add in to molnet (6be5b3a5) · Commits · 钟慕尧 / deepchem

deepchem/molnet/preset_hyper_parameters.py

+8 −0

+3 −2

Original line number	Diff line number	Diff line
		@@ -37,7 +37,7 @@ def run_benchmark(datasets,
		model: string or user-defined model stucture
		choice of which model to use, deepchem provides implementation of
		logistic regression, random forest, multitask network,
		bypass multitask network, irv, graph convolution;
		bypass multitask network, irv, graph convolution, deep tensor NN(only for QM);
		for user define model, it should include function: fit, evaluate
		split: string, optional (default=None)
		choice of splitter function, None = using the default splitter
		@@ -112,7 +112,8 @@ def run_benchmark(datasets,
		return
		if split in ['scaffold', 'butina']:
		return

		if model == 'DTNN' and not dataset in ['qm7', 'qm7b', 'qm9']:
		return
		if not split in [
		None, 'index', 'random', 'scaffold', 'butina', 'stratified'
		]:

+29 −2

Original line number	Diff line number	Diff line
		@@ -363,7 +363,7 @@ def benchmark_regression(
		n_eval=10,
		seed=seed)

		if model_name == 'graphconvreg':
		elif model_name == 'graphconvreg':
		# Initialize model folder

		# Loading hyper parameters
		@@ -400,7 +400,34 @@ def benchmark_regression(
		beta1=.9,
		beta2=.999)

		if model_name == 'rf_regression':
		elif model_name == 'DTNN':
		# Initialize model folder

		# Loading hyper parameters
		batch_size = hyper_parameters['batch_size']
		nb_epoch = hyper_parameters['nb_epoch']
		learning_rate = hyper_parameters['learning_rate']
		n_distance = hyper_parameters['n_distance']
		n_hidden = hyper_parameters['n_hidden']

		tf.set_random_seed(seed)
		graph_model = dc.nn.SequentialDTNNGraph(max_n_atoms=n_features[0],
		n_distance=n_distance)
		graph_model.add(dc.nn.DTNNEmbedding(n_features=n_hidden))
		graph_model.add(dc.nn.DTNNStep(n_features=n_hidden, n_distance=n_distance))
		graph_model.add(dc.nn.DTNNStep(n_features=n_hidden, n_distance=n_distance))
		graph_model.add(dc.nn.DTNNGather(n_tasks=len(tasks)))

		model = dc.models.DTNNRegressor(
		graph_model,
		n_tasks=len(tasks),
		batch_size=batch_size,
		learning_rate=learning_rate,
		optimizer_type="adam",
		beta1=.9,
		beta2=.999)

		elif model_name == 'rf_regression':
		# Loading hyper parameters
		n_estimators = hyper_parameters['n_estimators']
		nb_epoch = None