From 54c3694f81267e9e97dc9526f1893db1f7ff25e7 Mon Sep 17 00:00:00 2001 From: James Powell Date: Mon, 4 Jul 2022 02:00:16 -0400 Subject: [PATCH] ensure unit tests succeed; small fixes --- pattern/vector/__init__.py | 32 +++++++++---------- pattern/vector/svm/liblinear.py | 4 +-- pattern/vector/svm/liblinearutil.py | 49 +++++++++++++++-------------- pattern/vector/svm/libsvm.py | 4 +-- pattern/vector/svm/libsvmutil.py | 25 ++++++++------- 5 files changed, 58 insertions(+), 56 deletions(-) diff --git a/pattern/vector/__init__.py b/pattern/vector/__init__.py index 76a368ab..e0f8a1d6 100644 --- a/pattern/vector/__init__.py +++ b/pattern/vector/__init__.py @@ -280,9 +280,9 @@ def pop(self, i): stopwords = _stopwords = {} for f in glob.glob(os.path.join(MODULE, "stopwords-*.txt")): language = os.path.basename(f)[-6:-4] # stopwords-[en].txt - w = codecs.open(f, encoding="utf-8") - w = (w.strip() for w in w.read().split(",")) - stopwords[language] = dict.fromkeys(w, True) + with codecs.open(f, encoding="utf-8") as w: + w = (w.strip() for w in w.read().split(",")) + stopwords[language] = dict.fromkeys(w, True) # The following English words could also be meaningful nouns: @@ -514,7 +514,8 @@ def load(cls, path): The given text file must be generated with Document.save(). """ # Open unicode file. - s = open(path, "rb").read() + with open(path, "rb") as f: + s = f.read() s = s.lstrip(codecs.BOM_UTF8) s = decode_utf8(s) a = {} @@ -571,10 +572,9 @@ def save(self, path): s = "\n".join(s) s = encode_utf8(s) # Save unicode file. - f = open(path, "wb") - f.write(codecs.BOM_UTF8) - f.write(s) - f.close() + with open(path, "wb") as f: + f.write(codecs.BOM_UTF8) + f.write(s) def _get_model(self): return self._model @@ -1068,9 +1068,8 @@ def load(cls, path): # Deserialize Model.classifier. if model.classifier: p = path + ".tmp" - f = open(p, "wb") - f.write(model.classifier) - f.close() + with open(p, "wb") as f: + f.write(model.classifier) model._classifier = Classifier.load(p) os.remove(p) return model @@ -1095,7 +1094,8 @@ def save(self, path, update=False, final=False): if self._classifier: p = path + ".tmp" self._classifier.save(p, final) - self._classifier = open(p, "rb").read() + with open(p, "rb") as f: + self._classifier = f.read() os.remove(p) f = gzip.GzipFile(path, "wb") f.write(pickle.dumps(self, 1)) # 1 = binary @@ -1130,9 +1130,8 @@ def export(self, path, format=ORANGE, **kwargs): v = "%s,%s" % (v, document.type or "") s.append(v) s = "\n".join(s) - f = open(path, "w", encoding="utf-8") - f.write(decode_utf8(s)) - f.close() + with open(path, "w", encoding="utf-8") as f: + f.write(decode_utf8(s)) def _update(self): # Ensures that all document vectors are recalculated @@ -3633,7 +3632,8 @@ def save(self, path, final=False): # Unlink LIBSVM/LIBLINEAR binaries for cPickle. svm, model = self._svm, self._model self._svm = None - self._model = (open(path, "rb").read(),) + model[1:] + with open(path, "rb") as f: + self._model = (f.read(),) + model[1:] Classifier.save(self, path, final) self._svm = svm self._model = model diff --git a/pattern/vector/svm/liblinear.py b/pattern/vector/svm/liblinear.py index 00ee4d24..e9b0235c 100644 --- a/pattern/vector/svm/liblinear.py +++ b/pattern/vector/svm/liblinear.py @@ -29,9 +29,9 @@ try: dirname = path.dirname(path.abspath(__file__)) if sys.platform == 'win32': - liblinear = CDLL(path.join(dirname, 'windows\liblinear-2.20\liblinear.dll')) + liblinear = CDLL(path.join(dirname, r'windows\liblinear-2.20\liblinear.dll')) else: - liblinear = CDLL(path.join(dirname, 'macos/liblinear-2.20/liblinear.so.3')) + liblinear = CDLL(path.join(dirname, r'macos/liblinear-2.20/liblinear.so.3')) except: # For unix the prefix 'lib' is not considered. if find_library('linear'): diff --git a/pattern/vector/svm/liblinearutil.py b/pattern/vector/svm/liblinearutil.py index e705460b..0d5c2c40 100644 --- a/pattern/vector/svm/liblinearutil.py +++ b/pattern/vector/svm/liblinearutil.py @@ -33,30 +33,31 @@ def svm_read_problem(data_file_name, return_scipy=False): prob_x = [] row_ptr = [0] col_idx = [] - for i, line in enumerate(open(data_file_name)): - line = line.split(None, 1) - # In case an instance with all zero features - if len(line) == 1: - line += [''] - label, features = line - prob_y += [float(label)] - if scipy is not None and return_scipy: - nz = 0 - for e in features.split(): - ind, val = e.split(":") - val = float(val) - if val != 0: - col_idx += [int(ind) - 1] - prob_x += [val] - nz += 1 - row_ptr += [row_ptr[-1] + nz] - else: - xi = {} - for e in features.split(): - ind, val = e.split(":") - if val != 0: - xi[int(ind)] = float(val) - prob_x += [xi] + with open(data_file_name) as f: + for i, line in enumerate(f): + line = line.split(None, 1) + # In case an instance with all zero features + if len(line) == 1: + line += [''] + label, features = line + prob_y += [float(label)] + if scipy is not None and return_scipy: + nz = 0 + for e in features.split(): + ind, val = e.split(":") + val = float(val) + if val != 0: + col_idx += [int(ind) - 1] + prob_x += [val] + nz += 1 + row_ptr += [row_ptr[-1] + nz] + else: + xi = {} + for e in features.split(): + ind, val = e.split(":") + if val != 0: + xi[int(ind)] = float(val) + prob_x += [xi] if scipy is not None and return_scipy: prob_y = scipy.array(prob_y) prob_x = scipy.array(prob_x) diff --git a/pattern/vector/svm/libsvm.py b/pattern/vector/svm/libsvm.py index ac992d2b..50678aa5 100644 --- a/pattern/vector/svm/libsvm.py +++ b/pattern/vector/svm/libsvm.py @@ -22,9 +22,9 @@ try: dirname = path.dirname(path.abspath(__file__)) if sys.platform == 'win32': - libsvm = CDLL(path.join(dirname, 'windows\libsvm-3.22\libsvm.dll')) + libsvm = CDLL(path.join(dirname, r'windows\libsvm-3.22\libsvm.dll')) else: - libsvm = CDLL(path.join(dirname, 'macos/libsvm-3.22/libsvm.so.2')) + libsvm = CDLL(path.join(dirname, r'macos/libsvm-3.22/libsvm.so.2')) except: # For unix the prefix 'lib' is not considered. diff --git a/pattern/vector/svm/libsvmutil.py b/pattern/vector/svm/libsvmutil.py index 29110b00..1e3a3a5a 100644 --- a/pattern/vector/svm/libsvmutil.py +++ b/pattern/vector/svm/libsvmutil.py @@ -29,18 +29,19 @@ def svm_read_problem(data_file_name): """ prob_y = [] prob_x = [] - for line in open(data_file_name): - line = line.split(None, 1) - # In case an instance with all zero features - if len(line) == 1: - line += [''] - label, features = line - xi = {} - for e in features.split(): - ind, val = e.split(":") - xi[int(ind)] = float(val) - prob_y += [float(label)] - prob_x += [xi] + with open(data_file_name) as f: + for line in f: + line = line.split(None, 1) + # In case an instance with all zero features + if len(line) == 1: + line += [''] + label, features = line + xi = {} + for e in features.split(): + ind, val = e.split(":") + xi[int(ind)] = float(val) + prob_y += [float(label)] + prob_x += [xi] return (prob_y, prob_x)