Spaces:
Running
on
Zero
Running
on
Zero
update
Browse files- model_utils.py +3 -4
model_utils.py
CHANGED
@@ -7,7 +7,6 @@ from NN_classifier.simple_binary_classifier import Medium_Binary_Network
|
|
7 |
from feature_extraction import extract_features
|
8 |
import pandas as pd
|
9 |
|
10 |
-
DEVICE = torch.device('cuda' if torch.cuda.is_available() else 'cpu')
|
11 |
|
12 |
def load_model(model_dir='models/medium_binary_classifier'):
|
13 |
model_path = os.path.join(model_dir, 'nn_model.pt')
|
@@ -29,8 +28,8 @@ def load_model(model_dir='models/medium_binary_classifier'):
|
|
29 |
|
30 |
input_size = scaler.n_features_in_
|
31 |
|
32 |
-
model = Medium_Binary_Network(input_size, hidden_sizes=[256, 192, 128, 64], dropout=0.3)
|
33 |
-
model.load_state_dict(torch.load(model_path
|
34 |
model.eval()
|
35 |
|
36 |
if imputer is not None:
|
@@ -76,7 +75,7 @@ def classify_text(text, model, scaler, label_encoder, imputer=None, scores=None)
|
|
76 |
|
77 |
features_scaled = scaler.transform(features)
|
78 |
|
79 |
-
features_tensor = torch.FloatTensor(features_scaled)
|
80 |
|
81 |
with torch.no_grad():
|
82 |
outputs = model(features_tensor)
|
|
|
7 |
from feature_extraction import extract_features
|
8 |
import pandas as pd
|
9 |
|
|
|
10 |
|
11 |
def load_model(model_dir='models/medium_binary_classifier'):
|
12 |
model_path = os.path.join(model_dir, 'nn_model.pt')
|
|
|
28 |
|
29 |
input_size = scaler.n_features_in_
|
30 |
|
31 |
+
model = Medium_Binary_Network(input_size, hidden_sizes=[256, 192, 128, 64], dropout=0.3)
|
32 |
+
model.load_state_dict(torch.load(model_path))
|
33 |
model.eval()
|
34 |
|
35 |
if imputer is not None:
|
|
|
75 |
|
76 |
features_scaled = scaler.transform(features)
|
77 |
|
78 |
+
features_tensor = torch.FloatTensor(features_scaled)
|
79 |
|
80 |
with torch.no_grad():
|
81 |
outputs = model(features_tensor)
|