John Graham Reynolds
commited on
Commit
β’
9230b05
1
Parent(s):
f2422c4
updated app description
Browse files
app.py
CHANGED
@@ -4,20 +4,30 @@ from fixed_recall import FixedRecall
|
|
4 |
|
5 |
import gradio as gr
|
6 |
|
7 |
-
title = "π€ Evaluate Fix!"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
8 |
|
9 |
-
description = """
|
10 |
-
This space is designed to show off a workaround for a current bug in the π€ Evaluate library as we introduce ourselves to the entirety of the π€ ecosystem.
|
11 |
"""
|
12 |
|
13 |
-
article = "Check out [original repo](https://
|
14 |
-
classicifcation model](https://github.com/johngrahamreynolds/RoBERTa-base-DReiFT/tree/main) that makes use of it during evaluation
|
15 |
|
16 |
def show_off(input):
|
17 |
f1 = FixedF1()
|
18 |
precision = FixedPrecision()
|
19 |
recall = FixedRecall()
|
20 |
|
|
|
|
|
21 |
return "Checking this out! Here's what you put in: " + f"""{input} """
|
22 |
|
23 |
|
|
|
4 |
|
5 |
import gradio as gr
|
6 |
|
7 |
+
title = "'Combine' multiple metrics with this π€ Evaluate πͺ² Fix!"
|
8 |
+
|
9 |
+
description = """<p style='text-align: center'>
|
10 |
+
As I introduce myself to the entirety of the π€ ecosystem, I've put together this space to show off a workaround for a current πͺ² in the π€ Evaluate library. \n
|
11 |
+
|
12 |
+
\tCheck out the original, longstanding issue [here](https://github.com/huggingface/evaluate/issues/234). This details how it is currently impossible to \
|
13 |
+
'evaluate.combine()' multiple metrics related to multilabel text classification. Particularly, one cannot 'combine()' the f1, precision, and recall scores for \
|
14 |
+
evaluation. I encountered this issue specifically while training [RoBERTa-base-DReiFT](https://huggingface.co/MarioBarbeque/RoBERTa-base-DReiFT) for multilabel \
|
15 |
+
text classification of 805 labeled medical conditions based on drug reviews for treatment received for the same underlying conditio. Use the space below for \
|
16 |
+
a preview of the workaround! </p>
|
17 |
+
|
18 |
|
|
|
|
|
19 |
"""
|
20 |
|
21 |
+
article = "<p style='text-align: center'>Check out the [original repo](https://github.com/johngrahamreynolds/FixedMetricsForHF) housing this code, and a quickly \
|
22 |
+
trained [multilabel text classicifcation model](https://github.com/johngrahamreynolds/RoBERTa-base-DReiFT/tree/main) that makes use of it during evaluation.</p>"
|
23 |
|
24 |
def show_off(input):
|
25 |
f1 = FixedF1()
|
26 |
precision = FixedPrecision()
|
27 |
recall = FixedRecall()
|
28 |
|
29 |
+
|
30 |
+
|
31 |
return "Checking this out! Here's what you put in: " + f"""{input} """
|
32 |
|
33 |
|