Update app.py

app.py

@@ -30,10 +30,10 @@ matplotlib.rcParams['savefig.dpi'] = 400
 # Initialize models and other global variables
 device = 'cuda'
 
-mtcnn = MTCNN(keep_all=False, device=device, thresholds=[0.98, 0.98, 0.98], min_face_size=100)
+mtcnn = MTCNN(keep_all=False, device=device, thresholds=[0.985, 0.985, 0.985], min_face_size=80)
 model = InceptionResnetV1(pretrained='vggface2').eval().to(device)
 mp_face_mesh = mp.solutions.face_mesh
-face_mesh = mp_face_mesh.FaceMesh(static_image_mode=False, max_num_faces=1, min_detection_confidence=0.7)
+face_mesh = mp_face_mesh.FaceMesh(static_image_mode=False, max_num_faces=1, min_detection_confidence=0.8)
 emotion_detector = FER(mtcnn=False)
 
 
@@ -277,7 +277,7 @@ class LSTMAutoencoder(nn.Module):
         return out
 
 
-def lstm_anomaly_detection(X, feature_columns, raw_embedding_columns, epochs=100, batch_size=64):
+def lstm_anomaly_detection(X, feature_columns, raw_embedding_columns, epochs=100, batch_size=8):
     device = 'cuda'
     X = torch.FloatTensor(X).to(device)
     if X.dim() == 2:
@@ -322,7 +322,7 @@ def lstm_anomaly_detection(X, feature_columns, raw_embedding_columns, epochs=100
 
     return mse_all, mse_comp, mse_raw
 
-def embedding_anomaly_detection(embeddings, epochs=100, batch_size=64):
+def embedding_anomaly_detection(embeddings, epochs=100, batch_size=8):
     device = 'cuda'
     X = torch.FloatTensor(embeddings).to(device)
     if X.dim() == 2:
@@ -349,7 +349,7 @@ def embedding_anomaly_detection(embeddings, epochs=100, batch_size=64):
     mse = np.mean(np.power(X.squeeze(0).cpu().numpy() - reconstructed, 2), axis=1)
     return mse
 
-def determine_anomalies(mse_values, threshold=4):
+def determine_anomalies(mse_values, threshold=5):
     mean = np.mean(mse_values)
     std = np.std(mse_values)
     anomalies = mse_values > (mean + threshold * std)
@@ -558,7 +558,7 @@ iface = gr.Interface(
     inputs=[
         gr.Video(),
         gr.Slider(minimum=1, maximum=20, step=1, value=10, label="Desired FPS"),
-        gr.Slider(minimum=1, maximum=32, step=1, value=16, label="Batch Size")
+        gr.Slider(minimum=1, maximum=32, step=1, value=8, label="Batch Size")
     ],
     outputs=[
         gr.Textbox(label="Anomaly Detection Results"),