Commit
·
4ee9c74
1
Parent(s):
9d55375
Update app.py
Browse files
app.py
CHANGED
|
@@ -14,16 +14,17 @@ model_name = "valurank/finetuned-distilbert-news-article-categorization"
|
|
| 14 |
model = AutoModelForSequenceClassification.from_pretrained(model_name)
|
| 15 |
tokenizer = AutoTokenizer.from_pretrained(model_name)
|
| 16 |
|
|
|
|
| 17 |
#Reading in the text file
|
| 18 |
def read_in_text(url):
|
| 19 |
with open(url, 'r') as file:
|
| 20 |
article = file.read()
|
| 21 |
|
| 22 |
return article
|
| 23 |
-
|
| 24 |
-
|
| 25 |
-
|
| 26 |
-
text =
|
| 27 |
"ascii"
|
| 28 |
) # remove non-ascii, Chinese characters
|
| 29 |
|
|
@@ -41,8 +42,8 @@ def clean_text(url):
|
|
| 41 |
return text
|
| 42 |
|
| 43 |
#Defining a function to get the category of the news article
|
| 44 |
-
def get_category(
|
| 45 |
-
text = clean_text(
|
| 46 |
|
| 47 |
input_tensor = tokenizer.encode(text, return_tensors="pt", truncation=True)
|
| 48 |
logits = model(input_tensor).logits
|
|
|
|
| 14 |
model = AutoModelForSequenceClassification.from_pretrained(model_name)
|
| 15 |
tokenizer = AutoTokenizer.from_pretrained(model_name)
|
| 16 |
|
| 17 |
+
"""
|
| 18 |
#Reading in the text file
|
| 19 |
def read_in_text(url):
|
| 20 |
with open(url, 'r') as file:
|
| 21 |
article = file.read()
|
| 22 |
|
| 23 |
return article
|
| 24 |
+
"""
|
| 25 |
+
|
| 26 |
+
def clean_text(raw_text):
|
| 27 |
+
text = raw_text.encode("ascii", errors="ignore").decode(
|
| 28 |
"ascii"
|
| 29 |
) # remove non-ascii, Chinese characters
|
| 30 |
|
|
|
|
| 42 |
return text
|
| 43 |
|
| 44 |
#Defining a function to get the category of the news article
|
| 45 |
+
def get_category(text):
|
| 46 |
+
text = clean_text(text)
|
| 47 |
|
| 48 |
input_tensor = tokenizer.encode(text, return_tensors="pt", truncation=True)
|
| 49 |
logits = model(input_tensor).logits
|