Rename variables (ME-ICA#1145)

tedana/decomposition/ica.py

@@ -54,7 +54,7 @@ def tedica(
 
     Returns
     -------
-    mmix : (T x C) :obj:`numpy.ndarray`
+    mixing : (T x C) :obj:`numpy.ndarray`
         Z-scored mixing matrix for converting input data to component space,
         where `C` is components and `T` is the same as in `data`
     fixed_seed : :obj:`int`
@@ -69,15 +69,15 @@ def tedica(
     ica_method = ica_method.lower()
 
     if ica_method == "robustica":
-        mmix, fixed_seed = r_ica(
+        mixing, fixed_seed = r_ica(
             data,
             n_components=n_components,
             fixed_seed=fixed_seed,
             n_robust_runs=n_robust_runs,
             max_it=maxit,
         )
     elif ica_method == "fastica":
-        mmix, fixed_seed = f_ica(
+        mixing, fixed_seed = f_ica(
             data,
             n_components=n_components,
             fixed_seed=fixed_seed,
@@ -87,7 +87,7 @@ def tedica(
     else:
         raise ValueError("The selected ICA method is invalid!")
 
-    return mmix, fixed_seed
+    return mixing, fixed_seed
 
 
 def r_ica(data, n_components, fixed_seed, n_robust_runs, max_it):
@@ -109,7 +109,7 @@ def r_ica(data, n_components, fixed_seed, n_robust_runs, max_it):
 
     Returns
     -------
-    mmix : (T x C) :obj:`numpy.ndarray`
+    mixing : (T x C) :obj:`numpy.ndarray`
         Z-scored mixing matrix for converting input data to component space,
         where `C` is components and `T` is the same as in `data`
     fixed_seed : :obj:`int`
@@ -140,7 +140,7 @@ def r_ica(data, n_components, fixed_seed, n_robust_runs, max_it):
                 robust_method=robust_method,
             )
 
-            s, mmix = robust_ica.fit_transform(data)
+            s, mixing = robust_ica.fit_transform(data)
             q = robust_ica.evaluate_clustering(
                 robust_ica.S_all,
                 robust_ica.clustering.labels_,
@@ -151,7 +151,7 @@ def r_ica(data, n_components, fixed_seed, n_robust_runs, max_it):
 
         except Exception:
             if robust_method == "DBSCAN":
-                # if RobustICA failed wtih DBSCAN, run again wtih AgglomerativeClustering
+                # if RobustICA failed wtih DBSCAN, run again with AgglomerativeClustering
                 robust_method = "AgglomerativeClustering"
                 LGR.warning(
                     "DBSCAN clustering method did not converge. "
@@ -165,20 +165,18 @@ def r_ica(data, n_components, fixed_seed, n_robust_runs, max_it):
         f"components across different runs"
     )
 
-    iq = np.array(
-        np.mean(q[q["cluster_id"] >= 0].iq)
-    )  # Excluding outliers (cluster -1) from the index quality calculation
+    # Excluding outliers (cluster -1) from the index quality calculation
+    iq = np.array(np.mean(q[q["cluster_id"] >= 0].iq))
 
     if iq < WARN_IQ:
         LGR.warning(
             f"The resultant mean Index Quality is low ({iq}). It is recommended to rerun the "
             "process with a different seed."
         )
 
-    mmix = mmix[
-        :, q["cluster_id"] >= 0
-    ]  # Excluding outliers (cluster -1) when calculating the mixing matrix
-    mmix = stats.zscore(mmix, axis=0)
+    # Excluding outliers (cluster -1) when calculating the mixing matrix
+    mixing = mixing[:, q["cluster_id"] >= 0]
+    mixing = stats.zscore(mixing, axis=0)
 
     LGR.info(
         f"RobustICA with {n_robust_runs} robust runs and seed {fixed_seed} was used. "
@@ -194,7 +192,7 @@ def r_ica(data, n_components, fixed_seed, n_robust_runs, max_it):
             f"decomposition."
         )
 
-    return mmix, fixed_seed
+    return mixing, fixed_seed
 
 
 def f_ica(data, n_components, fixed_seed, maxit, maxrestart):
@@ -218,7 +216,7 @@ def f_ica(data, n_components, fixed_seed, maxit, maxrestart):
 
     Returns
     -------
-    mmix : (T x C) :obj:`numpy.ndarray`
+    mixing : (T x C) :obj:`numpy.ndarray`
         Z-scored mixing matrix for converting input data to component space,
         where `C` is components and `T` is the same as in `data`
     fixed_seed : :obj:`int`
@@ -262,6 +260,6 @@ def f_ica(data, n_components, fixed_seed, maxit, maxrestart):
                 )
                 break
 
-    mmix = ica.mixing_
-    mmix = stats.zscore(mmix, axis=0)
-    return mmix, fixed_seed
+    mixing = ica.mixing_
+    mixing = stats.zscore(mixing, axis=0)
+    return mixing, fixed_seed

tedana/decomposition/pca.py

@@ -50,7 +50,7 @@ def low_mem_pca(data):
 
 def tedpca(
     data_cat,
-    data_oc,
+    data_optcom,
     mask,
     adaptive_mask,
     io_generator,
@@ -66,7 +66,7 @@ def tedpca(
     ----------
     data_cat : (S x E x T) array_like
         Input functional data
-    data_oc : (S x T) array_like
+    data_optcom : (S x T) array_like
         Optimally combined time series data
     mask : (S,) array_like
         Boolean mask array
@@ -203,7 +203,7 @@ def tedpca(
     LGR.info(
         f"Computing PCA of optimally combined multi-echo data with selection criteria: {algorithm}"
     )
-    data = data_oc[mask, :]
+    data = data_optcom[mask, :]
 
     data_z = ((data.T - data.T.mean(axis=0)) / data.T.std(axis=0)).T  # var normalize ts
     data_z = (data_z - data_z.mean()) / data_z.std()  # var normalize everything
@@ -349,9 +349,9 @@ def tedpca(
         "d_table_score",
     ]
     # Even if user inputted, don't fit external_regressors to PCA components
-    comptable, _ = metrics.collect.generate_metrics(
+    component_table, _ = metrics.collect.generate_metrics(
         data_cat=data_cat,
-        data_optcom=data_oc,
+        data_optcom=data_optcom,
         mixing=comp_ts,
         adaptive_mask=adaptive_mask,
         tes=tes,
@@ -363,27 +363,27 @@ def tedpca(
 
     # varex_norm from PCA overrides varex_norm from dependence_metrics,
     # but we retain the original
-    comptable["estimated normalized variance explained"] = comptable[
+    component_table["estimated normalized variance explained"] = component_table[
         "normalized variance explained"
     ]
-    comptable["normalized variance explained"] = varex_norm
+    component_table["normalized variance explained"] = varex_norm
 
     # write component maps to 4D image
-    comp_maps = utils.unmask(computefeats2(data_oc, comp_ts, mask), mask)
+    comp_maps = utils.unmask(computefeats2(data_optcom, comp_ts, mask), mask)
     io_generator.save_file(comp_maps, "z-scored PCA components img")
 
     # Select components using decision tree
     if algorithm == "kundu":
-        comptable, metric_metadata = kundu_tedpca(
-            comptable,
+        component_table, metric_metadata = kundu_tedpca(
+            component_table,
             n_echos,
             kdaw,
             rdaw,
             stabilize=False,
         )
     elif algorithm == "kundu-stabilize":
-        comptable, metric_metadata = kundu_tedpca(
-            comptable,
+        component_table, metric_metadata = kundu_tedpca(
+            component_table,
             n_echos,
             kdaw,
             rdaw,
@@ -397,25 +397,26 @@ def tedpca(
         else:
             alg_str = algorithm
         LGR.info(
-            f"Selected {comptable.shape[0]} components with {round(100 * varex_norm.sum(), 2)}% "
+            f"Selected {component_table.shape[0]} components with "
+            f"{round(100 * varex_norm.sum(), 2)}% "
             f"normalized variance explained using {alg_str} dimensionality estimate"
         )
-        comptable["classification"] = "accepted"
-        comptable["rationale"] = ""
+        component_table["classification"] = "accepted"
+        component_table["rationale"] = ""
 
     # Save decomposition files
     comp_names = [
-        io.add_decomp_prefix(comp, prefix="pca", max_value=comptable.index.max())
-        for comp in comptable.index.values
+        io.add_decomp_prefix(comp, prefix="pca", max_value=component_table.index.max())
+        for comp in component_table.index.values
     ]
 
     mixing_df = pd.DataFrame(data=comp_ts, columns=comp_names)
     io_generator.save_file(mixing_df, "PCA mixing tsv")
 
     # Save component table and associated json
-    io_generator.save_file(comptable, "PCA metrics tsv")
+    io_generator.save_file(component_table, "PCA metrics tsv")
 
-    metric_metadata = metrics.collect.get_metadata(comptable)
+    metric_metadata = metrics.collect.get_metadata(component_table)
     io_generator.save_file(metric_metadata, "PCA metrics json")
 
     decomp_metadata = {
@@ -431,7 +432,7 @@ def tedpca(
         }
     io_generator.save_file(decomp_metadata, "PCA decomposition json")
 
-    acc = comptable[comptable.classification == "accepted"].index.values
+    acc = component_table[component_table.classification == "accepted"].index.values
     n_components = acc.size
     voxel_kept_comp_weighted = voxel_comp_weights[:, acc] * varex[None, acc]
     kept_data = np.dot(voxel_kept_comp_weighted, comp_ts[:, acc].T)

tedana/gscontrol.py

@@ -145,7 +145,7 @@ def minimum_image_regression(
     data_optcom: np.ndarray,
     mixing: np.ndarray,
     mask: np.ndarray,
-    comptable: pd.DataFrame,
+    component_table: pd.DataFrame,
     classification_tags: list,
     io_generator: io.OutputGenerator,
 ):
@@ -163,7 +163,7 @@ def minimum_image_regression(
         is components and ``T`` is the same as in ``data_optcom``
     mask : (S,) array_like
         Boolean mask array
-    comptable : (C x X) :obj:`pandas.DataFrame`
+    component_table : (C x X) :obj:`pandas.DataFrame`
         Component metric table. One row for each component, with a column for
         each metric. The index should be the component number.
     classification_tags : :obj:`list` of :obj:`str`
@@ -210,7 +210,7 @@ def minimum_image_regression(
         "diffuse noise \\citep{kundu2013integrated}."
     )
 
-    all_comps = comptable.index.values
+    all_comps = component_table.index.values
     # Get accepted and ignored components
     # Tedana has removed the "ignored" classification,
     # so we must separate "accepted" components based on the classification tag(s).
@@ -225,11 +225,11 @@ def minimum_image_regression(
         pattern = "|".join(ignore_tags)  # Create a pattern that matches any of the ignore tags
 
         # Select rows where the 'classification_tags' column contains any of the ignore tags
-        ign = comptable[
-            comptable.classification_tags.str.contains(pattern, na=False, regex=True)
+        ign = component_table[
+            component_table.classification_tags.str.contains(pattern, na=False, regex=True)
         ].index.values
 
-    acc = comptable[comptable.classification == "accepted"].index.values
+    acc = component_table[component_table.classification == "accepted"].index.values
     # Ignored components are classified as "accepted", so we need to remove them from the list
     acc = sorted(np.setdiff1d(acc, ign))
     not_ign = sorted(np.setdiff1d(all_comps, ign))
@@ -273,7 +273,7 @@ def minimum_image_regression(
     mixing_not1gs_z = np.vstack((np.atleast_2d(np.ones(max(gs_ts.shape))), gs_ts, mixing_not1gs_z))
 
     # Write T1-corrected components and mixing matrix
-    mixing_df = pd.DataFrame(data=mixing_not1gs.T, columns=comptable["Component"].values)
+    mixing_df = pd.DataFrame(data=mixing_not1gs.T, columns=component_table["Component"].values)
     io_generator.save_file(mixing_df, "ICA MIR mixing tsv")
 
     if io_generator.verbose: