Merge pull request #43 from MatthewSZhang/beam

FEAT add tol to minibatch

Author: MatthewSZhang

.github/workflows/ci.yml

@@ -13,7 +13,7 @@ jobs:
 
     steps:
       - uses: actions/checkout@v4
-      - uses: prefix-dev/[email protected].2
+      - uses: prefix-dev/[email protected].3
         with:
           environments: default
           cache: true

examples/plot_pruning.py

@@ -72,7 +72,7 @@ def _fastcan_pruning(
     ).fit(X)
     atoms = kmeans.cluster_centers_
     ids_fastcan = minibatch(
-        X.T, atoms.T, n_samples_to_select, batch_size=batch_size, verbose=0
+        X.T, atoms.T, n_samples_to_select, batch_size=batch_size, tol=1e-6, verbose=0
     )
     pruned_lr = LogisticRegression(max_iter=110).fit(X[ids_fastcan], y[ids_fastcan])
     return pruned_lr.coef_, pruned_lr.intercept_

examples/plot_redundancy.py

@@ -173,8 +173,11 @@ def get_n_missed(dep_info_ids, indep_info_ids, redundant_ids, selected_ids):
 # Run test
 # --------
 
+import time
+
 N_SELECTORS = len(selector_dict)
 n_missed = np.zeros((N_REPEATED, N_SELECTORS), dtype=int)
+elapsed_time = np.zeros((N_REPEATED, N_SELECTORS), dtype=float)
 
 for i in range(N_REPEATED):
     data, target = make_redundant(
@@ -186,7 +189,9 @@ def get_n_missed(dep_info_ids, indep_info_ids, redundant_ids, selected_ids):
         random_seed=i,
     )
     for j, selector in enumerate(selector_dict.values()):
+        start_time = time.time()
         result_ids = selector.fit(data, target).get_support(indices=True)
+        elapsed_time[i, j] = time.time() - start_time
         n_missed[i, j] = get_n_missed(
             dep_info_ids=DEP_INFO_IDS,
             indep_info_ids=INDEP_INFO_IDS,
@@ -202,10 +207,14 @@ def get_n_missed(dep_info_ids, indep_info_ids, redundant_ids, selected_ids):
 
 import matplotlib.pyplot as plt
 
-fig, ax = plt.subplots(figsize=(8, 5))
-rects = ax.bar(selector_dict.keys(), n_missed.sum(0), width=0.5)
-ax.bar_label(rects, n_missed.sum(0), padding=3)
+fig = plt.figure(figsize=(8, 5))
+ax1 = fig.add_subplot()
+ax2 = ax1.twinx()
+ax1.set_ylabel("No. of missed features")
+ax2.set_ylabel("Elapsed time (s)")
+rects = ax1.bar(selector_dict.keys(), n_missed.sum(0), width=0.5)
+ax1.bar_label(rects, n_missed.sum(0), padding=3)
+ax2.semilogy(selector_dict.keys(), elapsed_time.mean(0), marker="o", color="tab:orange")
 plt.xlabel("Selector")
-plt.ylabel("No. of missed features")
 plt.title("Performance of selectors on datasets with linearly redundant features")
 plt.show()

fastcan/_minibatch.py

@@ -6,7 +6,7 @@
 # SPDX-License-Identifier: MIT
 
 from copy import deepcopy
-from numbers import Integral
+from numbers import Integral, Real
 
 import numpy as np
 from sklearn.utils._openmp_helpers import _openmp_effective_n_threads
@@ -27,11 +27,12 @@
         "batch_size": [
             Interval(Integral, 1, None, closed="left"),
         ],
+        "tol": [Interval(Real, 0, None, closed="neither")],
         "verbose": ["verbose"],
     },
     prefer_skip_nested_validation=True,
 )
-def minibatch(X, y, n_features_to_select=1, batch_size=1, verbose=1):
+def minibatch(X, y, n_features_to_select=1, batch_size=1, tol=0.01, verbose=1):
     """Feature selection using :class:`fastcan.FastCan` with mini batches.
 
     It is suitable for selecting a very large number of features
@@ -60,6 +61,9 @@ def minibatch(X, y, n_features_to_select=1, batch_size=1, verbose=1):
         The upper bound of the number of features in a mini-batch.
         It is recommended that batch_size be less than n_samples.
 
+    tol : float, default=0.01
+        Tolerance for linear dependence check.
+
     verbose : int, default=1
         The verbosity level.
 
@@ -118,7 +122,7 @@ def minibatch(X, y, n_features_to_select=1, batch_size=1, verbose=1):
                     X=deepcopy(X_transformed_),
                     V=y_i,
                     t=batch_size_temp,
-                    tol=0.01,
+                    tol=tol,
                     num_threads=n_threads,
                     verbose=0,
                     mask=mask,