c4b9,0xc00994a6,3
92
+ np.float32,0x3e73bffc,0xc0048709,3
93
+ np.float32,0x3dfef8e5,0xc0405f16,3
94
+ np.float32,0x403750,0xc2fdfd83,3
95
+ np.float32,0x3ebedf17,0xbfb636a4,3
96
+ np.float32,0x15cae6,0xc3008de2,3
97
+ np.float32,0x3edf4d4e,0xbf993c24,3
98
+ np.float32,0x3f7cc41e,0xbc963fb3,3
99
+ np.float32,0x3e9e12a4,0xbfd907ee,3
100
+ np.float32,0x7ded7b59,0x42f9c889,3
101
+ np.float32,0x7f034878,0x42fe12b5,3
102
+ np.float32,0x7ddce43f,0x42f9930b,3
103
+ np.float32,0x3d82b257,0xc07e1333,3
104
+ np.float32,0x3dae89c1,0xc0635dd4,3
105
+ np.float32,0x6b1d00,0xc2fc8396,3
106
+ np.float32,0x449a5a,0xc2fdccb3,3
107
+ np.float32,0x4e89d2,0xc2fd68cb,3
108
+ np.float32,0x7e1ae83f,0x42fa8cef,3
109
+ np.float32,0x7e4bb22c,0x42fb572e,3
110
+ np.float32,0x3de308ea,0xc04b1634,3
111
+ np.float32,0x7f238c7a,0x42feb508,3
112
+ np.float32,0x3f6c62a3,0xbdeb86f3,3
113
+ np.float32,0x3e58cba6,0xc00f5908,3
114
+ np.float32,0x7f7dd91f,0x42fff9c4,3
115
+ np.float32,0x3d989376,0xc06fc88d,3
116
+ np.float32,0x3dd013c5,0xc0532339,3
117
+ np.float32,0x4b17e6,0xc2fd89ed,3
118
+ np.float32,0x7f67f287,0x42ffb71e,3
119
+ np.float32,0x3f69365e,0xbe09ba3c,3
120
+ np.float32,0x3e4b8b21,0xc0152bf1,3
121
+ np.float32,0x3a75b,0xc3032171,3
122
+ np.float32,0x7f303676,0x42feec1f,3
123
+ np.float32,0x7f6570e5,0x42ffaf18,3
124
+ np.float32,0x3f5ed61e,0xbe4cf676,3
125
+ np.float32,0x3e9b22f9,0xbfdc7e4f,3
126
+ np.float32,0x2c095e,0xc2ff1428,3
127
+ np.float32,0x3f1b17c1,0xbf391754,3
128
+ np.float32,0x422dc6,0xc2fde746,3
129
+ np.float32,0x3f677c8d,0xbe14b365,3
130
+ np.float32,0x3ef85d0c,0xbf8597a9,3
131
+ np.float32,0x3ecaaa6b,0xbfab2430,3
132
+ np.float32,0x3f0607d1,0xbf6eff3d,3
133
+ np.float32,0x3f011fdb,0xbf7cc50d,3
134
+ np.float32,0x6ed7c1,0xc2fc6a4e,3
135
+ np.float32,0x7ec2d1a2,0x42fd3644,3
136
+ np.float32,0x3f75b7fe,0xbd7238a2,3
137
+ np.float32,0x3ef2d146,0xbf89c344,3
138
+ np.float32,0x7ec2cd27,0x42fd3633,3
139
+ np.float32,0x7ee1e55a,0x42fda397,3
140
+ np.float32,0x7f464d6a,0x42ff435c,3
141
+ np.float32,0x7f469a93,0x42ff447b,3
142
+ np.float32,0x7ece752f,0x42fd6121,3
143
+ np.float32,0x2ed878,0xc2fee67b,3
144
+ np.float32,0x75b23,0xc3021eff,3
145
+ np.float32,0x3e0f4be4,0xc03593b8,3
146
+ np.float32,0x2778e1,0xc2ff64fc,3
147
+ np.float32,0x5fe2b7,0xc2fcd561,3
148
+ np.float32,0x19b8a9,0xc30050ab,3
149
+ np.float32,0x7df303e5,0x42f9d98d,3
150
+ np.float32,0x608b8d,0xc2fcd051,3
151
+ np.float32,0x588f46,0xc2fd1017,3
152
+ np.float32,0x3eec6a11,0xbf8eb2a1,3
153
+ np.float32,0x3f714121,0xbdaf4906,3
154
+ np.float32,0x7f4f7b9e,0x42ff64c9,3
155
+ np.float32,0x3c271606,0xc0d3b29c,3
156
+ np.float32,0x3f002fe0,0xbf7f75f6,3
157
+ np.float32,0x7efa4798,0x42fdef4f,3
158
+ np.float32,0x3f61a865,0xbe3a601a,3
159
+ np.float32,0x7e8087aa,0x42fc030d,3
160
+ np.float32,0x3f70f0c7,0xbdb321ba,3
161
+ np.float32,0x5db898,0xc2fce63f,3
162
+ np.float32,0x7a965f,0xc2fc1fea,3
163
+ np.float32,0x7f68b112,0x42ffb97c,3
164
+ np.float32,0x7ef0ed3d,0x42fdd32d,3
165
+ np.float32,0x7f3156a1,0x42fef0d3,3
166
+ np.float32,0x3f1d405f,0xbf33fc6e,3
167
+ np.float32,0x3e3494cf,0xc0203945,3
168
+ np.float32,0x6018de,0xc2fcd3c1,3
169
+ np.float32,0x623e49,0xc2fcc370,3
170
+ np.float32,0x3ea29f0f,0xbfd3cad4,3
171
+ np.float32,0xa514,0xc305a20c,3
172
+ np.float32,0x3e1b2ab1,0xc02e3a8f,3
173
+ np.float32,0x3f450b6f,0xbec1578f,3
174
+ np.float32,0x7eb12908,0x42fcf015,3
175
+ np.float32,0x3f10b720,0xbf52ab48,3
176
+ np.float32,0x3e0a93,0xc2fe16f6,3
177
+ np.float32,0x93845,0xc301cb96,3
178
+ np.float32,0x7f4e9ce3,0x42ff61af,3
179
+ np.float32,0x3f6d4296,0xbde09ceb,3
180
+ np.float32,0x6ddede,0xc2fc70d0,3
181
+ np.float32,0x3f4fb6fd,0xbe9a636d,3
182
+ np.float32,0x3f6d08de,0xbde36c0b,3
183
+ np.float32,0x3f56f057,0xbe8122ad,3
184
+ np.float32,0x334e95,0xc2fea349,3
185
+ np.float32,0x7efadbcd,0x42fdf104,3
186
+ np.float32,0x3db02e88,0xc0628046,3
187
+ np.float32,0x3f3309d1,0xbf041066,3
188
+ np.float32,0x2d8722,0xc2fefb8f,3
189
+ np.float32,0x7e926cac,0x42fc6356,3
190
+ np.float32,0x3e3674ab,0xc01f452e,3
191
+ np.float32,0x1b46ce,0xc3003afc,3
192
+ np.float32,0x3f06a338,0xbf6d53fc,3
193
+ np.float32,0x1b1ba7,0xc3003d46,3
194
+ np.float32,0x319dfb,0xc2febc06,3
195
+ np.float32,0x3e2f126a,0xc02315a5,3
196
+ np.float32,0x3f40fe65,0xbed0af9e,3
197
+ np.float32,0x3f1d842f,0xbf335d4b,3
198
+ np.float32,0x3d044e4f,0xc09e78f8,3
199
+ np.float32,0x7f272674,0x42fec51f,3
200
+ np.float32,0x3cda6d8f,0xc0a753db,3
201
+ np.float32,0x3eb92f12,0xbfbbccbb,3
202
+ np.float32,0x7e4318f4,0x42fb3752,3
203
+ np.float32,0x3c5890,0xc2fe2b6d,3
204
+ np.float32,0x3d1993c9,0xc09796f8,3
205
+ np.float32,0x7f18ef24,0x42fe8377,3
206
+ np.float32,0x3e30c3a0,0xc0223244,3
207
+ np.float32,0x3f27cd27,0xbf1c00ef,3
208
+ np.float32,0x3f150957,0xbf47cd6c,3
209
+ np.float32,0x7e7178a3,0x42fbd4d8,3
210
+ np.float32,0x3f298db8,0xbf182ac3,3
211
+ np.float32,0x7cb3be,0xc2fc1348,3
212
+ np.float32,0x3ef64266,0xbf8729de,3
213
+ np.float32,0x3eeb06ce,0xbf8fc8f2,3
214
+ np.float32,0x3f406e36,0xbed2d845,3
215
+ np.float32,0x7f1e1bd3,0x42fe9c0b,3
216
+ np.float32,0x478dcc,0xc2fdad97,3
217
+ np.float32,0x7f7937b5,0x42ffec2b,3
218
+ np.float32,0x3f20f350,0xbf2b6624,3
219
+ np.float32,0x7f13661a,0x42fe683c,3
220
+ np.float32,0x208177,0xc2fff46b,3
221
+ np.float32,0x263cfb,0xc2ff7c72,3
222
+ np.float32,0x7f0bd28c,0x42fe4141,3
223
+ np.float32,0x7230d8,0xc2fc5453,3
224
+ np.float32,0x3f261bbf,0xbf1fbfb4,3
225
+ np.float32,0x737b56,0xc2fc4c05,3
226
+ np.float32,0x3ef88f33,0xbf857263,3
227
+ np.float32,0x7e036464,0x42fa1352,3
228
+ np.float32,0x4b5c4f,0xc2fd874d,3
229
+ np.float32,0x3f77984d,0xbd454596,3
230
+ np.float32,0x3f674202,0xbe162932,3
231
+ np.float32,0x3e7157d9,0xc0057197,3
232
+ np.float32,0x3f3f21da,0xbed7d861,3
233
+ np.float32,0x7f1fb40f,0x42fea375,3
234
+ np.float32,0x7ef0157f,0x42fdd096,3
235
+ np.float32,0x3f71e88d,0xbda74962,3
236
+ np.float32,0x3f174855,0xbf424728,3
237
+ np.float32,0x3f3fdd2c,0xbed505d5,3
238
+ np.float32,0x7b95d1,0xc2fc19ed,3
239
+ np.float32,0x7f23f4e5,0x42feb6df,3
240
+ np.float32,0x7d741925,0x42f7dcd6,3
241
+ np.float32,0x60f81d,0xc2fccd14,3
242
+ np.float32,0x3f17d267,0xbf40f6ae,3
243
+ np.float32,0x3f036fc8,0xbf7636f8,3
244
+ np.float32,0x167653,0xc30082b5,3
245
+ np.float32,0x256d05,0xc2ff8c4f,3
246
+ np.float32,0x3eccc63d,0xbfa93adb,3
247
+ np.float32,0x7f6c91ea,0x42ffc5b2,3
248
+ np.float32,0x2ee52a,0xc2fee5b3,3
249
+ np.float32,0x3dc3579e,0xc058f80d,3
250
+ np.float32,0x4c7170,0xc2fd7cc4,3
251
+ np.float32,0x7f737f20,0x42ffdb03,3
252
+ np.float32,0x3f2f9dbf,0xbf0b3119,3
253
+ np.float32,0x3f4d0c54,0xbea3eec5,3
254
+ np.float32,0x7e380862,0x42fb0c32,3
255
+ np.float32,0x5d637f,0xc2fce8df,3
256
+ np.float32,0x3f0aa623,0xbf627c27,3
257
+ np.float32,0x3e4d5896,0xc0145b88,3
258
+ np.float32,0x3f6cacdc,0xbde7e7ca,3
259
+ np.float32,0x63a2c3,0xc2fcb90a,3
260
+ np.float32,0x6c138c,0xc2fc7cfa,3
261
+ np.float32,0x2063c,0xc303fb88,3
262
+ np.float32,0x7e9e5a3e,0x42fc9d2f,3
263
+ np.float32,0x56ec64,0xc2fd1ddd,3
264
+ np.float32,0x7f1d6a35,0x42fe98cc,3
265
+ np.float32,0x73dc96,0xc2fc4998,3
266
+ np.float32,0x3e5d74e5,0xc00d6238,3
267
+ np.float32,0x7f033cbb,0x42fe1273,3
268
+ np.float32,0x3f5143fc,0xbe94e4e7,3
269
+ np.float32,0x1d56d9,0xc3002010,3
270
+ np.float32,0x2bf3e4,0xc2ff1591,3
271
+ np.float32,0x3f2a6ef1,0xbf164170,3
272
+ np.float32,0x3f33238b,0xbf03db58,3
273
+ np.float32,0x22780e,0xc2ffc91a,3
274
+ np.float32,0x7f00b873,0x42fe0425,3
275
+ np.float32,0x3f7f6145,0xbb654706,3
276
+ np.float32,0x7fc00000,0x7fc00000,3
277
+ np.float32,0x63895a,0xc2fcb9c7,3
278
+ np.float32,0x18a1b2,0xc30060a8,3
279
+ np.float32,0x7e43c6a6,0x42fb39e3,3
280
+ np.float32,0x78676e,0xc2fc2d30,3
281
+ np.float32,0x3f16d839,0xbf435940,3
282
+ np.float32,0x7eff78ba,0x42fdfe79,3
283
+ np.float32,0x3f2e152c,0xbf0e6e54,3
284
+ np.float32,0x3db20ced,0xc06186e1,3
285
+ np.float32,0x3f0cd1d8,0xbf5cbf57,3
286
+ np.float32,0x3fd7a8,0xc2fe01d2,3
287
+ np.float32,0x3ebb075e,0xbfb9f816,3
288
+ np.float32,0x7f94ef,0xc2fc026b,3
289
+ np.float32,0x3d80ba0e,0xc07f7a2b,3
290
+ np.float32,0x7f227e15,0x42feb03f,3
291
+ np.float32,0x792264bf,0x42e6afcc,3
292
+ np.float32,0x7f501576,0x42ff66ec,3
293
+ np.float32,0x223629,0xc2ffcea3,3
294
+ np.float32,0x40a79e,0xc2fdf87b,3
295
+ np.float32,0x449483,0xc2fdccf2,3
296
+ np.float32,0x3f4fa978,0xbe9a9382,3
297
+ np.float32,0x7f148c53,0x42fe6df9,3
298
+ np.float32,0x3ec98b3c,0xbfac2a98,3
299
+ np.float32,0x3e4da320,0xc0143a0a,3
300
+ np.float32,0x3d1d94bb,0xc09666d0,3
301
+ np.float32,0x3c8e624e,0xc0bb155b,3
302
+ np.float32,0x66a9af,0xc2fca2ef,3
303
+ np.float32,0x3ec76ed7,0xbfae1c57,3
304
+ np.float32,0x3f4b52f3,0xbeaa2b81,3
305
+ np.float32,0x7e99bbb5,0x42fc8750,3
306
+ np.float32,0x3f69a46b,0xbe0701be,3
307
+ np.float32,0x3f775400,0xbd4ba495,3
308
+ np.float32,0x131e56,0xc300be3c,3
309
+ np.float32,0x3f30abb4,0xbf08fb10,3
310
+ np.float32,0x7f7e528c,0x42fffb25,3
311
+ np.float32,0x3eb89515,0xbfbc668a,3
312
+ np.float32,0x7e9191b6,0x42fc5f02,3
313
+ np.float32,0x7e80c7e9,0x42fc047e,3
314
+ np.float32,0x3f77ef58,0xbd3d2995,3
315
+ np.float32,0x7ddb1f8a,0x42f98d1b,3
316
+ np.float32,0x7ebc6c4f,0x42fd1d9c,3
317
+ np.float32,0x3f6638e0,0xbe1ccab8,3
318
+ np.float32,0x7f4c45,0xc2fc0410,3
319
+ np.float32,0x3e7d8aad,0xc000e414,3
320
+ np.float32,0x3f4d148b,0xbea3d12e,3
321
+ np.float32,0x3e98c45c,0xbfdf55f4,3
322
+ np.float32,0x3d754c78,0xc081f8a9,3
323
+ np.float32,0x17e4cf,0xc3006be3,3
324
+ np.float32,0x7eb65814,0x42fd0563,3
325
+ np.float32,0x3f65e0d8,0xbe1f0008,3
326
+ np.float32,0x3e99541f,0xbfdea87e,3
327
+ np.float32,0x3f3cb80e,0xbee13b27,3
328
+ np.float32,0x3e99f0c0,0xbfddec3b,3
329
+ np.float32,0x3f43903e,0xbec6ea66,3
330
+ np.float32,0x7e211cd4,0x42faa9f2,3
331
+ np.float32,0x824af,0xc301f971,3
332
+ np.float32,0x3e16a56e,0xc030f56c,3
333
+ np.float32,0x542b3b,0xc2fd35a6,3
334
+ np.float32,0x3eeea2d1,0xbf8cf873,3
335
+ np.float32,0x232e93,0xc2ffb9fa,3
336
+ np.float32,0x3e8c52b9,0xbfef06aa,3
337
+ np.float32,0x7f69c7e3,0x42ffbcef,3
338
+ np.float32,0x3f573e43,0xbe801714,3
339
+ np.float32,0x43b009,0xc2fdd69f,3
340
+ np.float32,0x3ee571ab,0xbf943966,3
341
+ np.float32,0x3ee3d5d8,0xbf958604,3
342
+ np.float32,0x338b12,0xc2fe9fe4,3
343
+ np.float32,0x29cb1f,0xc2ff3ac6,3
344
+ np.float32,0x3f0892b4,0xbf680e7a,3
345
+ np.float32,0x3e8c4f7f,0xbfef0ae9,3
346
+ np.float32,0x7c9d3963,0x42f497e6,3
347
+ np.float32,0x3f26ba84,0xbf1e5f59,3
348
+ np.float32,0x3dd0acc0,0xc052df6f,3
349
+ np.float32,0x3e43fbda,0xc018aa8c,3
350
+ np.float32,0x3ec4fd0f,0xbfb0635d,3
351
+ np.float32,0x3f52c8c6,0xbe8f8d85,3
352
+ np.float32,0x3f5fdc5d,0xbe462fdb,3
353
+ np.float32,0x3f461920,0xbebd6743,3
354
+ np.float32,0x6161ff,0xc2fcc9ef,3
355
+ np.float32,0x7f7ed306,0x42fffc9a,3
356
+ np.float32,0x3d212263,0xc0955f46,3
357
+ np.float32,0x3eca5826,0xbfab6f36,3
358
+ np.float32,0x7d6317ac,0x42f7a77e,3
359
+ np.float32,0x3eb02063,0xbfc50f60,3
360
+ np.float32,0x7f71a6f8,0x42ffd565,3
361
+ np.float32,0x1a3efe,0xc3004935,3
362
+ np.float32,0x3dc599c9,0xc057e856,3
363
+ np.float32,0x3f3e1301,0xbedbf205,3
364
+ np.float32,0xf17d4,0xc301158d,3
365
+ np.float32,0x3f615f84,0xbe3c3d85,3
366
+ np.float32,0x3de63be1,0xc049cb77,3
367
+ np.float32,0x3e8d2f51,0xbfede541,3
368
+ np.float32,0x3a5cdd,0xc2fe441c,3
369
+ np.float32,0x3f443ec0,0xbec4586a,3
370
+ np.float32,0x3eacbd00,0xbfc8a5ad,3
371
+ np.float32,0x3f600f6a,0xbe44df1b,3
372
+ np.float32,0x5f77a6,0xc2fcd89c,3
373
+ np.float32,0x476706,0xc2fdaf28,3
374
+ np.float32,0x2f469,0xc3036fde,3
375
+ np.float32,0x7dc4ba24,0x42f93d77,3
376
+ np.float32,0x3e2d6080,0xc023fb9b,3
377
+ np.float32,0x7e8d7135,0x42fc49c3,3
378
+ np.float32,0x3f589065,0xbe77247b,3
379
+ np.float32,0x3f59e210,0xbe6e2c05,3
380
+ np.float32,0x7f51d388,0x42ff6d15,3
381
+ np.float32,0x7d9a5fda,0x42f88a63,3
382
+ np.float32,0x3e67d5bc,0xc00927ab,3
383
+ np.float32,0x61d72c,0xc2fcc679,3
384
+ np.float32,0x3ef3351d,0xbf897766,3
385
+ np.float32,0x1,0xc3150000,3
386
+ np.float32,0x7f653429,0x42ffae54,3
387
+ np.float32,0x7e1ad3e5,0x42fa8c8e,3
388
+ np.float32,0x3f4ca01d,0xbea57500,3
389
+ np.float32,0x3f7606db,0xbd6ad13e,3
390
+ np.float32,0x7ec4a27d,0x42fd3d1f,3
391
+ np.float32,0x3efe4fd5,0xbf8138c7,3
392
+ np.float32,0x77c2f1,0xc2fc3124,3
393
+ np.float32,0x7e4d3251,0x42fb5c9a,3
394
+ np.float32,0x3f543ac7,0xbe8a8154,3
395
+ np.float32,0x7c3dbe29,0x42f322c4,3
396
+ np.float32,0x408e01,0xc2fdf9a0,3
397
+ np.float32,0x45069b,0xc2fdc829,3
398
+ np.float32,0x3d7ecab7,0xc08037e8,3
399
+ np.float32,0xf8c22,0xc3010a99,3
400
+ np.float32,0x7f69af63,0x42ffbca2,3
401
+ np.float32,0x7ec7d228,0x42fd48fe,3
402
+ np.float32,0xff800000,0xffc00000,3
403
+ np.float32,0xdd7c5,0xc301357c,3
404
+ np.float32,0x143f38,0xc300a90e,3
405
+ np.float32,0x7e65c176,0x42fbb01b,3
406
+ np.float32,0x2c1a9e,0xc2ff1307,3
407
+ np.float32,0x7f6e9224,0x42ffcbeb,3
408
+ np.float32,0x3d32ab39,0xc0909a77,3
409
+ np.float32,0x3e150b42,0xc031f22b,3
410
+ np.float32,0x1f84b4,0xc300059a,3
411
+ np.float32,0x3f71ce21,0xbda88c2a,3
412
+ np.float32,0x2625c4,0xc2ff7e33,3
413
+ np.float32,0x3dd0b293,0xc052dcdc,3
414
+ np.float32,0x625c11,0xc2fcc290,3
415
+ np.float32,0x3f610297,0xbe3e9f24,3
416
+ np.float32,0x7ebdd5e5,0x42fd2320,3
417
+ np.float32,0x3e883458,0xbff486ff,3
418
+ np.float32,0x782313,0xc2fc2ed4,3
419
+ np.float32,0x7f39c843,0x42ff132f,3
420
+ np.float32,0x7f326aa7,0x42fef54d,3
421
+ np.float32,0x4d2c71,0xc2fd75be,3
422
+ np.float32,0x3f55747c,0xbe86409e,3
423
+ np.float32,0x7f7f0867,0x42fffd34,3
424
+ np.float32,0x321316,0xc2feb53f,3
425
+ np.float32,0x3e1b37ed,0xc02e32b0,3
426
+ np.float32,0x80edf,0xc301fd54,3
427
+ np.float32,0x3f0b08ad,0xbf617607,3
428
+ np.float32,0x7f3f4174,0x42ff28a2,3
429
+ np.float32,0x3d79306d,0xc0813eb0,3
430
+ np.float32,0x3f5f657a,0xbe49413d,3
431
+ np.float32,0x3f56c63a,0xbe81b376,3
432
+ np.float32,0x7f667123,0x42ffb24f,3
433
+ np.float32,0x3f71021b,0xbdb24d43,3
434
+ np.float32,0x7f434ab1,0x42ff380f,3
435
+ np.float32,0x3dcae496,0xc055779c,3
436
+ np.float32,0x3f5a7d88,0xbe6a0f5b,3
437
+ np.float32,0x3cdf5c32,0xc0a64bf5,3
438
+ np.float32,0x3e56222c,0xc0107d11,3
439
+ np.float32,0x561a3a,0xc2fd24df,3
440
+ np.float32,0x7ddd953c,0x42f9955a,3
441
+ np.float32,0x7e35d839,0x42fb035c,3
442
+ np.float32,0x3ec1816c,0xbfb3aeb2,3
443
+ np.float32,0x7c87cfcd,0x42f42bc2,3
444
+ np.float32,0xd9cd,0xc3053baf,3
445
+ np.float32,0x3f388234,0xbef1e5b7,3
446
+ np.float32,0x3edfcaca,0xbf98d47b,3
447
+ np.float32,0x3ef28852,0xbf89fac8,3
448
+ np.float32,0x7f7525df,0x42ffe001,3
449
+ np.float32,0x7f6c33ef,0x42ffc48c,3
450
+ np.float32,0x3ea4a881,0xbfd17e61,3
451
+ np.float32,0x3f3e379f,0xbedb63c6,3
452
+ np.float32,0x3f0524c1,0xbf717301,3
453
+ np.float32,0x3db3e7f0,0xc06091d3,3
454
+ np.float32,0x800000,0xc2fc0000,3
455
+ np.float32,0x3f2f2897,0xbf0c27ce,3
456
+ np.float32,0x7eb1776d,0x42fcf15c,3
457
+ np.float32,0x3f039018,0xbf75dc37,3
458
+ np.float32,0x3c4055,0xc2fe2c96,3
459
+ np.float32,0x3f603653,0xbe43dea5,3
460
+ np.float32,0x7f700d24,0x42ffd07c,3
461
+ np.float32,0x3f4741a3,0xbeb918dc,3
462
+ np.float32,0x3f5fe959,0xbe45da2d,3
463
+ np.float32,0x3f3e4401,0xbedb33b1,3
464
+ np.float32,0x7f0705ff,0x42fe2775,3
465
+ np.float32,0x3ea85662,0xbfcd69b0,3
466
+ np.float32,0x3f15f49f,0xbf458829,3
467
+ np.float32,0x3f17c50e,0xbf411728,3
468
+ np.float32,0x3e483f60,0xc016add2,3
469
+ np.float32,0x3f1ab9e5,0xbf39f71b,3
470
+ np.float32,0x3de0b6fb,0xc04c08fe,3
471
+ np.float32,0x7e671225,0x42fbb452,3
472
+ np.float32,0x80800000,0xffc00000,3
473
+ np.float32,0xe2df3,0xc3012c9d,3
474
+ np.float32,0x3ede1e3c,0xbf9a3770,3
475
+ np.float32,0x3df2ffde,0xc044cfec,3
476
+ np.float32,0x3eed8da5,0xbf8dcf6c,3
477
+ np.float32,0x3ead15c3,0xbfc846e1,3
478
+ np.float32,0x7ef3750a,0x42fddae4,3
479
+ np.float32,0x7e6ab7c0,0x42fbbfe4,3
480
+ np.float32,0x7ea4bbe5,0x42fcba5d,3
481
+ np.float32,0x3f227706,0xbf27f0a1,3
482
+ np.float32,0x3ef39bfd,0xbf89295a,3
483
+ np.float32,0x3f289a20,0xbf1a3edd,3
484
+ np.float32,0x7f225f82,0x42feafb4,3
485
+ np.float32,0x768963,0xc2fc38bc,3
486
+ np.float32,0x3f493c00,0xbeb1ccfc,3
487
+ np.float32,0x3f4e7249,0xbe9ee9a7,3
488
+ np.float32,0x1d0c3a,0xc30023c0,3
489
+ np.float32,0x7f3c5f78,0x42ff1d6a,3
490
+ np.float32,0xff7fffff,0xffc00000,3
491
+ np.float32,0x3ee7896a,0xbf928c2a,3
492
+ np.float32,0x3e788479,0xc002bd2e,3
493
+ np.float32,0x3ee4df17,0xbf94af84,3
494
+ np.float32,0x5e06d7,0xc2fce3d7,3
495
+ np.float32,0x3d7b2776,0xc080e1dc,3
496
+ np.float32,0x3e3d39d3,0xc01be7fd,3
497
+ np.float32,0x7c81dece,0x42f40ab7,3
498
+ np.float32,0x3f7d2085,0xbc856255,3
499
+ np.float32,0x7f7f6627,0x42fffe44,3
500
+ np.float32,0x7f5f2e94,0x42ff9aaa,3
501
+ np.float32,0x7f5835f2,0x42ff8339,3
502
+ np.float32,0x3f6a0e32,0xbe046580,3
503
+ np.float32,0x7e16f586,0x42fa79dd,3
504
+ np.float32,0x3f04a2f2,0xbf72dbc5,3
505
+ np.float32,0x3f35e334,0xbefc7740,3
506
+ np.float32,0x3f0d056e,0xbf5c3824,3
507
+ np.float32,0x7ebeb95e,0x42fd2693,3
508
+ np.float32,0x3c6192,0xc2fe2aff,3
509
+ np.float32,0x3e892b4f,0xbff33958,3
510
+ np.float32,0x3f61d694,0xbe3931df,3
511
+ np.float32,0x29d183,0xc2ff3a56,3
512
+ np.float32,0x7f0b0598,0x42fe3d04,3
513
+ np.float32,0x7f743b28,0x42ffdd3d,3
514
+ np.float32,0x3a2ed6,0xc2fe4663,3
515
+ np.float32,0x3e27403a,0xc0274de8,3
516
+ np.float32,0x3f58ee78,0xbe74a349,3
517
+ np.float32,0x3eaa4b,0xc2fe0f92,3
518
+ np.float32,0x3ecb613b,0xbfaa7de8,3
519
+ np.float32,0x7f637d81,0x42ffa8c9,3
520
+ np.float32,0x3f026e96,0xbf790c73,3
521
+ np.float32,0x386cdf,0xc2fe5d0c,3
522
+ np.float32,0x35abd1,0xc2fe8202,3
523
+ np.float32,0x3eac3cd1,0xbfc92ee8,3
524
+ np.float32,0x3f567869,0xbe82bf47,3
525
+ np.float32,0x3f65c643,0xbe1faae6,3
526
+ np.float32,0x7f5422b9,0x42ff752b,3
527
+ np.float32,0x7c26e9,0xc2fc168c,3
528
+ np.float32,0x7eff5cfd,0x42fdfe29,3
529
+ np.float32,0x3f728e7f,0xbd9f6142,3
530
+ np.float32,0x3f10fd43,0xbf51f874,3
531
+ np.float32,0x7e7ada08,0x42fbf0fe,3
532
+ np.float32,0x3e82a611,0xbffc37be,3
533
+ np.float32,0xbf800000,0xffc00000,3
534
+ np.float32,0x3dbe2e12,0xc05b711c,3
535
+ np.float32,0x7e768fa9,0x42fbe440,3
536
+ np.float32,0x5e44e8,0xc2fce1f0,3
537
+ np.float32,0x7f25071a,0x42febbae,3
538
+ np.float32,0x3f54db5e,0xbe885339,3
539
+ np.float32,0x3f0f2c26,0xbf56a0b8,3
540
+ np.float32,0x22f9a7,0xc2ffbe55,3
541
+ np.float32,0x7ed63dcb,0x42fd7c77,3
542
+ np.float32,0x7ea4fae2,0x42fcbb78,3
543
+ np.float32,0x3f1d7766,0xbf337b47,3
544
+ np.float32,0x7f16d59f,0x42fe7941,3
545
+ np.float32,0x3f3a1bb6,0xbeeb855c,3
546
+ np.float32,0x3ef57128,0xbf87c709,3
547
+ np.float32,0xb24ff,0xc3018591,3
548
+ np.float32,0x3ef99e27,0xbf84a983,3
549
+ np.float32,0x3eac2ccf,0xbfc94013,3
550
+ np.float32,0x3e9d3e1e,0xbfda00dc,3
551
+ np.float32,0x718213,0xc2fc58c1,3
552
+ np.float32,0x7edbf509,0x42fd8fea,3
553
+ np.float32,0x70c7f1,0xc2fc5d80,3
554
+ np.float32,0x3f7012f5,0xbdbdc6cd,3
555
+ np.float32,0x12cba,0xc304c487,3
556
+ np.float32,0x7f5d445d,0x42ff944c,3
557
+ np.float32,0x7f3e30bd,0x42ff2481,3
558
+ np.float32,0x63b110,0xc2fcb8a0,3
559
+ np.float32,0x3f39f728,0xbeec1680,3
560
+ np.float32,0x3f5bea58,0xbe6074b1,3
561
+ np.float32,0x3f350749,0xbefff679,3
562
+ np.float32,0x3e91ab2c,0xbfe81f3e,3
563
+ np.float32,0x7ec53fe0,0x42fd3f6d,3
564
+ np.float32,0x3f6cbbdc,0xbde72c8e,3
565
+ np.float32,0x3f4df49f,0xbea0abcf,3
566
+ np.float32,0x3e9c9638,0xbfdac674,3
567
+ np.float32,0x7f3b82ec,0x42ff1a07,3
568
+ np.float32,0x7f612a09,0x42ffa132,3
569
+ np.float32,0x7ea26650,0x42fcafd3,3
570
+ np.float32,0x3a615138,0xc122f26d,3
571
+ np.float32,0x3f1108bd,0xbf51db39,3
572
+ np.float32,0x6f80f6,0xc2fc65ea,3
573
+ np.float32,0x3f7cb578,0xbc98ecb1,3
574
+ np.float32,0x7f54d31a,0x42ff7790,3
575
+ np.float32,0x196868,0xc3005532,3
576
+ np.float32,0x3f01ee0a,0xbf7a7925,3
577
+ np.float32,0x3e184013,0xc02ffb11,3
578
+ np.float32,0xadde3,0xc3018ee3,3
579
+ np.float32,0x252a91,0xc2ff9173,3
580
+ np.float32,0x3f0382c2,0xbf7601a9,3
581
+ np.float32,0x6d818c,0xc2fc7345,3
582
+ np.float32,0x3bfbfd,0xc2fe2fdd,3
583
+ np.float32,0x7f3cad19,0x42ff1e9a,3
584
+ np.float32,0x4169a7,0xc2fdefdf,3
585
+ np.float32,0x3f615d96,0xbe3c4a2b,3
586
+ np.float32,0x3f036480,0xbf7656ac,3
587
+ np.float32,0x7f5fbda3,0x42ff9c83,3
588
+ np.float32,0x3d202d,0xc2fe21f1,3
589
+ np.float32,0x3d0f5e5d,0xc09ac3e9,3
590
+ np.float32,0x3f0fff6e,0xbf548142,3
591
+ np.float32,0x7f11ed32,0x42fe60d2,3
592
+ np.float32,0x3e6f856b,0xc00624b6,3
593
+ np.float32,0x7f7c4dd7,0x42fff542,3
594
+ np.float32,0x3e76fb86,0xc0034fa0,3
595
+ np.float32,0x3e8a0d6e,0xbff209e7,3
596
+ np.float32,0x3eacad19,0xbfc8b6ad,3
597
+ np.float32,0xa7776,0xc3019cbe,3
598
+ np.float32,0x3dc84d74,0xc056a754,3
599
+ np.float32,0x3efb8052,0xbf834626,3
600
+ np.float32,0x3f0e55fc,0xbf58cacc,3
601
+ np.float32,0x7e0e71e3,0x42fa4efb,3
602
+ np.float32,0x3ed5a800,0xbfa1639c,3
603
+ np.float32,0x3f33335b,0xbf03babf,3
604
+ np.float32,0x38cad7,0xc2fe5842,3
605
+ np.float32,0x3bc21256,0xc0ecc927,3
606
+ np.float32,0x3f09522d,0xbf660a19,3
607
+ np.float32,0xcbd5d,0xc3015428,3
608
+ np.float32,0x492752,0xc2fd9d42,3
609
+ np.float32,0x3f2b9b32,0xbf13b904,3
610
+ np.float32,0x6544ac,0xc2fcad09,3
611
+ np.float32,0x52eb12,0xc2fd40b5,3
612
+ np.float32,0x3f66a7c0,0xbe1a03e8,3
613
+ np.float32,0x7ab289,0xc2fc1f41,3
614
+ np.float32,0x62af5e,0xc2fcc020,3
615
+ np.float32,0x7f73e9cf,0x42ffdc46,3
616
+ np.float32,0x3e5eca,0xc2fe130e,3
617
+ np.float32,0x3e3a10f4,0xc01d7602,3
618
+ np.float32,0x3f04db46,0xbf723f0d,3
619
+ np.float32,0x18fc4a,0xc3005b63,3
620
+ np.float32,0x525bcb,0xc2fd45b6,3
621
+ np.float32,0x3f6b9108,0xbdf5c769,3
622
+ np.float32,0x3e992e8c,0xbfded5c5,3
623
+ np.float32,0x7efea647,0x42fdfc18,3
624
+ np.float32,0x7e8371db,0x42fc139e,3
625
+ np.float32,0x3f397cfb,0xbeedfc69,3
626
+ np.float32,0x7e46d233,0x42fb454a,3
627
+ np.float32,0x7d5281ad,0x42f76f79,3
628
+ np.float32,0x7f4c1878,0x42ff58a1,3
629
+ np.float32,0x3e96ca5e,0xbfe1bd97,3
630
+ np.float32,0x6a2743,0xc2fc8a3d,3
631
+ np.float32,0x7f688781,0x42ffb8f8,3
632
+ np.float32,0x7814b7,0xc2fc2f2d,3
633
+ np.float32,0x3f2ffdc9,0xbf0a6756,3
634
+ np.float32,0x3f766fa8,0xbd60fe24,3
635
+ np.float32,0x4dc64e,0xc2fd7003,3
636
+ np.float32,0x3a296f,0xc2fe46a8,3
637
+ np.float32,0x3f2af942,0xbf15162e,3
638
+ np.float32,0x7f702c32,0x42ffd0dc,3
639
+ np.float32,0x7e61e318,0x42fba390,3
640
+ np.float32,0x7f7d3bdb,0x42fff7fa,3
641
+ np.float32,0x3ee87f3f,0xbf91c881,3
642
+ np.float32,0x2bbc28,0xc2ff193c,3
643
+ np.float32,0x3e01f918,0xc03e966e,3
644
+ np.float32,0x7f0b39f4,0x42fe3e1a,3
645
+ np.float32,0x3eaa4d64,0xbfcb4516,3
646
+ np.float32,0x3e53901e,0xc0119a88,3
647
+ np.float32,0x603cb,0xc3026957,3
648
+ np.float32,0x7e81f926,0x42fc0b4d,3
649
+ np.float32,0x5dab7c,0xc2fce6a6,3
650
+ np.float32,0x3f46fefd,0xbeba1018,3
651
+ np.float32,0x648448,0xc2fcb28a,3
652
+ np.float32,0x3ec49470,0xbfb0c58b,3
653
+ np.float32,0x3e8a5393,0xbff1ac2b,3
654
+ np.float32,0x3f27ccfc,0xbf1c014e,3
655
+ np.float32,0x3ed886e6,0xbf9eeca8,3
656
+ np.float32,0x7cfbe06e,0x42f5f401,3
657
+ np.float32,0x3f5aa7ba,0xbe68f229,3
658
+ np.float32,0x9500d,0xc301c7e3,3
659
+ np.float32,0x3f4861,0xc2fe0853,3
660
+ np.float32,0x3e5ae104,0xc00e76f5,3
661
+ np.float32,0x71253a,0xc2fc5b1e,3
662
+ np.float32,0xcf7b8,0xc3014d9c,3
663
+ np.float32,0x7f7edd2d,0x42fffcb7,3
664
+ np.float32,0x3e9039ee,0xbfe9f5ab,3
665
+ np.float32,0x2fd54e,0xc2fed712,3
666
+ np.float32,0x3f600752,0xbe45147a,3
667
+ np.float32,0x3f4da8f6,0xbea1bb5c,3
668
+ np.float32,0x3f2d34a9,0xbf104bd9,3
669
+ np.float32,0x3e1e66dd,0xc02c52d2,3
670
+ np.float32,0x798276,0xc2fc2670,3
671
+ np.float32,0xd55e2,0xc3014347,3
672
+ np.float32,0x80000001,0xffc00000,3
673
+ np.float32,0x3e7a5ead,0xc0020da6,3
674
+ np.float32,0x7ec4c744,0x42fd3da9,3
675
+ np.float32,0x597e00,0xc2fd085a,3
676
+ np.float32,0x3dff6bf4,0xc0403575,3
677
+ np.float32,0x5d6f1a,0xc2fce883,3
678
+ np.float32,0x7e21faff,0x42faadea,3
679
+ np.float32,0x3e570fea,0xc01016c6,3
680
+ np.float32,0x28e6b6,0xc2ff4ab7,3
681
+ np.float32,0x7e77062d,0x42fbe5a3,3
682
+ np.float32,0x74cac4,0xc2fc43b0,3
683
+ np.float32,0x3f707273,0xbdb93078,3
684
+ np.float32,0x228e96,0xc2ffc737,3
685
+ np.float32,0x686ac1,0xc2fc966b,3
686
+ np.float32,0x3d76400d,0xc081cae8,3
687
+ np.float32,0x3e9f502f,0xbfd7966b,3
688
+ np.float32,0x3f6bc656,0xbdf32b1f,3
689
+ np.float32,0x3edb828b,0xbf9c65d4,3
690
+ np.float32,0x6c6e56,0xc2fc7a8e,3
691
+ np.float32,0x3f04552e,0xbf73b48f,3
692
+ np.float32,0x3f39cb69,0xbeecc457,3
693
+ np.float32,0x7f681c44,0x42ffb7a3,3
694
+ np.float32,0x7f5b44ee,0x42ff8d99,3
695
+ np.float32,0x3e71430a,0xc005798d,3
696
+ np.float32,0x3edcfde3,0xbf9b27c6,3
697
+ np.float32,0x3f616a5a,0xbe3bf67f,3
698
+ np.float32,0x3f523936,0xbe918548,3
699
+ np.float32,0x3f39ce3a,0xbeecb925,3
700
+ np.float32,0x3eac589a,0xbfc91120,3
701
+ np.float32,0x7efc8d3d,0x42fdf5fc,3
702
+ np.float32,0x5704b0,0xc2fd1d0f,3
703
+ np.float32,0x7e7972e9,0x42fbecda,3
704
+ np.float32,0x3eb0811c,0xbfc4aa13,3
705
+ np.float32,0x7f1efcbb,0x42fea023,3
706
+ np.float32,0x3e0b9e32,0xc037fa6b,3
707
+ np.float32,0x7eef6a48,0x42fdce87,3
708
+ np.float32,0x3cc0a373,0xc0ad20c0,3
709
+ np.float32,0x3f2a75bb,0xbf1632ba,3
710
+ np.float32,0x0,0xff800000,3
711
+ np.float32,0x7ecdb6f4,0x42fd5e77,3
712
+ np.float32,0x7f2e2dfd,0x42fee38d,3
713
+ np.float32,0x3ee17f6e,0xbf976d8c,3
714
+ np.float32,0x3f51e7ee,0xbe92a319,3
715
+ np.float32,0x3f06942f,0xbf6d7d3c,3
716
+ np.float32,0x3f7ba528,0xbccac6f1,3
717
+ np.float32,0x3f413787,0xbecfd513,3
718
+ np.float32,0x3e085e48,0xc03a2716,3
719
+ np.float32,0x7e4c5e0e,0x42fb599c,3
720
+ np.float32,0x306f76,0xc2fecdd4,3
721
+ np.float32,0x7f5c2203,0x42ff9081,3
722
+ np.float32,0x3d5355b4,0xc088da05,3
723
+ np.float32,0x9a2a,0xc305bb4f,3
724
+ np.float32,0x3db93a1f,0xc05de0db,3
725
+ np.float32,0x4e50c6,0xc2fd6ae4,3
726
+ np.float32,0x7ec4afed,0x42fd3d51,3
727
+ np.float32,0x3a8f27,0xc2fe41a0,3
728
+ np.float32,0x7f213caf,0x42feaa84,3
729
+ np.float32,0x7e7b5f00,0x42fbf286,3
730
+ np.float32,0x7e367194,0x42fb05ca,3
731
+ np.float32,0x7f56e6de,0x42ff7ebd,3
732
+ np.float32,0x3ed7383e,0xbfa00aef,3
733
+ np.float32,0x7e844752,0x42fc184a,3
734
+ np.float32,0x15157,0xc3049a19,3
735
+ np.float32,0x3f78cd92,0xbd28824a,3
736
+ np.float32,0x7ecddb16,0x42fd5ef9,3
737
+ np.float32,0x3e479f16,0xc016f7d8,3
738
+ np.float32,0x3f5cb418,0xbe5b2bd3,3
739
+ np.float32,0x7c0934cb,0x42f2334e,3
740
+ np.float32,0x3ebe5505,0xbfb6bc69,3
741
+ np.float32,0x3eb1335a,0xbfc3eff5,3
742
+ np.float32,0x3f2488a3,0xbf234444,3
743
+ np.float32,0x642906,0xc2fcb52a,3
744
+ np.float32,0x3da635fa,0xc067e15a,3
745
+ np.float32,0x7e0d80db,0x42fa4a15,3
746
+ np.float32,0x4f0b9d,0xc2fd640a,3
747
+ np.float32,0x7e083806,0x42fa2df8,3
748
+ np.float32,0x7f77f8c6,0x42ffe877,3
749
+ np.float32,0x3e7bb46a,0xc0018ff5,3
750
+ np.float32,0x3f06eb2e,0xbf6c8eca,3
751
+ np.float32,0x7eae8f7c,0x42fce52a,3
752
+ np.float32,0x3de481a0,0xc04a7d7f,3
753
+ np.float32,0x3eed4311,0xbf8e096f,3
754
+ np.float32,0x3f7b0300,0xbce8903d,3
755
+ np.float32,0x3811b,0xc30330dd,3
756
+ np.float32,0x3eb6f8e1,0xbfbe04bc,3
757
+ np.float32,0x3ec35210,0xbfb1f55a,3
758
+ np.float32,0x3d386916,0xc08f24a5,3
759
+ np.float32,0x3f1fa197,0xbf2e704d,3
760
+ np.float32,0x7f2020a5,0x42fea56a,3
761
+ np.float32,0x7e1ea53f,0x42fa9e8c,3
762
+ np.float32,0x3f148903,0xbf490bf9,3
763
+ np.float32,0x3f2f56a0,0xbf0bc6c9,3
764
+ np.float32,0x7da9fc,0xc2fc0d9b,3
765
+ np.float32,0x3d802134,0xc07fe810,3
766
+ np.float32,0x3f6cb927,0xbde74e57,3
767
+ np.float32,0x7e05b125,0x42fa2023,3
768
+ np.float32,0x3f3307f9,0xbf041433,3
769
+ np.float32,0x5666bf,0xc2fd2250,3
770
+ np.float32,0x3f51c93b,0xbe930f28,3
771
+ np.float32,0x3eb5dcfe,0xbfbf241e,3
772
+ np.float32,0xb2773,0xc301853f,3
773
+ np.float32,0x7f4dee96,0x42ff5f3f,3
774
+ np.float32,0x3e3f5c33,0xc01adee1,3
775
+ np.float32,0x3f2ed29a,0xbf0cdd4a,3
776
+ np.float32,0x3e3c01ef,0xc01c80ab,3
777
+ np.float32,0x3ec2236e,0xbfb31458,3
778
+ np.float32,0x7e841dc4,0x42fc1761,3
779
+ np.float32,0x3df2cd8e,0xc044e30c,3
780
+ np.float32,0x3f010901,0xbf7d0670,3
781
+ np.float32,0x3c05ceaa,0xc0ddf39b,3
782
+ np.float32,0x3f517226,0xbe944206,3
783
+ np.float32,0x3f23c83d,0xbf24f522,3
784
+ np.float32,0x7fc9da,0xc2fc0139,3
785
+ np.float32,0x7f1bde53,0x42fe9181,3
786
+ np.float32,0x3ea3786c,0xbfd2d4a5,3
787
+ np.float32,0x3e83a71b,0xbffacdd2,3
788
+ np.float32,0x3f6f0d4f,0xbdca61d5,3
789
+ np.float32,0x7f5ab613,0x42ff8bb7,3
790
+ np.float32,0x3ab1ec,0xc2fe3fea,3
791
+ np.float32,0x4fbf58,0xc2fd5d82,3
792
+ np.float32,0x3dea141b,0xc0484403,3
793
+ np.float32,0x7d86ad3b,0x42f8258f,3
794
+ np.float32,0x7f345315,0x42fefd29,3
795
+ np.float32,0x3f3752fe,0xbef6a780,3
796
+ np.float32,0x64830d,0xc2fcb293,3
797
+ np.float32,0x3d9dc1eb,0xc06cb32a,3
798
+ np.float32,0x3f2f935a,0xbf0b46f6,3
799
+ np.float32,0xb90a4,0xc30177e3,3
800
+ np.float32,0x4111dd,0xc2fdf3c1,3
801
+ np.float32,0x3d4cd078,0xc08a4c68,3
802
+ np.float32,0x3e95c3f1,0xbfe30011,3
803
+ np.float32,0x3ec9f356,0xbfabcb4e,3
804
+ np.float32,0x1b90d5,0xc3003717,3
805
+ np.float32,0xee70f,0xc3011a3e,3
806
+ np.float32,0x7fa00000,0x7fe00000,3
807
+ np.float32,0x3f74cdb6,0xbd8422af,3
808
+ np.float32,0x3d9b56fe,0xc06e2037,3
809
+ np.float32,0x3f1853df,0xbf3fbc40,3
810
+ np.float32,0x7d86a011,0x42f82547,3
811
+ np.float32,0x3dff9629,0xc0402634,3
812
+ np.float32,0x46f8c9,0xc2fdb39f,3
813
+ np.float32,0x3e9b410b,0xbfdc5a87,3
814
+ np.float32,0x3f5aed42,0xbe671cac,3
815
+ np.float32,0x3b739886,0xc101257f,3
816
+ np.float64,0x3fe2f58d6565eb1b,0xbfe82a641138e19a,2
817
+ np.float64,0x3fee7f0642fcfe0d,0xbfb1c702f6974932,2
818
+ np.float64,0x25b71f244b6e5,0xc090030d3b3c5d2b,2
819
+ np.float64,0x8c9cc8e1193b,0xc0900b752a678fa8,2
820
+ np.float64,0x3fd329b5d326536c,0xbffbd607f6db945c,2
821
+ np.float64,0x3fb5109b3a2a2136,0xc00cd36bd15dfb18,2
822
+ np.float64,0x3fd5393ae12a7276,0xbff97a7e4a157154,2
823
+ np.float64,0x3fd374d1b926e9a3,0xbffb7c3e1a3a7ed3,2
824
+ np.float64,0x3fe2c7f4e2658fea,0xbfe899f15ca78fcb,2
825
+ np.float64,0x7fe3d6b81ee7ad6f,0x408ffa7b63d407ee,2
826
+ np.float64,0x3fe086d097e10da1,0xbfee81456ce8dd03,2
827
+ np.float64,0x7fd374a64ca6e94c,0x408ff241c7306d39,2
828
+ np.float64,0x3fc0709a5b20e135,0xc007afdede31b29c,2
829
+ np.float64,0x3fd4218f4b28431f,0xbffab2c696966e2d,2
830
+ np.float64,0x143134c828628,0xc09006a8372c4d8a,2
831
+ np.float64,0x3f8bd0aa0037a154,0xc018cf0e8b9c3107,2
832
+ np.float64,0x7fe0ce905ee19d20,0x408ff8915e71bd67,2
833
+ np.float64,0x3fda0f5f32b41ebe,0xbff4bd5e0869e820,2
834
+ np.float64,0x7fe9ae63d0b35cc7,0x408ffd760ca4f292,2
835
+ np.float64,0x3fe75abd9eeeb57b,0xbfdd1476fc8b3089,2
836
+ np.float64,0x786c3110f0d87,0xc08ff8b44cedbeea,2
837
+ np.float64,0x22c5fe80458d,0xc09013853591c2f2,2
838
+ np.float64,0x3fdc250797384a0f,0xbff2f6a02c961f0b,2
839
+ np.float64,0x3fa2b367b02566cf,0xc013199238485054,2
840
+ np.float64,0x3fd26a910ca4d522,0xbffcc0e2089b1c0c,2
841
+ np.float64,0x8068d3b300d1b,0xc08ff7f690210aac,2
842
+ np.float64,0x3fe663bfa9ecc77f,0xbfe07cd95a43a5ce,2
843
+ np.float64,0x3fd0ddb07321bb61,0xbffec886665e895e,2
844
+ np.float64,0x3f91c730b0238e61,0xc0176452badc8d22,2
845
+ np.float64,0x4dd10d309ba22,0xc08ffdbe738b1d8d,2
846
+ np.float64,0x7fe322afa4a6455e,0x408ffa10c038f9de,2
847
+ np.float64,0x7fdf7f7c42befef8,0x408ff7d147ddaad5,2
848
+ np.float64,0x7fd673f386ace7e6,0x408ff3e920d00eef,2
849
+ np.float64,0x3feaebfcadb5d7f9,0xbfcfe8ec27083478,2
850
+ np.float64,0x3fdc6dc23738db84,0xbff2bb46794f07b8,2
851
+ np.float64,0xcd8819599b103,0xc08ff288c5b2cf0f,2
852
+ np.float64,0xfda00e77fb402,0xc08ff01b895d2236,2
853
+ np.float64,0x840b02ff08161,0xc08ff7a41e41114c,2
854
+ np.float64,0x3fbdce3a383b9c74,0xc008d1e61903a289,2
855
+ np.float64,0x3fd24ed3c4a49da8,0xbffce3c12136b6d3,2
856
+ np.float64,0x3fe8d0834131a107,0xbfd77b194e7051d4,2
857
+ np.float64,0x3fdd0cb11aba1962,0xbff23b9dbd554455,2
858
+ np.float64,0x1a32d97e3465c,0xc090052781a37271,2
859
+ np.float64,0x3fdb09d2b1b613a5,0xbff3e396b862bd83,2
860
+ np.float64,0x3fe04c848aa09909,0xbfef2540dd90103a,2
861
+ np.float64,0x3fce0c48613c1891,0xc000b9f76877d744,2
862
+ np.float64,0x3fc37109a226e213,0xc005c05d8b2b9a2f,2
863
+ np.float64,0x81cf3837039e7,0xc08ff7d686517dff,2
864
+ np.float64,0xd9342c29b2686,0xc08ff1e591c9a895,2
865
+ np.float64,0x7fec731b0638e635,0x408ffea4884550a9,2
866
+ np.float64,0x3fba0fc138341f82,0xc00a5e839b085f64,2
867
+ np.float64,0x7fdda893b03b5126,0x408ff71f7c5a2797,2
868
+ np.float64,0xd2a4bb03a5498,0xc08ff2402f7a907c,2
869
+ np.float64,0x3fea61fb0d34c3f6,0xbfd1d293fbe76183,2
870
+ np.float64,0x3fed5cf486fab9e9,0xbfbfc2e01a7ffff1,2
871
+ np.float64,0x3fcbabc2bf375785,0xc001ad7750c9dbdf,2
872
+ np.float64,0x3fdb5fff53b6bfff,0xbff39a7973a0c6a5,2
873
+ np.float64,0x7feef05a00bde0b3,0x408fff9c5cbc8651,2
874
+ np.float64,0xb1cf24f1639e5,0xc08ff434de10fffb,2
875
+ np.float64,0x3fa583989c2b0731,0xc0124a8a3bbf18ce,2
876
+ np.float64,0x7feae90bf9f5d217,0x408ffe002e7bbbea,2
877
+ np.float64,0x3fe9ef41c4b3de84,0xbfd367878ae4528e,2
878
+ np.float64,0x9be24ce337c4a,0xc08ff5b9b1c31cf9,2
879
+ np.float64,0x3fe916894cb22d13,0xbfd677f915d58503,2
880
+ np.float64,0x3fec1bab20f83756,0xbfc7f2777aabe8ee,2
881
+ np.float64,0x3feaabf2873557e5,0xbfd0d11f28341233,2
882
+ np.float64,0x3fd4d3c3b529a787,0xbff9e9e47acc8ca9,2
883
+ np.float64,0x3fe4cfe96c699fd3,0xbfe3dc53fa739169,2
884
+ np.float64,0xccfdb97399fb7,0xc08ff2908d893400,2
885
+ np.float64,0x3fec7598be78eb31,0xbfc5a750f8f3441a,2
886
+ np.float64,0x355be5fc6ab7e,0xc090010ca315b50b,2
887
+ np.float64,0x3fba9f9074353f21,0xc00a1f80eaf5e581,2
888
+ np.float64,0x7fdcaff189395fe2,0x408ff6bd1c5b90d9,2
889
+ np.float64,0x3fd94d3b64b29a77,0xbff56be1b43d25f3,2
890
+ np.float64,0x4e5f29949cbe6,0xc08ffda972da1d73,2
891
+ np.float64,0x3fe654e2d9aca9c6,0xbfe09b88dcd8f15d,2
892
+ np.float64,0x7fdc130190b82602,0x408ff67d496c1a27,2
893
+ np.float64,0x3fbcd4701e39a8e0,0xc009343e36627e80,2
894
+ np.float64,0x7fdaa4d38f3549a6,0x408ff5e2c6d8678f,2
895
+ np.float64,0x3febe95e5237d2bd,0xbfc93e16d453fe3a,2
896
+ np.float64,0x9ef5ca553deba,0xc08ff57ff4f7883d,2
897
+ np.float64,0x7fe878e91170f1d1,0x408ffce795868fc8,2
898
+ np.float64,0x3fe63dff466c7bff,0xbfe0caf2b79c9e5f,2
899
+ np.float64,0x6561446ccac29,0xc08ffab0e383834c,2
900
+ np.float64,0x30c6c2ae618d9,0xc09001914b30381b,2
901
+ np.float64,0x7ff0000000000000,0x7ff0000000000000,2
902
+ np.float64,0x3fe5c9daf1ab93b6,0xbfe1be81baf4dbdb,2
903
+ np.float64,0x3fe0a03e24a1407c,0xbfee3a73c4c0e8f8,2
904
+ np.float64,0xff2a2cf3fe546,0xc08ff009a7e6e782,2
905
+ np.float64,0x7fcf0332213e0663,0x408fefa36235e210,2
906
+ np.float64,0x3fb612affc2c2560,0xc00c494be9c8c33b,2
907
+ np.float64,0x3fd2b259702564b3,0xbffc67967f077e75,2
908
+ np.float64,0x7fcb63685d36c6d0,0x408fee343343f913,2
909
+ np.float64,0x3fe369f1d5a6d3e4,0xbfe71251139939ad,2
910
+ np.float64,0x3fdd17c618ba2f8c,0xbff232d11c986251,2
911
+ np.float64,0x3f92cc8040259901,0xc01711d8e06b52ee,2
912
+ np.float64,0x69a81dc2d3504,0xc08ffa36cdaf1141,2
913
+ np.float64,0x3fea0fad99b41f5b,0xbfd2f4625a652645,2
914
+ np.float64,0xd1cd5799a39ab,0xc08ff24c02b90d26,2
915
+ np.float64,0x324e59ce649cc,0xc0900163ad091c76,2
916
+ np.float64,0x3fc3d460a227a8c1,0xc00585f903dc7a7f,2
917
+ np.float64,0xa7185ec74e30c,0xc08ff4ec7d65ccd9,2
918
+ np.float64,0x3fa254eaac24a9d5,0xc01337053963321a,2
919
+ np.float64,0x3feaeb112435d622,0xbfcfef3be17f81f6,2
920
+ np.float64,0x60144c3ac028a,0xc08ffb4f8eb94595,2
921
+ np.float64,0x7fa4d2ec6829a5d8,0x408fdb0a9670ab83,2
922
+ np.float64,0x3fed1372f97a26e6,0xbfc1b1fe50d48a55,2
923
+ np.float64,0x3fd5ade5972b5bcb,0xbff8fcf28f525031,2
924
+ np.float64,0x7fe72e335bee5c66,0x408ffc4759236437,2
925
+ np.float64,0x7fdfafab143f5f55,0x408ff7e2e22a8129,2
926
+ np.float64,0x3fe90d0db9321a1b,0xbfd69ae5fe10eb9e,2
927
+ np.float64,0x7fe20a59072414b1,0x408ff962a2492484,2
928
+ np.float64,0x3fed853690bb0a6d,0xbfbdc9dc5f199d2b,2
929
+ np.float64,0x3fd709d469ae13a9,0xbff795a218deb700,2
930
+ np.float64,0x3fe21c35f5e4386c,0xbfea47d71789329b,2
931
+ np.float64,0x9ea5ec053d4be,0xc08ff585c2f6b7a3,2
932
+ np.float64,0x3fc0580f9e20b01f,0xc007c1268f49d037,2
933
+ np.float64,0xd99127abb3225,0xc08ff1e0a1ff339d,2
934
+ np.float64,0x3fdc8c9bbfb91937,0xbff2a2478354effb,2
935
+ np.float64,0x3fe15fc6b162bf8d,0xbfec323ac358e008,2
936
+ np.float64,0xffefffffffffffff,0x7ff8000000000000,2
937
+ np.float64,0x3fee341afb3c6836,0xbfb556b6faee9a84,2
938
+ np.float64,0x3fe4b64c56296c99,0xbfe4154835ad2afe,2
939
+ np.float64,0x85de22810bbc5,0xc08ff77b914fe5b5,2
940
+ np.float64,0x3fd22c72e3a458e6,0xbffd0f4269d20bb9,2
941
+ np.float64,0xc090e5218123,0xc09009a4a65a8a8f,2
942
+ np.float64,0x7fd9641692b2c82c,0x408ff5547782bdfc,2
943
+ np.float64,0x3fd9b9cb28b37396,0xbff509a8fb59a9f1,2
944
+ np.float64,0x3fcd2726f93a4e4e,0xc001135059a22117,2
945
+ np.float64,0x3fa4b493d4296928,0xc0128323c7a55f4a,2
946
+ np.float64,0x47455e788e8ac,0xc08ffec2101c1e82,2
947
+ np.float64,0x3fe0d7e2e261afc6,0xbfeda0f1e2d0f4bd,2
948
+ np.float64,0x3fe860fc5b70c1f9,0xbfd91dc42eaf72c2,2
949
+ np.float64,0xa5d7805b4baf0,0xc08ff502bc819ff6,2
950
+ np.float64,0xd83395b1b0673,0xc08ff1f33c3f94c2,2
951
+ np.float64,0x3f865972e02cb2e6,0xc01a1243651565c8,2
952
+ np.float64,0x52fc6952a5f8e,0xc08ffd006b158179,2
953
+ np.float64,0x7fecac6c793958d8,0x408ffebbb1c09a70,2
954
+ np.float64,0x7fe621ff606c43fe,0x408ffbbeb2b1473a,2
955
+ np.float64,0x3fdb9f3f9db73e7f,0xbff365610c52bda7,2
956
+ np.float64,0x7feab92992757252,0x408ffdeb92a04813,2
957
+ np.float64,0xcc46c79f988d9,0xc08ff29adf03fb7c,2
958
+ np.float64,0x3fe3156a03262ad4,0xbfe7dd0f598781c7,2
959
+ np.float64,0x3fc00e3a61201c75,0xc007f5c121a87302,2
960
+ np.float64,0x3fdce8e9f739d1d4,0xbff2581d41ef50ef,2
961
+ np.float64,0x0,0xfff0000000000000,2
962
+ np.float64,0x7d373ac4fa6e8,0xc08ff840fa8beaec,2
963
+ np.float64,0x3fee41e0653c83c1,0xbfb4ae786f2a0d54,2
964
+ np.float64,0x3ff0000000000000,0x0,2
965
+ np.float64,0x7feca6fff9794dff,0x408ffeb982a70556,2
966
+ np.float64,0x7fc532716d2a64e2,0x408feb3f0f6c095b,2
967
+ np.float64,0x3fe4ec2954a9d853,0xbfe39dd44aa5a040,2
968
+ np.float64,0x7fd3321d52a6643a,0x408ff21a0ab9cd85,2
969
+ np.float64,0x7fd8f1b2dfb1e365,0x408ff52001fa7922,2
970
+ np.float64,0x3fee5e58cabcbcb2,0xbfb3539734a24d8b,2
971
+ np.float64,0x3feebf6e7dfd7edd,0xbfad7c648f025102,2
972
+ np.float64,0x6008026ec0101,0xc08ffb5108b54a93,2
973
+ np.float64,0x3fea06f5e2340dec,0xbfd3134a48283360,2
974
+ np.float64,0x41cad13c8395b,0xc08fffae654b2426,2
975
+ np.float64,0x7fedb5c9353b6b91,0x408fff249f1f32b6,2
976
+ np.float64,0xe00c5af9c018c,0xc08ff189e68c655f,2
977
+ np.float64,0x7feac398ddf58731,0x408ffdf01374de9f,2
978
+ np.float64,0x3fed21127c7a4225,0xbfc15b8cf55628fa,2
979
+ np.float64,0x3fd3446711a688ce,0xbffbb5f7252a9fa3,2
980
+ np.float64,0x7fe75fa07a6ebf40,0x408ffc5fdb096018,2
981
+ np.float64,0x3feeb1618cbd62c3,0xbfaece3bd0863070,2
982
+ np.float64,0x7f5226e180244dc2,0x408fb174d506e52f,2
983
+ np.float64,0x3fcd67deca3acfbe,0xc000f9cd7a490749,2
984
+ np.float64,0xdc6f30efb8de6,0xc08ff1b9f2a22d2e,2
985
+ np.float64,0x9c14931338293,0xc08ff5b5f975ec5d,2
986
+ np.float64,0x7fe93e802df27cff,0x408ffd4354eba0e0,2
987
+ np.float64,0x3feb92ae5077255d,0xbfcb7f2084e44dbb,2
988
+ np.float64,0xd78dbfddaf1b8,0xc08ff1fc19fa5a13,2
989
+ np.float64,0x7fe14c301fa2985f,0x408ff8e666cb6592,2
990
+ np.float64,0xbda3d8b77b47b,0xc08ff37689f4b2e5,2
991
+ np.float64,0x8a42953b14853,0xc08ff71c2db3b8cf,2
992
+ np.float64,0x7fe4ca7e186994fb,0x408ffb05e94254a7,2
993
+ np.float64,0x7fe92ffc5e325ff8,0x408ffd3cb0265b12,2
994
+ np.float64,0x91b262912364d,0xc08ff681619be214,2
995
+ np.float64,0x33fe2b0667fc6,0xc0900132f3fab55e,2
996
+ np.float64,0x3fde10e9183c21d2,0xbff17060fb4416c7,2
997
+ np.float64,0xb6b811cb6d702,0xc08ff3e46303b541,2
998
+ np.float64,0x3fe4a7bda0a94f7b,0xbfe435c6481cd0e3,2
999
+ np.float64,0x7fd9fe6057b3fcc0,0x408ff599c79a822c,2
1000
+ np.float64,0x3fef44bf917e897f,0xbfa11484e351a6e9,2
1001
+ np.float64,0x3fe57d701daafae0,0xbfe2618ab40fc01b,2
1002
+ np.float64,0x7fe52d2adbaa5a55,0x408ffb3c2fb1c99d,2
1003
+ np.float64,0xb432f66d6865f,0xc08ff40d6b4084fe,2
1004
+ np.float64,0xbff0000000000000,0x7ff8000000000000,2
1005
+ np.float64,0x7fecd2292bf9a451,0x408ffecad860de6f,2
1006
+ np.float64,0x3fddd2ae153ba55c,0xbff1a059adaca33e,2
1007
+ np.float64,0x3fee55d6e5bcabae,0xbfb3bb1c6179d820,2
1008
+ np.float64,0x7fc1d0085623a010,0x408fe93d16ada7a7,2
1009
+ np.float64,0x829b000105360,0xc08ff7c47629a68f,2
1010
+ np.float64,0x7fe1e0257523c04a,0x408ff94782cf0717,2
1011
+ np.float64,0x7fd652f9ad2ca5f2,0x408ff3d820ec892e,2
1012
+ np.float64,0x3fef2246203e448c,0xbfa444ab6209d8cd,2
1013
+ np.float64,0x3fec6c0ae178d816,0xbfc5e559ebd4e790,2
1014
+ np.float64,0x3fe6ddfee92dbbfe,0xbfdf06dd7d3fa7a8,2
1015
+ np.float64,0x3fb7fbcbea2ff798,0xc00b5404d859d148,2
1016
+ np.float64,0x7feb9a154d37342a,0x408ffe4b26c29e55,2
1017
+ np.float64,0x3fe4db717aa9b6e3,0xbfe3c2c6b3ef13bc,2
1018
+ np.float64,0x3fbae17dda35c2fc,0xc00a030f7f4b37e7,2
1019
+ np.float64,0x7fd632b9082c6571,0x408ff3c76826ef19,2
1020
+ np.float64,0x7fc4184a15283093,0x408feaa14adf00be,2
1021
+ np.float64,0x3fe052d19920a5a3,0xbfef136b5df81a3e,2
1022
+ np.float64,0x7fe38b872b67170d,0x408ffa4f51aafc86,2
1023
+ np.float64,0x3fef9842d03f3086,0xbf92d3d2a21d4be2,2
1024
+ np.float64,0x9cea662139d4d,0xc08ff5a634810daa,2
1025
+ np.float64,0x3fe35f0855e6be11,0xbfe72c4b564e62aa,2
1026
+ np.float64,0x3fecee3d3779dc7a,0xbfc29ee942f8729e,2
1027
+ np.float64,0x3fe7903fd72f2080,0xbfdc41db9b5f4048,2
1028
+ np.float64,0xb958889572b11,0xc08ff3ba366cf84b,2
1029
+ np.float64,0x3fcb3a67c53674d0,0xc001dd21081ad1ea,2
1030
+ np.float64,0xe3b1b53fc7637,0xc08ff15a3505e1ce,2
1031
+ np.float64,0xe5954ae9cb2aa,0xc08ff141cbbf0ae4,2
1032
+ np.float64,0x3fe394af74e7295f,0xbfe6ad1d13f206e8,2
1033
+ np.float64,0x7fe21dd704643bad,0x408ff96f13f80c1a,2
1034
+ np.float64,0x3fd23a7cf02474fa,0xbffcfd7454117a05,2
1035
+ np.float64,0x7fe257515e24aea2,0x408ff99378764d52,2
1036
+ np.float64,0x7fe4c5d0a6e98ba0,0x408ffb03503cf939,2
1037
+ np.float64,0x3fadc2c1603b8583,0xc0106b2c17550e3a,2
1038
+ np.float64,0x3fc0f7f02421efe0,0xc007525ac446864c,2
1039
+ np.float64,0x3feaf0b27275e165,0xbfcfc8a03eaa32ad,2
1040
+ np.float64,0x5ce7503cb9ceb,0xc08ffbb2de365fa8,2
1041
+ np.float64,0x2a0014f654003,0xc090026e41761a0d,2
1042
+ np.float64,0x7fe2c848a8e59090,0x408ff9d9b723ee89,2
1043
+ np.float64,0x7f66f54bc02dea97,0x408fbc2ae0ec5623,2
1044
+ np.float64,0xa35a890146b6,0xc0900a97b358ddbd,2
1045
+ np.float64,0x7fee267ded7c4cfb,0x408fff501560c9f5,2
1046
+ np.float64,0x3fe07c328520f865,0xbfee9ef7c3435b58,2
1047
+ np.float64,0x3fe67122cf6ce246,0xbfe06147001932ba,2
1048
+ np.float64,0x3fdacc8925359912,0xbff41824cece219e,2
1049
+ np.float64,0xffa3047fff461,0xc08ff00431ec9be3,2
1050
+ np.float64,0x3e1af43e7c35f,0xc090002c6573d29b,2
1051
+ np.float64,0x86fa94590df53,0xc08ff7632525ed92,2
1052
+ np.float64,0x7fec4c76227898eb,0x408ffe94d032c657,2
1053
+ np.float64,0x7fe2274ce1e44e99,0x408ff975194cfdff,2
1054
+ np.float64,0x7fe670e1b4ace1c2,0x408ffbe78cc451de,2
1055
+ np.float64,0x7fe853871db0a70d,0x408ffcd5e6a6ff47,2
1056
+ np.float64,0x3fcbf265db37e4cc,0xc0019026336e1176,2
1057
+ np.float64,0x3fef033cef3e067a,0xbfa726712eaae7f0,2
1058
+ np.float64,0x5d74973abae94,0xc08ffba15e6bb992,2
1059
+ np.float64,0x7fdd9c99b6bb3932,0x408ff71ad24a7ae0,2
1060
+ np.float64,0xbdc8e09b7b91c,0xc08ff3744939e9a3,2
1061
+ np.float64,0xdbfcff71b7fa0,0xc08ff1bfeecc9dfb,2
1062
+ np.float64,0xf9b38cf5f3672,0xc08ff0499af34a43,2
1063
+ np.float64,0x3fea820aa6b50415,0xbfd162a38e1927b1,2
1064
+ np.float64,0x3fe67f59a12cfeb3,0xbfe04412adca49dc,2
1065
+ np.float64,0x3feb301d9c76603b,0xbfce17e6edeb92d5,2
1066
+ np.float64,0x828ce00b0519c,0xc08ff7c5b5c57cde,2
1067
+ np.float64,0x4f935e229f26c,0xc08ffd7c67c1c54f,2
1068
+ np.float64,0x7fcd139e023a273b,0x408feee4f12ff11e,2
1069
+ np.float64,0x666a9944ccd54,0xc08ffa92d5e5cd64,2
1070
+ np.float64,0x3fe792f0fa6f25e2,0xbfdc374fda28f470,2
1071
+ np.float64,0xe996029bd32c1,0xc08ff10eb9b47a11,2
1072
+ np.float64,0x3fe7b0dd1eef61ba,0xbfdbc2676dc77db0,2
1073
+ np.float64,0x7fd3ec0127a7d801,0x408ff287bf47e27d,2
1074
+ np.float64,0x3fe793a8ea6f2752,0xbfdc347f7717e48d,2
1075
+ np.float64,0x7fdb89d15e3713a2,0x408ff64457a13ea2,2
1076
+ np.float64,0x3fe35b3cbbe6b679,0xbfe73557c8321b70,2
1077
+ np.float64,0x66573c94ccae8,0xc08ffa9504af7eb5,2
1078
+ np.float64,0x3fc620a2302c4144,0xc00442036b944a67,2
1079
+ np.float64,0x49b2fe0693660,0xc08ffe5f131c3c7e,2
1080
+ np.float64,0x7fda936cdfb526d9,0x408ff5db3ab3f701,2
1081
+ np.float64,0xc774ceef8ee9a,0xc08ff2e16d082fa1,2
1082
+ np.float64,0x4da9f8a09b55,0xc0900ee2206d0c88,2
1083
+ np.float64,0x3fe2ca5d5ae594bb,0xbfe89406611a5f1a,2
1084
+ np.float64,0x7fe0832497e10648,0x408ff85d1de6056e,2
1085
+ np.float64,0x3fe6a9e3222d53c6,0xbfdfda35a9bc2de1,2
1086
+ np.float64,0x3fed3d92c8ba7b26,0xbfc0a73620db8b98,2
1087
+ np.float64,0x3fdd2ec093ba5d81,0xbff2209cf78ce3f1,2
1088
+ np.float64,0x62fcb968c5f98,0xc08ffaf775a593c7,2
1089
+ np.float64,0xfcfb019ff9f60,0xc08ff0230e95bd16,2
1090
+ np.float64,0x3fd7a63e8f2f4c7d,0xbff6faf4fff7dbe0,2
1091
+ np.float64,0x3fef23b0ec3e4762,0xbfa4230cb176f917,2
1092
+ np.float64,0x340d1e6a681a5,0xc09001314b68a0a2,2
1093
+ np.float64,0x7fc0b85ba02170b6,0x408fe8821487b802,2
1094
+ np.float64,0x7fe9976e84f32edc,0x408ffd6bb6aaf467,2
1095
+ np.float64,0x329a0e9e65343,0xc090015b044e3270,2
1096
+ np.float64,0x3fea4928d3f49252,0xbfd2299b05546eab,2
1097
+ np.float64,0x3f188c70003118e0,0xc02ac3ce23bc5d5a,2
1098
+ np.float64,0x3fecce5020b99ca0,0xbfc36b23153d5f50,2
1099
+ np.float64,0x3fe203873e24070e,0xbfea86edb3690830,2
1100
+ np.float64,0x3fe02d9eaa205b3d,0xbfef7d18c54a76d2,2
1101
+ np.float64,0xef7537ebdeea7,0xc08ff0c55e9d89e7,2
1102
+ np.float64,0x3fedf7572efbeeae,0xbfb840af357cf07c,2
1103
+ np.float64,0xd1a97a61a354,0xc0900926fdfb96cc,2
1104
+ np.float64,0x7fe6a0daeced41b5,0x408ffc001edf1407,2
1105
+ np.float64,0x3fe5063625aa0c6c,0xbfe3647cfb949d62,2
1106
+ np.float64,0x7fe9b28d31736519,0x408ffd77eb4a922b,2
1107
+ np.float64,0x7feea90d033d5219,0x408fff81a4bbff62,2
1108
+ np.float64,0x3fe9494d17f2929a,0xbfd5bde02eb5287a,2
1109
+ np.float64,0x7feee17a8cbdc2f4,0x408fff96cf0dc16a,2
1110
+ np.float64,0xb2ad18ef655a3,0xc08ff4267eda8af8,2
1111
+ np.float64,0x3fad3b52683a76a5,0xc01085ab75b797ce,2
1112
+ np.float64,0x2300a65846016,0xc090037b81ce9500,2
1113
+ np.float64,0x3feb1041f9b62084,0xbfcef0c87d8b3249,2
1114
+ np.float64,0x3fdd887d3e3b10fa,0xbff1da0e1ede6db2,2
1115
+ np.float64,0x3fd3e410eb27c822,0xbffaf9b5fc9cc8cc,2
1116
+ np.float64,0x3fe0aa53e3e154a8,0xbfee1e7b5c486578,2
1117
+ np.float64,0x7fe33e389aa67c70,0x408ffa214fe50961,2
1118
+ np.float64,0x3fd27e3a43a4fc75,0xbffca84a79e8adeb,2
1119
+ np.float64,0x3fb309e0082613c0,0xc00dfe407b77a508,2
1120
+ np.float64,0x7feaf2ed8cf5e5da,0x408ffe046a9d1ba9,2
1121
+ np.float64,0x1e76167a3cec4,0xc0900448cd35ec67,2
1122
+ np.float64,0x3fe0a18e1721431c,0xbfee36cf1165a0d4,2
1123
+ np.float64,0x3fa73b78c02e76f2,0xc011d9069823b172,2
1124
+ np.float64,0x3fef6d48287eda90,0xbf9ab2d08722c101,2
1125
+ np.float64,0x8fdf0da31fbe2,0xc08ff6a6a2accaa1,2
1126
+ np.float64,0x3fc3638db826c71b,0xc005c86191688826,2
1127
+ np.float64,0xaa9c09c555381,0xc08ff4aefe1d9473,2
1128
+ np.float64,0x7fccb0f4523961e8,0x408feebd84773f23,2
1129
+ np.float64,0xede75dcfdbcec,0xc08ff0d89ba887d1,2
1130
+ np.float64,0x7f8a051520340a29,0x408fcd9cc17f0d95,2
1131
+ np.float64,0x3fef5ca2babeb945,0xbf9dc221f3618e6a,2
1132
+ np.float64,0x7fea0ff4bcf41fe8,0x408ffda193359f22,2
1133
+ np.float64,0x7fe05c53fd20b8a7,0x408ff841dc7123e8,2
1134
+ np.float64,0x3fc625664b2c4acd,0xc0043f8749b9a1d8,2
1135
+ np.float64,0x7fed58f98f7ab1f2,0x408fff00585f48c2,2
1136
+ np.float64,0x3fb3e5e51427cbca,0xc00d7bcb6528cafe,2
1137
+ np.float64,0x3fe728bd3d6e517a,0xbfdddafa72bd0f60,2
1138
+ np.float64,0x3fe3f005dd27e00c,0xbfe5d7b3ec93bca0,2
1139
+ np.float64,0x3fd74fbd1a2e9f7a,0xbff750001b63ce81,2
1140
+ np.float64,0x3fd3af6d85a75edb,0xbffb371d678d11b4,2
1141
+ np.float64,0x7fa690ad8c2d215a,0x408fdbf7db9c7640,2
1142
+ np.float64,0x3fbdfd38e23bfa72,0xc008bfc1c5c9b89e,2
1143
+ np.float64,0x3fe2374684a46e8d,0xbfea030c4595dfba,2
1144
+ np.float64,0x7fc0806c372100d7,0x408fe85b36fee334,2
1145
+ np.float64,0x3fef3ac47b7e7589,0xbfa2007195c5213f,2
1146
+ np.float64,0x3fb55473922aa8e7,0xc00cae7af8230e0c,2
1147
+ np.float64,0x7fe018dc152031b7,0x408ff811e0d712fa,2
1148
+ np.float64,0x3fe3b3fca56767f9,0xbfe6638ae2c99c62,2
1149
+ np.float64,0x7fac79818c38f302,0x408fdea720b39c3c,2
1150
+ np.float64,0x7fefffffffffffff,0x4090000000000000,2
1151
+ np.float64,0xd2b290cba5652,0xc08ff23f6d7152a6,2
1152
+ np.float64,0x7fc5848eb52b091c,0x408feb6b6f8b77d0,2
1153
+ np.float64,0xf399f62de733f,0xc08ff092ae319ad8,2
1154
+ np.float64,0x7fdec56c12bd8ad7,0x408ff78c4ddbc667,2
1155
+ np.float64,0x3fca640f1e34c81e,0xc0023969c5cbfa4c,2
1156
+ np.float64,0x3fd55225db2aa44c,0xbff95f7442a2189e,2
1157
+ np.float64,0x7fefa009a97f4012,0x408fffdd2f42ef9f,2
1158
+ np.float64,0x4a3b70609478,0xc0900f24e449bc3d,2
1159
+ np.float64,0x7fe3738b1ba6e715,0x408ffa411f2cb5e7,2
1160
+ np.float64,0x7fe5e53f0b6bca7d,0x408ffb9ed8d95cea,2
1161
+ np.float64,0x3fe274dd24a4e9ba,0xbfe967fb114b2a83,2
1162
+ np.float64,0x3fcbc58b8c378b17,0xc001a2bb1e158bcc,2
1163
+ np.float64,0x3fefc2c0043f8580,0xbf862c9b464dcf38,2
1164
+ np.float64,0xc2c4fafd858a0,0xc08ff327aecc409b,2
1165
+ np.float64,0x3fd8bc39a9b17873,0xbff5f1ad46e5a51c,2
1166
+ np.float64,0x3fdf341656be682d,0xbff094f41e7cb4c4,2
1167
+ np.float64,0x3fef8495c13f092c,0xbf966cf6313bae4c,2
1168
+ np.float64,0x3fe14e0f05229c1e,0xbfec6166f26b7161,2
1169
+ np.float64,0x3fed42d3b2ba85a7,0xbfc0860b773d35d8,2
1170
+ np.float64,0x7fd92bbac5b25775,0x408ff53abcb3fe0c,2
1171
+ np.float64,0xb1635b6f62c6c,0xc08ff43bdf47accf,2
1172
+ np.float64,0x4a3a2dbc94746,0xc08ffe49fabddb36,2
1173
+ np.float64,0x87d831290fb06,0xc08ff750419dc6fb,2
1174
+ np.float64,0x3fec4713f7f88e28,0xbfc6d6217c9f5cf9,2
1175
+ np.float64,0x7fed43ba2d3a8773,0x408ffef7fa2fc303,2
1176
+ np.float64,0x7fd1ec5b56a3d8b6,0x408ff14f62615f1e,2
1177
+ np.float64,0x3fee534b6c7ca697,0xbfb3da1951aa3e68,2
1178
+ np.float64,0x3febb564c2b76aca,0xbfca9737062e55e7,2
1179
+ np.float64,0x943e6b0f287ce,0xc08ff64e2d09335c,2
1180
+ np.float64,0xf177d957e2efb,0xc08ff0acab2999fa,2
1181
+ np.float64,0x7fb5b881a82b7102,0x408fe3872b4fde5e,2
1182
+ np.float64,0x3fdb2b4a97b65695,0xbff3c715c91359bc,2
1183
+ np.float64,0x3fac0a17e4381430,0xc010c330967309fb,2
1184
+ np.float64,0x7fd8057990b00af2,0x408ff4b0a287a348,2
1185
+ np.float64,0x1f9026a23f206,0xc09004144f3a19dd,2
1186
+ np.float64,0x3fdb2977243652ee,0xbff3c8a2fd05803d,2
1187
+ np.float64,0x3fe0f6e74b21edcf,0xbfed4c3bb956bae0,2
1188
+ np.float64,0xde9cc3bbbd399,0xc08ff19ce5c1e762,2
1189
+ np.float64,0x3fe72ce106ae59c2,0xbfddca7ab14ceba2,2
1190
+ np.float64,0x3fa8ee14e031dc2a,0xc01170d54ca88e86,2
1191
+ np.float64,0x3fe0b09bbb216137,0xbfee0d189a95b877,2
1192
+ np.float64,0x7fdfdcb157bfb962,0x408ff7f33cf2afea,2
1193
+ np.float64,0x3fef84d5f53f09ac,0xbf966134e2a154f4,2
1194
+ np.float64,0x3fea0e0b1bb41c16,0xbfd2fa2d36637d19,2
1195
+ np.float64,0x1ab76fd6356ef,0xc090050a9616ffbd,2
1196
+ np.float64,0x7fd0ccf79a2199ee,0x408ff09045af2dee,2
1197
+ np.float64,0x7fea929345f52526,0x408ffddadc322b07,2
1198
+ np.float64,0x3fe9ef629cf3dec5,0xbfd367129c166838,2
1199
+ np.float64,0x3feedf0ea2fdbe1d,0xbfaa862afca44c00,2
1200
+ np.float64,0x7fce725f723ce4be,0x408fef6cfd2769a8,2
1201
+ np.float64,0x7fe4313b3ca86275,0x408ffaaf9557ef8c,2
1202
+ np.float64,0xe2d46463c5a8d,0xc08ff165725c6b08,2
1203
+ np.float64,0x7fbacb4ace359695,0x408fe5f3647bd0d5,2
1204
+ np.float64,0x3fbafd009635fa01,0xc009f745a7a5c5d5,2
1205
+ np.float64,0x3fe3cea66ce79d4d,0xbfe6253b895e2838,2
1206
+ np.float64,0x7feaa71484354e28,0x408ffde3c0bad2a6,2
1207
+ np.float64,0x3fd755b8b42eab71,0xbff74a1444c6e654,2
1208
+ np.float64,0x3fc313e2172627c4,0xc005f830e77940c3,2
1209
+ np.float64,0x12d699a225ad4,0xc090070ec00f2338,2
1210
+ np.float64,0x3fa975fe8432ebfd,0xc01151b3da48b3f9,2
1211
+ np.float64,0x7fdce3103b39c61f,0x408ff6d19b3326fa,2
1212
+ np.float64,0x7fd341cbba268396,0x408ff2237490fdca,2
1213
+ np.float64,0x3fd8405885b080b1,0xbff6666d8802a7d5,2
1214
+ np.float64,0x3fe0f0cca3a1e199,0xbfed5cdb3e600791,2
1215
+ np.float64,0x7fbd56680c3aaccf,0x408fe6ff55bf378d,2
1216
+ np.float64,0x3f939c4f3027389e,0xc016d364dd6313fb,2
1217
+ np.float64,0x3fe9e87fac73d0ff,0xbfd37f9a2be4fe38,2
1218
+ np.float64,0x7fc93c6a883278d4,0x408fed4260e614f1,2
1219
+ np.float64,0x7fa88c0ff031181f,0x408fdcf09a46bd3a,2
1220
+ np.float64,0xd5487f99aa910,0xc08ff21b6390ab3b,2
1221
+ np.float64,0x3fe34acc96e69599,0xbfe75c9d290428fb,2
1222
+ np.float64,0x3fd17f5964a2feb3,0xbffdef50b524137b,2
1223
+ np.float64,0xe23dec0dc47be,0xc08ff16d1ce61dcb,2
1224
+ np.float64,0x3fec8bd64fb917ad,0xbfc5173941614b8f,2
1225
+ np.float64,0x3fc81d97d7303b30,0xc00343ccb791401d,2
1226
+ np.float64,0x7fe79ad18e2f35a2,0x408ffc7cf0ab0f2a,2
1227
+ np.float64,0x3f96306b402c60d7,0xc0161ce54754cac1,2
1228
+ np.float64,0xfb09fc97f6140,0xc08ff039d1d30123,2
1229
+ np.float64,0x3fec9c4afa793896,0xbfc4ace43ee46079,2
1230
+ np.float64,0x3f9262dac824c5b6,0xc01732a3a7eeb598,2
1231
+ np.float64,0x3fa5cd33f42b9a68,0xc01236ed4d315a3a,2
1232
+ np.float64,0x3fe7bb336caf7667,0xbfdb9a268a82e267,2
1233
+ np.float64,0xc6c338f98d867,0xc08ff2ebb8475bbc,2
1234
+ np.float64,0x3fd50714482a0e29,0xbff9b14a9f84f2c2,2
1235
+ np.float64,0xfff0000000000000,0x7ff8000000000000,2
1236
+ np.float64,0x3fde2cd0f93c59a2,0xbff15afe35a43a37,2
1237
+ np.float64,0xf1719cb9e2e34,0xc08ff0acf77b06d3,2
1238
+ np.float64,0xfd3caaf9fa796,0xc08ff020101771bd,2
1239
+ np.float64,0x7f750d63a02a1ac6,0x408fc32ad0caa362,2
1240
+ np.float64,0x7fcc50f4e238a1e9,0x408fee96a5622f1a,2
1241
+ np.float64,0x421d1da0843a4,0xc08fff9ffe62d869,2
1242
+ np.float64,0x3fd9e17023b3c2e0,0xbff4e631d687ee8e,2
1243
+ np.float64,0x3fe4999a09693334,0xbfe4556b3734c215,2
1244
+ np.float64,0xd619ef03ac33e,0xc08ff21013c85529,2
1245
+ np.float64,0x3fc4da522229b4a4,0xc004f150b2c573aa,2
1246
+ np.float64,0x3feb04b053b60961,0xbfcf3fc9e00ebc40,2
1247
+ np.float64,0x3fbedec5ea3dbd8c,0xc0086a33dc22fab5,2
1248
+ np.float64,0x7fec3b217ab87642,0x408ffe8dbc8ca041,2
1249
+ np.float64,0xdb257d33b64b0,0xc08ff1cb42d3c182,2
1250
+ np.float64,0x7fa2d92ec025b25d,0x408fd9e414d11cb0,2
1251
+ np.float64,0x3fa425c550284b8b,0xc012ab7cbf83be12,2
1252
+ np.float64,0x10b4869021692,0xc09007c0487d648a,2
1253
+ np.float64,0x7f97918c902f2318,0x408fd47867806574,2
1254
+ np.float64,0x3fe4f91238e9f224,0xbfe38160b4e99919,2
1255
+ np.float64,0x3fc2b1af6125635f,0xc00634343bc58461,2
1256
+ np.float64,0x3fc2a98071255301,0xc0063942bc8301be,2
1257
+ np.float64,0x3fe4cfc585299f8b,0xbfe3dca39f114f34,2
1258
+ np.float64,0x3fd1ea75b3a3d4eb,0xbffd63acd02c5406,2
1259
+ np.float64,0x3fd6bf48492d7e91,0xbff7e0cd249f80f9,2
1260
+ np.float64,0x76643d36ecc88,0xc08ff8e68f13b38c,2
1261
+ np.float64,0x7feeabab3e7d5755,0x408fff82a0fd4501,2
1262
+ np.float64,0x46c0d4a68d81b,0xc08ffed79abaddc9,2
1263
+ np.float64,0x3fd088d57ca111ab,0xbfff3dd0ed7128ea,2
1264
+ np.float64,0x3fed25887cba4b11,0xbfc13f47639bd645,2
1265
+ np.float64,0x7fd90984b4b21308,0x408ff52b022c7fb4,2
1266
+ np.float64,0x3fe6ef31daadde64,0xbfdec185760cbf21,2
1267
+ np.float64,0x3fe48dbe83291b7d,0xbfe47005b99920bd,2
1268
+ np.float64,0x3fdce8422f39d084,0xbff258a33a96cc8e,2
1269
+ np.float64,0xb8ecdef771d9c,0xc08ff3c0eca61b10,2
1270
+ np.float64,0x3fe9bbf9a03377f3,0xbfd41ecfdcc336b9,2
1271
+ np.float64,0x7fe2565339a4aca5,0x408ff992d8851eaf,2
1272
+ np.float64,0x3fe1693e3822d27c,0xbfec1919da2ca697,2
1273
+ np.float64,0x3fd3680488a6d009,0xbffb8b7330275947,2
1274
+ np.float64,0x7fbe4f3d2c3c9e79,0x408fe75fa3f4e600,2
1275
+ np.float64,0x7fd4cfef3ca99fdd,0x408ff308ee3ab50f,2
1276
+ np.float64,0x3fd9c9a51cb3934a,0xbff4fb7440055ce6,2
1277
+ np.float64,0x3fe08a9640a1152d,0xbfee76bd1bfbf5c2,2
1278
+ np.float64,0x3fef012c41fe0259,0xbfa757a2da7f9707,2
1279
+ np.float64,0x3fee653fe2fcca80,0xbfb2ffae0c95025c,2
1280
+ np.float64,0x7fd0776933a0eed1,0x408ff054e7b43d41,2
1281
+ np.float64,0x4c94e5c09929d,0xc08ffdedb7f49e5e,2
1282
+ np.float64,0xca3e3d17947c8,0xc08ff2b86dce2f7a,2
1283
+ np.float64,0x3fb528e1342a51c2,0xc00cc626c8e2d9ba,2
1284
+ np.float64,0xd774df81aee9c,0xc08ff1fd6f0a7548,2
1285
+ np.float64,0x3fc47a9b6128f537,0xc00526c577b80849,2
1286
+ np.float64,0x3fe29a6f6a6534df,0xbfe90a5f83644911,2
1287
+ np.float64,0x3fecda4f59f9b49f,0xbfc31e4a80c4cbb6,2
1288
+ np.float64,0x7fe51d44f5aa3a89,0x408ffb3382437426,2
1289
+ np.float64,0x3fd677fc412ceff9,0xbff82999086977e7,2
1290
+ np.float64,0x3fe2a3c7e7254790,0xbfe8f33415cdba9d,2
1291
+ np.float64,0x3fe6d8d1dc6db1a4,0xbfdf1bc61bc24dff,2
1292
+ np.float64,0x7febb32d8ef7665a,0x408ffe55a043ded1,2
1293
+ np.float64,0x60677860c0d0,0xc0900da2caa7d571,2
1294
+ np.float64,0x7390c2e0e7219,0xc08ff92df18bb5d2,2
1295
+ np.float64,0x3fca53711b34a6e2,0xc00240b07a9b529b,2
1296
+ np.float64,0x7fe7ce6dd8ef9cdb,0x408ffc961164ead9,2
1297
+ np.float64,0x7fc0c9de0d2193bb,0x408fe88e245767f6,2
1298
+ np.float64,0xc0ee217981dc4,0xc08ff343b77ea770,2
1299
+ np.float64,0x72bd4668e57a9,0xc08ff94323fd74fc,2
1300
+ np.float64,0x7fd6970e252d2e1b,0x408ff3fb1e2fead2,2
1301
+ np.float64,0x7fdcb61040396c20,0x408ff6bf926bc98f,2
1302
+ np.float64,0xda4faa25b49f6,0xc08ff1d68b3877f0,2
1303
+ np.float64,0x3feb344749f6688f,0xbfcdfba2d66c72c5,2
1304
+ np.float64,0x3fe2aa4284e55485,0xbfe8e32ae0683f57,2
1305
+ np.float64,0x3f8e8fcfd03d1fa0,0xc01843efb2129908,2
1306
+ np.float64,0x8000000000000000,0xfff0000000000000,2
1307
+ np.float64,0x3fd8e01155b1c023,0xbff5d0529dae9515,2
1308
+ np.float64,0x3fe8033f3370067e,0xbfda837c80b87e7c,2
1309
+ np.float64,0x7fc5bf831e2b7f05,0x408feb8ae3b039a0,2
1310
+ np.float64,0x3fd8dcdf5331b9bf,0xbff5d349e1ed422a,2
1311
+ np.float64,0x3fe58b4e302b169c,0xbfe243c9cbccde44,2
1312
+ np.float64,0x3fea8a2e47b5145d,0xbfd1464e37221894,2
1313
+ np.float64,0x75cd1e88eb9a4,0xc08ff8f553ef0475,2
1314
+ np.float64,0x7fcfc876e23f90ed,0x408fefebe6cc95e6,2
1315
+ np.float64,0x7f51aceb002359d5,0x408fb1263f9003fb,2
1316
+ np.float64,0x7fc2a1b877254370,0x408fe9c1ec52f8b9,2
1317
+ np.float64,0x7fd495810e292b01,0x408ff2e859414d31,2
1318
+ np.float64,0x7fd72048632e4090,0x408ff440690cebdb,2
1319
+ np.float64,0x7fd7aafaffaf6,0xc08ff803a390779f,2
1320
+ np.float64,0x7fe18067d4a300cf,0x408ff9090a02693f,2
1321
+ np.float64,0x3fdc1080f8b82102,0xbff3077bf44a89bd,2
1322
+ np.float64,0x3fc34a462f26948c,0xc005d777b3cdf139,2
1323
+ np.float64,0x3fe21e4a1fe43c94,0xbfea428acfbc6ea9,2
1324
+ np.float64,0x1f0d79083e1b0,0xc090042c65a7abf2,2
1325
+ np.float64,0x3fe8d0d15931a1a3,0xbfd779f6bbd4db78,2
1326
+ np.float64,0x3fe74578022e8af0,0xbfdd68b6c15e9f5e,2
1327
+ np.float64,0x50995dd0a132c,0xc08ffd56a5c8accf,2
1328
+ np.float64,0x3f9a6342b034c685,0xc0151ce1973c62bd,2
1329
+ np.float64,0x3f30856a00210ad4,0xc027e852f4d1fcbc,2
1330
+ np.float64,0x3febcf7646b79eed,0xbfc9e9cc9d12425c,2
1331
+ np.float64,0x8010000000000000,0x7ff8000000000000,2
1332
+ np.float64,0x3fdf520c02bea418,0xbff07ed5013f3062,2
1333
+ np.float64,0x3fe5433ecbea867e,0xbfe2df38968b6d14,2
1334
+ np.float64,0x3fb933a84e326751,0xc00ac1a144ad26c5,2
1335
+ np.float64,0x7b6d72c2f6daf,0xc08ff86b7a67f962,2
1336
+ np.float64,0xaef5dae75debc,0xc08ff46496bb2932,2
1337
+ np.float64,0x522d869aa45b1,0xc08ffd1d55281e98,2
1338
+ np.float64,0xa2462b05448c6,0xc08ff542fe0ac5fd,2
1339
+ np.float64,0x3fe2b71dd6e56e3c,0xbfe8c3690cf15415,2
1340
+ np.float64,0x3fe5778231aaef04,0xbfe26e495d09b783,2
1341
+ np.float64,0x3fe9b8d564f371ab,0xbfd42a161132970d,2
1342
+ np.float64,0x3f89ebc34033d787,0xc019373f90bfc7f1,2
1343
+ np.float64,0x3fe438ddc6e871bc,0xbfe53039341b0a93,2
1344
+ np.float64,0x873c75250e78f,0xc08ff75d8478dccd,2
1345
+ np.float64,0x807134cb00e27,0xc08ff7f5cf59c57a,2
1346
+ np.float64,0x3fac459878388b31,0xc010b6fe803bcdc2,2
1347
+ np.float64,0xca9dc7eb953b9,0xc08ff2b2fb480784,2
1348
+ np.float64,0x7feb38587bb670b0,0x408ffe21ff6d521e,2
1349
+ np.float64,0x7fd70e9b782e1d36,0x408ff437936b393a,2
1350
+ np.float64,0x3fa4037bbc2806f7,0xc012b55744c65ab2,2
1351
+ np.float64,0x3fd3d4637427a8c7,0xbffb0beebf4311ef,2
1352
+ np.float64,0x7fdabbda5db577b4,0x408ff5ecbc0d4428,2
1353
+ np.float64,0x7fda9be0a2b537c0,0x408ff5dee5d03d5a,2
1354
+ np.float64,0x7fe9c74396338e86,0x408ffd813506a18a,2
1355
+ np.float64,0x3fd058243e20b048,0xbfff822ffd8a7f21,2
1356
+ np.float64,0x3fe6aa6ca9ed54d9,0xbfdfd805629ff49e,2
1357
+ np.float64,0x3fd91431d5322864,0xbff5a025eea8c78b,2
1358
+ np.float64,0x7fe4d7f02329afdf,0x408ffb0d5d9b7878,2
1359
+ np.float64,0x3fe2954a12252a94,0xbfe917266e3e22d5,2
1360
+ np.float64,0x3fb25f7c8224bef9,0xc00e6764c81b3718,2
1361
+ np.float64,0x3fda4bddeeb497bc,0xbff4880638908c81,2
1362
+ np.float64,0x55dfd12eabbfb,0xc08ffc9b54ff4002,2
1363
+ np.float64,0x3fe8f399e031e734,0xbfd6f8e5c4dcd93f,2
1364
+ np.float64,0x3fd954a24832a945,0xbff56521f4707a06,2
1365
+ np.float64,0x3fdea911f2bd5224,0xbff0fcb2d0c2b2e2,2
1366
+ np.float64,0x3fe6b4ff8a2d69ff,0xbfdfacfc85cafeab,2
1367
+ np.float64,0x3fc7fa02042ff404,0xc00354e13b0767ad,2
1368
+ np.float64,0x3fe955088c72aa11,0xbfd593130f29949e,2
1369
+ np.float64,0xd7e74ec1afcea,0xc08ff1f74f61721c,2
1370
+ np.float64,0x3fe9d69c1ab3ad38,0xbfd3bf710a337e06,2
1371
+ np.float64,0x3fd85669a2b0acd3,0xbff65176143ccc1e,2
1372
+ np.float64,0x3fea99b285353365,0xbfd11062744783f2,2
1373
+ np.float64,0x3fe2c79f80a58f3f,0xbfe89ac33f990289,2
1374
+ np.float64,0x3f8332ba30266574,0xc01af2cb7b635783,2
1375
+ np.float64,0x30d0150061a1,0xc090119030f74c5d,2
1376
+ np.float64,0x3fdbf4cb06b7e996,0xbff31e5207aaa754,2
1377
+ np.float64,0x3fe6b56c216d6ad8,0xbfdfab42fb2941c5,2
1378
+ np.float64,0x7fc4dc239829b846,0x408feb0fb0e13fbe,2
1379
+ np.float64,0x3fd0ab85ef21570c,0xbfff0d95d6c7a35c,2
1380
+ np.float64,0x7fe13d75e5e27aeb,0x408ff8dc8efa476b,2
1381
+ np.float64,0x3fece3b832f9c770,0xbfc2e21b165d583f,2
1382
+ np.float64,0x3fe3a279c4e744f4,0xbfe68ca4fbb55dbf,2
1383
+ np.float64,0x3feb64659ef6c8cb,0xbfccb6204b6bf724,2
1384
+ np.float64,0x2279a6bc44f36,0xc0900391eeeb3e7c,2
1385
+ np.float64,0xb88046d571009,0xc08ff3c7b5b45300,2
1386
+ np.float64,0x7ff4000000000000,0x7ffc000000000000,2
1387
+ np.float64,0x3fe49af059a935e1,0xbfe4526c294f248f,2
1388
+ np.float64,0xa3e5508147cc,0xc0900a92ce5924b1,2
1389
+ np.float64,0x7fc56def3d2adbdd,0x408feb5f46c360e8,2
1390
+ np.float64,0x7fd99f3574333e6a,0x408ff56f3807987c,2
1391
+ np.float64,0x3fdc38d56fb871ab,0xbff2e667cad8f36a,2
1392
+ np.float64,0xd0b03507a1607,0xc08ff25bbcf8aa9d,2
1393
+ np.float64,0xc493f9078927f,0xc08ff30c5fa4e759,2
1394
+ np.float64,0x3fc86ddbcb30dbb8,0xc0031da1fcb56d75,2
1395
+ np.float64,0x7fe75dc395aebb86,0x408ffc5eef841491,2
1396
+ np.float64,0x1647618a2c8ed,0xc0900616ef9479c1,2
1397
+ np.float64,0xdf144763be289,0xc08ff196b527f3c9,2
1398
+ np.float64,0x3fe0b29da6a1653b,0xbfee078b5f4d7744,2
1399
+ np.float64,0x3feb055852b60ab1,0xbfcf3b4db5779a7a,2
1400
+ np.float64,0x3fe8bc1625f1782c,0xbfd7c739ade904bc,2
1401
+ np.float64,0x7fd19bfb8ea337f6,0x408ff11b2b55699c,2
1402
+ np.float64,0x3fed1d80d1ba3b02,0xbfc1722e8d3ce094,2
1403
+ np.float64,0x2d9c65925b38e,0xc09001f46bcd3bc5,2
1404
+ np.float64,0x7fed6f4d857ade9a,0x408fff091cf6a3b4,2
1405
+ np.float64,0x3fd070cd6ba0e19b,0xbfff5f7609ca29e8,2
1406
+ np.float64,0x7fea3508b8f46a10,0x408ffdb1f30bd6be,2
1407
+ np.float64,0x508b897ca1172,0xc08ffd58a0eb3583,2
1408
+ np.float64,0x7feba367b07746ce,0x408ffe4f0bf4bd4e,2
1409
+ np.float64,0x3fefebd5c4bfd7ac,0xbf6d20b4fcf21b69,2
1410
+ np.float64,0x3fd8ef07b8b1de0f,0xbff5c2745c0795a5,2
1411
+ np.float64,0x3fd38ed518271daa,0xbffb5d75f00f6900,2
1412
+ np.float64,0x6de0fecedbc20,0xc08ff9c307bbc647,2
1413
+ np.float64,0xafc0ffc35f820,0xc08ff45737e5d6b4,2
1414
+ np.float64,0x7fd282097ca50412,0x408ff1ae3b27bf3b,2
1415
+ np.float64,0x3fe2f2d50b65e5aa,0xbfe831042e6a1e99,2
1416
+ np.float64,0x3faa437bac3486f7,0xc01123d8d962205a,2
1417
+ np.float64,0x3feea54434fd4a88,0xbfaff202cc456647,2
1418
+ np.float64,0x3fc9e65b8633ccb7,0xc00270e77ffd19da,2
1419
+ np.float64,0x7fee15af61fc2b5e,0x408fff49a49154a3,2
1420
+ np.float64,0x7fefe670a73fcce0,0x408ffff6c44c1005,2
1421
+ np.float64,0x3fc0832d0f21065a,0xc007a2dc2f25384a,2
1422
+ np.float64,0x3fecfc96bcb9f92d,0xbfc24367c3912620,2
1423
+ np.float64,0x3feb705682b6e0ad,0xbfcc65b1bb16f9c5,2
1424
+ np.float64,0x3fe185c4f9630b8a,0xbfebcdb401af67a4,2
1425
+ np.float64,0x3fb0a5a9f6214b54,0xc00f8ada2566a047,2
1426
+ np.float64,0x7fe2908cdda52119,0x408ff9b744861fb1,2
1427
+ np.float64,0x7fee776e183ceedb,0x408fff6ee7c2f86e,2
1428
+ np.float64,0x3fce1d608f3c3ac1,0xc000b3685d006474,2
1429
+ np.float64,0x7fecf92aa339f254,0x408ffeda6c998267,2
1430
+ np.float64,0xce13cb519c27a,0xc08ff280f02882a9,2
1431
+ np.float64,0x1,0xc090c80000000000,2
1432
+ np.float64,0x3fe485a8afa90b51,0xbfe4823265d5a50a,2
1433
+ np.float64,0x3feea60908bd4c12,0xbfafdf7ad7fe203f,2
1434
+ np.float64,0x3fd2253033a44a60,0xbffd187d0ec8d5b9,2
1435
+ np.float64,0x435338fc86a68,0xc08fff6a591059dd,2
1436
+ np.float64,0x7fce8763a73d0ec6,0x408fef74f1e715ff,2
1437
+ np.float64,0x3fbe5ddb783cbbb7,0xc0089acc5afa794b,2
1438
+ np.float64,0x7fe4cf19ada99e32,0x408ffb0877ca302b,2
1439
+ np.float64,0x3fe94c9ea1b2993d,0xbfd5b1c2e867b911,2
1440
+ np.float64,0x3fe75541c72eaa84,0xbfdd2a27aa117699,2
1441
+ np.float64,0x8000000000000001,0x7ff8000000000000,2
1442
+ np.float64,0x7fdbec7f2c37d8fd,0x408ff66d69a7f818,2
1443
+ np.float64,0x8ef10d091de22,0xc08ff6b9ca5094f8,2
1444
+ np.float64,0x3fea69025b74d205,0xbfd1b9fe2c252c70,2
1445
+ np.float64,0x562376d0ac46f,0xc08ffc924111cd31,2
1446
+ np.float64,0x8e8097ab1d013,0xc08ff6c2e2706f67,2
1447
+ np.float64,0x3fca6803ed34d008,0xc00237aef808825b,2
1448
+ np.float64,0x7fe8fe9067b1fd20,0x408ffd25f459a7d1,2
1449
+ np.float64,0x3f918e8c7f233,0xc0900009fe011d54,2
1450
+ np.float64,0x3fdfe773833fcee7,0xbff011bc1af87bb9,2
1451
+ np.float64,0xefffef6fdfffe,0xc08ff0beb0f09eb0,2
1452
+ np.float64,0x7fe64610282c8c1f,0x408ffbd17209db18,2
1453
+ np.float64,0xe66be8c1ccd7d,0xc08ff13706c056e1,2
1454
+ np.float64,0x2837e570506fd,0xc09002ae4dae0c1a,2
1455
+ np.float64,0x3febe3a081f7c741,0xbfc964171f2a5a47,2
1456
+ np.float64,0x3fe21ed09a243da1,0xbfea41342d29c3ff,2
1457
+ np.float64,0x3fe1596c8162b2d9,0xbfec431eee30823a,2
1458
+ np.float64,0x8f2b9a131e574,0xc08ff6b51104ed4e,2
1459
+ np.float64,0x3fe88ed179711da3,0xbfd870d08a4a4b0c,2
1460
+ np.float64,0x34159bc2682b4,0xc09001305a885f94,2
1461
+ np.float64,0x1ed31e543da65,0xc0900437481577f8,2
1462
+ np.float64,0x3feafbe9de75f7d4,0xbfcf7bcdbacf1c61,2
1463
+ np.float64,0xfb16fb27f62e0,0xc08ff03938e682a2,2
1464
+ np.float64,0x3fe5cd5ba7eb9ab7,0xbfe1b7165771af3c,2
1465
+ np.float64,0x7fe72905e76e520b,0x408ffc44c4e7e80c,2
1466
+ np.float64,0x7fb7136e2e2e26db,0x408fe439fd383fb7,2
1467
+ np.float64,0x8fa585e11f4c,0xc0900b55a08a486b,2
1468
+ np.float64,0x7fed985ce47b30b9,0x408fff192b596821,2
1469
+ np.float64,0x3feaaf0869755e11,0xbfd0c671571b3764,2
1470
+ np.float64,0x3fa40fd4ec281faa,0xc012b1c8dc0b9e5f,2
1471
+ np.float64,0x7fda2a70993454e0,0x408ff5ad47b0c68a,2
1472
+ np.float64,0x3fe5f7e931abefd2,0xbfe15d52b3605abf,2
1473
+ np.float64,0x3fe9fc6d3533f8da,0xbfd338b06a790994,2
1474
+ np.float64,0x3fe060649420c0c9,0xbfeeed1756111891,2
1475
+ np.float64,0x3fce8435e33d086c,0xc0008c41cea9ed40,2
1476
+ np.float64,0x7ff8000000000000,0x7ff8000000000000,2
1477
+ np.float64,0x617820aec2f05,0xc08ffb251e9af0f0,2
1478
+ np.float64,0x7fcc4ab6ee38956d,0x408fee9419c8f77d,2
1479
+ np.float64,0x7fdefda2fc3dfb45,0x408ff7a15063bc05,2
1480
+ np.float64,0x7fe5138ccaaa2719,0x408ffb2e30f3a46e,2
1481
+ np.float64,0x3fe3817a836702f5,0xbfe6da7c2b25e35a,2
1482
+ np.float64,0x3fb8a7dafa314fb6,0xc00b025bc0784ebe,2
1483
+ np.float64,0x349dc420693d,0xc09011215825d2c8,2
1484
+ np.float64,0x6b0e504ad61cb,0xc08ffa0fee9c5cd6,2
1485
+ np.float64,0x273987644e732,0xc09002d34294ed79,2
1486
+ np.float64,0x3fc0bd8a6e217b15,0xc0077a5828b4d2f5,2
1487
+ np.float64,0x758b48c4eb16a,0xc08ff8fbc8fbe46a,2
1488
+ np.float64,0x3fc8a9a52631534a,0xc00301854ec0ef81,2
1489
+ np.float64,0x7fe79d29a76f3a52,0x408ffc7e1607a4c1,2
1490
+ np.float64,0x3fd7d3ebce2fa7d8,0xbff6ce8a94aebcda,2
1491
+ np.float64,0x7fd1cb68a52396d0,0x408ff13a17533b2b,2
1492
+ np.float64,0x7fda514a5d34a294,0x408ff5be5e081578,2
1493
+ np.float64,0x3fc40b4382281687,0xc0056632c8067228,2
1494
+ np.float64,0x7feff1208c3fe240,0x408ffffaa180fa0d,2
1495
+ np.float64,0x8f58739f1eb0f,0xc08ff6b17402689d,2
1496
+ np.float64,0x1fdbe9a23fb7e,0xc090040685b2d24f,2
1497
+ np.float64,0xcb1d0e87963a2,0xc08ff2abbd903b82,2
1498
+ np.float64,0x3fc45a6a1a28b4d4,0xc00538f86c4aeaee,2
1499
+ np.float64,0x3fe61885b1ac310b,0xbfe118fd2251d2ec,2
1500
+ np.float64,0x3fedf584c8fbeb0a,0xbfb8572433ff67a9,2
1501
+ np.float64,0x7fb0bddd1a217bb9,0x408fe085e0d621db,2
1502
+ np.float64,0x72d8d3e0e5b3,0xc0900ca02f68c7a1,2
1503
+ np.float64,0x5cca6ff6b994f,0xc08ffbb6751fda01,2
1504
+ np.float64,0x7fe3197839a632ef,0x408ffa0b2fccfb68,2
1505
+ np.float64,0x3fcce4d9c139c9b4,0xc0012dae05baa91b,2
1506
+ np.float64,0x3fe76d00f62eda02,0xbfdccc5f12799be1,2
1507
+ np.float64,0x3fc53c22f72a7846,0xc004bbaa9cbc7958,2
1508
+ np.float64,0x7fdda02f1ebb405d,0x408ff71c37c71659,2
1509
+ np.float64,0x3fe0844eaba1089d,0xbfee884722762583,2
1510
+ np.float64,0x3febb438dc776872,0xbfca9f05e1c691f1,2
1511
+ np.float64,0x3fdf4170cdbe82e2,0xbff08b1561c8d848,2
1512
+ np.float64,0x3fce1b8d6f3c371b,0xc000b41b69507671,2
1513
+ np.float64,0x8370e60706e1d,0xc08ff7b19ea0b4ca,2
1514
+ np.float64,0x7fa5bf92382b7f23,0x408fdb8aebb3df87,2
1515
+ np.float64,0x7fe4a59979a94b32,0x408ffaf15c1358cd,2
1516
+ np.float64,0x3faa66086034cc11,0xc0111c466b7835d6,2
1517
+ np.float64,0x7fb7a958262f52af,0x408fe48408b1e093,2
1518
+ np.float64,0x3fdaacc5f635598c,0xbff43390d06b5614,2
1519
+ np.float64,0x3fd2825b9e2504b7,0xbffca3234264f109,2
1520
+ np.float64,0x3fcede160a3dbc2c,0xc0006a759e29060c,2
1521
+ np.float64,0x7fd3b19603a7632b,0x408ff265b528371c,2
1522
+ np.float64,0x7fcf8a86ea3f150d,0x408fefd552e7f3b2,2
1523
+ np.float64,0xedbcc0f7db798,0xc08ff0daad12096b,2
1524
+ np.float64,0xf1e1683de3c2d,0xc08ff0a7a0a37e00,2
1525
+ np.float64,0xb6ebd9bf6dd7b,0xc08ff3e11e28378d,2
1526
+ np.float64,0x3fec8090d6f90122,0xbfc56031b72194cc,2
1527
+ np.float64,0x3fd3e10e37a7c21c,0xbffafd34a3ebc933,2
1528
+ np.float64,0x7fbb1c96aa36392c,0x408fe616347b3342,2
1529
+ np.float64,0x3fe2f3996f25e733,0xbfe82f25bc5d1bbd,2
1530
+ np.float64,0x7fe8709da870e13a,0x408ffce3ab6ce59a,2
1531
+ np.float64,0x7fea3233d1b46467,0x408ffdb0b3bbc6de,2
1532
+ np.float64,0x65fa4112cbf49,0xc08ffa9f85eb72b9,2
1533
+ np.float64,0x3fca2cae9f34595d,0xc00251bb275afb87,2
1534
+ np.float64,0x8135fd9f026c0,0xc08ff7e42e14dce7,2
1535
+ np.float64,0x7fe0a6f057e14de0,0x408ff876081a4bfe,2
1536
+ np.float64,0x10000000000000,0xc08ff00000000000,2
1537
+ np.float64,0x3fe1fd506263faa1,0xbfea96dd8c543b72,2
1538
+ np.float64,0xa5532c554aa66,0xc08ff50bf5bfc66d,2
1539
+ np.float64,0xc239d00b8473a,0xc08ff32ff0ea3f92,2
1540
+ np.float64,0x7fdb5314e336a629,0x408ff62d4ff60d82,2
1541
+ np.float64,0x3fe5f506e2abea0e,0xbfe16362a4682120,2
1542
+ np.float64,0x3fa20c60202418c0,0xc0134e08d82608b6,2
1543
+ np.float64,0x7fe03864b22070c8,0x408ff82866d65e9a,2
1544
+ np.float64,0x3fe72cf5656e59eb,0xbfddca298969effa,2
1545
+ np.float64,0x5c295386b852b,0xc08ffbca90b136c9,2
1546
+ np.float64,0x7fd71e5020ae3c9f,0x408ff43f6d58eb7c,2
1547
+ np.float64,0x3fd1905a842320b5,0xbffdd8ecd288159c,2
1548
+ np.float64,0x3fe6bddb256d7bb6,0xbfdf88fee1a820bb,2
1549
+ np.float64,0xe061b967c0c37,0xc08ff18581951561,2
1550
+ np.float64,0x3fe534f65cea69ed,0xbfe2fe45fe7d3040,2
1551
+ np.float64,0xdc7dae07b8fb6,0xc08ff1b93074ea76,2
1552
+ np.float64,0x3fd0425082a084a1,0xbfffa11838b21633,2
1553
+ np.float64,0xba723fc974e48,0xc08ff3a8b8d01c58,2
1554
+ np.float64,0x3fce42ffc73c8600,0xc000a5062678406e,2
1555
+ np.float64,0x3f2e6d3c7e5ce,0xc090001304cfd1c7,2
1556
+ np.float64,0x3fd4b2e5f7a965cc,0xbffa0e6e6bae0a68,2
1557
+ np.float64,0x3fe6db1d18edb63a,0xbfdf128158ee92d9,2
1558
+ np.float64,0x7fe4e5792f29caf1,0x408ffb14d9dbf133,2
1559
+ np.float64,0x3fc11cdf992239bf,0xc00739569619cd77,2
1560
+ np.float64,0x3fc05ea11220bd42,0xc007bc841b48a890,2
1561
+ np.float64,0x4bd592d497ab3,0xc08ffe0ab1c962e2,2
1562
+ np.float64,0x280068fc5000e,0xc09002b64955e865,2
1563
+ np.float64,0x7fe2f2637065e4c6,0x408ff9f379c1253a,2
1564
+ np.float64,0x3fefc38467ff8709,0xbf85e53e64b9a424,2
1565
+ np.float64,0x2d78ec5a5af1e,0xc09001f8ea8601e0,2
1566
+ np.float64,0x7feeef2b957dde56,0x408fff9bebe995f7,2
1567
+ np.float64,0x2639baf44c738,0xc09002f9618d623b,2
1568
+ np.float64,0x3fc562964d2ac52d,0xc004a6d76959ef78,2
1569
+ np.float64,0x3fe21b071fe4360e,0xbfea4adb2cd96ade,2
1570
+ np.float64,0x7fe56aa6802ad54c,0x408ffb5d81d1a898,2
1571
+ np.float64,0x4296b452852d7,0xc08fff8ad7fbcbe1,2
1572
+ np.float64,0x7fe3fac4ff27f589,0x408ffa9049eec479,2
1573
+ np.float64,0x7fe7a83e6caf507c,0x408ffc837f436604,2
1574
+ np.float64,0x3fc4ac5b872958b7,0xc0050add72381ac3,2
1575
+ np.float64,0x3fd6d697c02dad30,0xbff7c931a3eefb01,2
1576
+ np.float64,0x3f61e391c023c724,0xc021ad91e754f94b,2
1577
+ np.float64,0x10817f9c21031,0xc09007d20434d7bc,2
1578
+ np.float64,0x3fdb9c4c4cb73899,0xbff367d8615c5ece,2
1579
+ np.float64,0x3fe26ead6b64dd5b,0xbfe977771def5989,2
1580
+ np.float64,0x3fc43ea5c3287d4c,0xc00548c2163ae631,2
1581
+ np.float64,0x3fe05bd8bba0b7b1,0xbfeef9ea0db91abc,2
1582
+ np.float64,0x3feac78369358f07,0xbfd071e2b0aeab39,2
1583
+ np.float64,0x7fe254922ca4a923,0x408ff991bdd4e5d3,2
1584
+ np.float64,0x3fe5a2f5842b45eb,0xbfe21135c9a71666,2
1585
+ np.float64,0x3fd5daf98c2bb5f3,0xbff8cd24f7c07003,2
1586
+ np.float64,0x3fcb2a1384365427,0xc001e40f0d04299a,2
1587
+ np.float64,0x3fe073974360e72f,0xbfeeb7183a9930b7,2
1588
+ np.float64,0xcf3440819e688,0xc08ff270d3a71001,2
1589
+ np.float64,0x3fd35656cda6acae,0xbffba083fba4939d,2
1590
+ np.float64,0x7fe6c59b4ded8b36,0x408ffc12ce725425,2
1591
+ np.float64,0x3fba896f943512df,0xc00a291cb6947701,2
1592
+ np.float64,0x7fe54917e86a922f,0x408ffb4b5e0fb848,2
1593
+ np.float64,0x7fed2a3f51ba547e,0x408ffeede945a948,2
1594
+ np.float64,0x3fdc72bd5038e57b,0xbff2b73b7e93e209,2
1595
+ np.float64,0x7fefdb3f9f3fb67e,0x408ffff2b702a768,2
1596
+ np.float64,0x3fb0184430203088,0xc00fee8c1351763c,2
1597
+ np.float64,0x7d6c3668fad87,0xc08ff83c195f2cca,2
1598
+ np.float64,0x3fd5aa254aab544b,0xbff900f16365991b,2
1599
+ np.float64,0x3f963daab02c7b55,0xc0161974495b1b71,2
1600
+ np.float64,0x3fa7a9c5982f538b,0xc011bde0f6052a89,2
1601
+ np.float64,0xb3a5a74b674b5,0xc08ff4167bc97c81,2
1602
+ np.float64,0x7fad0c14503a1828,0x408fdee1f2d56cd7,2
1603
+ np.float64,0x43e0e9d887c1e,0xc08fff522837b13b,2
1604
+ np.float64,0x3fe513b20aea2764,0xbfe346ea994100e6,2
1605
+ np.float64,0x7fe4e10393e9c206,0x408ffb12630f6a06,2
1606
+ np.float64,0x68b286e2d1651,0xc08ffa51c0d795d4,2
1607
+ np.float64,0x7fe8de453331bc89,0x408ffd17012b75ac,2
1608
+ np.float64,0x1b3d77d4367b0,0xc09004edea60aa36,2
1609
+ np.float64,0x3fd351cbc326a398,0xbffba5f0f4d5fdba,2
1610
+ np.float64,0x3fd264951b24c92a,0xbffcc8636788b9bf,2
1611
+ np.float64,0xd2465761a48cb,0xc08ff2455c9c53e5,2
1612
+ np.float64,0x7fe46a0ef028d41d,0x408ffacfe32c6f5d,2
1613
+ np.float64,0x3fafd8ac4c3fb159,0xc010071bf33195d0,2
1614
+ np.float64,0x902aec5d2055e,0xc08ff6a08e28aabc,2
1615
+ np.float64,0x3fcea61bb03d4c37,0xc0007f76e509b657,2
1616
+ np.float64,0x7fe8d90f9571b21e,0x408ffd1495f952e7,2
1617
+ np.float64,0x7fa650c9442ca192,0x408fdbd6ff22fdd8,2
1618
+ np.float64,0x3fe8ecfdf171d9fc,0xbfd7115df40e8580,2
1619
+ np.float64,0x7fd4e6fe7f29cdfc,0x408ff315b0dae183,2
1620
+ np.float64,0x77df4c52efbea,0xc08ff8c1d5c1df33,2
1621
+ np.float64,0xe200b0cfc4016,0xc08ff1703cfb8e79,2
1622
+ np.float64,0x3fe230ea7e2461d5,0xbfea132d2385160e,2
1623
+ np.float64,0x7fd1f7ced723ef9d,0x408ff156bfbf92a4,2
1624
+ np.float64,0x3fea762818f4ec50,0xbfd18c12a88e5f79,2
1625
+ np.float64,0x7feea4ba7c7d4974,0x408fff8004164054,2
1626
+ np.float64,0x833ec605067d9,0xc08ff7b606383841,2
1627
+ np.float64,0x7fd0c2d7fea185af,0x408ff0894f3a0cf4,2
1628
+ np.float64,0x3fe1d7d61d23afac,0xbfeaf76fee875d3e,2
1629
+ np.float64,0x65adecb0cb5be,0xc08ffaa82cb09d68,2
evalkit_tf446/lib/python3.10/site-packages/networkx/__pycache__/__init__.cpython-310.pyc ADDED
Binary file (1.29 kB). View file
 
evalkit_tf446/lib/python3.10/site-packages/networkx/__pycache__/conftest.cpython-310.pyc ADDED
Binary file (6.03 kB). View file
 
evalkit_tf446/lib/python3.10/site-packages/networkx/__pycache__/convert.cpython-310.pyc ADDED
Binary file (13.1 kB). View file
 
evalkit_tf446/lib/python3.10/site-packages/networkx/__pycache__/convert_matrix.cpython-310.pyc ADDED
Binary file (41.2 kB). View file
 
evalkit_tf446/lib/python3.10/site-packages/networkx/__pycache__/exception.cpython-310.pyc ADDED
Binary file (4.76 kB). View file
 
evalkit_tf446/lib/python3.10/site-packages/networkx/__pycache__/lazy_imports.cpython-310.pyc ADDED
Binary file (5.85 kB). View file
 
evalkit_tf446/lib/python3.10/site-packages/networkx/__pycache__/relabel.cpython-310.pyc ADDED
Binary file (10.2 kB). View file
 
evalkit_tf446/lib/python3.10/site-packages/networkx/algorithms/components/attracting.py ADDED
@@ -0,0 +1,115 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ """Attracting components."""
2
+
3
+ import networkx as nx
4
+ from networkx.utils.decorators import not_implemented_for
5
+
6
+ __all__ = [
7
+ "number_attracting_components",
8
+ "attracting_components",
9
+ "is_attracting_component",
10
+ ]
11
+
12
+
13
+ @not_implemented_for("undirected")
14
+ @nx._dispatchable
15
+ def attracting_components(G):
16
+ """Generates the attracting components in `G`.
17
+
18
+ An attracting component in a directed graph `G` is a strongly connected
19
+ component with the property that a random walker on the graph will never
20
+ leave the component, once it enters the component.
21
+
22
+ The nodes in attracting components can also be thought of as recurrent
23
+ nodes. If a random walker enters the attractor containing the node, then
24
+ the node will be visited infinitely often.
25
+
26
+ To obtain induced subgraphs on each component use:
27
+ ``(G.subgraph(c).copy() for c in attracting_components(G))``
28
+
29
+ Parameters
30
+ ----------
31
+ G : DiGraph, MultiDiGraph
32
+ The graph to be analyzed.
33
+
34
+ Returns
35
+ -------
36
+ attractors : generator of sets
37
+ A generator of sets of nodes, one for each attracting component of G.
38
+
39
+ Raises
40
+ ------
41
+ NetworkXNotImplemented
42
+ If the input graph is undirected.
43
+
44
+ See Also
45
+ --------
46
+ number_attracting_components
47
+ is_attracting_component
48
+
49
+ """
50
+ scc = list(nx.strongly_connected_components(G))
51
+ cG = nx.condensation(G, scc)
52
+ for n in cG:
53
+ if cG.out_degree(n) == 0:
54
+ yield scc[n]
55
+
56
+
57
+ @not_implemented_for("undirected")
58
+ @nx._dispatchable
59
+ def number_attracting_components(G):
60
+ """Returns the number of attracting components in `G`.
61
+
62
+ Parameters
63
+ ----------
64
+ G : DiGraph, MultiDiGraph
65
+ The graph to be analyzed.
66
+
67
+ Returns
68
+ -------
69
+ n : int
70
+ The number of attracting components in G.
71
+
72
+ Raises
73
+ ------
74
+ NetworkXNotImplemented
75
+ If the input graph is undirected.
76
+
77
+ See Also
78
+ --------
79
+ attracting_components
80
+ is_attracting_component
81
+
82
+ """
83
+ return sum(1 for ac in attracting_components(G))
84
+
85
+
86
+ @not_implemented_for("undirected")
87
+ @nx._dispatchable
88
+ def is_attracting_component(G):
89
+ """Returns True if `G` consists of a single attracting component.
90
+
91
+ Parameters
92
+ ----------
93
+ G : DiGraph, MultiDiGraph
94
+ The graph to be analyzed.
95
+
96
+ Returns
97
+ -------
98
+ attracting : bool
99
+ True if `G` has a single attracting component. Otherwise, False.
100
+
101
+ Raises
102
+ ------
103
+ NetworkXNotImplemented
104
+ If the input graph is undirected.
105
+
106
+ See Also
107
+ --------
108
+ attracting_components
109
+ number_attracting_components
110
+
111
+ """
112
+ ac = list(attracting_components(G))
113
+ if len(ac) == 1:
114
+ return len(ac[0]) == len(G)
115
+ return False
evalkit_tf446/lib/python3.10/site-packages/networkx/algorithms/components/strongly_connected.py ADDED
@@ -0,0 +1,351 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ """Strongly connected components."""
2
+
3
+ import networkx as nx
4
+ from networkx.utils.decorators import not_implemented_for
5
+
6
+ __all__ = [
7
+ "number_strongly_connected_components",
8
+ "strongly_connected_components",
9
+ "is_strongly_connected",
10
+ "kosaraju_strongly_connected_components",
11
+ "condensation",
12
+ ]
13
+
14
+
15
+ @not_implemented_for("undirected")
16
+ @nx._dispatchable
17
+ def strongly_connected_components(G):
18
+ """Generate nodes in strongly connected components of graph.
19
+
20
+ Parameters
21
+ ----------
22
+ G : NetworkX Graph
23
+ A directed graph.
24
+
25
+ Returns
26
+ -------
27
+ comp : generator of sets
28
+ A generator of sets of nodes, one for each strongly connected
29
+ component of G.
30
+
31
+ Raises
32
+ ------
33
+ NetworkXNotImplemented
34
+ If G is undirected.
35
+
36
+ Examples
37
+ --------
38
+ Generate a sorted list of strongly connected components, largest first.
39
+
40
+ >>> G = nx.cycle_graph(4, create_using=nx.DiGraph())
41
+ >>> nx.add_cycle(G, [10, 11, 12])
42
+ >>> [
43
+ ... len(c)
44
+ ... for c in sorted(nx.strongly_connected_components(G), key=len, reverse=True)
45
+ ... ]
46
+ [4, 3]
47
+
48
+ If you only want the largest component, it's more efficient to
49
+ use max instead of sort.
50
+
51
+ >>> largest = max(nx.strongly_connected_components(G), key=len)
52
+
53
+ See Also
54
+ --------
55
+ connected_components
56
+ weakly_connected_components
57
+ kosaraju_strongly_connected_components
58
+
59
+ Notes
60
+ -----
61
+ Uses Tarjan's algorithm[1]_ with Nuutila's modifications[2]_.
62
+ Nonrecursive version of algorithm.
63
+
64
+ References
65
+ ----------
66
+ .. [1] Depth-first search and linear graph algorithms, R. Tarjan
67
+ SIAM Journal of Computing 1(2):146-160, (1972).
68
+
69
+ .. [2] On finding the strongly connected components in a directed graph.
70
+ E. Nuutila and E. Soisalon-Soinen
71
+ Information Processing Letters 49(1): 9-14, (1994)..
72
+
73
+ """
74
+ preorder = {}
75
+ lowlink = {}
76
+ scc_found = set()
77
+ scc_queue = []
78
+ i = 0 # Preorder counter
79
+ neighbors = {v: iter(G[v]) for v in G}
80
+ for source in G:
81
+ if source not in scc_found:
82
+ queue = [source]
83
+ while queue:
84
+ v = queue[-1]
85
+ if v not in preorder:
86
+ i = i + 1
87
+ preorder[v] = i
88
+ done = True
89
+ for w in neighbors[v]:
90
+ if w not in preorder:
91
+ queue.append(w)
92
+ done = False
93
+ break
94
+ if done:
95
+ lowlink[v] = preorder[v]
96
+ for w in G[v]:
97
+ if w not in scc_found:
98
+ if preorder[w] > preorder[v]:
99
+ lowlink[v] = min([lowlink[v], lowlink[w]])
100
+ else:
101
+ lowlink[v] = min([lowlink[v], preorder[w]])
102
+ queue.pop()
103
+ if lowlink[v] == preorder[v]:
104
+ scc = {v}
105
+ while scc_queue and preorder[scc_queue[-1]] > preorder[v]:
106
+ k = scc_queue.pop()
107
+ scc.add(k)
108
+ scc_found.update(scc)
109
+ yield scc
110
+ else:
111
+ scc_queue.append(v)
112
+
113
+
114
+ @not_implemented_for("undirected")
115
+ @nx._dispatchable
116
+ def kosaraju_strongly_connected_components(G, source=None):
117
+ """Generate nodes in strongly connected components of graph.
118
+
119
+ Parameters
120
+ ----------
121
+ G : NetworkX Graph
122
+ A directed graph.
123
+
124
+ Returns
125
+ -------
126
+ comp : generator of sets
127
+ A generator of sets of nodes, one for each strongly connected
128
+ component of G.
129
+
130
+ Raises
131
+ ------
132
+ NetworkXNotImplemented
133
+ If G is undirected.
134
+
135
+ Examples
136
+ --------
137
+ Generate a sorted list of strongly connected components, largest first.
138
+
139
+ >>> G = nx.cycle_graph(4, create_using=nx.DiGraph())
140
+ >>> nx.add_cycle(G, [10, 11, 12])
141
+ >>> [
142
+ ... len(c)
143
+ ... for c in sorted(
144
+ ... nx.kosaraju_strongly_connected_components(G), key=len, reverse=True
145
+ ... )
146
+ ... ]
147
+ [4, 3]
148
+
149
+ If you only want the largest component, it's more efficient to
150
+ use max instead of sort.
151
+
152
+ >>> largest = max(nx.kosaraju_strongly_connected_components(G), key=len)
153
+
154
+ See Also
155
+ --------
156
+ strongly_connected_components
157
+
158
+ Notes
159
+ -----
160
+ Uses Kosaraju's algorithm.
161
+
162
+ """
163
+ post = list(nx.dfs_postorder_nodes(G.reverse(copy=False), source=source))
164
+
165
+ seen = set()
166
+ while post:
167
+ r = post.pop()
168
+ if r in seen:
169
+ continue
170
+ c = nx.dfs_preorder_nodes(G, r)
171
+ new = {v for v in c if v not in seen}
172
+ seen.update(new)
173
+ yield new
174
+
175
+
176
+ @not_implemented_for("undirected")
177
+ @nx._dispatchable
178
+ def number_strongly_connected_components(G):
179
+ """Returns number of strongly connected components in graph.
180
+
181
+ Parameters
182
+ ----------
183
+ G : NetworkX graph
184
+ A directed graph.
185
+
186
+ Returns
187
+ -------
188
+ n : integer
189
+ Number of strongly connected components
190
+
191
+ Raises
192
+ ------
193
+ NetworkXNotImplemented
194
+ If G is undirected.
195
+
196
+ Examples
197
+ --------
198
+ >>> G = nx.DiGraph(
199
+ ... [(0, 1), (1, 2), (2, 0), (2, 3), (4, 5), (3, 4), (5, 6), (6, 3), (6, 7)]
200
+ ... )
201
+ >>> nx.number_strongly_connected_components(G)
202
+ 3
203
+
204
+ See Also
205
+ --------
206
+ strongly_connected_components
207
+ number_connected_components
208
+ number_weakly_connected_components
209
+
210
+ Notes
211
+ -----
212
+ For directed graphs only.
213
+ """
214
+ return sum(1 for scc in strongly_connected_components(G))
215
+
216
+
217
+ @not_implemented_for("undirected")
218
+ @nx._dispatchable
219
+ def is_strongly_connected(G):
220
+ """Test directed graph for strong connectivity.
221
+
222
+ A directed graph is strongly connected if and only if every vertex in
223
+ the graph is reachable from every other vertex.
224
+
225
+ Parameters
226
+ ----------
227
+ G : NetworkX Graph
228
+ A directed graph.
229
+
230
+ Returns
231
+ -------
232
+ connected : bool
233
+ True if the graph is strongly connected, False otherwise.
234
+
235
+ Examples
236
+ --------
237
+ >>> G = nx.DiGraph([(0, 1), (1, 2), (2, 3), (3, 0), (2, 4), (4, 2)])
238
+ >>> nx.is_strongly_connected(G)
239
+ True
240
+ >>> G.remove_edge(2, 3)
241
+ >>> nx.is_strongly_connected(G)
242
+ False
243
+
244
+ Raises
245
+ ------
246
+ NetworkXNotImplemented
247
+ If G is undirected.
248
+
249
+ See Also
250
+ --------
251
+ is_weakly_connected
252
+ is_semiconnected
253
+ is_connected
254
+ is_biconnected
255
+ strongly_connected_components
256
+
257
+ Notes
258
+ -----
259
+ For directed graphs only.
260
+ """
261
+ if len(G) == 0:
262
+ raise nx.NetworkXPointlessConcept(
263
+ """Connectivity is undefined for the null graph."""
264
+ )
265
+
266
+ return len(next(strongly_connected_components(G))) == len(G)
267
+
268
+
269
+ @not_implemented_for("undirected")
270
+ @nx._dispatchable(returns_graph=True)
271
+ def condensation(G, scc=None):
272
+ """Returns the condensation of G.
273
+
274
+ The condensation of G is the graph with each of the strongly connected
275
+ components contracted into a single node.
276
+
277
+ Parameters
278
+ ----------
279
+ G : NetworkX DiGraph
280
+ A directed graph.
281
+
282
+ scc: list or generator (optional, default=None)
283
+ Strongly connected components. If provided, the elements in
284
+ `scc` must partition the nodes in `G`. If not provided, it will be
285
+ calculated as scc=nx.strongly_connected_components(G).
286
+
287
+ Returns
288
+ -------
289
+ C : NetworkX DiGraph
290
+ The condensation graph C of G. The node labels are integers
291
+ corresponding to the index of the component in the list of
292
+ strongly connected components of G. C has a graph attribute named
293
+ 'mapping' with a dictionary mapping the original nodes to the
294
+ nodes in C to which they belong. Each node in C also has a node
295
+ attribute 'members' with the set of original nodes in G that
296
+ form the SCC that the node in C represents.
297
+
298
+ Raises
299
+ ------
300
+ NetworkXNotImplemented
301
+ If G is undirected.
302
+
303
+ Examples
304
+ --------
305
+ Contracting two sets of strongly connected nodes into two distinct SCC
306
+ using the barbell graph.
307
+
308
+ >>> G = nx.barbell_graph(4, 0)
309
+ >>> G.remove_edge(3, 4)
310
+ >>> G = nx.DiGraph(G)
311
+ >>> H = nx.condensation(G)
312
+ >>> H.nodes.data()
313
+ NodeDataView({0: {'members': {0, 1, 2, 3}}, 1: {'members': {4, 5, 6, 7}}})
314
+ >>> H.graph["mapping"]
315
+ {0: 0, 1: 0, 2: 0, 3: 0, 4: 1, 5: 1, 6: 1, 7: 1}
316
+
317
+ Contracting a complete graph into one single SCC.
318
+
319
+ >>> G = nx.complete_graph(7, create_using=nx.DiGraph)
320
+ >>> H = nx.condensation(G)
321
+ >>> H.nodes
322
+ NodeView((0,))
323
+ >>> H.nodes.data()
324
+ NodeDataView({0: {'members': {0, 1, 2, 3, 4, 5, 6}}})
325
+
326
+ Notes
327
+ -----
328
+ After contracting all strongly connected components to a single node,
329
+ the resulting graph is a directed acyclic graph.
330
+
331
+ """
332
+ if scc is None:
333
+ scc = nx.strongly_connected_components(G)
334
+ mapping = {}
335
+ members = {}
336
+ C = nx.DiGraph()
337
+ # Add mapping dict as graph attribute
338
+ C.graph["mapping"] = mapping
339
+ if len(G) == 0:
340
+ return C
341
+ for i, component in enumerate(scc):
342
+ members[i] = component
343
+ mapping.update((n, i) for n in component)
344
+ number_of_components = i + 1
345
+ C.add_nodes_from(range(number_of_components))
346
+ C.add_edges_from(
347
+ (mapping[u], mapping[v]) for u, v in G.edges() if mapping[u] != mapping[v]
348
+ )
349
+ # Add a list of members (ie original nodes) to each node (ie scc) in C.
350
+ nx.set_node_attributes(C, members, "members")
351
+ return C