fixing more circular imports

dynamo/prediction/utils.py

@@ -8,7 +8,6 @@
 from tqdm import tqdm
 
 from ..dynamo_logger import main_warning
-from ..tools.utils import log1p_, nearest_neighbors
 from ..utils import isarray, normalize
 
 # import scipy.sparse as sp
@@ -335,6 +334,8 @@ def estimate_sigma(
     Returns:
         The estimated diffusion matrix.
     """
+    from ..tools.utils import nearest_neighbors
+
     if nbr_idx is None:
         nbr_idx = nearest_neighbors(X, X, k=num_nbrs)
 
@@ -450,6 +451,7 @@ def fetch_exprs(
     Returns:
         The expression data for the given genes and time points.
     """
+    from ..tools.utils import log1p_
     if type(genes) != list:
         genes = list(genes)
 

dynamo/simulation/evaluation.py

@@ -1,8 +1,6 @@
 import numpy as np
 from sklearn.metrics import mean_squared_error
 
-from ..tools.utils import einsum_correlation
-
 
 def evaluate(reference: np.ndarray, prediction: np.ndarray, metric: str = "cosine") -> float:
     """Function to evaluate the vector field related reference quantities vs. that from vector field prediction.
@@ -17,7 +15,7 @@ def evaluate(reference: np.ndarray, prediction: np.ndarray, metric: str = "cosin
     Returns:
         res: The score between the reference vs. reconstructed quantities based on the metric.
     """
-
+    from ..tools.utils import einsum_correlation
     if metric == "cosine":
         true_normalized = reference / (np.linalg.norm(reference, axis=1).reshape(-1, 1) + 1e-20)
         predict_normalized = prediction / (np.linalg.norm(prediction, axis=1).reshape(-1, 1) + 1e-20)

dynamo/vectorfield/stochastic_process.py

@@ -10,7 +10,7 @@
     generate_neighbor_keys,
     k_nearest_neighbors,
 )
-from ..tools.utils import log1p_
+
 from .utils import VecFldDict, vecfld_from_adata, vector_field_function
 
 
@@ -47,6 +47,8 @@ def diffusionMatrix(
             the diffusion matrix for each cell. A column `diffusion` corresponds to the square root of the sum of all
             elements for each cell's diffusion matrix will also be added.
     """
+
+    from ..tools.utils import log1p_
 
     if X_data is None or V_data is not None:
         if genes is not None: