word2vecIncrementalLearning/run_multiclass.sh at master · tangzhenyu/word2vecIncrementalLearning · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
make clean
make

alphas="0.01 0.02 0.03"

for alpha1 in ${alphas}
do

iterations1=1
dim1=600
windows1=5
save_binary1=2
threads1=35
#alpha1="0.05"
input_file1="/data/tangzhenyu/base_words_content_disturb_order"
output_file1=${input_file1}"_vector_"${iterations1}"_"${dim1}"_"${windows1}_${alpha1}"_increment_alpha.bin.1"
save_vocab_file1=${input_file1}"_words_"${iterations1}"_"${dim1}"_"${windows1}_${alpha1}"_increment_alpha.vocab.1"
model_output_file1=${input_file1}"_model_"${iterations1}"_"${dim1}"_"${windows1}_${alpha1}"_increment_alpha.model.1"
class_file1=

command1="./word2vec -train ${input_file1} -output ${output_file1} -model-output ${model_output_file1} -cbow 0 -size ${dim1} -window 5 -negative 8 -hs 0 -sample 1e-5 -threads ${threads1} -save-binary ${save_binary1} -iterations ${iterations1} -update 0 -save-vocab ${save_vocab_file1} -min-count 5 -alpha ${alpha1} -negative-classes "
echo ${command1}
${command1}
done

iterations2=1
dim2=600
windows2=5
binary2=0
read_binary2=0
save_binary2=2
threads2=35
#alpha2="0.05"
vec_input_file2=${output_file1}".txt"
model_input_file2=${model_output_file1}".txt"
read_vocab_file2=${save_vocab_file1}
input_file2="/data/tangzhenyu/sports_words_content_disturb_order"
output_file2=${input_file2}"_vector_"${iterations2}"_"${dim2}"_"${windows2}_${alpha2}"_increment_alpha.bin"
save_vocab_file2=${input_file2}"_words_"${iterations2}"_"${dim2}"_"${windows2}_${alpha2}"_increment_alpha.vocab"
model_output_file2=${input_file2}"_model_"${iterations2}"_"${dim2}"_"${windows2}_${alpha2}"_increment_alpha.model"
#command2="./word2vec -train ${input_file2} -output ${output_file2} -model-output ${model_output_file2} -cbow 0 -size ${dim2} -window 5 -negative 8 -hs 0 -sample 1e-5 -threads ${threads2} -read-binary ${read_binary2} -save-binary ${save_binary2} -iterations ${iterations2} -update 1 -read-vocab ${read_vocab_file2} -save-vocab ${save_vocab_file2} -model-input ${model_input_file2} -min-count 5 -vec-input-file ${vec_input_file2} -alpha ${alpha2}"
#echo ${command2}
#${command2}

#./word2vec -train ${input_file1} -output ${output_file1} -model-output ${model_output_file1} -cbow 0 -size ${dim1} -window 5 -hs 1 -sample 1e-5 -threads 35 -binary 2 -iterations ${iterations1} -update 0 -save-vocab ${save_vocab_file1} -min-count 1000
#./word2vec -train ${input_file2} -output ${output_file2} -model-output ${model_output_file2} -cbow 0 -size ${dim2} -window 5 -hs 1 -sample 1e-5 -threads 35 -binary 2 -iterations ${iterations2} -update 1 -read-vocab ${read_vocab_file2} -save-vocab ${save_vocab_file2} -model-input ${model_input_file2} -min-count 100