From 72b15c371d657c3f1f98ec5bfe74493daea6fc3e Mon Sep 17 00:00:00 2001
From: yunfei-shi <30384335+yunfei-shi@users.noreply.github.com>
Date: Thu, 5 Sep 2019 16:29:35 +0800
Subject: [PATCH] Create to_upper.py

create new capitalized paths to a new folder data_lists_upper.
---
 to_upper.py | 24 ++++++++++++++++++++++++
 1 file changed, 24 insertions(+)
 create mode 100644 to_upper.py

diff --git a/to_upper.py b/to_upper.py
new file mode 100644
index 00000000..8aced288
--- /dev/null
+++ b/to_upper.py
@@ -0,0 +1,24 @@
+import os
+import pandas as pd
+import numpy as np
+
+if not os.path.exists('data_lists_upper'):
+    os.mkdir('data_lists_upper')
+
+all_data = pd.read_csv('data_lists/TIMIT_all.scp', header=None, names=['path'])
+all_data['path'] = all_data.path.str.upper()
+all_data.to_csv('data_lists_upper/TIMIT_all.scp', header=False, index=False)
+
+train_data = pd.read_csv('data_lists/TIMIT_train.scp', header=None, names=['path'])
+train_data['path'] = train_data.path.str.upper()
+train_data.to_csv('data_lists_upper/TIMIT_train.scp', header=False, index=False)
+
+test_data = pd.read_csv('data_lists/TIMIT_test.scp', header=None, names=['path'])
+test_data['path'] = test_data.path.str.upper()
+test_data.to_csv('data_lists_upper/TIMIT_test.scp', header=False, index=False)
+
+labels = np.load('data_lists/TIMIT_labels.npy', allow_pickle=True).item()
+labels = pd.DataFrame(list(labels.items()), columns=['path', 'label'])
+labels['path'] = labels.path.str.upper()
+label_dict = dict(labels.values.tolist())
+np.save('data_lists_upper/TIMIT_labels.npy', label_dict)