fix LV

Author: arrjon

bayesflow/simulators/benchmark_simulators/lotka_volterra.py

@@ -116,5 +116,5 @@ def observation_model(self, params: np.ndarray) -> np.ndarray:
         # Add noise, decide whether to flatten and return
         x = self.rng.lognormal(np.log(pp), sigma=self.obs_noise)
         if self.flatten:
-            return x.flatten()
+            return x.T.flatten()
         return x