Benchmark (003ba2ae) · Commits · 钟慕尧 / deepchem

examples/Benchmark.py

+26 −23

Original line number	Diff line number	Diff line
		@@ -63,10 +63,10 @@ def benchmark_loading_datasets(base_dir_o, n_features = 1024,
		number of features, or length of binary fingerprints

		dataset_name, string, optional (default='all')
		choice of which dataset to use, 'all' means evaluating performance of all the datasets
		choice of which dataset to use, 'all' = computing all the datasets

		model string, optional (default='all')
		choice of which model to use, 'all' means running all models on the dataset
		choice of which model to use, 'all' = running all models on the dataset

		out_path, string, optional(default='/tmp')
		path of result file
		@@ -97,12 +97,14 @@ def benchmark_loading_datasets(base_dir_o, n_features = 1024,
		#running model
		tox21_train,tox21_valid = benchmark_train_and_valid(base_dir,train_dataset,
		valid_dataset,
		tasks_tox21,transformers_tox21,
		n_features, model, verbosity)
		tasks_tox21,
		transformers_tox21,
		n_features, model,
		verbosity)
		time_finish_running = time.time()
		#time_finish_running-time_finish_loading is the time(s) used for fitting and evaluating

		with open(out_path+'/results.csv','a') as f:
		with open(os.path.join(out_path,'results.csv'),'a') as f:
		f.write ('\n'+'tox21,train')
		for i in tox21_train:
		f.write(','+i+','+str(tox21_train[i])) #output train score
		@@ -139,7 +141,7 @@ def benchmark_loading_datasets(base_dir_o, n_features = 1024,
		n_features,model,verbosity)
		time_finish_running = time.time()

		with open(out_path+'/results.csv','a') as f:
		with open(os.path.join(out_path,'results.csv'),'a') as f:
		f.write ('\n'+'muv,train')
		for i in muv_train:
		f.write(','+i+','+str(muv_train[i]))
		@@ -179,11 +181,13 @@ def benchmark_loading_datasets(base_dir_o, n_features = 1024,
		#running model
		pcba_train,pcba_valid = benchmark_train_and_valid(base_dir,train_dataset,
		valid_dataset,
		tasks_pcba,transformers_pcba,
		n_features, model, verbosity)
		tasks_pcba,
		transformers_pcba,
		n_features, model,
		verbosity)
		time_finish_running = time.time()

		with open(out_path+'/results.csv','a') as f:
		with open(os.path.join(out_path,'results.csv'),'a') as f:
		f.write ('\n'+'pcba,train')
		for i in pcba_train:
		f.write(','+i+','+str(pcba_train[i]))
		@@ -226,7 +230,7 @@ def benchmark_loading_datasets(base_dir_o, n_features = 1024,
		n_features,model,verbosity)
		time_finish_running = time.time()

		with open(out_path+'/results.csv','a') as f:
		with open(os.path.join(out_path,'results.csv'),'a') as f:
		f.write ('\n'+'nci,train')
		for i in nci_train:
		f.write(','+i+','+str(nci_train[i]))
		@@ -271,7 +275,7 @@ def benchmark_train_and_valid(base_dir,train_dataset,valid_dataset,tasks,
		number of features, or length of binary fingerprints

		model, string, optional (default='all')
		choice of which model to use, 'all' means running all models on the dataset
		choice of which model to use, 'all' = running all models on the dataset

		"""
		train_scores = {}
		@@ -340,8 +344,7 @@ if __name__ == '__main__':
		verbosity = 'high'

		#Working folder initialization
		base_dir = "/tmp/benchmark_test_"+time.strftime(
		"%Y_%m_%d", time.localtime())
		base_dir = "/tmp/benchmark_test_"+time.strftime("%Y_%m_%d", time.localtime())
		if os.path.exists(base_dir):
		shutil.rmtree(base_dir)
		os.makedirs(base_dir)

examples/pcba/pcba_datasets.py

+0 −18

Original line number	Diff line number	Diff line
		@@ -91,22 +91,4 @@ def load_pcba(base_dir, reload=True, frac_train=.8):
		for transformer in transformers:
		transformer.transform(dataset)

		#print("About to perform train/valid/test split.")
		#num_train = int(frac_train * len(dataset))
		#X, y, w, ids = (dataset.X, dataset.y, dataset.w, dataset.ids)
		#num_tasks = 120
		#PCBA_tasks = PCBA_tasks[:num_tasks]
		#print("Using following tasks")
		#print(PCBA_tasks)
		#X_train, X_valid = X[:num_train], X[num_train:]
		#y_train, y_valid = y[:num_train, :num_tasks], y[num_train:, :num_tasks]
		#w_train, w_valid = w[:num_train, :num_tasks], w[num_train:, :num_tasks]
		#ids_train, ids_valid = ids[:num_train], ids[num_train:]

		#train_dataset = DiskDataset.from_numpy(train_dir, X_train, y_train,
		# w_train, ids_train, PCBA_tasks)
		#valid_dataset = DiskDataset.from_numpy(valid_dir, X_valid, y_valid,
		# w_valid, ids_valid, PCBA_tasks)


		return PCBA_tasks, dataset, transformers

Admin message