Parent Directory
| 
Revision Log
Revision 1.1.1.1 - (view) (download)
| 1 : | efrank | 1.1 | Version 23.7 January 2005 |
| 2 : | [1075953] added franf to fix problems on AMD64. | ||
| 3 : | [1027480] this should have been fixed before. | ||
| 4 : | [scipy] fixed reshape to convert dimensions to PyArray_LONG to avoid typecasting problem. | ||
| 5 : | [eric jones] added empty to generate unitialized array | ||
| 6 : | [user ] added support for vecLib on MacOS X in setup.py | ||
| 7 : | [unreported] added UfuncType to Numeric to Parallel ArrayType | ||
| 8 : | |||
| 9 : | Version 23.6 | ||
| 10 : | [unreported] fixed argmin to work with Unsigned Array Types | ||
| 11 : | [unreported] fixed UINT_to_XXX methods so input is assumed unsigned int | ||
| 12 : | [unreported] fixed UINT conversion rules and inappropriate casting to unsigned integers | ||
| 13 : | from signed integers | ||
| 14 : | [build ] fixed tarball so same tarball for multiple versions. | ||
| 15 : | |||
| 16 : | |||
| 17 : | Version 23.5 Sept 2004 | ||
| 18 : | [unreported] changed arrayfnsmodule to look for PyArray_LONG on input not PyArray_INT | ||
| 19 : | [perez ] fixed so that matrixmultiply is always dot (even if BLAS optimized) | ||
| 20 : | |||
| 21 : | Version 23.4 Sept 2004 | ||
| 22 : | [unreported] fixed setup.py file so ATLAS not default | ||
| 23 : | [unreported] altered coercion so that LONG is not cast to INT safely | ||
| 24 : | |||
| 25 : | Version 23.3 June 2004 | ||
| 26 : | [unreported] misspelled UFunc_Type in ufunc API | ||
| 27 : | [ 843510 ] ubyte abs giving incorrect results | ||
| 28 : | [ 919540 ] added weakreferences to Numeric Object | ||
| 29 : | [unreported] fixed segfault due to unchecked NULL in get_type | ||
| 30 : | [unreported] added faster array(a) when a is already an array | ||
| 31 : | [unreported] a can now be interpreted as a scalar as long as it only | ||
| 32 : | has one element (instead of just zero-dimensions). | ||
| 33 : | changed compress keyword from dimension to axis to match | ||
| 34 : | everyother function in the library. | ||
| 35 : | |||
| 36 : | Version 23.1 August 2003 | ||
| 37 : | [unreported] fixed assignment to matrix. | ||
| 38 : | [unreported] added check so that non-character array cannot be interpreted | ||
| 39 : | as a character buffer. | ||
| 40 : | [unreported] array([m,m,m]) when m is a rank-0 array now works as expected. | ||
| 41 : | Patches: | ||
| 42 : | [ 782001 ] Add true and floor divide to MA | ||
| 43 : | [ 781215 ] true_divide bug | ||
| 44 : | [ 776991 ] multiplying small UINT values gives random results | ||
| 45 : | [ 776467 ] remove string exceptions | ||
| 46 : | |||
| 47 : | Version 23 March 2003 | ||
| 48 : | |||
| 49 : | Important notice: | ||
| 50 : | |||
| 51 : | Two packages have been removed from optional ones: PropertiedClasses, kinds. | ||
| 52 : | MA has been rewritten to use standard property and will not work for | ||
| 53 : | ancient Pythons. (Pre 2.1, I think). Use the MA / Propertied Classes | ||
| 54 : | from Numeric 22 if you can't use this one. | ||
| 55 : | |||
| 56 : | The kinds package (subject of PEP-0242) will be released as a separate | ||
| 57 : | package shortly. PEP-0242 was withdrawn because this facility did not seem | ||
| 58 : | to be worth putting in the standard library, but kinds is correct as is. | ||
| 59 : | |||
| 60 : | [ 695200 ] Richard Everson (R.M.Everson@exeter.ac.uk) has donated a dotblas | ||
| 61 : | Package that gets Numeric to use optimized BLAS libraries for dot | ||
| 62 : | innerproduct, and vdot --- a conjugate vector dot product he introduced. | ||
| 63 : | setup.py must be altered by the user in a manner similar to the alterations | ||
| 64 : | to use optimized BLAS for LinearAlgebra | ||
| 65 : | |||
| 66 : | [675777] new-style classes as objects in a sequence were not being detected | ||
| 67 : | correctly by array_objecttype. Corrected array_objecttype to handle | ||
| 68 : | them. (Oliphant) | ||
| 69 : | |||
| 70 : | [contribution] Fernando Perez has donated a revised version of the tutorial | ||
| 71 : | file view.py that seems to be less likely to hang the interpreter. | ||
| 72 : | |||
| 73 : | [68392923] dimensions+scalar -> crash (jneb) | ||
| 74 : | Found divergent value of MAX_DIMS in ufuncobject.c; | ||
| 75 : | The value was 20 there and 40 in two other places. But 40 is ludicrous, | ||
| 76 : | we would never have that much memory available. Changed | ||
| 77 : | them all to 30. | ||
| 78 : | |||
| 79 : | [ unreported ] Changed PY_VERSION_HEX check for version 2.2 | ||
| 80 : | to 0x0202000 as it should have been | ||
| 81 : | so that true_division numeric ops can be supported | ||
| 82 : | |||
| 83 : | [ 614808 ] Inconsistent use of tabs and spaces | ||
| 84 : | Fixed as suggested by Jimmy Retzlaff | ||
| 85 : | LinearAlgebra.py | ||
| 86 : | Matrix.py | ||
| 87 : | RNG/__init__.py | ||
| 88 : | RNG/Statistics.py | ||
| 89 : | |||
| 90 : | [ 621032 ] needless work in multiarraymodule.c | ||
| 91 : | Fixes suggested by Greg Smith applied. | ||
| 92 : | Also recoded OBJECT_DotProduct to eliminate a warning error. | ||
| 93 : | |||
| 94 : | [ 630584 ] generalized_inverse of complex array | ||
| 95 : | Fix suggested by Greg Smith applied. | ||
| 96 : | |||
| 97 : | [ 652061 ] PyArray_As2D doesn't check pointer. | ||
| 98 : | Fix suggested by Andrea Riciputi applied. | ||
| 99 : | |||
| 100 : | [ 655512 ] inverse_real_fft incorrect many sizes | ||
| 101 : | Fix given by mbriest applied. | ||
| 102 : | |||
| 103 : | [unreported] a.real increased reference count of a and raised error when | ||
| 104 : | a is not complex. | ||
| 105 : | Fixed to apparent intended behavior of returning an array | ||
| 106 : | with the same data. (Oliphant) | ||
| 107 : | |||
| 108 : | Patch for 64-bit machines applied. Appears to work ok on 32 bit but don't | ||
| 109 : | have the machine to test the patch.(Dubois) | ||
| 110 : | |||
| 111 : | [ 627771 ] matrixmultiply broken for non-contig (fixed, test case added) | ||
| 112 : | (Greg Smith) | ||
| 113 : | |||
| 114 : | [unreported] Fixed ArrayPrinter when NaN's show up in Float32 | ||
| 115 : | precision. | ||
| 116 : | |||
| 117 : | [ 545336 ] Bug in RandomArray.randint | ||
| 118 : | Changed the function ranf to type double (Chuck Harris) | ||
| 119 : | Harris is probably right that all floats should be double in this module | ||
| 120 : | but it may be this has performance or storage consequences that would | ||
| 121 : | bite somebody. I support RNG, not this one. -- Dubois | ||
| 122 : | |||
| 123 : | Version 22.0 August , 2002 | ||
| 124 : | a. Changed multiarraymodule functions to accept keywords where documentation implies | ||
| 125 : | it through the use of optional variables. | ||
| 126 : | Specifically in multiarray: zeros, take, transpose, repeat, | ||
| 127 : | set_string_function, cross_correlate. | ||
| 128 : | in ufuncobject: reduce and accumulate now take keyword arguments | ||
| 129 : | for the optional axis argument. | ||
| 130 : | b. Added support for unsigned shorts 'w' and unsigned ints 'u' | ||
| 131 : | -- Travis Oliphant with help from Darren Hart and F. Oliver Gathmann. | ||
| 132 : | Increased max permissible iterations in SVD for supplied lapack. -- Dubois | ||
| 133 : | Recoded RandomArray.randint to try to see if we can work around bug | ||
| 134 : | on some platforms. -- Dubois | ||
| 135 : | |||
| 136 : | Version 21.3 June 8, 2002 | ||
| 137 : | Fixed bugs: | ||
| 138 : | [ #557927 ] fixed matrix slice assignment | ||
| 139 : | [ #552922 ] added check for correct datatype in .astype() method. | ||
| 140 : | Created new API PyArray_ValidType to handle this check | ||
| 141 : | here as well as in multiarraymodule.c | ||
| 142 : | [ #551808 ] fixed segfault with unicode array (Travis O.) | ||
| 143 : | [ #559511 ] MLab.std now works for axis != 0 (Travis O.) | ||
| 144 : | [ #542979 ] sum returns exception tuple | ||
| 145 : | |||
| 146 : | [ #528328 ] true division operators used to return single precision on division of | ||
| 147 : | integers and longs --- now defaults to double precision (but only on int and | ||
| 148 : | long division --- still single-precision for ubyte, short, and byte division. | ||
| 149 : | [ none ] arange(start, end, step) slightly different near end points than | ||
| 150 : | start + arange(0, N)*step where N is the length. | ||
| 151 : | [ none ] a = zeros(2,'D'); a[0] = array(0.0+0.6j) would not work. (Assigning a | ||
| 152 : | rank-0 array did not work for CFLOAT_setitem or CDOUBLE_setitem. | ||
| 153 : | [ 530688 ] Python crash when transposing array (Walter Moreira) | ||
| 154 : | |||
| 155 : | Version 21.0 March 13, 2002 | ||
| 156 : | Fixed bugs: | ||
| 157 : | [ #482603 ] Memory leak in MA/Numeric/Python | ||
| 158 : | Reported by Reggie Dugard. Turned out to be | ||
| 159 : | *two* memory leaks in one case in a routine in Numeric, | ||
| 160 : | array_objectype. (Dubois) | ||
| 161 : | [ none ] if vals was a null-array array([]) putmask and put would | ||
| 162 : | crash. Fixed with check. | ||
| 163 : | [ #469951 ] n = n1[0] gives array which shares dimension of n1 array. | ||
| 164 : | This causes bugs if shape of n1 is changed (n didn't used | ||
| 165 : | to have it's own dimensions array (Travis Oliphant) | ||
| 166 : | [ #514588 ] MLab.cov(x,x) != MLab.cov(x) (Travis Oliphant) | ||
| 167 : | [ #518702 ] segfault when invalid typecode for asarray (Travis Oliphant) | ||
| 168 : | [ #497530 ] MA __getitem__ prevents 0 len arrays (Reggie Duggard) | ||
| 169 : | [ #508363 ] outerproduct of noncontiguous arrays (Martin Wiechert) | ||
| 170 : | [ #513010 ] memory leak in comparisons (Byran Nollett) | ||
| 171 : | [ #512223 ] Character typecode not defined (Jochen Kupper) | ||
| 172 : | [ #500784 ] MLab.py diff error (anonymous, fixed by Dubois) | ||
| 173 : | [ #503741 ] accuracy of MLab.std(x) (Katsunori Waragai) | ||
| 174 : | [ #507568 ] overlapping copy a[2:5] = a[3:6] | ||
| 175 : | Change uses of memcpy to memmove which allows overlaps. | ||
| 176 : | [ numpy-Patches-499722 ] size of buffer created from array is bad (Michel Sanner). | ||
| 177 : | [ #502186 ] a BUG in RandomArray.normal (introduced by last bug fix in 20.3) | ||
| 178 : | (Katsunori Waragai). | ||
| 179 : | |||
| 180 : | Fixed errors for Mac (Jack Jensen). | ||
| 181 : | |||
| 182 : | Make rpm's properly, better Windows installers. (Gerard Vermeulen) | ||
| 183 : | Added files setup.cfg; setup calculates rpm_install.sh to use current Python. | ||
| 184 : | New setup.py, eliminate setup_all.py. Use os.path.join everywhere. Revision in b6 | ||
| 185 : | added file README.RPM, further improvements. | ||
| 186 : | |||
| 187 : | Implement true division operations for Python 2.2. (Bruce Sherwood) | ||
| 188 : | Note: true division of all integer types results in an array of floats, | ||
| 189 : | not doubles. This decision is arbitrary and there are arguments either way, | ||
| 190 : | so users of this new feature should be aware that the decision may | ||
| 191 : | change in the future. | ||
| 192 : | |||
| 193 : | New functions in Numeric; they work on any sequence a that can be converted to a | ||
| 194 : | Numeric array. Similar change to average in MA. (Dubois) | ||
| 195 : | |||
| 196 : | def rank (a): | ||
| 197 : | "Get the rank of a (the number of dimensions, not a matrix rank)" | ||
| 198 : | |||
| 199 : | def shape (a): | ||
| 200 : | "Get the shape of a" | ||
| 201 : | |||
| 202 : | def size (a, axis=None): | ||
| 203 : | "Get the number of elements in a, or along a certain axis." | ||
| 204 : | |||
| 205 : | def average (a, axis=0, weights=None, returned = 0): | ||
| 206 : | """average(a, axis=0, weights=None) | ||
| 207 : | Computes average along indicated axis. | ||
| 208 : | If axis is None, average over the entire array. | ||
| 209 : | Inputs can be integer or floating types; result is type Float. | ||
| 210 : | |||
| 211 : | If weights are given, result is: | ||
| 212 : | sum(a*weights)/(sum(weights)) | ||
| 213 : | weights must have a's shape or be the 1-d with length the size | ||
| 214 : | of a in the given axis. Integer weights are converted to Float. | ||
| 215 : | |||
| 216 : | Not supplying weights is equivalent to supply weights that are | ||
| 217 : | all 1. | ||
| 218 : | |||
| 219 : | If returned, return a tuple: the result and the sum of the weights | ||
| 220 : | or count of values. The shape of these two results will be the same. | ||
| 221 : | |||
| 222 : | raises ZeroDivisionError if appropriate when result is scalar. | ||
| 223 : | (The version in MA does not -- it returns masked values). | ||
| 224 : | """ | ||
| 225 : | |||
| 226 : |
| MCS Webmaster | ViewVC Help |
| Powered by ViewVC 1.0.3 |