Support Vector Machine Regression

from sklearn.svm import SVR

steps = [
  ('scaler', RobustScaler()),
  ('svr', SVR()),
]
np.random.seed(0)
reg_svr = Pipeline(steps=steps)
reg_svr.fit(x_train, y_train)

predictions = reg_svr.predict(x_val)
plt.figure()
plt.scatter(y_val, predictions, s=5, alpha=0.2)
plt.xlabel('RUL (true)')
plt.ylabel('RUL')

predictions[predictions < 1] = 1
print('MSE', mean_squared_error(y_val, predictions))
print('R2', r2_score(y_val, predictions))
print('score', custom_loss(y_val, predictions))

      Linear    ElasticNet  SVR
===================================
MSE   1197.8    1167.3      736.0
R2    0.68183   0.68992     0.80450
score inf       inf         inf