def evaluate_joined_models_seq(base_out_folder, base_in_folder, models_to_load,
cur_fold, epochs, s):
# set random seeds
set_seeds(s)
# load the projection testing and training entrez ids for this fold
training_entrez, testing_entrez = read_entrez_indexes(base_in_folder +
models_to_load[0] +
"/folds/fold_" +
str(cur_fold))
# for each projection
training_projections = [
base_in_folder + model_name + "/projections/fold_train_" +
str(cur_fold) + "_projection.csv" for model_name in models_to_load
]
training_datasets = merge_projections(training_projections,
training_entrez)
class_indexes = training_datasets.class_indexes
training_datasets = training_datasets.sample(frac=1).reset_index(drop=True)
training_datasets.class_indexes = class_indexes
internal_training_datasets = training_datasets.iloc[:int(training_datasets.
shape[0] * 0.7)]
internal_training_datasets.class_indexes = training_datasets.class_indexes
internal_testing_datasets = training_datasets.iloc[int(training_datasets.
shape[0] * 0.7):]
internal_testing_datasets.class_indexes = training_datasets.class_indexes
model = train_model.JoinedModel(epochs)
model.train(internal_training_datasets,
internal_testing_datasets,
lc_file=base_out_folder + "lcs/internal_lc_fold_" +
str(cur_fold))
ids, predictions, classes = model.evaluate(internal_testing_datasets)
auroc = metrics.roc_auc_score(classes, predictions, average="weighted")
return auroc
def run_joined_models_tree(base_out_folder, base_in_folder, models_to_load,
cur_fold, s):
create_dir_structure(base_out_folder)
# set random seeds
set_seeds(s)
# checks if result file already exists. if so, do not run the algorithm.
if cur_fold != -1:
f_to_check = base_out_folder + "/predictions/predictions_fold_" + str(
cur_fold)
if file_exists(f_to_check):
print("not running")
print("file exists", f_to_check)
return
# load the testing and training entrez ids for this fold
training_entrez, testing_entrez = read_entrez_indexes(
base_in_folder + models_to_load[0] + "/folds/fold_" +
str(cur_fold))
testing_predictions = [
base_in_folder + model_name + "/predictions/predictions_fold_" +
str(cur_fold) for model_name in models_to_load
]
testing_dataset = merge_predictions(testing_predictions,
models_to_load)
training_predictions = [
base_in_folder + model_name +
"/predictions_train/predictions_fold_" + str(cur_fold)
for model_name in models_to_load
]
training_dataset = merge_predictions(training_predictions,
models_to_load)
classes = pandas.read_csv(load_dataset.DatasetLoader.base_path +
"class_labels.csv",
sep=",",
header=0,
na_values=["?"])
classes.dropna(subset=["class_Brain.Alzheimer"], inplace=True)
class_indexes = classes.columns[1:]
classes.set_index("entrezId", inplace=True)
training_dataset = training_dataset.join(classes)
training_dataset.class_indexes = class_indexes
testing_dataset = testing_dataset.join(classes)
testing_dataset.class_indexes = class_indexes
model = train_model.BoostedTreeModel()
model.train(training_dataset, testing_dataset)
ids, predictions, classes = model.evaluate(testing_dataset)
CrossValidation.write_results(ids, predictions, classes, cur_fold,
base_out_folder + "/predictions/")
else:
if file_exists(base_out_folder + "/projections/_projection.csv"):
print("not running")
return
training_projections = [
base_in_folder + model_name +
"/projections/fold_train_-1_projection.csv"
for model_name in models_to_load
]
training_datasets = merge_projections(training_projections, None)
model = train_model.JoinedModel(epochs)
model.train(training_datasets,
training_datasets,
lc_file=base_out_folder + "lcs/lc_fold_" + str(cur_fold))
ids, projection, classes = model.get_projection(training_datasets)
cross_validation.save_projections(projection,
base_out_folder + "/projections/",
ids)
def run_joined_models(base_out_folder, base_in_folder, models_to_load,
cur_fold, epochs, s):
create_dir_structure(base_out_folder)
# set random seeds
set_seeds(s)
# load the projection testing and training entrez ids for this fold. If the results alread exist, do nothing.
if cur_fold != -1:
f_to_check = base_out_folder + "/predictions/predictions_fold_" + str(
cur_fold)
if file_exists(f_to_check):
print("not running")
print("file exists", f_to_check)
return
training_entrez, testing_entrez = read_entrez_indexes(
base_in_folder + models_to_load[0] + "/folds/fold_" +
str(cur_fold))
# gets the projection
training_projections = [
base_in_folder + model_name + "/projections/fold_train_" +
str(cur_fold) + "_projection.csv" for model_name in models_to_load
]
training_datasets = merge_projections(training_projections,
training_entrez)
testing_projections = [
base_in_folder + model_name + "/projections/fold_test_" +
str(cur_fold) + "_projection.csv" for model_name in models_to_load
]
testing_datasets = merge_projections(testing_projections,
testing_entrez)
# Trains the model and gets the predictions.
model = train_model.JoinedModel(epochs)
model.train(training_datasets,
testing_datasets,
lc_file=base_out_folder + "lcs/lc_fold_" + str(cur_fold))
ids, predictions, classes = model.evaluate(testing_datasets)
# Writes the results.
CrossValidation.write_results(ids, predictions, classes, cur_fold,
base_out_folder + "/predictions/")
# Runs the training algorithm to the whole dataset
else:
if file_exists(base_out_folder + "/projections/_projection.csv"):
print("not running")
return
training_projections = [
base_in_folder + model_name +
"/projections/fold_train_-1_projection.csv"
for model_name in models_to_load
]
training_datasets = merge_projections(training_projections, None)
model = train_model.JoinedModel(epochs)
model.train(training_datasets,
training_datasets,
lc_file=base_out_folder + "lcs/lc_fold_" + str(cur_fold))
ids, projection, classes = model.get_projection(training_datasets)
cross_validation.save_projections(projection,
base_out_folder + "/projections/",
ids)
for class_name in classes.columns:
ids, projection, classes = model.get_projection(
training_datasets, class_name)
cross_validation.save_projections(
projection, base_out_folder +
"/projections/class_fold_train_" + class_name + "_-1", ids)
ids, predictions, classes = model.evaluate(training_datasets)
CrossValidation.write_results(ids, predictions, classes, cur_fold,
base_out_folder + "/predictions/")