Forked from Witty-Kitty/alp_visualize_lg_rw_vectors.py
Created
November 8, 2023 21:20
-
-
Save EVANE-hub/a7871b702c51d2105e1bd96485f84fdb to your computer and use it in GitHub Desktop.
Read in word vectors and visualize using TSNE
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
import gensim | |
from sklearn.manifold import TSNE | |
import matplotlib.pyplot as plt | |
model = gensim.models.KeyedVectors.load_word2vec_format('MUSE/dumped/debug/8gywilp7r1/vectors-rw1.txt') | |
model2 = gensim.models.KeyedVectors.load_word2vec_format('MUSE/dumped/debug/8gywilp7r1/vectors-lg1.txt') | |
lg_dict = { | |
"katonda": "god", | |
"emu": "one", | |
"kabaka": "king", | |
"mutabani" : "son", | |
"mukazi":"son-in-law", | |
"bantu":"people", | |
"lukalu":"land", | |
"ttaka":"land", | |
"ettaka":"land", | |
"nsi":"land", | |
"olukalu":"land", | |
"nnyumba":"house", | |
"ennyumba":"house", | |
"omulenzi":"son", | |
"taata":"father", | |
"ekintu":"thing", | |
"engoye":"thing", | |
"kintu":"thing", | |
"mukono":"hand", | |
"omukono":"hand", | |
"kifo":"place", | |
"ekifo":"place", | |
"ekibuga":"city", | |
"kibuga":"city", | |
"ensi":"earth", | |
"erinnya":"name", | |
"lunaku":"day", | |
"misana":"day", | |
"emisana":"day", | |
"olunaku":"day", | |
"mutima":"heart", | |
"ekigambo":"word", | |
"kigambo":"word", | |
"byombi":"two", | |
"mannyo":"two", | |
"ekikoola":"two", | |
"erinnyo":"two", | |
"liiso":"eye", | |
"eriiso":"eye", | |
"nnyindo":"nose", | |
"ennyindo":"nose", | |
"omukwano":"love", | |
"mukwano":"love", | |
"mwanyina":"brother", | |
"kwagalana":"to-love-each-other", | |
"omuntu":"man", | |
"omusajja":"man", | |
"musajja":"man", | |
} | |
rw_dict = { | |
"imana": "god", | |
"rimwe": "one", | |
"umwami": "king", | |
"mwene" : "son", | |
"umukwe":"son-in-law", | |
"rubanda":"people", | |
"isambu":"land", | |
"igihugu":"land", | |
"ubutaka":"land", | |
"inzu":"house", | |
"akazu":"house", | |
"baba":"father", | |
"ikintu":"thing", | |
"akantu":"small-thing", | |
"ikiganza":"hand", | |
"umukono":"hand", | |
"ahantu":"place", | |
"igitaka":"earth", | |
"isi":"earth", | |
"izina":"name", | |
"umunsi":"day", | |
"umutima":"heart", | |
"ijambo":"word", | |
"ebyiri":"two", | |
"kabiri":"two", | |
"ijisho":"eye", | |
"izuru":"nose", | |
"ishyanga":"foreign-country", | |
"urukundo":"love", | |
"musaza":"brother-of-a-female", | |
"umugambi":"plan", | |
"umugambanyi":"traitor", | |
"umusaza":"old-man", | |
"umusore":"male-teenager", | |
"data":"paternal-uncle", | |
"sebukwe":"father-in-law-of-someone-else", | |
"sobukwe":"father-in-law-of-person-being-spoken-to", | |
"kurandata":"to-lead-by-the-hand", | |
"gukunda":"to-like-love", | |
"gukundana":"to-like-or-love-each-other", | |
"gukundwa":"to-be-loved" | |
} | |
def tsne_plot(model1, model2): | |
"Creates and TSNE model and plots it" | |
labels1 = [] | |
tokens1 = [] | |
labels2 = [] | |
tokens2 = [] | |
for word in model1.wv.vocab: | |
tokens1.append(model1[word]) | |
labels1.append(word) | |
for word in model2.wv.vocab: | |
tokens2.append(model2[word]) | |
labels2.append(word) | |
tsne_model1 = TSNE(perplexity=50, n_components=2, init='pca', n_iter=2500, random_state=23) | |
new_values1 = tsne_model1.fit_transform(tokens1) | |
tsne_model2 = TSNE(perplexity=50, n_components=2, init='pca', n_iter=2500, random_state=23) | |
new_values2 = tsne_model2.fit_transform(tokens2) | |
x1 = [] | |
y1 = [] | |
for value in new_values1: | |
x1.append(value[0]) | |
y1.append(value[1]) | |
x2 = [] | |
y2 = [] | |
for value in new_values2: | |
x2.append(value[0]) | |
y2.append(value[1]) | |
plt.figure(figsize=(16, 16)) | |
for i in range(len(x1)): | |
translation = '' | |
if (labels1[i]) in (rw_dict.keys()): | |
translation = rw_dict[labels1[i]] | |
else: | |
translation = "null" | |
plt.scatter(x1[i],y1[i],c='b') | |
plt.annotate(labels1[i] + " - " + translation, | |
xy=(x1[i], y1[i]), | |
xytext=(5, 2), | |
textcoords='offset points', | |
ha='right', | |
va='bottom') | |
for i in range(len(x2)): | |
translation = '' | |
if (labels2[i]) in (lg_dict.keys()): | |
translation = lg_dict[labels2[i]] | |
else: | |
translation = "null" | |
plt.scatter(x2[i],y2[i],c='g') | |
plt.annotate(labels2[i] + " - " + translation, | |
xy=(x2[i], y2[i]), | |
xytext=(5, 2), | |
textcoords='offset points', | |
ha='right', | |
va='bottom') | |
plt.savefig('shared-filtered.png') | |
plt.show() | |
tsne_plot(model, model2) |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment