Datasets:
ruanchaves
commited on
Commit
•
30a0c9e
1
Parent(s):
6bbffc9
timestamp
Browse files- b2w-reviews01.py +17 -3
b2w-reviews01.py
CHANGED
@@ -42,7 +42,7 @@ class Reviews(datasets.GeneratorBasedBuilder):
|
|
42 |
description=_DESCRIPTION,
|
43 |
features=datasets.Features(
|
44 |
dict(
|
45 |
-
submission_date=datasets.Value("
|
46 |
reviewer_id=datasets.Value("string"),
|
47 |
product_id=datasets.Value("string"),
|
48 |
product_name=datasets.Value("string"),
|
@@ -50,8 +50,8 @@ class Reviews(datasets.GeneratorBasedBuilder):
|
|
50 |
site_category_lv1=datasets.Value("string"),
|
51 |
site_category_lv2=datasets.Value("string"),
|
52 |
review_title=datasets.Value("string"),
|
53 |
-
overall_rating=datasets.Value("
|
54 |
-
recommend_to_a_friend=datasets.Value("
|
55 |
review_text=datasets.Value("string"),
|
56 |
reviewer_birth_year=datasets.Value("int32"),
|
57 |
reviewer_gender=datasets.Value("string"),
|
@@ -74,10 +74,24 @@ class Reviews(datasets.GeneratorBasedBuilder):
|
|
74 |
|
75 |
def _generate_examples(self, filepath):
|
76 |
def process_row(row):
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
77 |
try:
|
78 |
row["reviewer_birth_year"] = int(float(row["reviewer_birth_year"]))
|
79 |
except ValueError:
|
80 |
row["reviewer_birth_year"] = None
|
|
|
81 |
return row
|
82 |
records = pd.read_csv(filepath).to_dict("records")
|
83 |
for idx, row in enumerate(records):
|
|
|
42 |
description=_DESCRIPTION,
|
43 |
features=datasets.Features(
|
44 |
dict(
|
45 |
+
submission_date=datasets.Value("timestamp[(s)]"),
|
46 |
reviewer_id=datasets.Value("string"),
|
47 |
product_id=datasets.Value("string"),
|
48 |
product_name=datasets.Value("string"),
|
|
|
50 |
site_category_lv1=datasets.Value("string"),
|
51 |
site_category_lv2=datasets.Value("string"),
|
52 |
review_title=datasets.Value("string"),
|
53 |
+
overall_rating=datasets.Value("int32"),
|
54 |
+
recommend_to_a_friend=datasets.Value("bool"),
|
55 |
review_text=datasets.Value("string"),
|
56 |
reviewer_birth_year=datasets.Value("int32"),
|
57 |
reviewer_gender=datasets.Value("string"),
|
|
|
74 |
|
75 |
def _generate_examples(self, filepath):
|
76 |
def process_row(row):
|
77 |
+
|
78 |
+
recommend_key = {
|
79 |
+
"Yes": True,
|
80 |
+
"No": False,
|
81 |
+
"nan": None
|
82 |
+
}
|
83 |
+
|
84 |
+
row["recommend_to_a_friend"] = recommend_key[row["recommend_to_a_friend"]]
|
85 |
+
|
86 |
+
to_int_list = ["overall_rating", "product_id"]
|
87 |
+
for row_value in to_int_list:
|
88 |
+
row[row_value] = int(row[row_value])
|
89 |
+
|
90 |
try:
|
91 |
row["reviewer_birth_year"] = int(float(row["reviewer_birth_year"]))
|
92 |
except ValueError:
|
93 |
row["reviewer_birth_year"] = None
|
94 |
+
|
95 |
return row
|
96 |
records = pd.read_csv(filepath).to_dict("records")
|
97 |
for idx, row in enumerate(records):
|