temp save (7cb3989c) · Commits · 钟慕尧 / deepchem

deepchem/data/datasets.py

+1 −1

Original line number	Diff line number	Diff line
		@@ -793,7 +793,7 @@ class DiskDataset(Dataset):
		X, y, w, ids = (dataset.X, dataset.y, dataset.w, dataset.ids)
		yield (X, y, w, ids)

		return DiskDataset.create_dataset(generator(), data_dir=merge_dir)
		return DiskDataset.create_dataset(generator(), data_dir=merge_dir, tasks=datasets[0].tasks)

		def subset(self, shard_nums, subset_dir=None):
		"""Creates a subset of the original dataset on disk."""

deepchem/hyper/gaussian_process.py

+34 −3

Original line number	Diff line number	Diff line
		@@ -206,7 +206,7 @@ class GaussianProcessHyperparamOpt(HyperparamOpt):
		metric,
		self.model_class,
		hyper_parameters=hyper_parameters)
		return valid_scores[self.model_class][metric[0].name]
		score = valid_scores[self.model_class][metric[0].name]
		else:
		model_dir = tempfile.mkdtemp()
		model = self.model_class(hyper_parameters, model_dir)
		@@ -214,7 +214,11 @@ class GaussianProcessHyperparamOpt(HyperparamOpt):
		model.save()
		evaluator = Evaluator(model, valid_dataset, output_transformers)
		multitask_scores = evaluator.compute_model_performance([metric])
		return multitask_scores[metric.name]
		score = multitask_scores[metric.name]
		if direction:
		return score
		else:
		return -score

		cov = pyGPGO_covfunc.matern32()
		gp = pyGPGO_surrogates_GaussianProcess.GaussianProcess(cov)
		@@ -223,7 +227,6 @@ class GaussianProcessHyperparamOpt(HyperparamOpt):
		gpgo.run(max_iter=max_iter)

		hp_opt, valid_performance_opt = gpgo.getResult()

		# Readout best hyper parameters
		i = 0
		for hp in hp_list_single:
		@@ -239,4 +242,32 @@ class GaussianProcessHyperparamOpt(HyperparamOpt):
		hyper_parameters[hp[0]] = map(int, hyper_parameters[hp[0]])
		i = i + hp[1]

		if isinstance(self.model_class, str) or isinstance(
		self.model_class, unicode):
		try:
		train_scores, valid_scores, _ = benchmark_classification(
		train_dataset,
		valid_dataset,
		valid_dataset, ['task_placeholder'] * n_tasks,
		output_transformers,
		n_features,
		metric,
		self.model_class,
		hyper_parameters=params_dict)
		except AssertionError:
		train_scores, valid_scores, _ = benchmark_regression(
		train_dataset,
		valid_dataset,
		valid_dataset, ['task_placeholder'] * n_tasks,
		output_transformers,
		n_features,
		metric,
		self.model_class,
		hyper_parameters=params_dict)
		score = valid_scores[self.model_class][metric[0].name]
		if not direction:
		score = -score
		if score > valid_performance_opt:
		return params_dict, score

		return hyper_parameters, valid_performance_opt

deepchem/models/multitask.py

+2 −0

Original line number	Diff line number	Diff line
		@@ -56,6 +56,8 @@ class SingletaskToMultitask(Model):
		def _to_singletask(dataset, task_dirs):
		"""Transforms a multitask dataset to a collection of singletask datasets."""
		tasks = dataset.get_task_names()
		print(tasks)
		print(task_dirs)
		assert len(tasks) == len(task_dirs)
		log("Splitting multitask dataset into singletask datasets", dataset.verbose)
		task_datasets = [

deepchem/molnet/load_function/qm9_datasets.py

+2 −2

Original line number	Diff line number	Diff line
		@@ -51,9 +51,9 @@ def load_qm9(featurizer='CoulombMatrix', split='random', reload=True):

		if featurizer in ['CoulombMatrix', 'BPSymmetryFunction', 'MP', 'Raw']:
		if featurizer == 'CoulombMatrix':
		featurizer = deepchem.feat.CoulombMatrix(26)
		featurizer = deepchem.feat.CoulombMatrix(29)
		elif featurizer == 'BPSymmetryFunction':
		featurizer = deepchem.feat.BPSymmetryFunction(26)
		featurizer = deepchem.feat.BPSymmetryFunction(29)
		elif featurizer == 'Raw':
		featurizer = deepchem.feat.RawFeaturizer()
		elif featurizer == 'MP':

deepchem/splits/splitters.py

+3 −3

Original line number	Diff line number	Diff line
		@@ -458,9 +458,9 @@ class SingletaskStratifiedSplitter(Splitter):
		sortidx = np.argsort(y_s)

		split_cd = 10
		train_cutoff = int(frac_train * split_cd)
		valid_cutoff = int(frac_valid * split_cd) + train_cutoff
		test_cutoff = int(frac_test * split_cd) + valid_cutoff
		train_cutoff = int(np.round(frac_train * split_cd))
		valid_cutoff = int(np.round(frac_valid * split_cd)) + train_cutoff
		test_cutoff = int(np.round(frac_test * split_cd)) + valid_cutoff

		train_idx = np.array([])
		valid_idx = np.array([])

Admin message