Commit
•
9910446
1
Parent(s):
bb0fb0e
./...
Browse files- tabs/id_lang_tab.py +7 -5
- tabs/modelisation_seq2seq_tab.py +8 -36
- tabs/read_api_url.py +17 -0
tabs/id_lang_tab.py
CHANGED
@@ -13,6 +13,7 @@ import seaborn as sns
|
|
13 |
from sklearn import naive_bayes
|
14 |
from translate_app import tr
|
15 |
import requests
|
|
|
16 |
|
17 |
title = "Identification de langue"
|
18 |
sidebar_name = "Identification de langue"
|
@@ -280,11 +281,11 @@ def display_shapley(lang_sel):
|
|
280 |
def run():
|
281 |
global tokenizer, vectorizer, dict_token, dict_ids, nb_token, lan_to_language, clf_nb
|
282 |
global toggle_val, custom_sentence, lan_identified
|
283 |
-
global lang_exemples, exemples
|
284 |
|
285 |
-
|
286 |
tokenizer, dict_token, dict_ids, nb_token, lan_to_language, clf_nb, vectorizer = init_nb_identifier()
|
287 |
-
lan_identified = requests.get( "
|
288 |
lang_id_model_ext, dict_xlmr, sentence_test, lang_exemples, exemples= init_lang_id_external()
|
289 |
|
290 |
st.write("")
|
@@ -329,7 +330,7 @@ def run():
|
|
329 |
|
330 |
if custom_sentence!='':
|
331 |
|
332 |
-
url = "
|
333 |
params = {"sentence": custom_sentence}
|
334 |
response = requests.get(url, params=params)
|
335 |
|
@@ -409,7 +410,8 @@ def run():
|
|
409 |
st.write("<center><h5>"+tr("Architecture du modèle utilisé")+":</h5></center>", unsafe_allow_html=True)
|
410 |
col1, col2, col3 = st.columns([0.15,0.7,0.15])
|
411 |
with col2:
|
412 |
-
|
|
|
413 |
elif (chosen_id == "tab3"):
|
414 |
st.write("### **"+tr("Interpretabilité du classifieur Naïve Bayes sur 5 langues")+"**")
|
415 |
st.write("##### "+tr("..et un Training set réduit (15000 phrases et 94 tokens)"))
|
|
|
13 |
from sklearn import naive_bayes
|
14 |
from translate_app import tr
|
15 |
import requests
|
16 |
+
import read_api_url
|
17 |
|
18 |
title = "Identification de langue"
|
19 |
sidebar_name = "Identification de langue"
|
|
|
281 |
def run():
|
282 |
global tokenizer, vectorizer, dict_token, dict_ids, nb_token, lan_to_language, clf_nb
|
283 |
global toggle_val, custom_sentence, lan_identified
|
284 |
+
global lang_exemples, exemples, url_base
|
285 |
|
286 |
+
url_base = read_api_url()
|
287 |
tokenizer, dict_token, dict_ids, nb_token, lan_to_language, clf_nb, vectorizer = init_nb_identifier()
|
288 |
+
lan_identified = requests.get( url_base[0]+"/lan_identified").json()
|
289 |
lang_id_model_ext, dict_xlmr, sentence_test, lang_exemples, exemples= init_lang_id_external()
|
290 |
|
291 |
st.write("")
|
|
|
330 |
|
331 |
if custom_sentence!='':
|
332 |
|
333 |
+
url = url_base[0]+"/lang_id_dl"
|
334 |
params = {"sentence": custom_sentence}
|
335 |
response = requests.get(url, params=params)
|
336 |
|
|
|
410 |
st.write("<center><h5>"+tr("Architecture du modèle utilisé")+":</h5></center>", unsafe_allow_html=True)
|
411 |
col1, col2, col3 = st.columns([0.15,0.7,0.15])
|
412 |
with col2:
|
413 |
+
|
414 |
+
st.image(url_base[0]+"e/small_vocab/plot_model?&model_type=lang_id",use_column_width="auto")
|
415 |
elif (chosen_id == "tab3"):
|
416 |
st.write("### **"+tr("Interpretabilité du classifieur Naïve Bayes sur 5 langues")+"**")
|
417 |
st.write("##### "+tr("..et un Training set réduit (15000 phrases et 94 tokens)"))
|
tabs/modelisation_seq2seq_tab.py
CHANGED
@@ -13,33 +13,17 @@ import wavio
|
|
13 |
from gtts import gTTS
|
14 |
from extra_streamlit_components import tab_bar, TabBarItemData
|
15 |
from translate_app import tr
|
16 |
-
import csv
|
17 |
import requests
|
18 |
# from multiprocessing import Pool
|
19 |
import concurrent.futures
|
20 |
-
import time
|
21 |
-
|
22 |
|
23 |
title = "Traduction Sequence à Sequence"
|
24 |
sidebar_name = "Traduction Seq2Seq"
|
25 |
dataPath = st.session_state.DataPath
|
26 |
|
27 |
-
@st.cache_data
|
28 |
-
def read_api_url():
|
29 |
-
|
30 |
-
api_url = []
|
31 |
-
# Ouvrir le fichier CSV en mode lecture
|
32 |
-
with open("api-url.txt", newline='') as fichier_csv:
|
33 |
-
lecteur_csv = csv.reader(fichier_csv)
|
34 |
-
|
35 |
-
# Lire et afficher les trois premières lignes
|
36 |
-
for i in range(3):
|
37 |
-
ligne = next(lecteur_csv, None) # Lire la ligne suivante
|
38 |
-
if ligne is not None:
|
39 |
-
api_url.append(ligne[0])
|
40 |
-
else: return None
|
41 |
-
return api_url
|
42 |
-
|
43 |
@st.cache_data
|
44 |
def load_corpus(path):
|
45 |
input_file = os.path.join(path)
|
@@ -66,10 +50,10 @@ n1 = 0
|
|
66 |
df_data_en, df_data_fr, translation_en_fr, translation_fr_en, lang_classifier, model_speech, finetuned_translation_en_fr = load_all_data()
|
67 |
|
68 |
|
69 |
-
def
|
70 |
return requests.get(url)
|
71 |
|
72 |
-
def
|
73 |
global df_data_src, df_data_tgt, placeholder, url_base
|
74 |
|
75 |
n = 3
|
@@ -215,23 +199,11 @@ def run():
|
|
215 |
mode = 2 # Transformer
|
216 |
|
217 |
# Exécuter la fonction asynchrone
|
218 |
-
'''
|
219 |
-
t0 = time.time()
|
220 |
-
asyncio.run(display_translation1(n1, Lang, mode))
|
221 |
-
t1 = time.time()
|
222 |
-
st.write("Durée 1: "+str(t1-t0))
|
223 |
-
'''
|
224 |
# t0 = time.time()
|
225 |
-
|
226 |
# t1 = time.time()
|
227 |
-
# st.write("Durée
|
228 |
-
|
229 |
-
t0 = time.time()
|
230 |
-
display_translation3(n1, Lang,mode)
|
231 |
-
t1 = time.time()
|
232 |
-
st.write("Durée 3: "+str(t1-t0))
|
233 |
-
'''
|
234 |
-
|
235 |
st.write("## **"+tr("Details sur la méthode")+" :**\n")
|
236 |
if (chosen_id == "tab1"):
|
237 |
st.markdown(tr(
|
|
|
13 |
from gtts import gTTS
|
14 |
from extra_streamlit_components import tab_bar, TabBarItemData
|
15 |
from translate_app import tr
|
16 |
+
# import csv
|
17 |
import requests
|
18 |
# from multiprocessing import Pool
|
19 |
import concurrent.futures
|
20 |
+
# import time
|
21 |
+
import read_api_url
|
22 |
|
23 |
title = "Traduction Sequence à Sequence"
|
24 |
sidebar_name = "Traduction Seq2Seq"
|
25 |
dataPath = st.session_state.DataPath
|
26 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
27 |
@st.cache_data
|
28 |
def load_corpus(path):
|
29 |
input_file = os.path.join(path)
|
|
|
50 |
df_data_en, df_data_fr, translation_en_fr, translation_fr_en, lang_classifier, model_speech, finetuned_translation_en_fr = load_all_data()
|
51 |
|
52 |
|
53 |
+
def fetch_translation(url):
|
54 |
return requests.get(url)
|
55 |
|
56 |
+
def display_translation(n1, Lang, model_type):
|
57 |
global df_data_src, df_data_tgt, placeholder, url_base
|
58 |
|
59 |
n = 3
|
|
|
199 |
mode = 2 # Transformer
|
200 |
|
201 |
# Exécuter la fonction asynchrone
|
|
|
|
|
|
|
|
|
|
|
|
|
202 |
# t0 = time.time()
|
203 |
+
display_translation(n1, Lang,mode)
|
204 |
# t1 = time.time()
|
205 |
+
# st.write("Durée : "+str(t1-t0))
|
206 |
+
|
|
|
|
|
|
|
|
|
|
|
|
|
207 |
st.write("## **"+tr("Details sur la méthode")+" :**\n")
|
208 |
if (chosen_id == "tab1"):
|
209 |
st.markdown(tr(
|
tabs/read_api_url.py
ADDED
@@ -0,0 +1,17 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
import csv
|
2 |
+
|
3 |
+
@st.cache_data
|
4 |
+
def read_api_url():
|
5 |
+
|
6 |
+
api_url = []
|
7 |
+
# Ouvrir le fichier CSV en mode lecture
|
8 |
+
with open("api-url.txt", newline='') as fichier_csv:
|
9 |
+
lecteur_csv = csv.reader(fichier_csv)
|
10 |
+
|
11 |
+
# Lire toutes les lignes
|
12 |
+
ligne = next(lecteur_csv, None)
|
13 |
+
while ligne is not None:
|
14 |
+
api_url.append(ligne[0])
|
15 |
+
ligne = next(lecteur_csv, None)
|
16 |
+
|
17 |
+
return api_url
|