diff --git a/dabest/__init__.py b/dabest/__init__.py
index 6f7d114e..109a9822 100644
--- a/dabest/__init__.py
+++ b/dabest/__init__.py
@@ -1,6 +1,14 @@
 from ._api import load, prop_dataset
 from ._stats_tools import effsize as effsize
+from ._stats_tools import confint_2group_diff as ci_2g
 from ._effsize_objects import TwoGroupsEffectSize, PermutationTest
 from ._dabest_object import Dabest
 
-__version__ = "2024.03.29"
+
+import os
+if os.environ.get('SKIP_NUMBA_COMPILE') != '1':
+    from ._stats_tools.precompile import precompile_all, _NUMBA_COMPILED
+    if not _NUMBA_COMPILED:
+        precompile_all()
+
+__version__ = "2024.03.30"
\ No newline at end of file
diff --git a/dabest/_bootstrap_tools.py b/dabest/_bootstrap_tools.py
index 0951ffb5..7a3e979c 100644
--- a/dabest/_bootstrap_tools.py
+++ b/dabest/_bootstrap_tools.py
@@ -66,7 +66,9 @@ def __init__(
         reps: int = 5000,  # Number of bootstrap iterations to perform.
     ):
         # Turn to pandas series.
-        x1 = pd.Series(x1).dropna()
+        # x1 = pd.Series(x1).dropna()
+        x1 = x1[~np.isnan(x1)]
+
         diff = False
 
         # Initialise stat_function
@@ -89,7 +91,9 @@ def __init__(
             if x2 is None:
                 raise ValueError("Please specify x2.")
             
-            x2 = pd.Series(x2).dropna()
+            # x2 = pd.Series(x2).dropna()
+            x2 = x1[~np.isnan(x2)]
+
             if len(x1) != len(x2):
                 raise ValueError("x1 and x2 are not the same length.")
 
@@ -134,7 +138,8 @@ def __init__(
 
         elif x2 is not None and paired is None:
             diff = True
-            x2 = pd.Series(x2).dropna()
+            # x2 = pd.Series(x2).dropna()
+            x2 = x2[~np.isnan(x2)]
             # Generate statarrays for both arrays.
             ref_statarray = sns.algorithms.bootstrap(x1, **sns_bootstrap_kwargs)
             exp_statarray = sns.algorithms.bootstrap(x2, **sns_bootstrap_kwargs)
diff --git a/dabest/_dabest_object.py b/dabest/_dabest_object.py
index d5f0db01..8e260887 100644
--- a/dabest/_dabest_object.py
+++ b/dabest/_dabest_object.py
@@ -112,7 +112,7 @@ def __init__(
         # Determine the kind of estimation plot we need to produce.
         if all([isinstance(i, (str, int, float)) for i in idx]):
             # flatten out idx.
-            all_plot_groups = pd.unique(pd.Series([t for t in idx])).tolist()
+            all_plot_groups = pd.Series([t for t in idx]).unique().tolist()
             if len(idx) > len(all_plot_groups):
                 err0 = "`idx` contains duplicated groups. Please remove any duplicates and try again."
                 raise ValueError(err0)
@@ -122,7 +122,7 @@ def __init__(
             self.__idx = (idx,)
 
         elif all([isinstance(i, (tuple, list)) for i in idx]):
-            all_plot_groups = pd.unique(pd.Series([tt for t in idx for tt in t])).tolist()
+            all_plot_groups = pd.Series([tt for t in idx for tt in t]).unique().tolist()
 
             actual_groups_given = sum([len(i) for i in idx])
 
diff --git a/dabest/_delta_objects.py b/dabest/_delta_objects.py
index 1827c1b2..df896573 100644
--- a/dabest/_delta_objects.py
+++ b/dabest/_delta_objects.py
@@ -388,13 +388,14 @@ def __init__(self, effectsizedataframe, permutation_count,
         # compute the variances of each control group and each test group
         control_var=[]
         test_var=[]
+        grouped_data = {name: group[yvar].copy() for name, group in dat.groupby(xvar, observed=False)}
         for j, current_tuple in enumerate(idx):
             cname = current_tuple[0]
-            control = dat[dat[xvar] == cname][yvar].copy()
+            control = grouped_data[cname]
             control_var.append(np.var(control, ddof=1))
 
             tname = current_tuple[1]
-            test = dat[dat[xvar] == tname][yvar].copy()
+            test = grouped_data[tname]
             test_var.append(np.var(test, ddof=1))
         self.__control_var = np.array(control_var)
         self.__test_var    = np.array(test_var)
@@ -414,7 +415,7 @@ def __init__(self, effectsizedataframe, permutation_count,
                                                           self.__bootstraps)
 
         # Compute the weighted average mean difference based on the raw data
-        self.__difference = es.weighted_delta(self.__effsizedf["difference"],
+        self.__difference = es.weighted_delta(np.array(self.__effsizedf["difference"]),
                                                    self.__group_var)
 
         sorted_weighted_deltas = npsort(self.__bootstraps_weighted_delta)
diff --git a/dabest/_effsize_objects.py b/dabest/_effsize_objects.py
index b562b603..cbcb55ac 100644
--- a/dabest/_effsize_objects.py
+++ b/dabest/_effsize_objects.py
@@ -9,6 +9,7 @@
 import pandas as pd
 import lqrt
 from scipy.stats import norm
+import numpy as np
 from numpy import array, isnan, isinf, repeat, random, isin, abs, var
 from numpy import sort as npsort
 from numpy import nan as npnan
@@ -357,12 +358,17 @@ def _perform_statistical_test(self):
             # References:
             # https://en.wikipedia.org/wiki/McNemar%27s_test
 
-            df_temp = pd.DataFrame({"control": self.__control, "test": self.__test})
-            x1 = len(df_temp[(df_temp["control"] == 0) & (df_temp["test"] == 0)])
-            x2 = len(df_temp[(df_temp["control"] == 0) & (df_temp["test"] == 1)])
-            x3 = len(df_temp[(df_temp["control"] == 1) & (df_temp["test"] == 0)])
-            x4 = len(df_temp[(df_temp["control"] == 1) & (df_temp["test"] == 1)])
-            table = [[x1, x2], [x3, x4]]
+            # df_temp = pd.DataFrame({"control": self.__control, "test": self.__test})
+            # x1 = len(df_temp[(df_temp["control"] == 0) & (df_temp["test"] == 0)])
+            # x2 = len(df_temp[(df_temp["control"] == 0) & (df_temp["test"] == 1)])
+            # x3 = len(df_temp[(df_temp["control"] == 1) & (df_temp["test"] == 0)])
+            # x4 = len(df_temp[(df_temp["control"] == 1) & (df_temp["test"] == 1)])
+            # table = [[x1, x2], [x3, x4]]
+            x1 = np.sum((self.__control == 0) & (self.__test == 0))
+            x2 = np.sum((self.__control == 0) & (self.__test == 1))
+            x3 = np.sum((self.__control == 1) & (self.__test == 0))
+            x4 = np.sum((self.__control == 1) & (self.__test == 1))
+            table = np.array([[x1, x2], [x3, x4]])
             _mcnemar = mcnemar(table, exact=True, correction=True)
             self.__pvalue_mcnemar = _mcnemar.pvalue
             self.__statistic_mcnemar = _mcnemar.statistic
@@ -861,18 +867,19 @@ def __pre_calc(self):
         out = []
         reprs = []
 
+        grouped_data = {name: group[yvar].copy() for name, group in dat.groupby(xvar, observed=False)}
         if self.__delta2:
             mixed_data = []
             for j, current_tuple in enumerate(idx):
                 if self.__is_paired != "sequential":
                     cname = current_tuple[0]
-                    control = dat[dat[xvar] == cname][yvar].copy()
+                    control = grouped_data[cname]
 
                 for ix, tname in enumerate(current_tuple[1:]):
                     if self.__is_paired == "sequential":
                         cname = current_tuple[ix]
-                        control = dat[dat[xvar] == cname][yvar].copy()
-                    test = dat[dat[xvar] == tname][yvar].copy()
+                        control = grouped_data[cname]
+                    test = grouped_data[tname]
                     mixed_data.append(control)
                     mixed_data.append(test)
             bootstraps_delta_delta = ci2g.compute_delta2_bootstrapped_diff(
@@ -888,13 +895,13 @@ def __pre_calc(self):
         for j, current_tuple in enumerate(idx):
             if self.__is_paired != "sequential":
                 cname = current_tuple[0]
-                control = dat[dat[xvar] == cname][yvar].copy()
+                control = grouped_data[cname]
 
             for ix, tname in enumerate(current_tuple[1:]):
                 if self.__is_paired == "sequential":
                     cname = current_tuple[ix]
-                    control = dat[dat[xvar] == cname][yvar].copy()
-                test = dat[dat[xvar] == tname][yvar].copy()
+                    control = grouped_data[cname]
+                test = grouped_data[tname]
 
                 result = TwoGroupsEffectSize(
                     control,
@@ -1055,16 +1062,18 @@ def __calc_lqrt(self):
 
         out = []
 
+        grouped_data = {name:group[yvar].copy() for name, group in dat.groupby(xvar)}
+
         for j, current_tuple in enumerate(db_obj.idx):
             if self.__is_paired != "sequential":
                 cname = current_tuple[0]
-                control = dat[dat[xvar] == cname][yvar].copy()
+                control = grouped_data[cname]
 
             for ix, tname in enumerate(current_tuple[1:]):
                 if self.__is_paired == "sequential":
                     cname = current_tuple[ix]
-                    control = dat[dat[xvar] == cname][yvar].copy()
-                test = dat[dat[xvar] == tname][yvar].copy()
+                    control = grouped_data[cname]
+                test = grouped_data[tname]
 
                 if self.__is_paired:
                     # Refactored here in v0.3.0 for performance issues.
diff --git a/dabest/_modidx.py b/dabest/_modidx.py
index eb260d99..f483a4be 100644
--- a/dabest/_modidx.py
+++ b/dabest/_modidx.py
@@ -25,6 +25,8 @@
                                                                                                                         'dabest/_stats_tools/confint_2group_diff.py'),
                                                          'dabest._stats_tools.confint_2group_diff._create_two_group_jackknife_indexes': ( 'API/confint_2group_diff.html#_create_two_group_jackknife_indexes',
                                                                                                                                           'dabest/_stats_tools/confint_2group_diff.py'),
+                                                         'dabest._stats_tools.confint_2group_diff.bootstrap_indices': ( 'API/confint_2group_diff.html#bootstrap_indices',
+                                                                                                                        'dabest/_stats_tools/confint_2group_diff.py'),
                                                          'dabest._stats_tools.confint_2group_diff.calculate_group_var': ( 'API/confint_2group_diff.html#calculate_group_var',
                                                                                                                           'dabest/_stats_tools/confint_2group_diff.py'),
                                                          'dabest._stats_tools.confint_2group_diff.calculate_weighted_delta': ( 'API/confint_2group_diff.html#calculate_weighted_delta',
@@ -42,11 +44,17 @@
                                                          'dabest._stats_tools.confint_2group_diff.create_jackknife_indexes': ( 'API/confint_2group_diff.html#create_jackknife_indexes',
                                                                                                                                'dabest/_stats_tools/confint_2group_diff.py'),
                                                          'dabest._stats_tools.confint_2group_diff.create_repeated_indexes': ( 'API/confint_2group_diff.html#create_repeated_indexes',
-                                                                                                                              'dabest/_stats_tools/confint_2group_diff.py')},
-            'dabest._stats_tools.effsize': { 'dabest._stats_tools.effsize._compute_hedges_correction_factor': ( 'API/effsize.html#_compute_hedges_correction_factor',
+                                                                                                                              'dabest/_stats_tools/confint_2group_diff.py'),
+                                                         'dabest._stats_tools.confint_2group_diff.delta2_bootstrap_loop': ( 'API/confint_2group_diff.html#delta2_bootstrap_loop',
+                                                                                                                            'dabest/_stats_tools/confint_2group_diff.py')},
+            'dabest._stats_tools.effsize': { 'dabest._stats_tools.effsize._cliffs_delta_core': ( 'API/effsize.html#_cliffs_delta_core',
+                                                                                                 'dabest/_stats_tools/effsize.py'),
+                                             'dabest._stats_tools.effsize._compute_hedges_correction_factor': ( 'API/effsize.html#_compute_hedges_correction_factor',
                                                                                                                 'dabest/_stats_tools/effsize.py'),
                                              'dabest._stats_tools.effsize._compute_standardizers': ( 'API/effsize.html#_compute_standardizers',
                                                                                                      'dabest/_stats_tools/effsize.py'),
+                                             'dabest._stats_tools.effsize._mann_whitney_u': ( 'API/effsize.html#_mann_whitney_u',
+                                                                                              'dabest/_stats_tools/effsize.py'),
                                              'dabest._stats_tools.effsize.cliffs_delta': ( 'API/effsize.html#cliffs_delta',
                                                                                            'dabest/_stats_tools/effsize.py'),
                                              'dabest._stats_tools.effsize.cohens_d': ( 'API/effsize.html#cohens_d',
@@ -61,6 +69,8 @@
                                                                                                    'dabest/_stats_tools/effsize.py'),
                                              'dabest._stats_tools.effsize.weighted_delta': ( 'API/effsize.html#weighted_delta',
                                                                                              'dabest/_stats_tools/effsize.py')},
+            'dabest._stats_tools.precompile': { 'dabest._stats_tools.precompile.precompile_all': ( 'API/precompile.html#precompile_all',
+                                                                                                   'dabest/_stats_tools/precompile.py')},
             'dabest.forest_plot': { 'dabest.forest_plot.extract_plot_data': ( 'API/forest_plot.html#extract_plot_data',
                                                                               'dabest/forest_plot.py'),
                                     'dabest.forest_plot.forest_plot': ('API/forest_plot.html#forest_plot', 'dabest/forest_plot.py'),
diff --git a/dabest/_stats_tools/confint_2group_diff.py b/dabest/_stats_tools/confint_2group_diff.py
index c599e178..afdb44b2 100644
--- a/dabest/_stats_tools/confint_2group_diff.py
+++ b/dabest/_stats_tools/confint_2group_diff.py
@@ -3,9 +3,10 @@
 # AUTOGENERATED! DO NOT EDIT! File to edit: ../../nbs/API/confint_2group_diff.ipynb.
 
 # %% auto 0
-__all__ = ['create_jackknife_indexes', 'create_repeated_indexes', 'compute_meandiff_jackknife', 'compute_bootstrapped_diff',
-           'compute_delta2_bootstrapped_diff', 'compute_meandiff_bias_correction', 'compute_interval_limits',
-           'calculate_group_var', 'calculate_weighted_delta']
+__all__ = ['create_jackknife_indexes', 'create_repeated_indexes', 'compute_meandiff_jackknife', 'bootstrap_indices',
+           'compute_bootstrapped_diff', 'delta2_bootstrap_loop', 'compute_delta2_bootstrapped_diff',
+           'compute_meandiff_bias_correction', 'compute_interval_limits', 'calculate_group_var',
+           'calculate_weighted_delta']
 
 # %% ../../nbs/API/confint_2group_diff.ipynb 4
 import numpy as np
@@ -13,11 +14,12 @@
 from numpy import mean as npmean
 from numpy import sum as npsum
 from numpy.random import PCG64, RandomState
-import pandas as pd
+from numba import njit, prange
 from scipy.stats import norm
 from numpy import isnan
 
 # %% ../../nbs/API/confint_2group_diff.ipynb 5
+@njit(cache=True, parallel=True)
 def create_jackknife_indexes(data):
     """
     Given an array-like, creates a jackknife bootstrap.
@@ -34,18 +36,25 @@ def create_jackknife_indexes(data):
     Generator that yields all jackknife bootstrap samples.
     """
 
-    index_range = arange(0, len(data))
-    return (delete(index_range, i) for i in index_range)
+    n = len(data)
+    indexes = np.empty((n, n - 1), dtype=np.int64)
+    for i in prange(n):
+        indexes[i] = np.concatenate((np.arange(i), np.arange(i + 1, n)))
+    return indexes
 
 
+@njit(cache=True, parallel=True)
 def create_repeated_indexes(data):
     """
     Convenience function. Given an array-like with length N,
     returns a generator that yields N indexes [0, 1, ..., N].
     """
 
-    index_range = arange(0, len(data))
-    return (index_range for i in index_range)
+    n = len(data)
+    indexes = np.empty((n, n), dtype=np.int64)  # Pre-allocate the output array
+    for i in prange(n):
+        indexes[i, :] = np.arange(n)  # Fill each row with the full index range
+    return indexes
 
 
 def _create_two_group_jackknife_indexes(x0, x1, is_paired):
@@ -113,6 +122,20 @@ def _calc_accel(jack_dist):
         return numer / denom
 
 
+@njit(cache=True) # parallelization must be turned off for random number generation
+def bootstrap_indices(is_paired, x0_len, x1_len, resamples, random_seed):
+    np.random.seed(random_seed)
+    indices = np.empty((resamples, x0_len if is_paired else x0_len + x1_len), dtype=np.int64)
+    
+    for i in range(resamples):
+        if is_paired:
+            indices[i, :x0_len] = np.random.choice(x0_len, x0_len)
+        else:  
+            indices[i, :x0_len] = np.random.choice(x0_len, x0_len)
+            indices[i, x0_len:x0_len+x1_len] = np.random.choice(x1_len, x1_len)
+    return indices
+
+
 def compute_bootstrapped_diff(
     x0, x1, is_paired, effect_size, resamples=5000, random_seed=12345
 ):
@@ -120,27 +143,60 @@ def compute_bootstrapped_diff(
 
     from . import effsize as __es
 
-    rng = RandomState(PCG64(random_seed))
-
-    out = np.repeat(np.nan, resamples)
-    x0_len = len(x0)
-    x1_len = len(x1)
+    x0_len, x1_len = len(x0), len(x1)
+    indices = bootstrap_indices(is_paired, x0_len, x1_len, resamples, random_seed)
+    out = np.empty(resamples, dtype=np.float64)
 
-    for i in range(int(resamples)):
+    for i in range(resamples):
         if is_paired:
-            if x0_len != x1_len:
-                raise ValueError("The two arrays do not have the same length.")
-            random_idx = rng.choice(x0_len, x0_len, replace=True)
-            x0_sample = x0[random_idx]
-            x1_sample = x1[random_idx]
+            x0_sample = x0[indices[i, :x0_len]]
+            x1_sample = x1[indices[i, :x0_len]]
         else:
-            x0_sample = rng.choice(x0, x0_len, replace=True)
-            x1_sample = rng.choice(x1, x1_len, replace=True)
+            x0_sample = x0[indices[i, :x0_len]]
+            x1_sample = x1[indices[i, x0_len:x0_len+x1_len]]
 
         out[i] = __es.two_group_difference(x0_sample, x1_sample, is_paired, effect_size)
 
     return out
 
+@njit(cache=True) # parallelization must be turned off for random number generation
+def delta2_bootstrap_loop(x1, x2, x3, x4, resamples, pooled_sd, rng_seed, is_paired):
+    np.random.seed(rng_seed)
+    out_delta_g = np.empty(resamples)
+    deltadelta = np.empty(resamples)
+    
+    n1, n2, n3, n4 = len(x1), len(x2), len(x3), len(x4)
+    if is_paired:
+        if n1 != n2 or n3 != n4:
+            raise ValueError("Each control group must have the same length as its corresponding test group in paired analysis.")
+    
+
+    # Bootstrapping
+    for i in range(resamples):
+        # Paired or unpaired resampling
+        if is_paired:
+            indices_1 = np.random.choice(len(x1),len(x1))
+            indices_2 = np.random.choice(len(x3),len(x3))
+            x1_sample, x2_sample = x1[indices_1], x2[indices_1]
+            x3_sample, x4_sample = x3[indices_2], x4[indices_2]
+        else:
+            indices_1 = np.random.randint(0, len(x1), len(x1))
+            indices_2 = np.random.randint(0, len(x2), len(x2))
+            indices_3 = np.random.randint(0, len(x3), len(x3))
+            indices_4 = np.random.randint(0, len(x4), len(x4))
+            x1_sample, x2_sample = x1[indices_1], x2[indices_2]
+            x3_sample, x4_sample = x3[indices_3], x4[indices_4]
+
+        # Calculating deltas
+        delta_1 = np.mean(x2_sample) - np.mean(x1_sample)
+        delta_2 = np.mean(x4_sample) - np.mean(x3_sample)
+        delta_delta = delta_2 - delta_1
+
+        deltadelta[i] = delta_delta
+        out_delta_g[i] = delta_delta / pooled_sd
+
+    return out_delta_g, deltadelta
+
 
 def compute_delta2_bootstrapped_diff(
     x1: np.ndarray,  # Control group 1
@@ -158,8 +214,6 @@ def compute_delta2_bootstrapped_diff(
 
     """
 
-    rng = RandomState(PCG64(random_seed))
-
     x1, x2, x3, x4 = map(np.asarray, [x1, x2, x3, x4])
 
     # Calculating pooled sample standard deviation
@@ -179,33 +233,7 @@ def compute_delta2_bootstrapped_diff(
     if np.isnan(pooled_sample_sd) or pooled_sample_sd == 0:
         raise ValueError("Pooled sample standard deviation is NaN or zero.")
 
-    out_delta_g = np.empty(resamples)
-    deltadelta = np.empty(resamples)
-
-    # Bootstrapping
-    for i in range(resamples):
-        # Paired or unpaired resampling
-        if is_paired:
-            if len(x1) != len(x2) or len(x3) != len(x4):
-                raise ValueError("Each control group must have the same length as its corresponding test group in paired analysis.")
-            indices_1 = rng.choice(len(x1), len(x1), replace=True)
-            indices_2 = rng.choice(len(x3), len(x3), replace=True)
-
-            x1_sample, x2_sample = x1[indices_1], x2[indices_1]
-            x3_sample, x4_sample = x3[indices_2], x4[indices_2]
-        else:
-            x1_sample = rng.choice(x1, len(x1), replace=True)
-            x2_sample = rng.choice(x2, len(x2), replace=True)
-            x3_sample = rng.choice(x3, len(x3), replace=True)
-            x4_sample = rng.choice(x4, len(x4), replace=True)
-
-        # Calculating deltas
-        delta_1 = np.mean(x2_sample) - np.mean(x1_sample)
-        delta_2 = np.mean(x4_sample) - np.mean(x3_sample)
-        delta_delta = delta_2 - delta_1
-
-        deltadelta[i] = delta_delta
-        out_delta_g[i] = delta_delta / pooled_sample_sd
+    out_delta_g, deltadelta = delta2_bootstrap_loop(x1, x2, x3, x4, resamples, pooled_sample_sd, random_seed, is_paired)
 
     # Empirical delta_g calculation
     delta_g = ((np.mean(x4) - np.mean(x3)) - (np.mean(x2) - np.mean(x1))) / pooled_sample_sd
@@ -242,6 +270,7 @@ def _compute_alpha_from_ci(ci):
     return (100.0 - ci) / 100.0
 
 
+@njit(cache=True)
 def _compute_quantile(z, bias, acceleration):
     numer = bias + z
     denom = 1 - (acceleration * numer)
@@ -277,6 +306,7 @@ def compute_interval_limits(bias, acceleration, n_boots, ci=95):
     return low, high
 
 
+@njit(cache=True)
 def calculate_group_var(control_var, control_N, test_var, test_N):
     return control_var / control_N + test_var / test_N
 
@@ -288,6 +318,7 @@ def calculate_weighted_delta(group_var, differences):
 
     weight = 1 / group_var
     denom = np.sum(weight)
-    num = np.sum(weight[i] * differences[i] for i in range(0, len(weight)))
-
+    num = 0.0
+    for i in range(len(weight)):
+        num += weight[i] * differences[i]
     return num / denom
diff --git a/dabest/_stats_tools/effsize.py b/dabest/_stats_tools/effsize.py
index 402745cc..2711d021 100644
--- a/dabest/_stats_tools/effsize.py
+++ b/dabest/_stats_tools/effsize.py
@@ -5,10 +5,12 @@
 # %% ../../nbs/API/effsize.ipynb 4
 from __future__ import annotations
 import numpy as np
+from numba import njit
 import warnings
 from scipy.special import gamma
 from scipy.stats import mannwhitneyu
 
+
 # %% auto 0
 __all__ = ['two_group_difference', 'func_difference', 'cohens_d', 'cohens_h', 'hedges_g', 'cliffs_delta', 'weighted_delta']
 
@@ -62,6 +64,10 @@ def two_group_difference(control:list|tuple|np.ndarray, #Accepts lists, tuples,
 
     """
 
+    if ~isinstance(control, np.ndarray):
+        control = np.array(control)
+    if ~isinstance(test, np.ndarray):
+        test    = np.array(test)
 
     if effect_size == "mean_diff":
         return func_difference(control, test, np.mean, is_paired)
@@ -117,19 +123,11 @@ def func_difference(control:list|tuple|np.ndarray, # NaNs are automatically disc
             err = "The two arrays supplied do not have the same length."
             raise ValueError(err)
 
-        control_nan = np.where(np.isnan(control))[0]
-        test_nan    = np.where(np.isnan(test))[0]
+        non_nan_mask = ~np.isnan(control) & ~np.isnan(test)
+        control_non_nan = control[non_nan_mask]
+        test_non_nan = test[non_nan_mask]
 
-        indexes_to_drop = np.unique(np.concatenate([control_nan,
-                                                    test_nan]))
-
-        good_indexes = [i for i in range(0, len(control))
-                        if i not in indexes_to_drop]
-
-        control = control[good_indexes]
-        test    = test[good_indexes]
-
-        return func(test - control)
+        return func(test_non_nan - control_non_nan)
 
     
     control = control[~np.isnan(control)]
@@ -138,6 +136,7 @@ def func_difference(control:list|tuple|np.ndarray, # NaNs are automatically disc
 
 
 # %% ../../nbs/API/effsize.ipynb 7
+@njit(cache=True)
 def cohens_d(control:list|tuple|np.ndarray,
              test:list|tuple|np.ndarray,
              is_paired:str=None # If not None, the paired Cohen's d is returned.
@@ -182,12 +181,6 @@ def cohens_d(control:list|tuple|np.ndarray,
         - https://en.wikipedia.org/wiki/Standard_deviation#Corrected_sample_standard_deviation
     """
 
-    # Convert to numpy arrays for speed.
-    # NaNs are automatically dropped.
-    if ~isinstance(control, np.ndarray):
-        control = np.array(control)
-    if ~isinstance(test, np.ndarray):
-        test    = np.array(test)
     control = control[~np.isnan(control)]
     test    = test[~np.isnan(test)]
 
@@ -218,6 +211,7 @@ def cohens_d(control:list|tuple|np.ndarray,
     return M / divisor
 
 # %% ../../nbs/API/effsize.ipynb 8
+# @njit(cache=True) # It uses np.seterr which is not supported by Numba
 def cohens_h(control:list|tuple|np.ndarray, 
              test:list|tuple|np.ndarray
             )->float:
@@ -240,10 +234,6 @@ def cohens_h(control:list|tuple|np.ndarray,
     # Convert to numpy arrays for speed.
     # NaNs are automatically dropped.
     # Aligned with cohens_d calculation.
-    if ~isinstance(control, np.ndarray):
-        control = np.array(control)
-    if ~isinstance(test, np.ndarray):
-        test    = np.array(test)
     control = control[~np.isnan(control)]
     test = test[~np.isnan(test)]
 
@@ -272,10 +262,6 @@ def hedges_g(control:list|tuple|np.ndarray,
 
     # Convert to numpy arrays for speed.
     # NaNs are automatically dropped.
-    if ~isinstance(control, np.ndarray):
-        control = np.array(control)
-    if ~isinstance(test, np.ndarray):
-        test    = np.array(test)
     control = control[~np.isnan(control)]
     test    = test[~np.isnan(test)]
 
@@ -286,6 +272,29 @@ def hedges_g(control:list|tuple|np.ndarray,
     return correction_factor * d
 
 # %% ../../nbs/API/effsize.ipynb 10
+@njit(cache=True)
+def _mann_whitney_u(x, y):
+    """Numba-optimized Mann-Whitney U calculation"""
+    n1, n2 = len(x), len(y)
+    combined = np.concatenate((x, y))
+    
+    # Use numpy broadcasting for comparison
+    less_than = (combined.reshape(-1, 1) > combined).sum(axis=1)
+    equal_to = (combined.reshape(-1, 1) == combined).sum(axis=1)
+    
+    # Calculate ranks directly
+    ranks = less_than + (equal_to + 1) / 2
+    
+    R1 = np.sum(ranks[:n1])
+    U1 = R1 - (n1 * (n1 + 1)) / 2
+    return U1
+
+@njit(cache=True)
+def _cliffs_delta_core(control, test):
+    """Numba-optimized Cliff's delta calculation"""
+    U = _mann_whitney_u(test, control)
+    return ((2 * U) / (len(control) * len(test))) - 1
+
 def cliffs_delta(control:list|tuple|np.ndarray, 
                  test:list|tuple|np.ndarray
                 )->float:
@@ -293,28 +302,13 @@ def cliffs_delta(control:list|tuple|np.ndarray,
     Computes Cliff's delta for 2 samples.
     See [here](https://en.wikipedia.org/wiki/Effect_size#Effect_size_for_ordinal_data)
     """
-
-    # Convert to numpy arrays for speed.
-    # NaNs are automatically dropped.
-    if ~isinstance(control, np.ndarray):
-        control = np.array(control)
-    if ~isinstance(test, np.ndarray):
-        test    = np.array(test)
-
     c = control[~np.isnan(control)]
     t = test[~np.isnan(test)]
-
-    control_n = len(c)
-    test_n = len(t)
-
-    # Note the order of the control and test arrays.
-    U, _ = mannwhitneyu(t, c, alternative='two-sided')
-    cliffs_delta = ((2 * U) / (control_n * test_n)) - 1
-
-    return cliffs_delta
+    return _cliffs_delta_core(c, t)
 
 
 # %% ../../nbs/API/effsize.ipynb 11
+@njit(cache=True)
 def _compute_standardizers(control, test):
     """
     Computes the pooled and average standard deviations for two datasets.
@@ -348,9 +342,9 @@ def _compute_standardizers(control, test):
     control_n = len(control)
     test_n = len(test)
 
-    control_var = np.var(control, ddof=1) # use N-1 to compute the variance.
-    test_var = np.var(test, ddof=1)
-
+    # ddof parameter is not supported by numba.
+    control_var = np.var(control)*control_n/(control_n-1) # use N-1 to compute the variance.
+    test_var = np.var(test)*test_n/(test_n-1)
 
     # For unpaired 2-groups standardized mean difference.
     pooled = np.sqrt(((control_n - 1) * control_var + (test_n - 1) * test_var) /
@@ -379,6 +373,7 @@ def _compute_hedges_correction_factor(n1,
     """
 
     df = n1 + n2 - 2
+    # gamma function is not supported by numba.
     numer = gamma(df / 2)
     denom0 = gamma((df - 1) / 2)
     denom = np.sqrt(df / 2) * denom0
@@ -396,6 +391,7 @@ def _compute_hedges_correction_factor(n1,
     return out
 
 # %% ../../nbs/API/effsize.ipynb 13
+@njit(cache=True)
 def weighted_delta(difference, group_var):
     '''
     Compute the weighted deltas where the weight is the inverse of the
diff --git a/dabest/_stats_tools/precompile.py b/dabest/_stats_tools/precompile.py
new file mode 100644
index 00000000..46cc2bc4
--- /dev/null
+++ b/dabest/_stats_tools/precompile.py
@@ -0,0 +1,53 @@
+"""A tool to pre-compile Numba functions for speeding up DABEST bootstrapping"""
+
+# AUTOGENERATED! DO NOT EDIT! File to edit: ../../nbs/API/precompile.ipynb.
+
+# %% auto 0
+__all__ = ['precompile_all']
+
+# %% ../../nbs/API/precompile.ipynb 4
+import numpy as np
+from tqdm import tqdm
+from . import effsize
+from . import confint_2group_diff
+
+# %% ../../nbs/API/precompile.ipynb 5
+_NUMBA_COMPILED = False
+
+def precompile_all():
+    """Pre-compile all numba functions with dummy data"""
+    global _NUMBA_COMPILED
+    
+    if _NUMBA_COMPILED:
+        return
+        
+    print("Pre-compiling numba functions for DABEST...")
+    
+    # Create dummy data
+    dummy_control = np.array([1.0, 2.0, 3.0])
+    dummy_test = np.array([4.0, 5.0, 6.0])
+    
+    funcs = [
+        # effsize.py functions
+        (effsize.cohens_d, (dummy_control, dummy_test)),
+        (effsize._mann_whitney_u, (dummy_control, dummy_test)),
+        (effsize._cliffs_delta_core, (dummy_control, dummy_test)),
+        (effsize._compute_standardizers, (dummy_control, dummy_test)),
+        (effsize.weighted_delta, (np.array([1.0, 2.0]), np.array([0.1, 0.2]))),
+        
+        # confint_2group_diff.py functions
+        (confint_2group_diff.create_jackknife_indexes, (dummy_control,)),
+        (confint_2group_diff.create_repeated_indexes, (dummy_control,)),
+        (confint_2group_diff.bootstrap_indices, (True, 3, 3, 10, 12345)),
+        (confint_2group_diff.delta2_bootstrap_loop, 
+            (dummy_control, dummy_test, dummy_control, dummy_test, 10, 1.0, 12345, False)),
+        (confint_2group_diff._compute_quantile, (0.5, 0.1, 0.1)),
+        (confint_2group_diff.calculate_group_var, (1.0, 3, 1.0, 3))
+    ]
+    
+    for func, args in tqdm(funcs, desc="Compiling numba functions"):
+        func(*args)
+    
+    _NUMBA_COMPILED = True
+    
+    print("Numba compilation complete!")
diff --git a/dabest/misc_tools.py b/dabest/misc_tools.py
index 6836d336..90c4a6bc 100644
--- a/dabest/misc_tools.py
+++ b/dabest/misc_tools.py
@@ -762,8 +762,8 @@ def Gardner_Altman_Plot_Aesthetic_Adjustments(effect_size_type, plot_data, xvar,
             which_std = 1
         else:
             which_std = 0
-        temp_control = plot_data[plot_data[xvar] == current_control][yvar]
-        temp_test = plot_data[plot_data[xvar] == current_group][yvar]
+        temp_control = np.array(plot_data[plot_data[xvar] == current_control][yvar])
+        temp_test = np.array(plot_data[plot_data[xvar] == current_group][yvar])
 
         stds = _compute_standardizers(temp_control, temp_test)
         if is_paired:
diff --git a/dabest/plot_tools.py b/dabest/plot_tools.py
index 72437f38..40d49a91 100644
--- a/dabest/plot_tools.py
+++ b/dabest/plot_tools.py
@@ -369,7 +369,9 @@ def single_sankey(
 ):
     """
     Make a single Sankey diagram showing proportion flow from left to right
+
     Original code from: https://github.com/anazalea/pySankey
+    
     Changes are added to normalize each diagram's height to be 1
 
     """
diff --git a/nbs/API/bootstrap.ipynb b/nbs/API/bootstrap.ipynb
index eb33a083..79de8c0e 100644
--- a/nbs/API/bootstrap.ipynb
+++ b/nbs/API/bootstrap.ipynb
@@ -114,7 +114,9 @@
     "        reps: int = 5000,  # Number of bootstrap iterations to perform.\n",
     "    ):\n",
     "        # Turn to pandas series.\n",
-    "        x1 = pd.Series(x1).dropna()\n",
+    "        # x1 = pd.Series(x1).dropna()\n",
+    "        x1 = x1[~np.isnan(x1)]\n",
+    "\n",
     "        diff = False\n",
     "\n",
     "        # Initialise stat_function\n",
@@ -137,7 +139,9 @@
     "            if x2 is None:\n",
     "                raise ValueError(\"Please specify x2.\")\n",
     "            \n",
-    "            x2 = pd.Series(x2).dropna()\n",
+    "            # x2 = pd.Series(x2).dropna()\n",
+    "            x2 = x1[~np.isnan(x2)]\n",
+    "\n",
     "            if len(x1) != len(x2):\n",
     "                raise ValueError(\"x1 and x2 are not the same length.\")\n",
     "\n",
@@ -182,7 +186,8 @@
     "\n",
     "        elif x2 is not None and paired is None:\n",
     "            diff = True\n",
-    "            x2 = pd.Series(x2).dropna()\n",
+    "            # x2 = pd.Series(x2).dropna()\n",
+    "            x2 = x2[~np.isnan(x2)]\n",
     "            # Generate statarrays for both arrays.\n",
     "            ref_statarray = sns.algorithms.bootstrap(x1, **sns_bootstrap_kwargs)\n",
     "            exp_statarray = sns.algorithms.bootstrap(x2, **sns_bootstrap_kwargs)\n",
diff --git a/nbs/API/confint_2group_diff.ipynb b/nbs/API/confint_2group_diff.ipynb
index dd6477aa..c080d452 100644
--- a/nbs/API/confint_2group_diff.ipynb
+++ b/nbs/API/confint_2group_diff.ipynb
@@ -60,7 +60,7 @@
     "from numpy import mean as npmean\n",
     "from numpy import sum as npsum\n",
     "from numpy.random import PCG64, RandomState\n",
-    "import pandas as pd\n",
+    "from numba import njit, prange\n",
     "from scipy.stats import norm\n",
     "from numpy import isnan"
    ]
@@ -73,6 +73,7 @@
    "outputs": [],
    "source": [
     "#| export\n",
+    "@njit(cache=True, parallel=True)\n",
     "def create_jackknife_indexes(data):\n",
     "    \"\"\"\n",
     "    Given an array-like, creates a jackknife bootstrap.\n",
@@ -89,18 +90,25 @@
     "    Generator that yields all jackknife bootstrap samples.\n",
     "    \"\"\"\n",
     "\n",
-    "    index_range = arange(0, len(data))\n",
-    "    return (delete(index_range, i) for i in index_range)\n",
+    "    n = len(data)\n",
+    "    indexes = np.empty((n, n - 1), dtype=np.int64)\n",
+    "    for i in prange(n):\n",
+    "        indexes[i] = np.concatenate((np.arange(i), np.arange(i + 1, n)))\n",
+    "    return indexes\n",
     "\n",
     "\n",
+    "@njit(cache=True, parallel=True)\n",
     "def create_repeated_indexes(data):\n",
     "    \"\"\"\n",
     "    Convenience function. Given an array-like with length N,\n",
     "    returns a generator that yields N indexes [0, 1, ..., N].\n",
     "    \"\"\"\n",
     "\n",
-    "    index_range = arange(0, len(data))\n",
-    "    return (index_range for i in index_range)\n",
+    "    n = len(data)\n",
+    "    indexes = np.empty((n, n), dtype=np.int64)  # Pre-allocate the output array\n",
+    "    for i in prange(n):\n",
+    "        indexes[i, :] = np.arange(n)  # Fill each row with the full index range\n",
+    "    return indexes\n",
     "\n",
     "\n",
     "def _create_two_group_jackknife_indexes(x0, x1, is_paired):\n",
@@ -168,6 +176,20 @@
     "        return numer / denom\n",
     "\n",
     "\n",
+    "@njit(cache=True) # parallelization must be turned off for random number generation\n",
+    "def bootstrap_indices(is_paired, x0_len, x1_len, resamples, random_seed):\n",
+    "    np.random.seed(random_seed)\n",
+    "    indices = np.empty((resamples, x0_len if is_paired else x0_len + x1_len), dtype=np.int64)\n",
+    "    \n",
+    "    for i in range(resamples):\n",
+    "        if is_paired:\n",
+    "            indices[i, :x0_len] = np.random.choice(x0_len, x0_len)\n",
+    "        else:  \n",
+    "            indices[i, :x0_len] = np.random.choice(x0_len, x0_len)\n",
+    "            indices[i, x0_len:x0_len+x1_len] = np.random.choice(x1_len, x1_len)\n",
+    "    return indices\n",
+    "\n",
+    "\n",
     "def compute_bootstrapped_diff(\n",
     "    x0, x1, is_paired, effect_size, resamples=5000, random_seed=12345\n",
     "):\n",
@@ -175,27 +197,60 @@
     "\n",
     "    from . import effsize as __es\n",
     "\n",
-    "    rng = RandomState(PCG64(random_seed))\n",
-    "\n",
-    "    out = np.repeat(np.nan, resamples)\n",
-    "    x0_len = len(x0)\n",
-    "    x1_len = len(x1)\n",
+    "    x0_len, x1_len = len(x0), len(x1)\n",
+    "    indices = bootstrap_indices(is_paired, x0_len, x1_len, resamples, random_seed)\n",
+    "    out = np.empty(resamples, dtype=np.float64)\n",
     "\n",
-    "    for i in range(int(resamples)):\n",
+    "    for i in range(resamples):\n",
     "        if is_paired:\n",
-    "            if x0_len != x1_len:\n",
-    "                raise ValueError(\"The two arrays do not have the same length.\")\n",
-    "            random_idx = rng.choice(x0_len, x0_len, replace=True)\n",
-    "            x0_sample = x0[random_idx]\n",
-    "            x1_sample = x1[random_idx]\n",
+    "            x0_sample = x0[indices[i, :x0_len]]\n",
+    "            x1_sample = x1[indices[i, :x0_len]]\n",
     "        else:\n",
-    "            x0_sample = rng.choice(x0, x0_len, replace=True)\n",
-    "            x1_sample = rng.choice(x1, x1_len, replace=True)\n",
+    "            x0_sample = x0[indices[i, :x0_len]]\n",
+    "            x1_sample = x1[indices[i, x0_len:x0_len+x1_len]]\n",
     "\n",
     "        out[i] = __es.two_group_difference(x0_sample, x1_sample, is_paired, effect_size)\n",
     "\n",
     "    return out\n",
     "\n",
+    "@njit(cache=True) # parallelization must be turned off for random number generation\n",
+    "def delta2_bootstrap_loop(x1, x2, x3, x4, resamples, pooled_sd, rng_seed, is_paired):\n",
+    "    np.random.seed(rng_seed)\n",
+    "    out_delta_g = np.empty(resamples)\n",
+    "    deltadelta = np.empty(resamples)\n",
+    "    \n",
+    "    n1, n2, n3, n4 = len(x1), len(x2), len(x3), len(x4)\n",
+    "    if is_paired:\n",
+    "        if n1 != n2 or n3 != n4:\n",
+    "            raise ValueError(\"Each control group must have the same length as its corresponding test group in paired analysis.\")\n",
+    "    \n",
+    "\n",
+    "    # Bootstrapping\n",
+    "    for i in range(resamples):\n",
+    "        # Paired or unpaired resampling\n",
+    "        if is_paired:\n",
+    "            indices_1 = np.random.choice(len(x1),len(x1))\n",
+    "            indices_2 = np.random.choice(len(x3),len(x3))\n",
+    "            x1_sample, x2_sample = x1[indices_1], x2[indices_1]\n",
+    "            x3_sample, x4_sample = x3[indices_2], x4[indices_2]\n",
+    "        else:\n",
+    "            indices_1 = np.random.randint(0, len(x1), len(x1))\n",
+    "            indices_2 = np.random.randint(0, len(x2), len(x2))\n",
+    "            indices_3 = np.random.randint(0, len(x3), len(x3))\n",
+    "            indices_4 = np.random.randint(0, len(x4), len(x4))\n",
+    "            x1_sample, x2_sample = x1[indices_1], x2[indices_2]\n",
+    "            x3_sample, x4_sample = x3[indices_3], x4[indices_4]\n",
+    "\n",
+    "        # Calculating deltas\n",
+    "        delta_1 = np.mean(x2_sample) - np.mean(x1_sample)\n",
+    "        delta_2 = np.mean(x4_sample) - np.mean(x3_sample)\n",
+    "        delta_delta = delta_2 - delta_1\n",
+    "\n",
+    "        deltadelta[i] = delta_delta\n",
+    "        out_delta_g[i] = delta_delta / pooled_sd\n",
+    "\n",
+    "    return out_delta_g, deltadelta\n",
+    "\n",
     "\n",
     "def compute_delta2_bootstrapped_diff(\n",
     "    x1: np.ndarray,  # Control group 1\n",
@@ -213,8 +268,6 @@
     "\n",
     "    \"\"\"\n",
     "\n",
-    "    rng = RandomState(PCG64(random_seed))\n",
-    "\n",
     "    x1, x2, x3, x4 = map(np.asarray, [x1, x2, x3, x4])\n",
     "\n",
     "    # Calculating pooled sample standard deviation\n",
@@ -234,33 +287,7 @@
     "    if np.isnan(pooled_sample_sd) or pooled_sample_sd == 0:\n",
     "        raise ValueError(\"Pooled sample standard deviation is NaN or zero.\")\n",
     "\n",
-    "    out_delta_g = np.empty(resamples)\n",
-    "    deltadelta = np.empty(resamples)\n",
-    "\n",
-    "    # Bootstrapping\n",
-    "    for i in range(resamples):\n",
-    "        # Paired or unpaired resampling\n",
-    "        if is_paired:\n",
-    "            if len(x1) != len(x2) or len(x3) != len(x4):\n",
-    "                raise ValueError(\"Each control group must have the same length as its corresponding test group in paired analysis.\")\n",
-    "            indices_1 = rng.choice(len(x1), len(x1), replace=True)\n",
-    "            indices_2 = rng.choice(len(x3), len(x3), replace=True)\n",
-    "\n",
-    "            x1_sample, x2_sample = x1[indices_1], x2[indices_1]\n",
-    "            x3_sample, x4_sample = x3[indices_2], x4[indices_2]\n",
-    "        else:\n",
-    "            x1_sample = rng.choice(x1, len(x1), replace=True)\n",
-    "            x2_sample = rng.choice(x2, len(x2), replace=True)\n",
-    "            x3_sample = rng.choice(x3, len(x3), replace=True)\n",
-    "            x4_sample = rng.choice(x4, len(x4), replace=True)\n",
-    "\n",
-    "        # Calculating deltas\n",
-    "        delta_1 = np.mean(x2_sample) - np.mean(x1_sample)\n",
-    "        delta_2 = np.mean(x4_sample) - np.mean(x3_sample)\n",
-    "        delta_delta = delta_2 - delta_1\n",
-    "\n",
-    "        deltadelta[i] = delta_delta\n",
-    "        out_delta_g[i] = delta_delta / pooled_sample_sd\n",
+    "    out_delta_g, deltadelta = delta2_bootstrap_loop(x1, x2, x3, x4, resamples, pooled_sample_sd, random_seed, is_paired)\n",
     "\n",
     "    # Empirical delta_g calculation\n",
     "    delta_g = ((np.mean(x4) - np.mean(x3)) - (np.mean(x2) - np.mean(x1))) / pooled_sample_sd\n",
@@ -297,6 +324,7 @@
     "    return (100.0 - ci) / 100.0\n",
     "\n",
     "\n",
+    "@njit(cache=True)\n",
     "def _compute_quantile(z, bias, acceleration):\n",
     "    numer = bias + z\n",
     "    denom = 1 - (acceleration * numer)\n",
@@ -332,6 +360,7 @@
     "    return low, high\n",
     "\n",
     "\n",
+    "@njit(cache=True)\n",
     "def calculate_group_var(control_var, control_N, test_var, test_N):\n",
     "    return control_var / control_N + test_var / test_N\n",
     "\n",
@@ -343,18 +372,11 @@
     "\n",
     "    weight = 1 / group_var\n",
     "    denom = np.sum(weight)\n",
-    "    num = np.sum(weight[i] * differences[i] for i in range(0, len(weight)))\n",
-    "\n",
+    "    num = 0.0\n",
+    "    for i in range(len(weight)):\n",
+    "        num += weight[i] * differences[i]\n",
     "    return num / denom"
    ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": null,
-   "id": "87e0c164",
-   "metadata": {},
-   "outputs": [],
-   "source": []
   }
  ],
  "metadata": {
diff --git a/nbs/API/dabest_object.ipynb b/nbs/API/dabest_object.ipynb
index 35c8d5c3..75903f07 100644
--- a/nbs/API/dabest_object.ipynb
+++ b/nbs/API/dabest_object.ipynb
@@ -180,7 +180,7 @@
     "        # Determine the kind of estimation plot we need to produce.\n",
     "        if all([isinstance(i, (str, int, float)) for i in idx]):\n",
     "            # flatten out idx.\n",
-    "            all_plot_groups = pd.unique(pd.Series([t for t in idx])).tolist()\n",
+    "            all_plot_groups = pd.Series([t for t in idx]).unique().tolist()\n",
     "            if len(idx) > len(all_plot_groups):\n",
     "                err0 = \"`idx` contains duplicated groups. Please remove any duplicates and try again.\"\n",
     "                raise ValueError(err0)\n",
@@ -190,7 +190,7 @@
     "            self.__idx = (idx,)\n",
     "\n",
     "        elif all([isinstance(i, (tuple, list)) for i in idx]):\n",
-    "            all_plot_groups = pd.unique(pd.Series([tt for t in idx for tt in t])).tolist()\n",
+    "            all_plot_groups = pd.Series([tt for t in idx for tt in t]).unique().tolist()\n",
     "\n",
     "            actual_groups_given = sum([len(i) for i in idx])\n",
     "\n",
diff --git a/nbs/API/delta_objects.ipynb b/nbs/API/delta_objects.ipynb
index 358e45ad..3fb4aaef 100644
--- a/nbs/API/delta_objects.ipynb
+++ b/nbs/API/delta_objects.ipynb
@@ -528,13 +528,14 @@
     "        # compute the variances of each control group and each test group\n",
     "        control_var=[]\n",
     "        test_var=[]\n",
+    "        grouped_data = {name: group[yvar].copy() for name, group in dat.groupby(xvar, observed=False)}\n",
     "        for j, current_tuple in enumerate(idx):\n",
     "            cname = current_tuple[0]\n",
-    "            control = dat[dat[xvar] == cname][yvar].copy()\n",
+    "            control = grouped_data[cname]\n",
     "            control_var.append(np.var(control, ddof=1))\n",
     "\n",
     "            tname = current_tuple[1]\n",
-    "            test = dat[dat[xvar] == tname][yvar].copy()\n",
+    "            test = grouped_data[tname]\n",
     "            test_var.append(np.var(test, ddof=1))\n",
     "        self.__control_var = np.array(control_var)\n",
     "        self.__test_var    = np.array(test_var)\n",
@@ -554,7 +555,7 @@
     "                                                          self.__bootstraps)\n",
     "\n",
     "        # Compute the weighted average mean difference based on the raw data\n",
-    "        self.__difference = es.weighted_delta(self.__effsizedf[\"difference\"],\n",
+    "        self.__difference = es.weighted_delta(np.array(self.__effsizedf[\"difference\"]),\n",
     "                                                   self.__group_var)\n",
     "\n",
     "        sorted_weighted_deltas = npsort(self.__bootstraps_weighted_delta)\n",
diff --git a/nbs/API/effsize.ipynb b/nbs/API/effsize.ipynb
index 9d8dc5d2..5123e11d 100644
--- a/nbs/API/effsize.ipynb
+++ b/nbs/API/effsize.ipynb
@@ -56,9 +56,10 @@
     "#| export\n",
     "from __future__ import annotations\n",
     "import numpy as np\n",
+    "from numba import njit\n",
     "import warnings\n",
     "from scipy.special import gamma\n",
-    "from scipy.stats import mannwhitneyu"
+    "from scipy.stats import mannwhitneyu\n"
    ]
   },
   {
@@ -118,6 +119,10 @@
     "\n",
     "    \"\"\"\n",
     "\n",
+    "    if ~isinstance(control, np.ndarray):\n",
+    "        control = np.array(control)\n",
+    "    if ~isinstance(test, np.ndarray):\n",
+    "        test    = np.array(test)\n",
     "\n",
     "    if effect_size == \"mean_diff\":\n",
     "        return func_difference(control, test, np.mean, is_paired)\n",
@@ -180,19 +185,11 @@
     "            err = \"The two arrays supplied do not have the same length.\"\n",
     "            raise ValueError(err)\n",
     "\n",
-    "        control_nan = np.where(np.isnan(control))[0]\n",
-    "        test_nan    = np.where(np.isnan(test))[0]\n",
-    "\n",
-    "        indexes_to_drop = np.unique(np.concatenate([control_nan,\n",
-    "                                                    test_nan]))\n",
+    "        non_nan_mask = ~np.isnan(control) & ~np.isnan(test)\n",
+    "        control_non_nan = control[non_nan_mask]\n",
+    "        test_non_nan = test[non_nan_mask]\n",
     "\n",
-    "        good_indexes = [i for i in range(0, len(control))\n",
-    "                        if i not in indexes_to_drop]\n",
-    "\n",
-    "        control = control[good_indexes]\n",
-    "        test    = test[good_indexes]\n",
-    "\n",
-    "        return func(test - control)\n",
+    "        return func(test_non_nan - control_non_nan)\n",
     "\n",
     "    \n",
     "    control = control[~np.isnan(control)]\n",
@@ -208,6 +205,7 @@
    "outputs": [],
    "source": [
     "#| export\n",
+    "@njit(cache=True)\n",
     "def cohens_d(control:list|tuple|np.ndarray,\n",
     "             test:list|tuple|np.ndarray,\n",
     "             is_paired:str=None # If not None, the paired Cohen's d is returned.\n",
@@ -252,12 +250,6 @@
     "        - https://en.wikipedia.org/wiki/Standard_deviation#Corrected_sample_standard_deviation\n",
     "    \"\"\"\n",
     "\n",
-    "    # Convert to numpy arrays for speed.\n",
-    "    # NaNs are automatically dropped.\n",
-    "    if ~isinstance(control, np.ndarray):\n",
-    "        control = np.array(control)\n",
-    "    if ~isinstance(test, np.ndarray):\n",
-    "        test    = np.array(test)\n",
     "    control = control[~np.isnan(control)]\n",
     "    test    = test[~np.isnan(test)]\n",
     "\n",
@@ -296,6 +288,7 @@
    "outputs": [],
    "source": [
     "#| export\n",
+    "# @njit(cache=True) # It uses np.seterr which is not supported by Numba\n",
     "def cohens_h(control:list|tuple|np.ndarray, \n",
     "             test:list|tuple|np.ndarray\n",
     "            )->float:\n",
@@ -318,10 +311,6 @@
     "    # Convert to numpy arrays for speed.\n",
     "    # NaNs are automatically dropped.\n",
     "    # Aligned with cohens_d calculation.\n",
-    "    if ~isinstance(control, np.ndarray):\n",
-    "        control = np.array(control)\n",
-    "    if ~isinstance(test, np.ndarray):\n",
-    "        test    = np.array(test)\n",
     "    control = control[~np.isnan(control)]\n",
     "    test = test[~np.isnan(test)]\n",
     "\n",
@@ -357,10 +346,6 @@
     "\n",
     "    # Convert to numpy arrays for speed.\n",
     "    # NaNs are automatically dropped.\n",
-    "    if ~isinstance(control, np.ndarray):\n",
-    "        control = np.array(control)\n",
-    "    if ~isinstance(test, np.ndarray):\n",
-    "        test    = np.array(test)\n",
     "    control = control[~np.isnan(control)]\n",
     "    test    = test[~np.isnan(test)]\n",
     "\n",
@@ -379,6 +364,29 @@
    "outputs": [],
    "source": [
     "#| export\n",
+    "@njit(cache=True)\n",
+    "def _mann_whitney_u(x, y):\n",
+    "    \"\"\"Numba-optimized Mann-Whitney U calculation\"\"\"\n",
+    "    n1, n2 = len(x), len(y)\n",
+    "    combined = np.concatenate((x, y))\n",
+    "    \n",
+    "    # Use numpy broadcasting for comparison\n",
+    "    less_than = (combined.reshape(-1, 1) > combined).sum(axis=1)\n",
+    "    equal_to = (combined.reshape(-1, 1) == combined).sum(axis=1)\n",
+    "    \n",
+    "    # Calculate ranks directly\n",
+    "    ranks = less_than + (equal_to + 1) / 2\n",
+    "    \n",
+    "    R1 = np.sum(ranks[:n1])\n",
+    "    U1 = R1 - (n1 * (n1 + 1)) / 2\n",
+    "    return U1\n",
+    "\n",
+    "@njit(cache=True)\n",
+    "def _cliffs_delta_core(control, test):\n",
+    "    \"\"\"Numba-optimized Cliff's delta calculation\"\"\"\n",
+    "    U = _mann_whitney_u(test, control)\n",
+    "    return ((2 * U) / (len(control) * len(test))) - 1\n",
+    "\n",
     "def cliffs_delta(control:list|tuple|np.ndarray, \n",
     "                 test:list|tuple|np.ndarray\n",
     "                )->float:\n",
@@ -386,25 +394,9 @@
     "    Computes Cliff's delta for 2 samples.\n",
     "    See [here](https://en.wikipedia.org/wiki/Effect_size#Effect_size_for_ordinal_data)\n",
     "    \"\"\"\n",
-    "\n",
-    "    # Convert to numpy arrays for speed.\n",
-    "    # NaNs are automatically dropped.\n",
-    "    if ~isinstance(control, np.ndarray):\n",
-    "        control = np.array(control)\n",
-    "    if ~isinstance(test, np.ndarray):\n",
-    "        test    = np.array(test)\n",
-    "\n",
     "    c = control[~np.isnan(control)]\n",
     "    t = test[~np.isnan(test)]\n",
-    "\n",
-    "    control_n = len(c)\n",
-    "    test_n = len(t)\n",
-    "\n",
-    "    # Note the order of the control and test arrays.\n",
-    "    U, _ = mannwhitneyu(t, c, alternative='two-sided')\n",
-    "    cliffs_delta = ((2 * U) / (control_n * test_n)) - 1\n",
-    "\n",
-    "    return cliffs_delta\n"
+    "    return _cliffs_delta_core(c, t)\n"
    ]
   },
   {
@@ -415,6 +407,7 @@
    "outputs": [],
    "source": [
     "#| export\n",
+    "@njit(cache=True)\n",
     "def _compute_standardizers(control, test):\n",
     "    \"\"\"\n",
     "    Computes the pooled and average standard deviations for two datasets.\n",
@@ -448,9 +441,9 @@
     "    control_n = len(control)\n",
     "    test_n = len(test)\n",
     "\n",
-    "    control_var = np.var(control, ddof=1) # use N-1 to compute the variance.\n",
-    "    test_var = np.var(test, ddof=1)\n",
-    "\n",
+    "    # ddof parameter is not supported by numba.\n",
+    "    control_var = np.var(control)*control_n/(control_n-1) # use N-1 to compute the variance.\n",
+    "    test_var = np.var(test)*test_n/(test_n-1)\n",
     "\n",
     "    # For unpaired 2-groups standardized mean difference.\n",
     "    pooled = np.sqrt(((control_n - 1) * control_var + (test_n - 1) * test_var) /\n",
@@ -487,6 +480,7 @@
     "    \"\"\"\n",
     "\n",
     "    df = n1 + n2 - 2\n",
+    "    # gamma function is not supported by numba.\n",
     "    numer = gamma(df / 2)\n",
     "    denom0 = gamma((df - 1) / 2)\n",
     "    denom = np.sqrt(df / 2) * denom0\n",
@@ -512,6 +506,7 @@
    "outputs": [],
    "source": [
     "#| export\n",
+    "@njit(cache=True)\n",
     "def weighted_delta(difference, group_var):\n",
     "    '''\n",
     "    Compute the weighted deltas where the weight is the inverse of the\n",
diff --git a/nbs/API/effsize_objects.ipynb b/nbs/API/effsize_objects.ipynb
index da633fda..cfdc7e92 100644
--- a/nbs/API/effsize_objects.ipynb
+++ b/nbs/API/effsize_objects.ipynb
@@ -62,6 +62,7 @@
     "import pandas as pd\n",
     "import lqrt\n",
     "from scipy.stats import norm\n",
+    "import numpy as np\n",
     "from numpy import array, isnan, isinf, repeat, random, isin, abs, var\n",
     "from numpy import sort as npsort\n",
     "from numpy import nan as npnan\n",
@@ -417,12 +418,17 @@
     "            # References:\n",
     "            # https://en.wikipedia.org/wiki/McNemar%27s_test\n",
     "\n",
-    "            df_temp = pd.DataFrame({\"control\": self.__control, \"test\": self.__test})\n",
-    "            x1 = len(df_temp[(df_temp[\"control\"] == 0) & (df_temp[\"test\"] == 0)])\n",
-    "            x2 = len(df_temp[(df_temp[\"control\"] == 0) & (df_temp[\"test\"] == 1)])\n",
-    "            x3 = len(df_temp[(df_temp[\"control\"] == 1) & (df_temp[\"test\"] == 0)])\n",
-    "            x4 = len(df_temp[(df_temp[\"control\"] == 1) & (df_temp[\"test\"] == 1)])\n",
-    "            table = [[x1, x2], [x3, x4]]\n",
+    "            # df_temp = pd.DataFrame({\"control\": self.__control, \"test\": self.__test})\n",
+    "            # x1 = len(df_temp[(df_temp[\"control\"] == 0) & (df_temp[\"test\"] == 0)])\n",
+    "            # x2 = len(df_temp[(df_temp[\"control\"] == 0) & (df_temp[\"test\"] == 1)])\n",
+    "            # x3 = len(df_temp[(df_temp[\"control\"] == 1) & (df_temp[\"test\"] == 0)])\n",
+    "            # x4 = len(df_temp[(df_temp[\"control\"] == 1) & (df_temp[\"test\"] == 1)])\n",
+    "            # table = [[x1, x2], [x3, x4]]\n",
+    "            x1 = np.sum((self.__control == 0) & (self.__test == 0))\n",
+    "            x2 = np.sum((self.__control == 0) & (self.__test == 1))\n",
+    "            x3 = np.sum((self.__control == 1) & (self.__test == 0))\n",
+    "            x4 = np.sum((self.__control == 1) & (self.__test == 1))\n",
+    "            table = np.array([[x1, x2], [x3, x4]])\n",
     "            _mcnemar = mcnemar(table, exact=True, correction=True)\n",
     "            self.__pvalue_mcnemar = _mcnemar.pvalue\n",
     "            self.__statistic_mcnemar = _mcnemar.statistic\n",
@@ -1020,18 +1026,19 @@
     "        out = []\n",
     "        reprs = []\n",
     "\n",
+    "        grouped_data = {name: group[yvar].copy() for name, group in dat.groupby(xvar, observed=False)}\n",
     "        if self.__delta2:\n",
     "            mixed_data = []\n",
     "            for j, current_tuple in enumerate(idx):\n",
     "                if self.__is_paired != \"sequential\":\n",
     "                    cname = current_tuple[0]\n",
-    "                    control = dat[dat[xvar] == cname][yvar].copy()\n",
+    "                    control = grouped_data[cname]\n",
     "\n",
     "                for ix, tname in enumerate(current_tuple[1:]):\n",
     "                    if self.__is_paired == \"sequential\":\n",
     "                        cname = current_tuple[ix]\n",
-    "                        control = dat[dat[xvar] == cname][yvar].copy()\n",
-    "                    test = dat[dat[xvar] == tname][yvar].copy()\n",
+    "                        control = grouped_data[cname]\n",
+    "                    test = grouped_data[tname]\n",
     "                    mixed_data.append(control)\n",
     "                    mixed_data.append(test)\n",
     "            bootstraps_delta_delta = ci2g.compute_delta2_bootstrapped_diff(\n",
@@ -1047,13 +1054,13 @@
     "        for j, current_tuple in enumerate(idx):\n",
     "            if self.__is_paired != \"sequential\":\n",
     "                cname = current_tuple[0]\n",
-    "                control = dat[dat[xvar] == cname][yvar].copy()\n",
+    "                control = grouped_data[cname]\n",
     "\n",
     "            for ix, tname in enumerate(current_tuple[1:]):\n",
     "                if self.__is_paired == \"sequential\":\n",
     "                    cname = current_tuple[ix]\n",
-    "                    control = dat[dat[xvar] == cname][yvar].copy()\n",
-    "                test = dat[dat[xvar] == tname][yvar].copy()\n",
+    "                    control = grouped_data[cname]\n",
+    "                test = grouped_data[tname]\n",
     "\n",
     "                result = TwoGroupsEffectSize(\n",
     "                    control,\n",
@@ -1214,16 +1221,18 @@
     "\n",
     "        out = []\n",
     "\n",
+    "        grouped_data = {name:group[yvar].copy() for name, group in dat.groupby(xvar)}\n",
+    "\n",
     "        for j, current_tuple in enumerate(db_obj.idx):\n",
     "            if self.__is_paired != \"sequential\":\n",
     "                cname = current_tuple[0]\n",
-    "                control = dat[dat[xvar] == cname][yvar].copy()\n",
+    "                control = grouped_data[cname]\n",
     "\n",
     "            for ix, tname in enumerate(current_tuple[1:]):\n",
     "                if self.__is_paired == \"sequential\":\n",
     "                    cname = current_tuple[ix]\n",
-    "                    control = dat[dat[xvar] == cname][yvar].copy()\n",
-    "                test = dat[dat[xvar] == tname][yvar].copy()\n",
+    "                    control = grouped_data[cname]\n",
+    "                test = grouped_data[tname]\n",
     "\n",
     "                if self.__is_paired:\n",
     "                    # Refactored here in v0.3.0 for performance issues.\n",
diff --git a/nbs/API/misc_tools.ipynb b/nbs/API/misc_tools.ipynb
index 64eda8c4..499b9be1 100644
--- a/nbs/API/misc_tools.ipynb
+++ b/nbs/API/misc_tools.ipynb
@@ -815,8 +815,8 @@
     "            which_std = 1\n",
     "        else:\n",
     "            which_std = 0\n",
-    "        temp_control = plot_data[plot_data[xvar] == current_control][yvar]\n",
-    "        temp_test = plot_data[plot_data[xvar] == current_group][yvar]\n",
+    "        temp_control = np.array(plot_data[plot_data[xvar] == current_control][yvar])\n",
+    "        temp_test = np.array(plot_data[plot_data[xvar] == current_group][yvar])\n",
     "\n",
     "        stds = _compute_standardizers(temp_control, temp_test)\n",
     "        if is_paired:\n",
diff --git a/nbs/API/plot_tools.ipynb b/nbs/API/plot_tools.ipynb
index 2c431029..c5c3b311 100644
--- a/nbs/API/plot_tools.ipynb
+++ b/nbs/API/plot_tools.ipynb
@@ -420,7 +420,9 @@
     "):\n",
     "    \"\"\"\n",
     "    Make a single Sankey diagram showing proportion flow from left to right\n",
+    "\n",
     "    Original code from: https://github.com/anazalea/pySankey\n",
+    "    \n",
     "    Changes are added to normalize each diagram's height to be 1\n",
     "\n",
     "    \"\"\"\n",
diff --git a/nbs/API/precompile.ipynb b/nbs/API/precompile.ipynb
new file mode 100644
index 00000000..223c4ce9
--- /dev/null
+++ b/nbs/API/precompile.ipynb
@@ -0,0 +1,111 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "# precompile\n",
+    "\n",
+    "> A tool to pre-compile Numba functions for speeding up DABEST bootstrapping\n",
+    "\n",
+    "- order: 10"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "#| default_exp _stats_tools/precompile"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "#| hide\n",
+    "from __future__ import annotations"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "#| hide\n",
+    "from nbdev.showdoc import *\n",
+    "import nbdev\n",
+    "nbdev.nbdev_export()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "#| export\n",
+    "import numpy as np\n",
+    "from tqdm import tqdm\n",
+    "from dabest._stats_tools import effsize\n",
+    "from dabest._stats_tools import confint_2group_diff"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "#| export\n",
+    "\n",
+    "_NUMBA_COMPILED = False\n",
+    "\n",
+    "def precompile_all():\n",
+    "    \"\"\"Pre-compile all numba functions with dummy data\"\"\"\n",
+    "    global _NUMBA_COMPILED\n",
+    "    \n",
+    "    if _NUMBA_COMPILED:\n",
+    "        return\n",
+    "        \n",
+    "    print(\"Pre-compiling numba functions for DABEST...\")\n",
+    "    \n",
+    "    # Create dummy data\n",
+    "    dummy_control = np.array([1.0, 2.0, 3.0])\n",
+    "    dummy_test = np.array([4.0, 5.0, 6.0])\n",
+    "    \n",
+    "    funcs = [\n",
+    "        # effsize.py functions\n",
+    "        (effsize.cohens_d, (dummy_control, dummy_test)),\n",
+    "        (effsize._mann_whitney_u, (dummy_control, dummy_test)),\n",
+    "        (effsize._cliffs_delta_core, (dummy_control, dummy_test)),\n",
+    "        (effsize._compute_standardizers, (dummy_control, dummy_test)),\n",
+    "        (effsize.weighted_delta, (np.array([1.0, 2.0]), np.array([0.1, 0.2]))),\n",
+    "        \n",
+    "        # confint_2group_diff.py functions\n",
+    "        (confint_2group_diff.create_jackknife_indexes, (dummy_control,)),\n",
+    "        (confint_2group_diff.create_repeated_indexes, (dummy_control,)),\n",
+    "        (confint_2group_diff.bootstrap_indices, (True, 3, 3, 10, 12345)),\n",
+    "        (confint_2group_diff.delta2_bootstrap_loop, \n",
+    "            (dummy_control, dummy_test, dummy_control, dummy_test, 10, 1.0, 12345, False)),\n",
+    "        (confint_2group_diff._compute_quantile, (0.5, 0.1, 0.1)),\n",
+    "        (confint_2group_diff.calculate_group_var, (1.0, 3, 1.0, 3))\n",
+    "    ]\n",
+    "    \n",
+    "    for func, args in tqdm(funcs, desc=\"Compiling numba functions\"):\n",
+    "        func(*args)\n",
+    "    \n",
+    "    _NUMBA_COMPILED = True\n",
+    "    \n",
+    "    print(\"Numba compilation complete!\")"
+   ]
+  }
+ ],
+ "metadata": {},
+ "nbformat": 4,
+ "nbformat_minor": 2
+}
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_01_gardner_altman_unpaired_meandiff.png b/nbs/tests/mpl_image_tests/baseline_images/test_01_gardner_altman_unpaired_meandiff.png
index d6374405..3abf1737 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_01_gardner_altman_unpaired_meandiff.png and b/nbs/tests/mpl_image_tests/baseline_images/test_01_gardner_altman_unpaired_meandiff.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_02_gardner_altman_unpaired_mediandiff.png b/nbs/tests/mpl_image_tests/baseline_images/test_02_gardner_altman_unpaired_mediandiff.png
index 7dc4d313..ae655349 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_02_gardner_altman_unpaired_mediandiff.png and b/nbs/tests/mpl_image_tests/baseline_images/test_02_gardner_altman_unpaired_mediandiff.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_03_gardner_altman_unpaired_hedges_g.png b/nbs/tests/mpl_image_tests/baseline_images/test_03_gardner_altman_unpaired_hedges_g.png
index 689b26ce..4d63bbcc 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_03_gardner_altman_unpaired_hedges_g.png and b/nbs/tests/mpl_image_tests/baseline_images/test_03_gardner_altman_unpaired_hedges_g.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_04_gardner_altman_paired_hedges_g.png b/nbs/tests/mpl_image_tests/baseline_images/test_04_gardner_altman_paired_hedges_g.png
index 37c2cfd6..1f1372aa 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_04_gardner_altman_paired_hedges_g.png and b/nbs/tests/mpl_image_tests/baseline_images/test_04_gardner_altman_paired_hedges_g.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_04_gardner_altman_paired_meandiff.png b/nbs/tests/mpl_image_tests/baseline_images/test_04_gardner_altman_paired_meandiff.png
index 6d3f0f6d..ecf1b062 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_04_gardner_altman_paired_meandiff.png and b/nbs/tests/mpl_image_tests/baseline_images/test_04_gardner_altman_paired_meandiff.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_05_cummings_two_group_unpaired_meandiff.png b/nbs/tests/mpl_image_tests/baseline_images/test_05_cummings_two_group_unpaired_meandiff.png
index 895782b9..9b248613 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_05_cummings_two_group_unpaired_meandiff.png and b/nbs/tests/mpl_image_tests/baseline_images/test_05_cummings_two_group_unpaired_meandiff.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_06_cummings_two_group_paired_meandiff.png b/nbs/tests/mpl_image_tests/baseline_images/test_06_cummings_two_group_paired_meandiff.png
index 2c172249..003cee58 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_06_cummings_two_group_paired_meandiff.png and b/nbs/tests/mpl_image_tests/baseline_images/test_06_cummings_two_group_paired_meandiff.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_07_cummings_multi_group_unpaired.png b/nbs/tests/mpl_image_tests/baseline_images/test_07_cummings_multi_group_unpaired.png
index 5dabd6ba..4c1e3287 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_07_cummings_multi_group_unpaired.png and b/nbs/tests/mpl_image_tests/baseline_images/test_07_cummings_multi_group_unpaired.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_08_cummings_multi_group_paired.png b/nbs/tests/mpl_image_tests/baseline_images/test_08_cummings_multi_group_paired.png
index 5416a604..bf0299c1 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_08_cummings_multi_group_paired.png and b/nbs/tests/mpl_image_tests/baseline_images/test_08_cummings_multi_group_paired.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_09_cummings_shared_control.png b/nbs/tests/mpl_image_tests/baseline_images/test_09_cummings_shared_control.png
index 42484152..8932487d 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_09_cummings_shared_control.png and b/nbs/tests/mpl_image_tests/baseline_images/test_09_cummings_shared_control.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_101_gardner_altman_unpaired_propdiff.png b/nbs/tests/mpl_image_tests/baseline_images/test_101_gardner_altman_unpaired_propdiff.png
index 79cb7092..d163eb47 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_101_gardner_altman_unpaired_propdiff.png and b/nbs/tests/mpl_image_tests/baseline_images/test_101_gardner_altman_unpaired_propdiff.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_103_cummings_two_group_unpaired_propdiff.png b/nbs/tests/mpl_image_tests/baseline_images/test_103_cummings_two_group_unpaired_propdiff.png
index 4fdd936a..7fd3564d 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_103_cummings_two_group_unpaired_propdiff.png and b/nbs/tests/mpl_image_tests/baseline_images/test_103_cummings_two_group_unpaired_propdiff.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_105_cummings_multi_group_unpaired_propdiff.png b/nbs/tests/mpl_image_tests/baseline_images/test_105_cummings_multi_group_unpaired_propdiff.png
index 392abb53..29e798bd 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_105_cummings_multi_group_unpaired_propdiff.png and b/nbs/tests/mpl_image_tests/baseline_images/test_105_cummings_multi_group_unpaired_propdiff.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_106_cummings_shared_control_propdiff.png b/nbs/tests/mpl_image_tests/baseline_images/test_106_cummings_shared_control_propdiff.png
index 59d7ab71..1c9607ef 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_106_cummings_shared_control_propdiff.png and b/nbs/tests/mpl_image_tests/baseline_images/test_106_cummings_shared_control_propdiff.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_107_cummings_multi_groups_propdiff.png b/nbs/tests/mpl_image_tests/baseline_images/test_107_cummings_multi_groups_propdiff.png
index 606257de..e8da299c 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_107_cummings_multi_groups_propdiff.png and b/nbs/tests/mpl_image_tests/baseline_images/test_107_cummings_multi_groups_propdiff.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_109_gardner_altman_ylabel.png b/nbs/tests/mpl_image_tests/baseline_images/test_109_gardner_altman_ylabel.png
index 4641f875..252b4612 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_109_gardner_altman_ylabel.png and b/nbs/tests/mpl_image_tests/baseline_images/test_109_gardner_altman_ylabel.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_10_cummings_multi_groups.png b/nbs/tests/mpl_image_tests/baseline_images/test_10_cummings_multi_groups.png
index 751a187f..2e595360 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_10_cummings_multi_groups.png and b/nbs/tests/mpl_image_tests/baseline_images/test_10_cummings_multi_groups.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_110_change_fig_size.png b/nbs/tests/mpl_image_tests/baseline_images/test_110_change_fig_size.png
index 9d941ce2..52841a60 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_110_change_fig_size.png and b/nbs/tests/mpl_image_tests/baseline_images/test_110_change_fig_size.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_111_change_palette_b.png b/nbs/tests/mpl_image_tests/baseline_images/test_111_change_palette_b.png
index b73a9128..f0db92d1 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_111_change_palette_b.png and b/nbs/tests/mpl_image_tests/baseline_images/test_111_change_palette_b.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_112_change_palette_c.png b/nbs/tests/mpl_image_tests/baseline_images/test_112_change_palette_c.png
index bbedf6e9..aa95e180 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_112_change_palette_c.png and b/nbs/tests/mpl_image_tests/baseline_images/test_112_change_palette_c.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_113_desat.png b/nbs/tests/mpl_image_tests/baseline_images/test_113_desat.png
index eb015140..3cb1a98e 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_113_desat.png and b/nbs/tests/mpl_image_tests/baseline_images/test_113_desat.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_114_change_ylims.png b/nbs/tests/mpl_image_tests/baseline_images/test_114_change_ylims.png
index 9ccb46fe..56f87ef2 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_114_change_ylims.png and b/nbs/tests/mpl_image_tests/baseline_images/test_114_change_ylims.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_115_invert_ylim.png b/nbs/tests/mpl_image_tests/baseline_images/test_115_invert_ylim.png
index 33a9c89a..c6c789a2 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_115_invert_ylim.png and b/nbs/tests/mpl_image_tests/baseline_images/test_115_invert_ylim.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_116_ticker_gardner_altman.png b/nbs/tests/mpl_image_tests/baseline_images/test_116_ticker_gardner_altman.png
index fd5a79c4..72349492 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_116_ticker_gardner_altman.png and b/nbs/tests/mpl_image_tests/baseline_images/test_116_ticker_gardner_altman.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_117_err_color.png b/nbs/tests/mpl_image_tests/baseline_images/test_117_err_color.png
index 01184f12..7752500d 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_117_err_color.png and b/nbs/tests/mpl_image_tests/baseline_images/test_117_err_color.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_118_cummings_two_group_unpaired_meandiff_bar_width.png b/nbs/tests/mpl_image_tests/baseline_images/test_118_cummings_two_group_unpaired_meandiff_bar_width.png
index edad4e3c..a8302b4b 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_118_cummings_two_group_unpaired_meandiff_bar_width.png and b/nbs/tests/mpl_image_tests/baseline_images/test_118_cummings_two_group_unpaired_meandiff_bar_width.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_119_wide_df_nan.png b/nbs/tests/mpl_image_tests/baseline_images/test_119_wide_df_nan.png
index cd00b86d..0fc0b11b 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_119_wide_df_nan.png and b/nbs/tests/mpl_image_tests/baseline_images/test_119_wide_df_nan.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_11_inset_plots.png b/nbs/tests/mpl_image_tests/baseline_images/test_11_inset_plots.png
index 28a2bcd1..887e5283 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_11_inset_plots.png and b/nbs/tests/mpl_image_tests/baseline_images/test_11_inset_plots.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_120_long_df_nan.png b/nbs/tests/mpl_image_tests/baseline_images/test_120_long_df_nan.png
index cd00b86d..0fc0b11b 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_120_long_df_nan.png and b/nbs/tests/mpl_image_tests/baseline_images/test_120_long_df_nan.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_121_cohens_h_gardner_altman.png b/nbs/tests/mpl_image_tests/baseline_images/test_121_cohens_h_gardner_altman.png
index 7ae94529..44fe3297 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_121_cohens_h_gardner_altman.png and b/nbs/tests/mpl_image_tests/baseline_images/test_121_cohens_h_gardner_altman.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_122_cohens_h_cummings.png b/nbs/tests/mpl_image_tests/baseline_images/test_122_cohens_h_cummings.png
index c6f347da..73db4fe4 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_122_cohens_h_cummings.png and b/nbs/tests/mpl_image_tests/baseline_images/test_122_cohens_h_cummings.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_123_sankey_gardner_altman.png b/nbs/tests/mpl_image_tests/baseline_images/test_123_sankey_gardner_altman.png
index 52d6eee3..e7a11809 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_123_sankey_gardner_altman.png and b/nbs/tests/mpl_image_tests/baseline_images/test_123_sankey_gardner_altman.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_124_sankey_cummings.png b/nbs/tests/mpl_image_tests/baseline_images/test_124_sankey_cummings.png
index b23dd51e..0aa2f092 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_124_sankey_cummings.png and b/nbs/tests/mpl_image_tests/baseline_images/test_124_sankey_cummings.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_125_sankey_2paired_groups.png b/nbs/tests/mpl_image_tests/baseline_images/test_125_sankey_2paired_groups.png
index 1820a124..c2a17393 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_125_sankey_2paired_groups.png and b/nbs/tests/mpl_image_tests/baseline_images/test_125_sankey_2paired_groups.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_126_sankey_2sequential_groups.png b/nbs/tests/mpl_image_tests/baseline_images/test_126_sankey_2sequential_groups.png
index 1820a124..c2a17393 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_126_sankey_2sequential_groups.png and b/nbs/tests/mpl_image_tests/baseline_images/test_126_sankey_2sequential_groups.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_127_sankey_multi_group_paired.png b/nbs/tests/mpl_image_tests/baseline_images/test_127_sankey_multi_group_paired.png
index e4c7e8c2..b9d3169e 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_127_sankey_multi_group_paired.png and b/nbs/tests/mpl_image_tests/baseline_images/test_127_sankey_multi_group_paired.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_128_sankey_transparency.png b/nbs/tests/mpl_image_tests/baseline_images/test_128_sankey_transparency.png
index 83c2c9ae..ae6bd630 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_128_sankey_transparency.png and b/nbs/tests/mpl_image_tests/baseline_images/test_128_sankey_transparency.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_12_gardner_altman_ylabel.png b/nbs/tests/mpl_image_tests/baseline_images/test_12_gardner_altman_ylabel.png
index 33c4ebc6..914b8682 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_12_gardner_altman_ylabel.png and b/nbs/tests/mpl_image_tests/baseline_images/test_12_gardner_altman_ylabel.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_132_shared_control_sankey_off.png b/nbs/tests/mpl_image_tests/baseline_images/test_132_shared_control_sankey_off.png
index 7a4aeec3..e60f656c 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_132_shared_control_sankey_off.png and b/nbs/tests/mpl_image_tests/baseline_images/test_132_shared_control_sankey_off.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_133_shared_control_flow_off.png b/nbs/tests/mpl_image_tests/baseline_images/test_133_shared_control_flow_off.png
index 5c31d70c..ded4d4d2 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_133_shared_control_flow_off.png and b/nbs/tests/mpl_image_tests/baseline_images/test_133_shared_control_flow_off.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_134_separate_control_sankey_off.png b/nbs/tests/mpl_image_tests/baseline_images/test_134_separate_control_sankey_off.png
index 9609f767..66426292 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_134_separate_control_sankey_off.png and b/nbs/tests/mpl_image_tests/baseline_images/test_134_separate_control_sankey_off.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_135_separate_control_flow_off.png b/nbs/tests/mpl_image_tests/baseline_images/test_135_separate_control_flow_off.png
index 392cdfb6..e3eec96e 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_135_separate_control_flow_off.png and b/nbs/tests/mpl_image_tests/baseline_images/test_135_separate_control_flow_off.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_136_style_sheets.png b/nbs/tests/mpl_image_tests/baseline_images/test_136_style_sheets.png
index 54e02859..fd7aa500 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_136_style_sheets.png and b/nbs/tests/mpl_image_tests/baseline_images/test_136_style_sheets.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_13_multi_2group_color.png b/nbs/tests/mpl_image_tests/baseline_images/test_13_multi_2group_color.png
index 6ed27f79..032a4930 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_13_multi_2group_color.png and b/nbs/tests/mpl_image_tests/baseline_images/test_13_multi_2group_color.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_14_gardner_altman_paired_color.png b/nbs/tests/mpl_image_tests/baseline_images/test_14_gardner_altman_paired_color.png
index 2b9aea2f..7f3b1a0b 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_14_gardner_altman_paired_color.png and b/nbs/tests/mpl_image_tests/baseline_images/test_14_gardner_altman_paired_color.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_15_change_palette_a.png b/nbs/tests/mpl_image_tests/baseline_images/test_15_change_palette_a.png
index da2bc0d1..2eaf771d 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_15_change_palette_a.png and b/nbs/tests/mpl_image_tests/baseline_images/test_15_change_palette_a.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_16_change_palette_b.png b/nbs/tests/mpl_image_tests/baseline_images/test_16_change_palette_b.png
index aa4fc957..02536dc2 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_16_change_palette_b.png and b/nbs/tests/mpl_image_tests/baseline_images/test_16_change_palette_b.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_17_change_palette_c.png b/nbs/tests/mpl_image_tests/baseline_images/test_17_change_palette_c.png
index c448fc55..bb3c6d62 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_17_change_palette_c.png and b/nbs/tests/mpl_image_tests/baseline_images/test_17_change_palette_c.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_18_desat.png b/nbs/tests/mpl_image_tests/baseline_images/test_18_desat.png
index fa571711..685c2d43 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_18_desat.png and b/nbs/tests/mpl_image_tests/baseline_images/test_18_desat.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_19_dot_sizes.png b/nbs/tests/mpl_image_tests/baseline_images/test_19_dot_sizes.png
index 8eaf98df..39345266 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_19_dot_sizes.png and b/nbs/tests/mpl_image_tests/baseline_images/test_19_dot_sizes.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_201_forest_plot_no_colorpalette.png b/nbs/tests/mpl_image_tests/baseline_images/test_201_forest_plot_no_colorpalette.png
index 4f394293..907e6146 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_201_forest_plot_no_colorpalette.png and b/nbs/tests/mpl_image_tests/baseline_images/test_201_forest_plot_no_colorpalette.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_202_forest_plot_with_colorpalette.png b/nbs/tests/mpl_image_tests/baseline_images/test_202_forest_plot_with_colorpalette.png
index 50a85d7b..8bafdc53 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_202_forest_plot_with_colorpalette.png and b/nbs/tests/mpl_image_tests/baseline_images/test_202_forest_plot_with_colorpalette.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_203_horizontal_forest_plot_no_colorpalette.png b/nbs/tests/mpl_image_tests/baseline_images/test_203_horizontal_forest_plot_no_colorpalette.png
index 756567a6..814c3d4a 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_203_horizontal_forest_plot_no_colorpalette.png and b/nbs/tests/mpl_image_tests/baseline_images/test_203_horizontal_forest_plot_no_colorpalette.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_204_horizontal_forest_plot_with_colorpalette.png b/nbs/tests/mpl_image_tests/baseline_images/test_204_horizontal_forest_plot_with_colorpalette.png
index 5e457b96..f7f0533f 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_204_horizontal_forest_plot_with_colorpalette.png and b/nbs/tests/mpl_image_tests/baseline_images/test_204_horizontal_forest_plot_with_colorpalette.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_205_forest_mini_meta_horizontal.png b/nbs/tests/mpl_image_tests/baseline_images/test_205_forest_mini_meta_horizontal.png
index 9f296a66..ba7970ec 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_205_forest_mini_meta_horizontal.png and b/nbs/tests/mpl_image_tests/baseline_images/test_205_forest_mini_meta_horizontal.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_206_forest_mini_meta.png b/nbs/tests/mpl_image_tests/baseline_images/test_206_forest_mini_meta.png
index 864654a6..f003b4a4 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_206_forest_mini_meta.png and b/nbs/tests/mpl_image_tests/baseline_images/test_206_forest_mini_meta.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_207_gardner_altman_meandiff_empty_circle.png b/nbs/tests/mpl_image_tests/baseline_images/test_207_gardner_altman_meandiff_empty_circle.png
index 3abb704e..a59c9d51 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_207_gardner_altman_meandiff_empty_circle.png and b/nbs/tests/mpl_image_tests/baseline_images/test_207_gardner_altman_meandiff_empty_circle.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_208_cummings_two_group_unpaired_meandiff_empty_circle.png b/nbs/tests/mpl_image_tests/baseline_images/test_208_cummings_two_group_unpaired_meandiff_empty_circle.png
index 1cc00350..25e665cd 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_208_cummings_two_group_unpaired_meandiff_empty_circle.png and b/nbs/tests/mpl_image_tests/baseline_images/test_208_cummings_two_group_unpaired_meandiff_empty_circle.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_209_cummings_shared_control_meandiff_empty_circle.png b/nbs/tests/mpl_image_tests/baseline_images/test_209_cummings_shared_control_meandiff_empty_circle.png
index 2e975cf8..aad3e270 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_209_cummings_shared_control_meandiff_empty_circle.png and b/nbs/tests/mpl_image_tests/baseline_images/test_209_cummings_shared_control_meandiff_empty_circle.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_20_change_ylims.png b/nbs/tests/mpl_image_tests/baseline_images/test_20_change_ylims.png
index d43d7033..ed1adf17 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_20_change_ylims.png and b/nbs/tests/mpl_image_tests/baseline_images/test_20_change_ylims.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_210_cummings_multi_groups_meandiff_empty_circle.png b/nbs/tests/mpl_image_tests/baseline_images/test_210_cummings_multi_groups_meandiff_empty_circle.png
index 24bf0b02..f763947b 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_210_cummings_multi_groups_meandiff_empty_circle.png and b/nbs/tests/mpl_image_tests/baseline_images/test_210_cummings_multi_groups_meandiff_empty_circle.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_211_cummings_multi_2_group_meandiff_empty_circle.png b/nbs/tests/mpl_image_tests/baseline_images/test_211_cummings_multi_2_group_meandiff_empty_circle.png
index 76967000..09da92bf 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_211_cummings_multi_2_group_meandiff_empty_circle.png and b/nbs/tests/mpl_image_tests/baseline_images/test_211_cummings_multi_2_group_meandiff_empty_circle.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_212_cummings_unpaired_delta_delta_meandiff_empty_circle.png b/nbs/tests/mpl_image_tests/baseline_images/test_212_cummings_unpaired_delta_delta_meandiff_empty_circle.png
index cd688658..c05965c3 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_212_cummings_unpaired_delta_delta_meandiff_empty_circle.png and b/nbs/tests/mpl_image_tests/baseline_images/test_212_cummings_unpaired_delta_delta_meandiff_empty_circle.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_213_cummings_unpaired_mini_meta_meandiff_empty_circle.png b/nbs/tests/mpl_image_tests/baseline_images/test_213_cummings_unpaired_mini_meta_meandiff_empty_circle.png
index ae0484d8..4235aad6 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_213_cummings_unpaired_mini_meta_meandiff_empty_circle.png and b/nbs/tests/mpl_image_tests/baseline_images/test_213_cummings_unpaired_mini_meta_meandiff_empty_circle.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_214_change_idx_order_custom_palette_original.png b/nbs/tests/mpl_image_tests/baseline_images/test_214_change_idx_order_custom_palette_original.png
index 976ab458..ffcc5b7e 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_214_change_idx_order_custom_palette_original.png and b/nbs/tests/mpl_image_tests/baseline_images/test_214_change_idx_order_custom_palette_original.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_215_change_idx_order_custom_palette_new.png b/nbs/tests/mpl_image_tests/baseline_images/test_215_change_idx_order_custom_palette_new.png
index 65ff816a..9ce3bfc4 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_215_change_idx_order_custom_palette_new.png and b/nbs/tests/mpl_image_tests/baseline_images/test_215_change_idx_order_custom_palette_new.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_216_cummings_multi_groups_meandiff_show_baseline_ec.png b/nbs/tests/mpl_image_tests/baseline_images/test_216_cummings_multi_groups_meandiff_show_baseline_ec.png
index b5824a18..d5c6c451 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_216_cummings_multi_groups_meandiff_show_baseline_ec.png and b/nbs/tests/mpl_image_tests/baseline_images/test_216_cummings_multi_groups_meandiff_show_baseline_ec.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_217_cummings_multi_2_group_meandiff_show_baseline_ec.png b/nbs/tests/mpl_image_tests/baseline_images/test_217_cummings_multi_2_group_meandiff_show_baseline_ec.png
index b4bf63c4..5911ab83 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_217_cummings_multi_2_group_meandiff_show_baseline_ec.png and b/nbs/tests/mpl_image_tests/baseline_images/test_217_cummings_multi_2_group_meandiff_show_baseline_ec.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_21_invert_ylim.png b/nbs/tests/mpl_image_tests/baseline_images/test_21_invert_ylim.png
index 2379d990..3b5c2a76 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_21_invert_ylim.png and b/nbs/tests/mpl_image_tests/baseline_images/test_21_invert_ylim.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_22_ticker_gardner_altman.png b/nbs/tests/mpl_image_tests/baseline_images/test_22_ticker_gardner_altman.png
index 19b28ad0..19300870 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_22_ticker_gardner_altman.png and b/nbs/tests/mpl_image_tests/baseline_images/test_22_ticker_gardner_altman.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_23_ticker_cumming.png b/nbs/tests/mpl_image_tests/baseline_images/test_23_ticker_cumming.png
index 0a9db591..8f4ef09f 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_23_ticker_cumming.png and b/nbs/tests/mpl_image_tests/baseline_images/test_23_ticker_cumming.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_24_wide_df_nan.png b/nbs/tests/mpl_image_tests/baseline_images/test_24_wide_df_nan.png
index 1350981c..bbedf974 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_24_wide_df_nan.png and b/nbs/tests/mpl_image_tests/baseline_images/test_24_wide_df_nan.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_25_long_df_nan.png b/nbs/tests/mpl_image_tests/baseline_images/test_25_long_df_nan.png
index 1350981c..bbedf974 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_25_long_df_nan.png and b/nbs/tests/mpl_image_tests/baseline_images/test_25_long_df_nan.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_26_slopegraph_kwargs.png b/nbs/tests/mpl_image_tests/baseline_images/test_26_slopegraph_kwargs.png
index 87359f5f..cd91fb59 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_26_slopegraph_kwargs.png and b/nbs/tests/mpl_image_tests/baseline_images/test_26_slopegraph_kwargs.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_27_gardner_altman_reflines_kwargs.png b/nbs/tests/mpl_image_tests/baseline_images/test_27_gardner_altman_reflines_kwargs.png
index 73336ca7..f9928420 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_27_gardner_altman_reflines_kwargs.png and b/nbs/tests/mpl_image_tests/baseline_images/test_27_gardner_altman_reflines_kwargs.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_28_unpaired_cumming_reflines_kwargs.png b/nbs/tests/mpl_image_tests/baseline_images/test_28_unpaired_cumming_reflines_kwargs.png
index 038bbf35..337de71f 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_28_unpaired_cumming_reflines_kwargs.png and b/nbs/tests/mpl_image_tests/baseline_images/test_28_unpaired_cumming_reflines_kwargs.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_29_paired_cumming_slopegraph_reflines_kwargs.png b/nbs/tests/mpl_image_tests/baseline_images/test_29_paired_cumming_slopegraph_reflines_kwargs.png
index f9177fa2..d25bdc9e 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_29_paired_cumming_slopegraph_reflines_kwargs.png and b/nbs/tests/mpl_image_tests/baseline_images/test_29_paired_cumming_slopegraph_reflines_kwargs.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_30_sequential_cumming_slopegraph.png b/nbs/tests/mpl_image_tests/baseline_images/test_30_sequential_cumming_slopegraph.png
index b42f20e8..8acb151f 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_30_sequential_cumming_slopegraph.png and b/nbs/tests/mpl_image_tests/baseline_images/test_30_sequential_cumming_slopegraph.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_31_baseline_cumming_slopegraph.png b/nbs/tests/mpl_image_tests/baseline_images/test_31_baseline_cumming_slopegraph.png
index 512e3038..99f98af6 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_31_baseline_cumming_slopegraph.png and b/nbs/tests/mpl_image_tests/baseline_images/test_31_baseline_cumming_slopegraph.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_47_cummings_unpaired_delta_delta_meandiff.png b/nbs/tests/mpl_image_tests/baseline_images/test_47_cummings_unpaired_delta_delta_meandiff.png
index 300f43ca..cf2cf853 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_47_cummings_unpaired_delta_delta_meandiff.png and b/nbs/tests/mpl_image_tests/baseline_images/test_47_cummings_unpaired_delta_delta_meandiff.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_48_cummings_sequential_delta_delta_meandiff.png b/nbs/tests/mpl_image_tests/baseline_images/test_48_cummings_sequential_delta_delta_meandiff.png
index a9384f01..b895b264 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_48_cummings_sequential_delta_delta_meandiff.png and b/nbs/tests/mpl_image_tests/baseline_images/test_48_cummings_sequential_delta_delta_meandiff.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_49_cummings_baseline_delta_delta_meandiff.png b/nbs/tests/mpl_image_tests/baseline_images/test_49_cummings_baseline_delta_delta_meandiff.png
index a9384f01..b895b264 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_49_cummings_baseline_delta_delta_meandiff.png and b/nbs/tests/mpl_image_tests/baseline_images/test_49_cummings_baseline_delta_delta_meandiff.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_50_delta_plot_ylabel.png b/nbs/tests/mpl_image_tests/baseline_images/test_50_delta_plot_ylabel.png
index f64a02db..7b6724d3 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_50_delta_plot_ylabel.png and b/nbs/tests/mpl_image_tests/baseline_images/test_50_delta_plot_ylabel.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_51_delta_plot_change_palette_a.png b/nbs/tests/mpl_image_tests/baseline_images/test_51_delta_plot_change_palette_a.png
index 2bb76b44..fe0f671c 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_51_delta_plot_change_palette_a.png and b/nbs/tests/mpl_image_tests/baseline_images/test_51_delta_plot_change_palette_a.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_52_delta_specified.png b/nbs/tests/mpl_image_tests/baseline_images/test_52_delta_specified.png
index 89b23a8d..08ce8953 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_52_delta_specified.png and b/nbs/tests/mpl_image_tests/baseline_images/test_52_delta_specified.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_53_delta_change_ylims.png b/nbs/tests/mpl_image_tests/baseline_images/test_53_delta_change_ylims.png
index a188c8ad..6bee7f87 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_53_delta_change_ylims.png and b/nbs/tests/mpl_image_tests/baseline_images/test_53_delta_change_ylims.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_54_delta_invert_ylim.png b/nbs/tests/mpl_image_tests/baseline_images/test_54_delta_invert_ylim.png
index ebccec26..da12c98e 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_54_delta_invert_ylim.png and b/nbs/tests/mpl_image_tests/baseline_images/test_54_delta_invert_ylim.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_55_delta_median_diff.png b/nbs/tests/mpl_image_tests/baseline_images/test_55_delta_median_diff.png
index 0195fbad..f0c64745 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_55_delta_median_diff.png and b/nbs/tests/mpl_image_tests/baseline_images/test_55_delta_median_diff.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_56_delta_cohens_d.png b/nbs/tests/mpl_image_tests/baseline_images/test_56_delta_cohens_d.png
index 37a1fb10..00ea66e0 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_56_delta_cohens_d.png and b/nbs/tests/mpl_image_tests/baseline_images/test_56_delta_cohens_d.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_57_delta_show_delta2.png b/nbs/tests/mpl_image_tests/baseline_images/test_57_delta_show_delta2.png
index 6d88ef90..85b09805 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_57_delta_show_delta2.png and b/nbs/tests/mpl_image_tests/baseline_images/test_57_delta_show_delta2.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_58_delta_axes_invert_ylim.png b/nbs/tests/mpl_image_tests/baseline_images/test_58_delta_axes_invert_ylim.png
index d15b1724..35b72020 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_58_delta_axes_invert_ylim.png and b/nbs/tests/mpl_image_tests/baseline_images/test_58_delta_axes_invert_ylim.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_59_delta_axes_invert_ylim_not_showing_delta2.png b/nbs/tests/mpl_image_tests/baseline_images/test_59_delta_axes_invert_ylim_not_showing_delta2.png
index 6d88ef90..85b09805 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_59_delta_axes_invert_ylim_not_showing_delta2.png and b/nbs/tests/mpl_image_tests/baseline_images/test_59_delta_axes_invert_ylim_not_showing_delta2.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_60_cummings_unpaired_mini_meta_meandiff.png b/nbs/tests/mpl_image_tests/baseline_images/test_60_cummings_unpaired_mini_meta_meandiff.png
index c3cd87bb..5843eea5 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_60_cummings_unpaired_mini_meta_meandiff.png and b/nbs/tests/mpl_image_tests/baseline_images/test_60_cummings_unpaired_mini_meta_meandiff.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_61_cummings_sequential_mini_meta_meandiff.png b/nbs/tests/mpl_image_tests/baseline_images/test_61_cummings_sequential_mini_meta_meandiff.png
index 678c6462..0560f44f 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_61_cummings_sequential_mini_meta_meandiff.png and b/nbs/tests/mpl_image_tests/baseline_images/test_61_cummings_sequential_mini_meta_meandiff.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_62_cummings_baseline_mini_meta_meandiff.png b/nbs/tests/mpl_image_tests/baseline_images/test_62_cummings_baseline_mini_meta_meandiff.png
index 678c6462..0560f44f 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_62_cummings_baseline_mini_meta_meandiff.png and b/nbs/tests/mpl_image_tests/baseline_images/test_62_cummings_baseline_mini_meta_meandiff.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_63_mini_meta_plot_ylabel.png b/nbs/tests/mpl_image_tests/baseline_images/test_63_mini_meta_plot_ylabel.png
index f0a2ef78..e4c4e38c 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_63_mini_meta_plot_ylabel.png and b/nbs/tests/mpl_image_tests/baseline_images/test_63_mini_meta_plot_ylabel.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_64_mini_meta_plot_change_palette_a.png b/nbs/tests/mpl_image_tests/baseline_images/test_64_mini_meta_plot_change_palette_a.png
index 7a7013b6..030ed983 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_64_mini_meta_plot_change_palette_a.png and b/nbs/tests/mpl_image_tests/baseline_images/test_64_mini_meta_plot_change_palette_a.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_65_mini_meta_dot_sizes.png b/nbs/tests/mpl_image_tests/baseline_images/test_65_mini_meta_dot_sizes.png
index 38fa293c..a69768f7 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_65_mini_meta_dot_sizes.png and b/nbs/tests/mpl_image_tests/baseline_images/test_65_mini_meta_dot_sizes.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_66_mini_meta_change_ylims.png b/nbs/tests/mpl_image_tests/baseline_images/test_66_mini_meta_change_ylims.png
index de122d50..9e317831 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_66_mini_meta_change_ylims.png and b/nbs/tests/mpl_image_tests/baseline_images/test_66_mini_meta_change_ylims.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_67_mini_meta_invert_ylim.png b/nbs/tests/mpl_image_tests/baseline_images/test_67_mini_meta_invert_ylim.png
index afd67c8d..822a8511 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_67_mini_meta_invert_ylim.png and b/nbs/tests/mpl_image_tests/baseline_images/test_67_mini_meta_invert_ylim.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_68_mini_meta_median_diff.png b/nbs/tests/mpl_image_tests/baseline_images/test_68_mini_meta_median_diff.png
index 9808f220..86c6ed2c 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_68_mini_meta_median_diff.png and b/nbs/tests/mpl_image_tests/baseline_images/test_68_mini_meta_median_diff.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_69_mini_meta_cohens_d.png b/nbs/tests/mpl_image_tests/baseline_images/test_69_mini_meta_cohens_d.png
index f3c1d761..e5984237 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_69_mini_meta_cohens_d.png and b/nbs/tests/mpl_image_tests/baseline_images/test_69_mini_meta_cohens_d.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_70_mini_meta_not_show.png b/nbs/tests/mpl_image_tests/baseline_images/test_70_mini_meta_not_show.png
index 3c881c1c..249d78a6 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_70_mini_meta_not_show.png and b/nbs/tests/mpl_image_tests/baseline_images/test_70_mini_meta_not_show.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_71_unpaired_delta_g.png b/nbs/tests/mpl_image_tests/baseline_images/test_71_unpaired_delta_g.png
index 20c75b9c..98a099f1 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_71_unpaired_delta_g.png and b/nbs/tests/mpl_image_tests/baseline_images/test_71_unpaired_delta_g.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_72_sequential_delta_g.png b/nbs/tests/mpl_image_tests/baseline_images/test_72_sequential_delta_g.png
index a9384f01..b895b264 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_72_sequential_delta_g.png and b/nbs/tests/mpl_image_tests/baseline_images/test_72_sequential_delta_g.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_73_baseline_delta_g.png b/nbs/tests/mpl_image_tests/baseline_images/test_73_baseline_delta_g.png
index a9384f01..b895b264 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_73_baseline_delta_g.png and b/nbs/tests/mpl_image_tests/baseline_images/test_73_baseline_delta_g.png differ
diff --git a/nbs/tests/mpl_image_tests/baseline_images/test_99_style_sheets.png b/nbs/tests/mpl_image_tests/baseline_images/test_99_style_sheets.png
index 2be2ffe5..8428745c 100644
Binary files a/nbs/tests/mpl_image_tests/baseline_images/test_99_style_sheets.png and b/nbs/tests/mpl_image_tests/baseline_images/test_99_style_sheets.png differ
diff --git a/nbs/tests/test_01_effsizes_pvals.ipynb b/nbs/tests/test_01_effsizes_pvals.ipynb
index f2997a42..d499f31b 100644
--- a/nbs/tests/test_01_effsizes_pvals.ipynb
+++ b/nbs/tests/test_01_effsizes_pvals.ipynb
@@ -132,7 +132,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "cohens_d = effsize.cohens_d(wellbeing.control, wellbeing.expt,\n",
+    "cohens_d = effsize.cohens_d(np.array(wellbeing.control), np.array(wellbeing.expt),\n",
     "                            is_paired=False)\n",
     "assert np.round(cohens_d, 2) == pytest.approx(0.47)"
    ]
@@ -152,7 +152,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "hedges_g = effsize.hedges_g(wellbeing.control, wellbeing.expt,\n",
+    "hedges_g = effsize.hedges_g(np.array(wellbeing.control), np.array(wellbeing.expt),\n",
     "                                is_paired=False)\n",
     "assert np.round(hedges_g, 2) == pytest.approx(0.45)"
    ]
@@ -172,7 +172,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "cohens_d = effsize.cohens_d(paired_wellbeing.pre, paired_wellbeing.post,\n",
+    "cohens_d = effsize.cohens_d(np.array(paired_wellbeing.pre), np.array(paired_wellbeing.post),\n",
     "                                is_paired=\"baseline\")\n",
     "assert np.round(cohens_d, 2) == pytest.approx(0.34)\n"
    ]
@@ -192,7 +192,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "hedges_g = effsize.hedges_g(paired_wellbeing.pre, paired_wellbeing.post,\n",
+    "hedges_g = effsize.hedges_g(np.array(paired_wellbeing.pre), np.array(paired_wellbeing.post),\n",
     "                            is_paired=\"baseline\")\n",
     "assert np.round(hedges_g, 2) == pytest.approx(0.33)"
    ]
@@ -212,7 +212,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "cohens_h = effsize.cohens_h(smoke.low, smoke.high)\n",
+    "cohens_h = effsize.cohens_h(np.array(smoke.low), np.array(smoke.high))\n",
     "assert np.round(cohens_h, 2) == pytest.approx(0.17)"
    ]
   },
@@ -231,10 +231,10 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "likert_delta = effsize.cliffs_delta(likert_treatment, likert_control)\n",
+    "likert_delta = effsize.cliffs_delta(np.array(likert_treatment), np.array(likert_control))\n",
     "assert likert_delta == pytest.approx(-0.25)\n",
     "\n",
-    "scores_delta = effsize.cliffs_delta(b_scores, a_scores)\n",
+    "scores_delta = effsize.cliffs_delta(np.array(b_scores), np.array(a_scores))\n",
     "assert scores_delta == pytest.approx(0.65)"
    ]
   },
diff --git a/nbs/tests/test_02_edge_cases.ipynb b/nbs/tests/test_02_edge_cases.ipynb
index 27821eee..42fb6377 100644
--- a/nbs/tests/test_02_edge_cases.ipynb
+++ b/nbs/tests/test_02_edge_cases.ipynb
@@ -49,7 +49,7 @@
     "random_seed=12345\n",
     "\n",
     "# rng = RandomState(MT19937(random_seed))\n",
-    "rng = RandomState(PCG64(12345))\n",
+    "rng = RandomState(PCG64(random_seed))\n",
     "# rng = np.random.default_rng(seed=random_seed)\n",
     "\n",
     "df = pd.DataFrame(\n",
@@ -63,19 +63,10 @@
     "            idx=['Group 1', 'Group 2'])\n",
     "\n",
     "md = test.mean_diff.results\n",
-    "\n",
     "assert md.difference[0] == pytest.approx(-0.0322, abs=1e-4)\n",
-    "assert md.bca_low[0]    == pytest.approx(-0.2279, abs=1e-4)\n",
-    "assert md.bca_high[0]   == pytest.approx(0.1613, abs=1e-4)"
+    "assert md.bca_low[0]    == pytest.approx(-0.2268, abs=1e-4)\n",
+    "assert md.bca_high[0]   == pytest.approx(0.1524, abs=1e-4)"
    ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": null,
-   "id": "afc96b46",
-   "metadata": {},
-   "outputs": [],
-   "source": []
   }
  ],
  "metadata": {
diff --git a/nbs/tests/test_08_mini_meta_pvals.ipynb b/nbs/tests/test_08_mini_meta_pvals.ipynb
index 464d3524..dd3730be 100644
--- a/nbs/tests/test_08_mini_meta_pvals.ipynb
+++ b/nbs/tests/test_08_mini_meta_pvals.ipynb
@@ -110,10 +110,10 @@
    "source": [
     "difference = unpaired.mean_diff.mini_meta_delta.difference\n",
     "\n",
-    "np_means = [np.mean(rep1_yes)-np.mean(rep1_no), \n",
-    "            np.mean(rep2_yes)-np.mean(rep2_no)]\n",
-    "np_var   = [np.var(rep1_yes, ddof=1)/N+np.var(rep1_no, ddof=1)/N,\n",
-    "            np.var(rep2_yes, ddof=1)/N+np.var(rep2_no, ddof=1)/N]\n",
+    "np_means = np.array([np.mean(rep1_yes)-np.mean(rep1_no), \n",
+    "            np.mean(rep2_yes)-np.mean(rep2_no)])\n",
+    "np_var   = np.array([np.var(rep1_yes, ddof=1)/N+np.var(rep1_no, ddof=1)/N,\n",
+    "            np.var(rep2_yes, ddof=1)/N+np.var(rep2_no, ddof=1)/N])\n",
     "\n",
     "np_difference = effsize.weighted_delta(np_means, np_var)\n",
     "\n",
diff --git a/nbs/tests/test_99_confidence_intervals.ipynb b/nbs/tests/test_99_confidence_intervals.ipynb
index 2475793b..2926a5c7 100644
--- a/nbs/tests/test_99_confidence_intervals.ipynb
+++ b/nbs/tests/test_99_confidence_intervals.ipynb
@@ -55,8 +55,9 @@
     "             paired=\"baseline\", id_col=\"subject_id\")\n",
     "paired_mean_diff = ex_bp.mean_diff.results\n",
     "\n",
-    "assert pytest.approx(3.875) == paired_mean_diff.bca_low[0]\n",
-    "assert pytest.approx(9.5) == paired_mean_diff.bca_high[0]"
+    "\n",
+    "assert pytest.approx(3.625) == paired_mean_diff.bca_low[0]\n",
+    "assert pytest.approx(9.125) == paired_mean_diff.bca_high[0]"
    ]
   },
   {
@@ -198,14 +199,6 @@
     "assert error_count_median_diff  <= max_errors\n",
     "assert error_count_cliffs_delta <= max_errors\n"
    ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": null,
-   "id": "9da1b76d",
-   "metadata": {},
-   "outputs": [],
-   "source": []
   }
  ],
  "metadata": {
diff --git a/settings.ini b/settings.ini
index a6b36da8..614453d7 100644
--- a/settings.ini
+++ b/settings.ini
@@ -2,7 +2,7 @@
 ### Python library ###
 repo = DABEST-python
 lib_name = dabest
-version = 2024.03.29
+version = 2024.03.30
 min_python = 3.9
 license = apache2
 
@@ -37,7 +37,7 @@ language = English
 status = 3
 user = acclab
 
-requirements = fastcore pandas~=2.1.4 numpy~=1.26 matplotlib~=3.8.4 seaborn~=0.12.2 scipy~=1.12 datetime statsmodels lqrt
+requirements = fastcore pandas~=2.1.4 numpy~=1.26 matplotlib~=3.8.4 seaborn~=0.12.2 scipy~=1.12 datetime statsmodels lqrt numba tqdm
 dev_requirements = pytest~=7.2.1 pytest-mpl~=0.16.1
 
 ### Optional ###