follow upstream

devc/x64/glove/glove.c

@@ -48,6 +48,7 @@ typedef struct cooccur_rec {
 	real val;
 } CREC;
 
+int write_header=0; //0=no, 1=yes; writes vocab_size/vector_size as first line for use with some libraries, such as gensim.
 int verbose = 2; // 0, 1, or 2
 int use_unk_vec = 1; // 0 or 1
 int num_threads = 8; // pthreads
@@ -93,8 +94,7 @@ inline real check_nan(real update) {
 	if (isnan(update) || isinf(update)) {
 		fprintf(stderr, "\ncaught NaN in update");
 		return 0.;
-	}
-	else {
+	} else {
 		return update;
 	}
 }
@@ -231,6 +231,7 @@ int save_params(int nb_iter) {
 		fid = fopen(vocab_file, "r");
 		sprintf(format, "%%%ds", MAX_STRING_LENGTH);
 		if (fid == NULL) { fprintf(stderr, "Unable to open file %s.\n", vocab_file); return 1; }
+		if (write_header) fprintf(fout, "%ld %d\n", vocab_size, vector_size);
 		for (a = 0; a < vocab_size; a++) {
 			if (fscanf(fid, format, word) == 0) return 1;
 			// input vocab cannot contain special <unk> keyword
@@ -381,6 +382,8 @@ int main(int argc, char **argv) {
 		printf("Usage options:\n");
 		printf("\t-verbose <int>\n");
 		printf("\t\tSet verbosity: 0, 1, or 2 (default)\n");
+		printf("\t-write-header <int>\n");
+		printf("\t\tIf 1, write vocab_size/vector_size as first line. Do nothing if 0 (default).\n");
 		printf("\t-vector-size <int>\n");
 		printf("\t\tDimension of word vector representations (excluding bias term); default 50\n");
 		printf("\t-threads <int>\n");
@@ -415,8 +418,8 @@ int main(int argc, char **argv) {
 		printf("\nExample usage:\n");
 		printf("./glove -input-file cooccurrence.shuf.bin -vocab-file vocab.txt -save-file vectors -gradsq-file gradsq -verbose 2 -vector-size 100 -threads 16 -alpha 0.75 -x-max 100.0 -eta 0.05 -binary 2 -model 2\n\n");
 		result = 0;
-	}
-	else {
+	} else {
+		if ((i = find_arg((char *)"-write-header", argc, argv)) > 0) write_header = atoi(argv[i + 1]);
 		if ((i = find_arg((char *)"-verbose", argc, argv)) > 0) verbose = atoi(argv[i + 1]);
 		if ((i = find_arg((char *)"-vector-size", argc, argv)) > 0) vector_size = atoi(argv[i + 1]);
 		if ((i = find_arg((char *)"-iter", argc, argv)) > 0) num_iter = atoi(argv[i + 1]);

devc/x64/glove/glove.layout

@@ -2,7 +2,7 @@
 Order=0
 Focused=0
 [Editor_0]
-CursorCol=1
-CursorRow=453
-TopLine=427
+CursorCol=2
+CursorRow=462
+TopLine=409
 LeftChar=1

devc/x86/glove/glove.c

@@ -48,6 +48,7 @@ typedef struct cooccur_rec {
 	real val;
 } CREC;
 
+int write_header=0; //0=no, 1=yes; writes vocab_size/vector_size as first line for use with some libraries, such as gensim.
 int verbose = 2; // 0, 1, or 2
 int use_unk_vec = 1; // 0 or 1
 int num_threads = 8; // pthreads
@@ -93,8 +94,7 @@ inline real check_nan(real update) {
 	if (isnan(update) || isinf(update)) {
 		fprintf(stderr, "\ncaught NaN in update");
 		return 0.;
-	}
-	else {
+	} else {
 		return update;
 	}
 }
@@ -231,6 +231,7 @@ int save_params(int nb_iter) {
 		fid = fopen(vocab_file, "r");
 		sprintf(format, "%%%ds", MAX_STRING_LENGTH);
 		if (fid == NULL) { fprintf(stderr, "Unable to open file %s.\n", vocab_file); return 1; }
+		if (write_header) fprintf(fout, "%ld %d\n", vocab_size, vector_size);
 		for (a = 0; a < vocab_size; a++) {
 			if (fscanf(fid, format, word) == 0) return 1;
 			// input vocab cannot contain special <unk> keyword
@@ -381,6 +382,8 @@ int main(int argc, char **argv) {
 		printf("Usage options:\n");
 		printf("\t-verbose <int>\n");
 		printf("\t\tSet verbosity: 0, 1, or 2 (default)\n");
+		printf("\t-write-header <int>\n");
+		printf("\t\tIf 1, write vocab_size/vector_size as first line. Do nothing if 0 (default).\n");
 		printf("\t-vector-size <int>\n");
 		printf("\t\tDimension of word vector representations (excluding bias term); default 50\n");
 		printf("\t-threads <int>\n");
@@ -415,8 +418,8 @@ int main(int argc, char **argv) {
 		printf("\nExample usage:\n");
 		printf("./glove -input-file cooccurrence.shuf.bin -vocab-file vocab.txt -save-file vectors -gradsq-file gradsq -verbose 2 -vector-size 100 -threads 16 -alpha 0.75 -x-max 100.0 -eta 0.05 -binary 2 -model 2\n\n");
 		result = 0;
-	}
-	else {
+	} else {
+		if ((i = find_arg((char *)"-write-header", argc, argv)) > 0) write_header = atoi(argv[i + 1]);
 		if ((i = find_arg((char *)"-verbose", argc, argv)) > 0) verbose = atoi(argv[i + 1]);
 		if ((i = find_arg((char *)"-vector-size", argc, argv)) > 0) vector_size = atoi(argv[i + 1]);
 		if ((i = find_arg((char *)"-iter", argc, argv)) > 0) num_iter = atoi(argv[i + 1]);

devc/x86/glove/glove.layout

@@ -4,5 +4,5 @@ Focused=0
 [Editor_0]
 CursorCol=55
 CursorRow=19
-TopLine=7
+TopLine=10
 LeftChar=1

vs/x64/glove/glove/glove.c

@@ -48,6 +48,7 @@ typedef struct cooccur_rec {
 	real val;
 } CREC;
 
+int write_header=0; //0=no, 1=yes; writes vocab_size/vector_size as first line for use with some libraries, such as gensim.
 int verbose = 2; // 0, 1, or 2
 int use_unk_vec = 1; // 0 or 1
 int num_threads = 8; // pthreads
@@ -93,8 +94,7 @@ inline real check_nan(real update) {
 	if (isnan(update) || isinf(update)) {
 		fprintf(stderr, "\ncaught NaN in update");
 		return 0.;
-	}
-	else {
+	} else {
 		return update;
 	}
 }
@@ -231,6 +231,7 @@ int save_params(int nb_iter) {
 		fid = fopen(vocab_file, "r");
 		sprintf(format, "%%%ds", MAX_STRING_LENGTH);
 		if (fid == NULL) { fprintf(stderr, "Unable to open file %s.\n", vocab_file); return 1; }
+		if (write_header) fprintf(fout, "%ld %d\n", vocab_size, vector_size);
 		for (a = 0; a < vocab_size; a++) {
 			if (fscanf(fid, format, word) == 0) return 1;
 			// input vocab cannot contain special <unk> keyword
@@ -381,6 +382,8 @@ int main(int argc, char **argv) {
 		printf("Usage options:\n");
 		printf("\t-verbose <int>\n");
 		printf("\t\tSet verbosity: 0, 1, or 2 (default)\n");
+		printf("\t-write-header <int>\n");
+		printf("\t\tIf 1, write vocab_size/vector_size as first line. Do nothing if 0 (default).\n");
 		printf("\t-vector-size <int>\n");
 		printf("\t\tDimension of word vector representations (excluding bias term); default 50\n");
 		printf("\t-threads <int>\n");
@@ -415,8 +418,8 @@ int main(int argc, char **argv) {
 		printf("\nExample usage:\n");
 		printf("./glove -input-file cooccurrence.shuf.bin -vocab-file vocab.txt -save-file vectors -gradsq-file gradsq -verbose 2 -vector-size 100 -threads 16 -alpha 0.75 -x-max 100.0 -eta 0.05 -binary 2 -model 2\n\n");
 		result = 0;
-	}
-	else {
+	} else {
+		if ((i = find_arg((char *)"-write-header", argc, argv)) > 0) write_header = atoi(argv[i + 1]);
 		if ((i = find_arg((char *)"-verbose", argc, argv)) > 0) verbose = atoi(argv[i + 1]);
 		if ((i = find_arg((char *)"-vector-size", argc, argv)) > 0) vector_size = atoi(argv[i + 1]);
 		if ((i = find_arg((char *)"-iter", argc, argv)) > 0) num_iter = atoi(argv[i + 1]);

vs/x64/glove/glove/x64/Release/glove.log

@@ -29,24 +29,28 @@ glove.c(231): warning C4996: 'fopen': This function or variable may be unsafe. C
   C:\Program Files (x86)\Windows Kits\10\Include\10.0.15063.0\ucrt\stdio.h(207): note: 参见“fopen”的声明
 glove.c(232): warning C4996: 'sprintf': This function or variable may be unsafe. Consider using sprintf_s instead. To disable deprecation, use _CRT_SECURE_NO_WARNINGS. See online help for details.
   C:\Program Files (x86)\Windows Kits\10\Include\10.0.15063.0\ucrt\stdio.h(1772): note: 参见“sprintf”的声明
-glove.c(235): warning C4996: 'fscanf': This function or variable may be unsafe. Consider using fscanf_s instead. To disable deprecation, use _CRT_SECURE_NO_WARNINGS. See online help for details.
+glove.c(234): warning C4477: “fprintf”: 格式字符串“%ld”需要类型“long”的参数，但可变参数 1 拥有了类型“__int64”
+  glove.c(234): note: 请考虑在格式字符串中使用“%lld”
+  glove.c(234): note: 请考虑在格式字符串中使用“%Id”
+  glove.c(234): note: 请考虑在格式字符串中使用“%I64d”
+glove.c(236): warning C4996: 'fscanf': This function or variable may be unsafe. Consider using fscanf_s instead. To disable deprecation, use _CRT_SECURE_NO_WARNINGS. See online help for details.
   C:\Program Files (x86)\Windows Kits\10\Include\10.0.15063.0\ucrt\stdio.h(1195): note: 参见“fscanf”的声明
-glove.c(254): warning C4996: 'fscanf': This function or variable may be unsafe. Consider using fscanf_s instead. To disable deprecation, use _CRT_SECURE_NO_WARNINGS. See online help for details.
+glove.c(255): warning C4996: 'fscanf': This function or variable may be unsafe. Consider using fscanf_s instead. To disable deprecation, use _CRT_SECURE_NO_WARNINGS. See online help for details.
   C:\Program Files (x86)\Windows Kits\10\Include\10.0.15063.0\ucrt\stdio.h(1195): note: 参见“fscanf”的声明
-glove.c(262): warning C4244: “初始化”: 从“__int64”转换到“int”，可能丢失数据
-glove.c(303): warning C4996: 'fopen': This function or variable may be unsafe. Consider using fopen_s instead. To disable deprecation, use _CRT_SECURE_NO_WARNINGS. See online help for details.
+glove.c(263): warning C4244: “初始化”: 从“__int64”转换到“int”，可能丢失数据
+glove.c(304): warning C4996: 'fopen': This function or variable may be unsafe. Consider using fopen_s instead. To disable deprecation, use _CRT_SECURE_NO_WARNINGS. See online help for details.
   C:\Program Files (x86)\Windows Kits\10\Include\10.0.15063.0\ucrt\stdio.h(207): note: 参见“fopen”的声明
-glove.c(337): warning C4996: 'localtime': This function or variable may be unsafe. Consider using localtime_s instead. To disable deprecation, use _CRT_SECURE_NO_WARNINGS. See online help for details.
+glove.c(338): warning C4996: 'localtime': This function or variable may be unsafe. Consider using localtime_s instead. To disable deprecation, use _CRT_SECURE_NO_WARNINGS. See online help for details.
   C:\Program Files (x86)\Windows Kits\10\Include\10.0.15063.0\ucrt\time.h(505): note: 参见“localtime”的声明
-glove.c(432): warning C4996: 'strcpy': This function or variable may be unsafe. Consider using strcpy_s instead. To disable deprecation, use _CRT_SECURE_NO_WARNINGS. See online help for details.
-glove.c(433): warning C4996: 'strcpy': This function or variable may be unsafe. Consider using strcpy_s instead. To disable deprecation, use _CRT_SECURE_NO_WARNINGS. See online help for details.
-glove.c(434): warning C4996: 'strcpy': This function or variable may be unsafe. Consider using strcpy_s instead. To disable deprecation, use _CRT_SECURE_NO_WARNINGS. See online help for details.
 glove.c(435): warning C4996: 'strcpy': This function or variable may be unsafe. Consider using strcpy_s instead. To disable deprecation, use _CRT_SECURE_NO_WARNINGS. See online help for details.
+glove.c(436): warning C4996: 'strcpy': This function or variable may be unsafe. Consider using strcpy_s instead. To disable deprecation, use _CRT_SECURE_NO_WARNINGS. See online help for details.
 glove.c(437): warning C4996: 'strcpy': This function or variable may be unsafe. Consider using strcpy_s instead. To disable deprecation, use _CRT_SECURE_NO_WARNINGS. See online help for details.
+glove.c(438): warning C4996: 'strcpy': This function or variable may be unsafe. Consider using strcpy_s instead. To disable deprecation, use _CRT_SECURE_NO_WARNINGS. See online help for details.
 glove.c(440): warning C4996: 'strcpy': This function or variable may be unsafe. Consider using strcpy_s instead. To disable deprecation, use _CRT_SECURE_NO_WARNINGS. See online help for details.
-glove.c(441): warning C4996: 'strcpy': This function or variable may be unsafe. Consider using strcpy_s instead. To disable deprecation, use _CRT_SECURE_NO_WARNINGS. See online help for details.
-glove.c(442): warning C4996: 'strcpy': This function or variable may be unsafe. Consider using strcpy_s instead. To disable deprecation, use _CRT_SECURE_NO_WARNINGS. See online help for details.
-glove.c(446): warning C4996: 'fopen': This function or variable may be unsafe. Consider using fopen_s instead. To disable deprecation, use _CRT_SECURE_NO_WARNINGS. See online help for details.
+glove.c(443): warning C4996: 'strcpy': This function or variable may be unsafe. Consider using strcpy_s instead. To disable deprecation, use _CRT_SECURE_NO_WARNINGS. See online help for details.
+glove.c(444): warning C4996: 'strcpy': This function or variable may be unsafe. Consider using strcpy_s instead. To disable deprecation, use _CRT_SECURE_NO_WARNINGS. See online help for details.
+glove.c(445): warning C4996: 'strcpy': This function or variable may be unsafe. Consider using strcpy_s instead. To disable deprecation, use _CRT_SECURE_NO_WARNINGS. See online help for details.
+glove.c(449): warning C4996: 'fopen': This function or variable may be unsafe. Consider using fopen_s instead. To disable deprecation, use _CRT_SECURE_NO_WARNINGS. See online help for details.
   C:\Program Files (x86)\Windows Kits\10\Include\10.0.15063.0\ucrt\stdio.h(207): note: 参见“fopen”的声明
   正在生成代码
 e:\ano\workspace\glove-win\vs\x64\glove\glove\glove.c(175): warning C4716: “glove_thread”: 必须返回一个值