Error with weat_perms when using own items and vectors?

[conjugateprior]

Error with weat_perms when using own items and vectors?

[santoshbs]

Further to above, the results for effect size (d) obtained using weat_param is different from calculation done as follows:

list_avg_diffs= c()

avg_t1_a1= 0
avg_t1_a2= 0
avg_t1_diff= 0

for(t in t1){
    print(t)
    x= cbn_extract_word_vectors(t)
    x= as.vector(x)
    for(a in a1){
        y= cbn_extract_word_vectors(a)
        y= as.vector(y)
        avg_t1_a1= avg_t1_a1 + as.numeric(cosine(x,y))
    }
    avg_t1_a1= avg_t1_a1/length(a1)
    
    for(a in a2){
        y= cbn_extract_word_vectors(a)
        y= as.vector(y)
        avg_t1_a2= avg_t1_a2 + as.numeric(cosine(x,y))
    }
    avg_t1_a2= avg_t1_a2/length(a2)

    avg_t1_diff= avg_t1_diff + (avg_t1_a1 - avg_t1_a2)
    list_avg_diffs= c(list_avg_diffs, (avg_t1_a1 - avg_t1_a2))
}

avg_t1_diff= avg_t1_diff/length(t1)

avg_t2_a1= 0
avg_t2_a2= 0
avg_t2_diff= 0

for(t in t2){
    print(t)
    x= cbn_extract_word_vectors(t)
    x= as.vector(x)
    for(a in a1){
        y= cbn_extract_word_vectors(a)
        y= as.vector(y)
        avg_t2_a1= avg_t2_a1 + as.numeric(cosine(x,y))
    }
    avg_t2_a1= avg_t2_a1/length(a1)
    
    for(a in a2){
        y= cbn_extract_word_vectors(a)
        y= as.vector(y)
        avg_t2_a2= avg_t2_a2 + as.numeric(cosine(x,y))
    }
    avg_t2_a2= avg_t2_a2/length(a2)

    avg_t2_diff= avg_t2_diff + (avg_t2_a1 - avg_t2_a2)
    list_avg_diffs= c(list_avg_diffs, (avg_t2_a1 - avg_t2_a2))
}

avg_t2_diff= avg_t2_diff/length(t2)

diff_of_diff= avg_t1_diff - avg_t2_diff
sd= sd(list_avg_diffs)
weat_effect_size= diff_of_diff/sd