Merge branch 'main' of https://huggingface.co/spaces/imomayiz/DODa
Browse files
app.py
CHANGED
@@ -128,3 +128,131 @@ def main():
|
|
128 |
|
129 |
if __name__ == "__main__":
|
130 |
main()
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
128 |
|
129 |
if __name__ == "__main__":
|
130 |
main()
|
131 |
+
import streamlit as st
|
132 |
+
from datasets import load_dataset, Dataset
|
133 |
+
import csv
|
134 |
+
import datetime as dt
|
135 |
+
import random
|
136 |
+
import os
|
137 |
+
from huggingface_hub import Repository
|
138 |
+
|
139 |
+
|
140 |
+
HF_API_KEY = os.environ.get("HF_TOKEN", None)
|
141 |
+
|
142 |
+
REPO_ID = "imomayiz/darija-english"
|
143 |
+
DATASET_REPO_URL = f"https://huggingface.co/datasets/{REPO_ID}"
|
144 |
+
DATA_FILE = os.path.join("submissions", "submissions.csv")
|
145 |
+
|
146 |
+
submissions_repo = Repository(
|
147 |
+
local_dir="submissions", clone_from=DATASET_REPO_URL, use_auth_token=HF_API_KEY
|
148 |
+
)
|
149 |
+
|
150 |
+
def load_data(repo_id):
|
151 |
+
dataset = load_dataset(f'{repo_id}', split='sentences')
|
152 |
+
return dataset
|
153 |
+
|
154 |
+
def get_sentence_from_dataset(dataset, column_name="darija_ar"):
|
155 |
+
|
156 |
+
# Get a random sentence
|
157 |
+
random_sentence_index = random.randint(0, len(dataset) - 1)
|
158 |
+
random_sentence = dataset[random_sentence_index][column_name]
|
159 |
+
|
160 |
+
return random_sentence
|
161 |
+
|
162 |
+
def store_submission(sentence: str, translation: str, translation_fr: str):
|
163 |
+
if sentence and (translation or translation_fr):
|
164 |
+
with open(DATA_FILE, "a") as csvfile:
|
165 |
+
writer = csv.DictWriter(csvfile,
|
166 |
+
fieldnames=["darija", "eng", "darija_ar", "time"])
|
167 |
+
writer.writerow(
|
168 |
+
{"darija_ar": sentence,
|
169 |
+
"eng": translation,
|
170 |
+
"darija": translation_fr,
|
171 |
+
"time": str(dt.datetime.now())}
|
172 |
+
)
|
173 |
+
commit_url = submissions_repo.push_to_hub()
|
174 |
+
print(commit_url)
|
175 |
+
|
176 |
+
|
177 |
+
# Load the dataset
|
178 |
+
dataset = load_data(REPO_ID)
|
179 |
+
|
180 |
+
def fetch_sentence():
|
181 |
+
st.session_state.sentence = get_sentence_from_dataset(dataset)
|
182 |
+
st.session_state.translation_input = ""
|
183 |
+
|
184 |
+
|
185 |
+
def main():
|
186 |
+
|
187 |
+
if "sentence" not in st.session_state:
|
188 |
+
st.session_state.sentence = get_sentence_from_dataset(dataset)
|
189 |
+
if 'translation_input' not in st.session_state:
|
190 |
+
st.session_state.translation_input = ""
|
191 |
+
if 'translation_input_fr' not in st.session_state:
|
192 |
+
st.session_state.translation_input_fr = ""
|
193 |
+
if 'display_new' not in st.session_state:
|
194 |
+
st.session_state.display_new = False
|
195 |
+
|
196 |
+
st.title("Translate From Arabic to English")
|
197 |
+
|
198 |
+
st.markdown(
|
199 |
+
"""This mini-app allows you to contribute to the **darija-english** dataset
|
200 |
+
as part of [DODa](https://darija-open-dataset.github.io/)
|
201 |
+
project. To contribute, simply translate the given sentence from Arabic to English.
|
202 |
+
The translated sentence will be submitted to the dataset
|
203 |
+
[here](https://huggingface.co/datasets/imomayiz/darija-english)."""
|
204 |
+
)
|
205 |
+
|
206 |
+
st.text("")
|
207 |
+
|
208 |
+
st.write(f"""
|
209 |
+
<div style="
|
210 |
+
padding: 5px;
|
211 |
+
border: 1px solid #000000;
|
212 |
+
border-radius: 5px;
|
213 |
+
">
|
214 |
+
<p style="font-size: 20px;">{st.session_state.sentence}.</p>
|
215 |
+
</div>""", unsafe_allow_html=True)
|
216 |
+
|
217 |
+
|
218 |
+
# Display new sentence button
|
219 |
+
st.session_state.display_new = st.button("New Sentence",
|
220 |
+
on_click=fetch_sentence)
|
221 |
+
|
222 |
+
|
223 |
+
# Input field for translation
|
224 |
+
translation_input_placeholder = st.empty()
|
225 |
+
|
226 |
+
with translation_input_placeholder.container():
|
227 |
+
translation_input = st.text_input("Enter translation to english: ",
|
228 |
+
st.session_state.translation_input)
|
229 |
+
st.session_state.translation_input = translation_input
|
230 |
+
|
231 |
+
# Input field for translation
|
232 |
+
translation_input_placeholder_fr = st.empty()
|
233 |
+
|
234 |
+
with translation_input_placeholder_fr.container():
|
235 |
+
translation_input_fr = st.text_input(
|
236 |
+
"Enter translation to darija in latin characters: ",
|
237 |
+
st.session_state.translation_input
|
238 |
+
)
|
239 |
+
st.session_state.translation_input_fr = translation_input_fr
|
240 |
+
|
241 |
+
# Submit button
|
242 |
+
if st.button("Submit Translation"):
|
243 |
+
if translation_input:
|
244 |
+
st.success("Translation submitted successfully!")
|
245 |
+
|
246 |
+
elif translation_input_fr:
|
247 |
+
st.success("Translation submitted successfully!")
|
248 |
+
|
249 |
+
else:
|
250 |
+
st.warning("Please enter a translation before submitting.")
|
251 |
+
|
252 |
+
|
253 |
+
store_submission(st.session_state.sentence,
|
254 |
+
st.session_state.translation_input,
|
255 |
+
st.session_state.translation_input_fr)
|
256 |
+
|
257 |
+
if __name__ == "__main__":
|
258 |
+
main()
|