Spaces:
Runtime error
Runtime error
jkjk
Browse files
app.py
CHANGED
|
@@ -1,11 +1,10 @@
|
|
| 1 |
#Importing necessary libraries
|
| 2 |
import gradio as gr
|
| 3 |
-
#import scikit-learn as sklearn
|
| 4 |
from fastai.vision.all import *
|
| 5 |
from sklearn.metrics import roc_auc_score
|
| 6 |
|
| 7 |
#Define dependent functions
|
| 8 |
-
def get_x(row): return Path(str(path/f"{row['rootname']}_small"/f"{row['ID']}") + ".png")
|
| 9 |
def get_y(row): return row["LABEL"]
|
| 10 |
|
| 11 |
def auroc_score(input, target):
|
|
@@ -15,31 +14,43 @@ def auroc_score(input, target):
|
|
| 15 |
#Load the model
|
| 16 |
learn = load_learner("export.pkl")
|
| 17 |
|
| 18 |
-
#Identify labels from the dataloaders class
|
| 19 |
labels = ["Negative", "Positive"]
|
| 20 |
|
| 21 |
#Define function for making prediction
|
| 22 |
def predict(img):
|
| 23 |
-
img = PILImage.create(img)
|
| 24 |
pred, idx, probs = learn.predict(img)
|
| 25 |
return {labels[i]: float(probs[i]) for i in range(len(labels))}
|
| 26 |
|
| 27 |
-
#
|
| 28 |
-
|
|
|
|
|
|
|
|
|
|
| 29 |
|
| 30 |
-
|
|
|
|
|
|
|
| 31 |
|
| 32 |
-
|
|
|
|
|
|
|
|
|
|
| 33 |
|
| 34 |
-
|
|
|
|
|
|
|
| 35 |
|
|
|
|
| 36 |
enable_queue=True
|
| 37 |
|
| 38 |
-
|
| 39 |
-
|
| 40 |
-
|
| 41 |
-
outputs=gr.outputs.Label(num_top_classes=1),
|
| 42 |
title=title,
|
| 43 |
-
description=description,
|
| 44 |
examples=examples,
|
| 45 |
-
enable_queue=enable_queue
|
|
|
|
|
|
| 1 |
#Importing necessary libraries
|
| 2 |
import gradio as gr
|
|
|
|
| 3 |
from fastai.vision.all import *
|
| 4 |
from sklearn.metrics import roc_auc_score
|
| 5 |
|
| 6 |
#Define dependent functions
|
| 7 |
+
def get_x(row): return Path(str(path/f"{row['rootname']}_small"/f"{row['ID']}") + ".png")
|
| 8 |
def get_y(row): return row["LABEL"]
|
| 9 |
|
| 10 |
def auroc_score(input, target):
|
|
|
|
| 14 |
#Load the model
|
| 15 |
learn = load_learner("export.pkl")
|
| 16 |
|
| 17 |
+
#Identify labels from the dataloaders class
|
| 18 |
labels = ["Negative", "Positive"]
|
| 19 |
|
| 20 |
#Define function for making prediction
|
| 21 |
def predict(img):
|
| 22 |
+
img = PILImage.create(img)
|
| 23 |
pred, idx, probs = learn.predict(img)
|
| 24 |
return {labels[i]: float(probs[i]) for i in range(len(labels))}
|
| 25 |
|
| 26 |
+
#Custom CSS
|
| 27 |
+
custom_css = """
|
| 28 |
+
body, button {
|
| 29 |
+
background-color: #00FBB9;
|
| 30 |
+
}
|
| 31 |
|
| 32 |
+
button {
|
| 33 |
+
color: #333;
|
| 34 |
+
}
|
| 35 |
|
| 36 |
+
p, span {
|
| 37 |
+
color: #333;
|
| 38 |
+
}
|
| 39 |
+
"""
|
| 40 |
|
| 41 |
+
#Customizing the gradio interface
|
| 42 |
+
title = "Ethiopian based Tuberculosis(TB) model"
|
| 43 |
+
description = """Model to detect TB from chest x-rays"""
|
| 44 |
|
| 45 |
+
examples = ['patient1.png', 'patient2.png', 'patient3.png']
|
| 46 |
enable_queue=True
|
| 47 |
|
| 48 |
+
#Launching the gradio interface
|
| 49 |
+
gr.Interface(fn=predict,
|
| 50 |
+
inputs=gr.inputs.Image(shape=(512, 512)),
|
| 51 |
+
outputs=gr.outputs.Label(num_top_classes=1),
|
| 52 |
title=title,
|
| 53 |
+
description=description,
|
| 54 |
examples=examples,
|
| 55 |
+
enable_queue=enable_queue,
|
| 56 |
+
css=custom_css).launch(inline=False)
|