Merge branch 'master' into master

.travis.yml

@@ -1,7 +1,10 @@
 sudo: required
 language: python
 python:
+
  - "3.6"
+ - "3.7"
+ - "3.8"
 
 notifications:
  email:

CHANGELOG.md

@@ -1,5 +1,11 @@
 # dev
 
+- Dropped support for python 2.
+- Added testing for python 3.7 and 3.8
+- Updated compatibility for ASE 3.19.0
+- Updated compatibility for Pandas 0.24.0
+- Updated compatibility for Scikit-learn 0.22.0
+
 # Version 0.6.1 (April 2019)
 
 - Fixed compatibility issue with MLNEB and [GPAW](https://wiki.fysik.dtu.dk/gpaw/index.html)

catlearn/featurize/setup.py

@@ -136,7 +136,7 @@ def featurize_atomic_pairs(self, candidates):
  for atoms in tqdm(candidates):
  # One hot encode elements.
  dummies = pd.get_dummies(atoms.numbers)
- d = dummies.T.reindex(self.atom_types).fillna(0).T.as_matrix()
+ d = dummies.T.reindex(self.atom_types).fillna(0).T.to_numpy()
 
  # Number of pairs by 1 + number of element types data matrix.
  n_pairs = len(atoms) ** 2
@@ -272,9 +272,9 @@ def _concatenate_vec(self, atoms, vec_names):
  fingerprint_vector.dtype.type is not np.str_) or
  (arr.dtype.type is not np.str_ and
  fingerprint_vector.dtype.type is np.str_)):
-  raise AssertionError("Fingerprints should be float" +
-  " type, and returned separately" +
-  " from str types.")
+ raise AssertionError("Fingerprints should be float" +
+ " type, and returned separately" +
+ " from str types.")
  fingerprint_vector = np.concatenate((fingerprint_vector,
  arr))
 

catlearn/learning_curve/feature_selection.py

@@ -73,8 +73,8 @@ def feature_inspection(self, lower=0, upper=1, interval=10**2,
  if alpha_list is None:
  alpha_list = np.linspace(float(upper), float(lower), int(interval))
  for alpha in alpha_list:
- model = LassoCV(alphas=[alpha]).fit(X=self.train_features,
- y=self.train_targets)
+ model = LassoCV(alphas=[alpha], cv=3).fit(X=self.train_features,
+  y=self.train_targets)
  feat_vec.append(np.shape(np.nonzero(model.coef_))[1])
  alpha_vec.append(alpha)
  return feat_vec, alpha_vec, np.nonzero(model.coef_)

catlearn/optimize/mlneb.py

@@ -384,7 +384,7 @@ def run(self, fmax=0.05, unc_convergence=0.050, steps=500,
  prev_save_positions.append(i.get_positions())
 
  neb_opt.run(fmax=(fmax * 0.85), steps=1)
- neb_opt.steps = 0
+ neb_opt.nsteps = 0
 
  n_steps_performed += 1
  get_results_predicted_path(self)

catlearn/preprocess/clean_data.py

@@ -1,7 +1,7 @@
 """Functions to clean data."""
 import numpy as np
 from collections import defaultdict
-from sklearn.preprocessing import Imputer
+from sklearn.impute import SimpleImputer
 from scipy.stats import skew
 
 
@@ -147,7 +147,7 @@ def clean_infinite(train, test=None, targets=None, labels=None, mask=None,
 
  # Get the fraction of finite values in each column.
  if max_impute_fraction > 0:
- impute = Imputer(missing_values="NaN", strategy=strategy)
+ impute = SimpleImputer(strategy=strategy)
  impute_fraction = 1 - np.isfinite(train).mean(axis=0)
  to_impute = impute_fraction <= max_impute_fraction
  train[:, to_impute] = impute.fit_transform(train[:, to_impute])

catlearn/preprocess/feature_elimination.py

@@ -264,11 +264,11 @@ def _get_correlation(self, target, feature_matrix):
  d = feature_matrix[:, i:i + 1]
  if np.allclose(d, d[0]):
  corr.append(0.)
- elif self.correlation is 'pearson':
- corr.append(pearsonr(x=d, y=t)[0])
- elif self.correlation is 'spearman':
+ elif self.correlation != 'pearson':
+ corr.append(pearsonr(x=d.ravel(), y=t.ravel())[0])
+ elif self.correlation != 'spearman':
  corr.append(spearmanr(a=d, b=t)[0])
- elif self.correlation is 'kendall':
+ elif self.correlation != 'kendall':
  corr.append(kendalltau(x=d, y=t)[0])
 
  return corr, order

requirements.txt

@@ -10,7 +10,7 @@ kiwisolver>=1.0.1
 markupsafe>=1.0
 networkx>=2.1.0
 numpy>=1.14.3
-pandas>=0.23.0
+pandas>=0.24.0
 psutil>=5.4.3
 pyparsing>=2.2.0
 python-dateutil>=2.7.3

test/test_suite.py

@@ -77,20 +77,21 @@ def setup_suite(class_list):
  ])
 
  # Clean everything up.
- os.remove('vec_store.sqlite')
- os.remove('hierarchy.pickle')
- os.remove('test.sqlite')
- os.remove('cvsave.pickle')
- os.remove('cvsave.json')
- os.remove('gaWrite.json')
-
- os.remove('all_predicted_paths.traj')
- os.remove('evaluated_structures.traj')
- os.remove('final.traj')
- os.remove('final_optimized.traj')
- os.remove('initial.traj')
- os.remove('initial_optimized.traj')
- os.remove('results_neb.csv')
- os.remove('results_neb_interpolation.csv')
- os.remove('ML-NEB.traj')
- os.remove('warnings_and_errors.txt')
+ for fname in ['vec_store.sqlite',
+ 'hierarchy.pickle',
+ 'test.sqlite',
+ 'cvsave.pickle',
+ 'cvsave.json',
+ 'gaWrite.json',
+ 'all_predicted_paths.traj',
+ 'evaluated_structures.traj',
+ 'final.traj',
+ 'final_optimized.traj',
+ 'initial.traj',
+ 'initial_optimized.traj',
+ 'results_neb.csv',
+ 'results_neb_interpolation.csv',
+ 'ML-NEB.traj',
+ 'warnings_and_errors.txt']:
+ if fname in os.listdir():
+ os.remove(fname)

tutorials/09_bulk_fingerprints/Voronoi_FP.ipynb

@@ -29,10 +29,8 @@
  },
  {
  "cell_type": "code",
- "execution_count": 1,
- "metadata": {
- "collapsed": true
- },
+ "execution_count": null,
+ "metadata": {},
  "outputs": [],
  "source": [
  "# Import packages.\n",
@@ -52,17 +50,9 @@
  },
  {
  "cell_type": "code",
- "execution_count": 2,
+ "execution_count": null,
  "metadata": {},
- "outputs": [
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "pulled 10 molecules from db\n"
- ]
- }
- ],
+ "outputs": [],
  "source": [
  "# Connect the ase-db.\n",
  "db = ase.db.connect('../../data/cubic_perovskites.db')\n",
@@ -81,17 +71,9 @@
  },
  {
  "cell_type": "code",
- "execution_count": 3,
+ "execution_count": null,
  "metadata": {},
- "outputs": [
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "min: 5, mean: 5, max: 5 atoms size\n"
- ]
- }
- ],
+ "outputs": [],
  "source": [
  "# Check the size of the atomic strucures.\n",
  "size = []\n",
@@ -111,17 +93,9 @@
  },
  {
  "cell_type": "code",
- "execution_count": 4,
+ "execution_count": null,
  "metadata": {},
- "outputs": [
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "Generate Voronoi fingerprint of 10 structures\n"
- ]
- }
- ],
+ "outputs": [],
  "source": [
  "voro = VoronoiFingerprintGenerator(alist)\n",
  "data_frame = voro.generate()"
@@ -139,27 +113,11 @@
  },
  {
  "cell_type": "code",
- "execution_count": 5,
+ "execution_count": null,
  "metadata": {},
- "outputs": [
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "(10, 271)\n"
- ]
- },
- {
- "name": "stderr",
- "output_type": "stream",
- "text": [
- "/home/raul_desktop/anaconda2/envs/py36/lib/python3.6/site-packages/ipykernel_launcher.py:1: FutureWarning: Method .as_matrix will be removed in a future version. Use .values instead.\n",
- " \"\"\"Entry point for launching an IPython kernel.\n"
- ]
- }
- ],
+ "outputs": [],
  "source": [
- "matrix = data_frame.as_matrix()\n",
+ "matrix = data_frame.to_numpy()\n",
  "finite_numeric_data = clean_infinite(matrix)\n",
  "print(np.shape(finite_numeric_data['train']))"
  ]
@@ -173,17 +131,9 @@
  },
  {
  "cell_type": "code",
- "execution_count": 6,
+ "execution_count": null,
  "metadata": {},
- "outputs": [
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "(10, 215)\n"
- ]
- }
- ],
+ "outputs": [],
  "source": [
  "useful_data = clean_variance(finite_numeric_data['train'])\n",
  "print(np.shape(useful_data['train']))"
@@ -213,9 +163,9 @@
  "name": "python",
  "nbconvert_exporter": "python",
  "pygments_lexer": "ipython3",
- "version": "3.6.7"
+ "version": "3.7.4"
  }
  },
  "nbformat": 4,
- "nbformat_minor": 2
+ "nbformat_minor": 4
 }

tutorials/09_bulk_fingerprints/Voronoi_FP.py

@@ -83,7 +83,7 @@
 # In[5]:
 
 
-matrix = data_frame.as_matrix()
+matrix = data_frame.to_numpy()
 finite_numeric_data = clean_infinite(matrix)
 print(np.shape(finite_numeric_data['train']))