From 1fbdaa4d49acf90d2124aef8810d723ddbb0dad5 Mon Sep 17 00:00:00 2001 From: graehl Date: Sat, 24 Jul 2010 21:17:51 +0000 Subject: sparse_vector non-const [], consistently use default ctor, operator bool instead of explicit 0 git-svn-id: https://ws10smt.googlecode.com/svn/trunk@398 ec762483-ff6d-05da-a07a-a48fb63a330f --- decoder/feature_vector.h | 14 ++++ decoder/have_64_bits.h | 17 +++++ decoder/murmur_hash.h | 187 +++++++++++++++++++++++++++++++++++++++++++++++ 3 files changed, 218 insertions(+) create mode 100755 decoder/feature_vector.h create mode 100755 decoder/have_64_bits.h create mode 100755 decoder/murmur_hash.h diff --git a/decoder/feature_vector.h b/decoder/feature_vector.h new file mode 100755 index 00000000..1b272506 --- /dev/null +++ b/decoder/feature_vector.h @@ -0,0 +1,14 @@ +#ifndef _FEATURE_VECTOR_H_ +#define _FEATURE_VECTOR_H_ + +#include +#include "sparse_vector.h" +#include "fdict.h" + +typedef double Featval; +typedef SparseVectorList FeatureVectorList; +typedef SparseVector FeatureVector; +typedef SparseVector WeightVector; +typedef std::vector DenseWeightVector; + +#endif diff --git a/decoder/have_64_bits.h b/decoder/have_64_bits.h new file mode 100755 index 00000000..d1e6064f --- /dev/null +++ b/decoder/have_64_bits.h @@ -0,0 +1,17 @@ +#ifndef HAVE_64_BITS_H +#define HAVE_64_BITS_H + +#include + +#undef HAVE_64_BITS + +#if INTPTR_MAX == INT32_MAX +# define HAVE_64_BITS 0 +#elif INTPTR_MAX >= INT64_MAX +# define HAVE_64_BITS 1 +#else +# error "couldn't tell if HAVE_64_BITS from INTPTR_MAX INT32_MAX INT64_MAX" +#endif + + +#endif diff --git a/decoder/murmur_hash.h b/decoder/murmur_hash.h new file mode 100755 index 00000000..8ddad339 --- /dev/null +++ b/decoder/murmur_hash.h @@ -0,0 +1,187 @@ +#ifndef _MURMUR_HASH_H_ +#define _MURMUR_HASH_H_ + +//NOTE: quite fast, nice collision properties, but endian dependent hash values + +#include + +// MurmurHash2, by Austin Appleby + +static const uint32_t DEFAULT_SEED=2654435769U; + +#if sizeof(void *)==8 +typedef uint64_t MurmurInt; +//MurmurInt MurmurHash(void const *key, int len, uint32_t seed=DEFAULT_SEED); + +inline uint64_t MurmurHash64( const void * key, int len, unsigned int seed=DEFAULT_SEED ) +{ + const uint64_t m = 0xc6a4a7935bd1e995; + const int r = 47; + + uint64_t h = seed ^ (len * m); + + const uint64_t * data = (const uint64_t *)key; + const uint64_t * end = data + (len/8); + + while(data != end) + { + uint64_t k = *data++; + + k *= m; + k ^= k >> r; + k *= m; + + h ^= k; + h *= m; + } + + const unsigned char * data2 = (const unsigned char*)data; + + switch(len & 7) + { + case 7: h ^= uint64_t(data2[6]) << 48; + case 6: h ^= uint64_t(data2[5]) << 40; + case 5: h ^= uint64_t(data2[4]) << 32; + case 4: h ^= uint64_t(data2[3]) << 24; + case 3: h ^= uint64_t(data2[2]) << 16; + case 2: h ^= uint64_t(data2[1]) << 8; + case 1: h ^= uint64_t(data2[0]); + h *= m; + }; + + h ^= h >> r; + h *= m; + h ^= h >> r; + + return h; +} + +inline uint32_t MurmurHash2(void const *key, int len, uint32_t seed=DEFAULT_SEED) +{ + return (uint32_t) MurmurHash64(key,len,seed); +} + +typedef uint64_t MurmurInt; +inline MurmurInt MurmurHash(void const *key, int len, uint32_t seed=DEFAULT_SEED) +{ + return MurmurHash64(key,len,seed); +} + +#else // 32-bit +typedef uint32_t MurmurInt; + + +// Note - This code makes a few assumptions about how your machine behaves - +// 1. We can read a 4-byte value from any address without crashing +// 2. sizeof(int) == 4 +inline uint32_t MurmurHash2 ( const void * key, int len, uint32_t seed=DEFAULT_SEED) +{ + // 'm' and 'r' are mixing constants generated offline. + // They're not really 'magic', they just happen to work well. + + const uint32_t m = 0x5bd1e995; + const int r = 24; + + // Initialize the hash to a 'random' value + + uint32_t h = seed ^ len; + + // Mix 4 bytes at a time into the hash + + const unsigned char * data = (const unsigned char *)key; + + while(len >= 4) + { + uint32_t k = *(uint32_t *)data; + + k *= m; + k ^= k >> r; + k *= m; + + h *= m; + h ^= k; + + data += 4; + len -= 4; + } + + // Handle the last few bytes of the input array + + switch(len) + { + case 3: h ^= data[2] << 16; + case 2: h ^= data[1] << 8; + case 1: h ^= data[0]; + h *= m; + }; + + // Do a few final mixes of the hash to ensure the last few + // bytes are well-incorporated. + + h ^= h >> 13; + h *= m; + h ^= h >> 15; + + return h; +} + +inline MurmurInt MurmurHash ( const void * key, int len, uint32_t seed=DEFAULT_SEED) { + return MurmurHash32(key,len,seed); +} + +// 64-bit hash for 32-bit platforms + +inline uint64_t MurmurHash64 ( const void * key, int len, uint32_t seed=DEFAULT_SEED) +{ + const uint32_t m = 0x5bd1e995; + const int r = 24; + + uint32_t h1 = seed ^ len; + uint32_t h2 = 0; + + const uint32_t * data = (const uint32_t *)key; + + while(len >= 8) + { + uint32_t k1 = *data++; + k1 *= m; k1 ^= k1 >> r; k1 *= m; + h1 *= m; h1 ^= k1; + len -= 4; + + uint32_t k2 = *data++; + k2 *= m; k2 ^= k2 >> r; k2 *= m; + h2 *= m; h2 ^= k2; + len -= 4; + } + + if(len >= 4) + { + uint32_t k1 = *data++; + k1 *= m; k1 ^= k1 >> r; k1 *= m; + h1 *= m; h1 ^= k1; + len -= 4; + } + + switch(len) + { + case 3: h2 ^= ((unsigned char*)data)[2] << 16; + case 2: h2 ^= ((unsigned char*)data)[1] << 8; + case 1: h2 ^= ((unsigned char*)data)[0]; + h2 *= m; + }; + + h1 ^= h2 >> 18; h1 *= m; + h2 ^= h1 >> 22; h2 *= m; + h1 ^= h2 >> 17; h1 *= m; + h2 ^= h1 >> 19; h2 *= m; + + uint64_t h = h1; + + h = (h << 32) | h2; + + return h; +} + +#endif + +#endif -- cgit v1.2.3