Improved logic for generating weights for datasets (b02cb094) · Commits · 钟慕尧 / deepchem

deepchem/data/datasets.py

+23 −8

Original line number	Diff line number	Diff line
		@@ -317,16 +317,19 @@ class NumpyDataset(Dataset):
		if n_samples > 0:
		if y is None:
		# Set labels to be zero, with zero weights
		y = np.zeros((n_samples, n_tasks))
		w = np.zeros_like(y)
		y = np.zeros((n_samples, n_tasks), np.float32)
		w = np.zeros((n_samples, 1), np.float32)
		if ids is None:
		ids = np.arange(n_samples)
		if w is None:
		w = np.ones_like(y)
		if not isinstance(X, np.ndarray):
		X = np.array(X)
		if not isinstance(y, np.ndarray):
		y = np.array(y)
		if w is None:
		if len(y.shape) == 1:
		w = np.ones(y.shape[0], np.float32)
		else:
		w = np.ones((y.shape[0], 1), np.float32)
		if not isinstance(w, np.ndarray):
		w = np.array(w)
		self._X = X
		@@ -751,7 +754,10 @@ class DiskDataset(Dataset):
		if os.path.exists(w_filename):
		w = np.array(load_from_disk(w_filename))
		else:
		w = np.ones(y.shape)
		if len(y.shape) == 1:
		w = np.ones(y.shape[0], np.float32)
		else:
		w = np.ones((y.shape[0], 1), np.float32)
		else:
		w = None
		yield (X, y, w, ids)
		@@ -975,7 +981,10 @@ class DiskDataset(Dataset):

		if y is not None:
		if w is None:
		w = np.ones_like(y)
		if len(y.shape) == 1:
		w = np.ones(y.shape[0], np.float32)
		else:
		w = np.ones((y.shape[0], 1), np.float32)

		if tasks is None:
		if len(y.shape) > 1:
		@@ -1170,7 +1179,10 @@ class DiskDataset(Dataset):
		if os.path.exists(w_filename):
		w = np.array(load_from_disk(w_filename))
		else:
		w = np.ones(y.shape)
		if len(y.shape) == 1:
		w = np.ones(y.shape[0], np.float32)
		else:
		w = np.ones((y.shape[0], 1), np.float32)
		else:
		w = None

		@@ -1370,7 +1382,10 @@ class ImageDataset(Dataset):
		self._X_shape = self._find_array_shape(X)
		self._y_shape = self._find_array_shape(y)
		if w is None:
		w = np.ones(self._y_shape[:2])
		if len(self._y_shape) == 1:
		w = np.ones(self._y_shape[0], np.float32)
		else:
		w = np.ones((self._y_shape[0], 1), np.float32)
		if ids is None:
		if not isinstance(X, np.ndarray):
		ids = X

deepchem/data/tests/test_image_dataset.py

+1 −1

Original line number	Diff line number	Diff line
		@@ -45,7 +45,7 @@ class TestImageDataset(test_util.TensorFlowTestCase):
		x_shape, y_shape, w_shape, ids_shape = ds2.get_shape()
		np.testing.assert_array_equal([10], x_shape)
		np.testing.assert_array_equal([10, 28, 28], y_shape)
		np.testing.assert_array_equal([10, 28], w_shape)
		np.testing.assert_array_equal([10, 1], w_shape)
		np.testing.assert_array_equal([10], ids_shape)
		np.testing.assert_array_equal(ds2.X.shape, x_shape)
		np.testing.assert_array_equal(ds2.y.shape, y_shape)

deepchem/metrics/init.py

+2 −0

Original line number	Diff line number	Diff line
		@@ -304,6 +304,8 @@ class Metric(object):
		y_pred_task = y_pred[:, task]
		if len(w.shape) == 1:
		w_task = w
		elif w.shape[1] == 1:
		w_task = w[:, 0]
		else:
		w_task = w[:, task]

deepchem/models/multitask.py

+8 −2

Original line number	Diff line number	Diff line
		@@ -47,7 +47,8 @@ class SingletaskToMultitask(Model):
		task_data_dirs.append(task_data_dir)
		task_datasets = self._to_singletask(dataset, task_data_dirs)
		for task, task_dataset in zip(self.tasks, task_datasets):
		log("Dataset for task %s has shape %s" % (task,
		log(
		"Dataset for task %s has shape %s" % (task,
		str(task_dataset.get_shape())),
		self.verbose)
		return task_datasets
		@@ -68,6 +69,11 @@ class SingletaskToMultitask(Model):
		basename = "dataset-%d" % shard_num
		for task_num, task in enumerate(tasks):
		log("\tTask %s" % task, dataset.verbose)
		if len(w.shape) == 1:
		w_task = w
		elif w.shape[1] == 1:
		w_task = w[:, 0]
		else:
		w_task = w[:, task_num]
		y_task = y[:, task_num]

Admin message