forked from zihao-ai/Awesome-Backdoor-in-Deep-Learning
-
Notifications
You must be signed in to change notification settings - Fork 0
/
script.py
280 lines (238 loc) · 8.78 KB
/
script.py
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
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
import re
import json
import numpy as np
import pandas as pd
def compare(df1, df2):
"""Compare two DataFrame and return the difference"""
df = pd.concat([df1, df2])
df = df.drop_duplicates(keep=False).reset_index(drop=True)
return df
def compare_and_add(df1, df2):
"""Compare two DataFrame and return the difference and add one column with state"""
df_dict = dict(df1=df1, df2=df2)
df = pd.concat(df_dict)
df = df.drop_duplicates(keep=False)
column_state = []
print
for belong, index in df.index:
if belong == "df1":
column_state.append("Removed")
else:
column_state.append("Added")
df["State"] = column_state
df = df.reset_index(drop=True)
return df
def strip(df):
df.columns = df.columns.str.strip()
df_obj = df.select_dtypes(["object"])
df[df_obj.columns] = df_obj.apply(lambda x: x.str.strip())
return df
def write_md(fname, df):
with open(fname, "w", encoding="utf-8") as f:
df.to_markdown(f)
def read_md(fname):
# read markdown files
try:
df = pd.read_table(fname, sep="|", header=0,
index_col=1, skipinitialspace=True)
except (pd.errors.EmptyDataError, FileNotFoundError):
print("No data.")
return None
# drop the first and the last columns
df = df.iloc[1:].drop(df.columns[[0, -1]], axis=1)
# strip strings
df = strip(df)
# replace `NaN` with `None`
df = df.where(pd.notnull(df), None)
# replace datatypes
df = df.astype({"Year": "int64"})
return df
def parse_paper(line):
instance = {}
content = line.split("*, ")
title = venue = code = None
if len(content) == 3:
title, venue, code = content
else:
title, venue = content
end = venue.find("]")
year = re.findall(r"\d+", venue[max(end - 5, 0): end])
if year:
year = int(year[0])
if year < 2000:
year += 2000
instance["Title"] = title
instance["Venue"] = venue
instance["Year"] = year
instance["Code"] = code
return instance
if __name__ == '__main__':
print("#" * 10, "Begin", "#" * 10)
with open("README.md", "r", encoding="UTF-8") as f:
content = f.readlines()
# parsing
i = 0
papers = []
titles = []
while i < len(content):
line = content[i]
if line.strip() == "# 🔗Resource":
break
if line.startswith("# "):
this_level = 0
titles = []
titles.append(line[this_level + 2:].strip())
elif line.startswith("## "):
this_level = 1
while len(titles) > this_level:
titles.pop()
titles.append(line[this_level + 2:].strip())
elif line.startswith("### "):
this_level = 2
while len(titles) > this_level:
titles.pop()
titles.append(line[this_level + 2:].strip())
elif line.startswith("#### "):
this_level = 3
while len(titles) > this_level:
titles.pop()
titles.append(line[this_level + 2:].strip())
is_paper = line.startswith("+ ")
if is_paper:
line = line[2:].strip()
paper = parse_paper(line)
paper["belong"] = titles[:]
if not paper["Year"] and len(titles) > 1:
paper["Year"] = int(titles[-1])
paper["Type"] = titles[0].strip()
papers.append(paper)
i += 1
################################### Step1: To pandas DataFrame ######################################################################
tb = []
columns = ["Title", "Type", "Venue", "Code", "Year"]
for paper in papers:
tmp = []
for col in columns:
if isinstance(paper[col], str):
tmp.append(paper[col].strip("*"))
else:
tmp.append(paper[col])
tb.append(tmp)
tb = pd.DataFrame(tb, columns=columns)
# strip
tb = strip(tb)
################################### Step2: Find papers with code ######################################################################
tb_with_code = tb.drop_duplicates(subset=["Title"])
tb_with_code = tb_with_code[pd.notna(
tb_with_code["Code"])].reset_index(drop=True)
write_md("Categorized/papers_with_code.md", tb_with_code)
################################### Step3: Categorize papers by Title ######################################################################
# Sorted by Title
tb_by_title = (
tb.drop_duplicates(subset=["Title"]).sort_values(
"Title").reset_index(drop=True)
)
write_md("Categorized/alphabet.md", tb_by_title)
################################### Step4: Categorize papers by Year ######################################################################
tb_by_year = (
tb.drop_duplicates(subset=["Title"]).sort_values(
"Type").reset_index(drop=True)
)
tb_by_year = tb_by_year.sort_values(
"Year", kind="mergesort", ascending=False
).reset_index(drop=True)
# read before write
tb_before = read_md("Categorized/year.md")
write_md("Categorized/year.md", tb_by_year)
################################### Step5: find recently updated and outdated papers ######################################################################
tb_now = tb_by_year
# find recently updadted papers
diff = compare_and_add(tb_before, tb_now)
# check if recently updadted papers are outdated in 30 days
now = pd.Timestamp.today()
# check outdated if 30 days passed
outdate = now - pd.Timedelta(30, unit="D")
recent_tb = read_md("Categorized/recent.md")
if recent_tb is not None and len(recent_tb):
recent_tb["Date"] = pd.to_datetime(recent_tb["Date"])
recent_tb = recent_tb[recent_tb["Date"] > outdate]
# if there is some newly added papers
if len(diff):
arr = diff.to_numpy()
arr_out = []
for line in arr:
title, types, venue, code, year, state = line
arr_out.append(
np.array([title, types, venue, code, year, state, now]))
df_new = pd.DataFrame(
np.array(arr_out), columns=diff.columns.to_list() + ["Date"])
if recent_tb is not None and len(recent_tb):
recent_tb = df_new.append(recent_tb)
else:
recent_tb = df_new
recent_tb.drop_duplicates(subset=["Title"]).reset_index(drop=True)
recent_tb = recent_tb.sort_values(
"Date", kind="mergesort", ascending=False
).reset_index(drop=True)
# write back
if recent_tb is not None and len(recent_tb):
recent_tb["Date"] = recent_tb["Date"].dt.strftime("%Y-%m-%d")
write_md("Categorized/recent.md", recent_tb)
################################### Step6: Categorize papers by venue ######################################################################
arr = tb.to_numpy()
confs = [
"AAAI",
"IJCAI",
"ICLR",
"WWW",
"KDD",
"ICML",
"TKDE",
"CIKM",
"WSDM",
"NeurIPS",
"USENIX",
"ICDM",
"Arxiv",
"UAI",
"ICSE",
"ECAI",
"Others",
]
arr_out = []
for line in arr:
title, types, venue, code, year = line
pubs = None
for c in confs:
if c == "KDD" and c in venue and not ("PAKDD" in venue and "PKDD" in venue):
pubs = c
break
elif c in venue:
pubs = c
break
pubs = pubs or "Others"
arr_out.append(np.array([title, types, venue, code, year, pubs]))
tb_by_venue = pd.DataFrame(np.array(arr_out), columns=columns + ["Pubs"])
tb_by_venue = tb_by_venue.sort_values(
"Pubs", kind="mergesort").reset_index(drop=True)
for i, c in enumerate(confs):
t = tb_by_venue[tb_by_venue["Pubs"] == c].reset_index(drop=True)
t = t.drop('Pubs', axis=1)
mod = "w" if i == 0 else "a"
with open("Categorized/venue.md", mod, encoding="utf-8") as f:
f.writelines("# " + c + "\n")
t.to_markdown(f)
f.writelines("\n")
# ################################### Step7: Write README.md ######################################################################
# for i in range(100):
# if "(Updating " in content[i]:
# begin = content[i].index("(Updating ")
# end = content[i].index(")")
# content[i] = (
# content[i][:begin] + f"(Updating {len(tb)} papers)" + content[i][end + 1:]
# )
# with open("README.md", "w", encoding="utf-8") as f:
# for line in content:
# f.writelines(line)
print(f"{len(tb)} papers in total.")
print("#" * 10, "End", "#" * 10)