Spaces:
Runtime error
Runtime error
theresatvan
commited on
Commit
•
8da68fd
1
Parent(s):
81414ba
Add dropdown options
Browse files
app.py
CHANGED
@@ -1,10 +1,32 @@
|
|
1 |
import streamlit as st
|
|
|
|
|
|
|
2 |
from datasets import load_dataset
|
|
|
3 |
from transformers import DistilBertForSequenceClassification, DistilBertTokenizer
|
4 |
|
5 |
decision_to_str = {'REJECTED': 0, 'ACCEPTED': 1, 'PENDING': 2, 'CONT-REJECTED': 3, 'CONT-ACCEPTED': 4, 'CONT-PENDING': 5}
|
6 |
|
7 |
dataset_dict = load_dataset('HUPD/hupd',
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
8 |
name='all',
|
9 |
data_files="https://huggingface.co/datasets/HUPD/hupd/blob/main/hupd_metadata_2022-02-22.feather",
|
10 |
icpr_label=None,
|
@@ -22,6 +44,7 @@ tokenizer_abstract = DistilBertTokenizer('theresatvan/hupd-distilbert-abstract')
|
|
22 |
|
23 |
model_claims = DistilBertForSequenceClassification('theresatvan/hupd-distilbert-claims')
|
24 |
tokenizer_claims = DistilBertTokenizer('theresatvan/hupd-distilbert-claims')
|
|
|
25 |
|
26 |
|
27 |
def predict(model_abstract, model_claims, tokenizer_abstract, tokenizer_claims, input):
|
@@ -52,13 +75,21 @@ if __name__ == '__main__':
|
|
52 |
st.title = "Can I Patent This?"
|
53 |
|
54 |
form = st.form('patent-prediction-form')
|
|
|
|
|
|
|
55 |
dropdown = []
|
|
|
56 |
|
57 |
input_application = form.selectbox('Select a patent\'s application number', patents_dropdown)
|
58 |
submit = form.form_submit_button("Submit")
|
59 |
|
60 |
if submit:
|
|
|
|
|
|
|
61 |
input = dataset.filter(lambda e: e['application_number'] == input_application)
|
|
|
62 |
|
63 |
label, prob = predict(model_abstract, model_claims, tokenizer_abstract, tokenizer_claims, input)
|
64 |
|
|
|
1 |
import streamlit as st
|
2 |
+
<<<<<<< HEAD
|
3 |
+
from datasets import load_dataset, Features, Value, Sequence
|
4 |
+
=======
|
5 |
from datasets import load_dataset
|
6 |
+
>>>>>>> 81414ba96ac55f927033c62ee5c2db6c6a22349c
|
7 |
from transformers import DistilBertForSequenceClassification, DistilBertTokenizer
|
8 |
|
9 |
decision_to_str = {'REJECTED': 0, 'ACCEPTED': 1, 'PENDING': 2, 'CONT-REJECTED': 3, 'CONT-ACCEPTED': 4, 'CONT-PENDING': 5}
|
10 |
|
11 |
dataset_dict = load_dataset('HUPD/hupd',
|
12 |
+
<<<<<<< HEAD
|
13 |
+
name='sample',
|
14 |
+
data_files="https://huggingface.co/datasets/HUPD/hupd/blob/main/hupd_metadata_2022-02-22.feather",
|
15 |
+
icpr_label=None,
|
16 |
+
train_filing_start_date='2016-01-01',
|
17 |
+
train_filing_end_date='2016-01-21',
|
18 |
+
val_filing_start_date='2016-01-22',
|
19 |
+
val_filing_end_date='2016-01-31',
|
20 |
+
)
|
21 |
+
|
22 |
+
dataset = dataset_dict['validation']
|
23 |
+
|
24 |
+
model_abstract = DistilBertForSequenceClassification.from_pretrained('theresatvan/hupd-distilbert-abstract')
|
25 |
+
tokenizer_abstract = DistilBertTokenizer.from_pretrained('theresatvan/hupd-distilbert-abstract')
|
26 |
+
|
27 |
+
model_claims = DistilBertForSequenceClassification.from_pretrained('theresatvan/hupd-distilbert-claims')
|
28 |
+
tokenizer_claims = DistilBertTokenizer.from_pretrained('theresatvan/hupd-distilbert-claims')
|
29 |
+
=======
|
30 |
name='all',
|
31 |
data_files="https://huggingface.co/datasets/HUPD/hupd/blob/main/hupd_metadata_2022-02-22.feather",
|
32 |
icpr_label=None,
|
|
|
44 |
|
45 |
model_claims = DistilBertForSequenceClassification('theresatvan/hupd-distilbert-claims')
|
46 |
tokenizer_claims = DistilBertTokenizer('theresatvan/hupd-distilbert-claims')
|
47 |
+
>>>>>>> 81414ba96ac55f927033c62ee5c2db6c6a22349c
|
48 |
|
49 |
|
50 |
def predict(model_abstract, model_claims, tokenizer_abstract, tokenizer_claims, input):
|
|
|
75 |
st.title = "Can I Patent This?"
|
76 |
|
77 |
form = st.form('patent-prediction-form')
|
78 |
+
<<<<<<< HEAD
|
79 |
+
dropdown = [example['application_number'] for example in dataset]
|
80 |
+
=======
|
81 |
dropdown = []
|
82 |
+
>>>>>>> 81414ba96ac55f927033c62ee5c2db6c6a22349c
|
83 |
|
84 |
input_application = form.selectbox('Select a patent\'s application number', patents_dropdown)
|
85 |
submit = form.form_submit_button("Submit")
|
86 |
|
87 |
if submit:
|
88 |
+
<<<<<<< HEAD
|
89 |
+
input = dataset.filter(lambda e: e['patent_number'] == input_application)
|
90 |
+
=======
|
91 |
input = dataset.filter(lambda e: e['application_number'] == input_application)
|
92 |
+
>>>>>>> 81414ba96ac55f927033c62ee5c2db6c6a22349c
|
93 |
|
94 |
label, prob = predict(model_abstract, model_claims, tokenizer_abstract, tokenizer_claims, input)
|
95 |
|