add application files
Browse files- app.py +266 -0
- awesome-japanese-nlp-resources-search.json +0 -0
- requirements.txt +4 -0
app.py
ADDED
@@ -0,0 +1,266 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
import json
|
2 |
+
|
3 |
+
import altair as alt
|
4 |
+
import pandas as pd
|
5 |
+
import streamlit as st
|
6 |
+
|
7 |
+
|
8 |
+
def read_json(file_name):
|
9 |
+
with open(file_name, "r") as f:
|
10 |
+
json_data = json.load(f)
|
11 |
+
return json_data
|
12 |
+
|
13 |
+
|
14 |
+
# Load a json file
|
15 |
+
json_file = "awesome-japanese-nlp-resources-search.json"
|
16 |
+
json_data = read_json(json_file)
|
17 |
+
df = pd.DataFrame(json_data)
|
18 |
+
|
19 |
+
# Sorted by selected columns
|
20 |
+
df = df[
|
21 |
+
[
|
22 |
+
"project_name",
|
23 |
+
"description",
|
24 |
+
"url",
|
25 |
+
"stargazers_count",
|
26 |
+
"downloads",
|
27 |
+
"source",
|
28 |
+
"score",
|
29 |
+
"first_commit",
|
30 |
+
"latest_commit",
|
31 |
+
"languages",
|
32 |
+
"model_or_dataset",
|
33 |
+
]
|
34 |
+
]
|
35 |
+
df = df.sort_values(by="score", ascending=False)
|
36 |
+
|
37 |
+
|
38 |
+
# Convert DataFrame for Dashboard
|
39 |
+
df["first_commit"] = pd.to_datetime(df["first_commit"], errors="coerce")
|
40 |
+
df["latest_commit"] = pd.to_datetime(df["latest_commit"], errors="coerce")
|
41 |
+
df["activity_period"] = (df["latest_commit"] - df["first_commit"]).dt.days
|
42 |
+
df = df[df["first_commit"] >= "2009-01-01"]
|
43 |
+
df = df[df["latest_commit"] >= "2009-01-01"]
|
44 |
+
df["str_languages"] = df["languages"].apply(
|
45 |
+
lambda x: ",".join(x) if isinstance(x, list) else str(x)
|
46 |
+
)
|
47 |
+
df["year"] = df["first_commit"].dt.year
|
48 |
+
|
49 |
+
|
50 |
+
# Set streamlit page settings
|
51 |
+
title = "Awesome Japanese NLP resources Dashboard"
|
52 |
+
icon = "🔎"
|
53 |
+
|
54 |
+
st.set_page_config(
|
55 |
+
page_title=title,
|
56 |
+
page_icon=icon,
|
57 |
+
layout="wide",
|
58 |
+
initial_sidebar_state="expanded",
|
59 |
+
)
|
60 |
+
|
61 |
+
# Main streamlit page (sidebar)
|
62 |
+
alt.themes.enable("dark")
|
63 |
+
with st.sidebar:
|
64 |
+
st.title(f"{title} {icon}")
|
65 |
+
st.markdown(
|
66 |
+
"You can search for open-source software from [1250+ Japanese NLP repositories](https://github.com/taishi-i/awesome-japanese-nlp-resources)."
|
67 |
+
)
|
68 |
+
query = st.text_input(label="Search keyword")
|
69 |
+
|
70 |
+
source_type = ["GitHub", "Hugging Face"]
|
71 |
+
selected_source_type = st.selectbox(
|
72 |
+
"Choose a source type: GitHub or Hugging Face", source_type
|
73 |
+
)
|
74 |
+
|
75 |
+
# Filtering GitHub or Hugging Face
|
76 |
+
df = df[df["source"] == selected_source_type]
|
77 |
+
|
78 |
+
if selected_source_type == "GitHub":
|
79 |
+
selected_model_or_dataset = None
|
80 |
+
all_languages = (
|
81 |
+
df["languages"]
|
82 |
+
.dropna()
|
83 |
+
.apply(lambda x: x if isinstance(x, list) else [])
|
84 |
+
.explode()
|
85 |
+
.unique()
|
86 |
+
)
|
87 |
+
all_languages = [""] + all_languages.tolist()
|
88 |
+
selected_languges = st.selectbox(
|
89 |
+
"Choose a programming language", all_languages, index=0
|
90 |
+
)
|
91 |
+
|
92 |
+
min_stars = int(df["stargazers_count"].min())
|
93 |
+
max_stars = int(df["stargazers_count"].max())
|
94 |
+
|
95 |
+
stars_range = st.slider(
|
96 |
+
"Choose the range for the stargazer count",
|
97 |
+
min_value=min_stars,
|
98 |
+
max_value=max_stars,
|
99 |
+
value=(min_stars, max_stars),
|
100 |
+
)
|
101 |
+
else:
|
102 |
+
selected_languges = None
|
103 |
+
selected_model_or_dataset = st.selectbox(
|
104 |
+
"Choose a model or a dataset", ["", "model", "dataset"], index=0
|
105 |
+
)
|
106 |
+
|
107 |
+
min_downloads = int(df["downloads"].min())
|
108 |
+
max_downloads = int(df["downloads"].max())
|
109 |
+
|
110 |
+
downloads_range = st.slider(
|
111 |
+
"Choose the range for the number of downloads",
|
112 |
+
min_value=min_downloads,
|
113 |
+
max_value=max_downloads,
|
114 |
+
value=(min_downloads, max_downloads),
|
115 |
+
)
|
116 |
+
|
117 |
+
min_activity_period = int(df["activity_period"].min())
|
118 |
+
max_activity_period = int(df["activity_period"].max())
|
119 |
+
|
120 |
+
activity_period_range = st.slider(
|
121 |
+
"Select the range for activity periods (in days)",
|
122 |
+
min_value=min_activity_period,
|
123 |
+
max_value=max_activity_period,
|
124 |
+
value=(min_activity_period, max_activity_period),
|
125 |
+
)
|
126 |
+
years = sorted(list(set(df["year"].dropna().astype(int).tolist())))
|
127 |
+
|
128 |
+
selected_year_range = st.slider(
|
129 |
+
"Select a range for the years of the first commit",
|
130 |
+
min_value=min(years),
|
131 |
+
max_value=max(years),
|
132 |
+
value=(min(years), max(years)),
|
133 |
+
)
|
134 |
+
|
135 |
+
|
136 |
+
df = df[
|
137 |
+
(df["year"] >= selected_year_range[0])
|
138 |
+
& (df["year"] <= selected_year_range[1])
|
139 |
+
]
|
140 |
+
|
141 |
+
|
142 |
+
if selected_source_type == "GitHub":
|
143 |
+
df = df[
|
144 |
+
(df["stargazers_count"] >= stars_range[0])
|
145 |
+
& (df["stargazers_count"] <= stars_range[1])
|
146 |
+
]
|
147 |
+
else:
|
148 |
+
df = df[
|
149 |
+
(df["downloads"] >= downloads_range[0])
|
150 |
+
& (df["downloads"] <= downloads_range[1])
|
151 |
+
]
|
152 |
+
|
153 |
+
df = df[
|
154 |
+
(df["activity_period"] >= activity_period_range[0])
|
155 |
+
& (df["activity_period"] <= activity_period_range[1])
|
156 |
+
]
|
157 |
+
|
158 |
+
contained_description = df["description"].str.contains(
|
159 |
+
query, case=False, na=False
|
160 |
+
)
|
161 |
+
contained_project_name = df["project_name"].str.contains(
|
162 |
+
query, case=False, na=False
|
163 |
+
)
|
164 |
+
df = df[contained_description | contained_project_name]
|
165 |
+
|
166 |
+
if selected_languges:
|
167 |
+
df = df[
|
168 |
+
df["str_languages"].str.contains(
|
169 |
+
selected_languges, case=False, na=False
|
170 |
+
)
|
171 |
+
]
|
172 |
+
|
173 |
+
if selected_model_or_dataset:
|
174 |
+
df = df[
|
175 |
+
df["model_or_dataset"].str.contains(
|
176 |
+
selected_model_or_dataset, case=False, na=False
|
177 |
+
)
|
178 |
+
]
|
179 |
+
|
180 |
+
|
181 |
+
# Main streamlit page (columns)
|
182 |
+
col1, col2 = st.columns(2, gap="large")
|
183 |
+
|
184 |
+
with col1:
|
185 |
+
st.markdown("### DataFrame")
|
186 |
+
st.markdown(f"#### Number of repositories: {len(df)}")
|
187 |
+
st.dataframe(df, height=600)
|
188 |
+
|
189 |
+
projects_per_year = (
|
190 |
+
df.groupby("year").size().reset_index(name="project_count")
|
191 |
+
)
|
192 |
+
|
193 |
+
chart = (
|
194 |
+
alt.Chart(projects_per_year)
|
195 |
+
.mark_bar()
|
196 |
+
.encode(
|
197 |
+
x=alt.X("year:O", title="Year"),
|
198 |
+
y=alt.Y("project_count:Q", title="Number of repositories"),
|
199 |
+
tooltip=["year", "project_count"],
|
200 |
+
)
|
201 |
+
.properties(
|
202 |
+
title="Number of projects per year based on the uear of the first commit",
|
203 |
+
width=600,
|
204 |
+
height=400,
|
205 |
+
)
|
206 |
+
)
|
207 |
+
|
208 |
+
st.altair_chart(chart, use_container_width=True)
|
209 |
+
|
210 |
+
with col2:
|
211 |
+
if selected_source_type == "GitHub":
|
212 |
+
vs_type = "stargazers_count"
|
213 |
+
else:
|
214 |
+
vs_type = "downloads"
|
215 |
+
|
216 |
+
st.markdown(f"### First commit vs {vs_type}")
|
217 |
+
chart = (
|
218 |
+
alt.Chart(df)
|
219 |
+
.mark_circle(size=60)
|
220 |
+
.encode(
|
221 |
+
x="first_commit:T",
|
222 |
+
y=f"{vs_type}:Q",
|
223 |
+
tooltip=["first_commit", "project_name", f"{vs_type}"],
|
224 |
+
)
|
225 |
+
.properties(
|
226 |
+
title=f"Relationship between first commit date and {vs_type}",
|
227 |
+
)
|
228 |
+
.interactive()
|
229 |
+
)
|
230 |
+
st.altair_chart(chart, use_container_width=True)
|
231 |
+
|
232 |
+
st.markdown(f"### Latest commit vs {vs_type}")
|
233 |
+
chart = (
|
234 |
+
alt.Chart(df)
|
235 |
+
.mark_circle(size=60)
|
236 |
+
.encode(
|
237 |
+
x="latest_commit:T",
|
238 |
+
y=f"{vs_type}:Q",
|
239 |
+
tooltip=["project_name", "latest_commit", f"{vs_type}"],
|
240 |
+
)
|
241 |
+
.properties(
|
242 |
+
title=f"Relationship between latest commit date and {vs_type}",
|
243 |
+
)
|
244 |
+
.interactive()
|
245 |
+
)
|
246 |
+
st.altair_chart(chart, use_container_width=True)
|
247 |
+
|
248 |
+
st.markdown(f"### Activity period vs {vs_type}")
|
249 |
+
chart = (
|
250 |
+
alt.Chart(df)
|
251 |
+
.mark_circle(size=60)
|
252 |
+
.encode(
|
253 |
+
x=alt.X("activity_period:Q", title="Activity Period (Days)"),
|
254 |
+
y=alt.Y(f"{vs_type}:Q", title=f"{vs_type}"),
|
255 |
+
tooltip=[
|
256 |
+
"project_name",
|
257 |
+
"activity_period",
|
258 |
+
f"{vs_type}",
|
259 |
+
],
|
260 |
+
)
|
261 |
+
.properties(
|
262 |
+
title=f"Relationship between activity period and {vs_type}",
|
263 |
+
)
|
264 |
+
.interactive()
|
265 |
+
)
|
266 |
+
st.altair_chart(chart, use_container_width=True)
|
awesome-japanese-nlp-resources-search.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|
requirements.txt
ADDED
@@ -0,0 +1,4 @@
|
|
|
|
|
|
|
|
|
|
|
1 |
+
streamlit
|
2 |
+
pandas
|
3 |
+
altair
|
4 |
+
plotly
|