vane/word2vec
1
0
Fork 0
Code Issues Pull Requests Actions Projects Releases Wiki Activity

fixed minor bugs

Browse Source
This commit is contained in:
tmikolov 2014-09-08 20:14:01 +00:00
parent 891d84c630
commit ffd46fd134
4 changed files with 23 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

8
compute-accuracy.c
View File

@ -53,7 +53,13 @@ int main(int argc, char **argv)
return -1; return -1;
} }
for (b = 0; b < words; b++) { for (b = 0; b < words; b++) {
fscanf(f, "%s%c", &vocab[b * max_w], &ch); a = 0;
while (1) {
vocab[b * max_w + a] = fgetc(f);
if (feof(f) || (vocab[b * max_w + a] == ' ')) break;
if ((a < max_w) && (vocab[b * max_w + a] != '\n')) a++;
}
vocab[b * max_w + a] = 0;
for (a = 0; a < max_w; a++) vocab[b * max_w + a] = toupper(vocab[b * max_w + a]); for (a = 0; a < max_w; a++) vocab[b * max_w + a] = toupper(vocab[b * max_w + a]);
for (a = 0; a < size; a++) fread(&M[a + b * size], sizeof(float), 1, f); for (a = 0; a < size; a++) fread(&M[a + b * size], sizeof(float), 1, f);
len = 0; len = 0;

8
distance.c
View File

@ -51,7 +51,13 @@ int main(int argc, char **argv) {
return -1; return -1;
} }
for (b = 0; b < words; b++) { for (b = 0; b < words; b++) {
fscanf(f, "%s%c", &vocab[b * max_w], &ch); a = 0;
while (1) {
vocab[b * max_w + a] = fgetc(f);
if (feof(f) || (vocab[b * max_w + a] == ' ')) break;
if ((a < max_w) && (vocab[b * max_w + a] != '\n')) a++;
}
vocab[b * max_w + a] = 0;
for (a = 0; a < size; a++) fread(&M[a + b * size], sizeof(float), 1, f); for (a = 0; a < size; a++) fread(&M[a + b * size], sizeof(float), 1, f);
len = 0; len = 0;
for (a = 0; a < size; a++) len += M[a + b * size] * M[a + b * size]; for (a = 0; a < size; a++) len += M[a + b * size] * M[a + b * size];

8
word-analogy.c
View File

@ -50,7 +50,13 @@ int main(int argc, char **argv) {
return -1; return -1;
} }
for (b = 0; b < words; b++) { for (b = 0; b < words; b++) {
fscanf(f, "%s%c", &vocab[b * max_w], &ch); a = 0;
while (1) {
vocab[b * max_w + a] = fgetc(f);
if (feof(f) || (vocab[b * max_w + a] == ' ')) break;
if ((a < max_w) && (vocab[b * max_w + a] != '\n')) a++;
}
vocab[b * max_w + a] = 0;
for (a = 0; a < size; a++) fread(&M[a + b * size], sizeof(float), 1, f); for (a = 0; a < size; a++) fread(&M[a + b * size], sizeof(float), 1, f);
len = 0; len = 0;
for (a = 0; a < size; a++) len += M[a + b * size] * M[a + b * size]; for (a = 0; a < size; a++) len += M[a + b * size] * M[a + b * size];

4
word2vec.c
View File

@ -152,9 +152,9 @@ void SortVocab() {
train_words = 0; train_words = 0;
for (a = 0; a < size; a++) { for (a = 0; a < size; a++) {
// Words occuring less than min_count times will be discarded from the vocab // Words occuring less than min_count times will be discarded from the vocab
if (vocab[a].cn < min_count) { if ((vocab[a].cn < min_count) && (a != 0)) {
vocab_size--; vocab_size--;
free(vocab[vocab_size].word); free(vocab[a].word);
} else { } else {
// Hash will be re-computed, as after the sorting it is not actual // Hash will be re-computed, as after the sorting it is not actual
hash=GetWordHash(vocab[a].word); hash=GetWordHash(vocab[a].word);