def train_for_classification():
for each in list_vals:
classification_df = pd.read_csv(input_dict[each])
classification_df['u_id'] = classification_df.index
# classification_df = process_df(classification_df)
classification_df.rename(columns={
"STATUS": "desc",
each: "label"
},
inplace=True)
number_of_classes = len(list(classification_df['label'].unique()))
model_directory = os.path.join(root_dir, "classify_dict_" + each)
metrics_json = os.path.join(root_dir,
"accuracy_metrics_" + each + ".json")
training_loader, testing_loader = load_datasets(
classification_df,
train_size=0.8,
number_of_classes=number_of_classes)
unique_ids, val_targets, val_outputs = start_epochs(
training_loader,
testing_loader,
metrics_json,
model_directory,
epochs=5,
number_of_classes=number_of_classes)
out_numpy = np.concatenate((unique_ids.reshape(
-1, 1), val_targets.reshape(-1, 1), val_outputs.reshape(-1, 1)),
axis=1)
predicted_df = pd.DataFrame(out_numpy,
columns=['id', 'original', 'predicted'])
predicted_df.to_csv(os.path.join(root_dir,
"predicted_" + each + ".csv"),
index=False,
header=True)
def train_classification():
classification_df = pd.read_csv(final_data)
classification_df = setup_data(classification_df)
number_of_classes = max(list(classification_df['label'].unique()))+1
model_directory = os.path.join(root_dir, "classify_dict")
metrics_json = os.path.join(root_dir, "accuracy_metrics.json")
training_loader, testing_loader = load_datasets(classification_df, train_size=0.8,
number_of_classes=number_of_classes)
unique_ids, val_targets, val_outputs = start_epochs(training_loader, testing_loader, metrics_json, model_directory,
epochs=20, number_of_classes=number_of_classes)
out_numpy = np.concatenate((unique_ids.reshape(-1, 1), val_targets.reshape(-1, 1), val_outputs.reshape(-1, 1)),
axis=1)
predicted_df = pd.DataFrame(out_numpy, columns=['id', 'original', 'predicted'])
predicted_df.to_csv(os.path.join(root_dir, "predicted.csv"), index=False, header=True)