Spaces:
Running
Running
File size: 4,758 Bytes
cab3f4d 53c86f7 cab3f4d 53c86f7 cab3f4d 53c86f7 b6e02e5 53c86f7 b6e02e5 53c86f7 b6e02e5 53c86f7 b6e02e5 cab3f4d 53c86f7 cab3f4d 8143d78 cab3f4d |
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 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 |
import streamlit as st
from datasets import load_dataset
import pandas as pd
import plotly.graph_objects as go
from transformers import pipeline
@st.cache_data
def fetch_counts():
dataset = load_dataset("atlasia/darija-translation", split="train")
dataset = pd.DataFrame(dataset)
n_eng = len(dataset["en"].dropna())
n_fr = len(dataset["fr"].dropna())
n = len(dataset)
return {"n_eng": n_eng, "n_fr": n_fr, "n": n}
def terjman(input_text: str) -> str:
pipe = pipeline("text2text-generation", model="atlasia/Terjman-Large")
result = pipe(input_text, max_length=512)
return result[0]["generated_text"]
def transliterate(input_text: str) -> str:
pipe = pipeline("text2text-generation", model="atlasia/Transliteration-Moroccan-Darija")
result = pipe(input_text, max_length=50)
return result[0]["generated_text"]
if __name__ == "__main__":
st.image("atlasia_white_wtext_nobg.png")
counts = fetch_counts()
n_goal = 100000
total_submissions = counts["n"]
st.text("")
# center text
st.markdown(
"""
<h1 style='text-align: center; font-size: 20px;'>
Contribute now to help build a better Darija dataset for all Moroccans
Contribute here: <a href="https://atlasia.ma" target="_blank">https://atlasia.ma</a>
</h1>
""",
unsafe_allow_html=True,
)
st.divider()
# with st.container() as c:
# # add a block where users can input text and get a translation
# st.markdown(
# """
# <div style='text-align: center;'>
# <h3>🔠Keyboard: Transliterate Letters from Latin to Arabic</h3
# >
# </div>
# """,
# unsafe_allow_html=True,
# )
# col1, col2 = st.columns(2)
# with col1:
# input_text = st.text_area(":grey[Enter a word/letter in English ⬇]", "")
# button = st.button("Transliterate")
# with col2:
# if button:
# with st.spinner('Transliterating...'):
# translation = transliterate(input_text)
# st.text_area(
# ":grey[Transliteration]",
# translation,
# )
# st.divider()
with st.container() as c:
# add a block where users can input text and get a translation
st.markdown(
"""
<div style='text-align: center;'>
<h3>💬Terjman: Translate to Darija</h3
>
</div>
""",
unsafe_allow_html=True,
)
st.caption(
"""
<div style='text-align: center;'>
<h3>This model has been developed thanks to your contributions.
While it's not perfect yet, your continued input is key for making it better.</h3>
</div>
""",
unsafe_allow_html=True,
)
col1, col2 = st.columns(2)
with col1:
input_text = st.text_area(":grey[Enter a sentence in English ⬇]", "")
button = st.button("Translate")
with col2:
if button:
with st.spinner('Translating...'):
translation = terjman(input_text)
st.text_area(
":grey[Translation in Darija]",
translation,
)
# add a separator
st.divider()
st.markdown(
"""
<div style='text-align: center;'>
<h3>📊 Data statistics</h3
>
</div>
""",
unsafe_allow_html=True,
)
# make progress chart
fig = go.Figure(
go.Indicator(
domain={"x": [0, 1], "y": [0, 1]},
value=total_submissions,
mode="gauge+number+delta",
title={"text": "Number of translations"},
delta={"reference": 42000},
gauge={
"axis": {"range": [0, n_goal]},
"steps": [
{"range": [0, total_submissions], "color": "gray"},
],
"threshold": {
"line": {"color": "green", "width": 4},
"thickness": 0.75,
"value": n_goal / 2,
},
},
)
)
st.plotly_chart(fig, use_container_width=True)
labels = ["English", "French"]
values = [counts["n_eng"], counts["n_fr"]]
# change color to blue and white
fig = go.Figure(data=[go.Pie(labels=labels, values=values, pull=[0.2, 0])])
fig.update_traces(marker=dict(colors=["#46607b", "#FFFFFF"]))
st.plotly_chart(fig, use_container_width=True)
|