From bcb26536109e106f8a95c6b7a463f34a4164a79f Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ignacio=20Hern=C3=A1ndez=20Montilla?=
 <ignaciohrdz@gmail.com>
Date: Sun, 12 Mar 2023 22:23:52 +0100
Subject: [PATCH] Minor updates (and more Pexels images)

---
 Readme.md        |  2 +-
 prepare_data.py  | 10 +++++++---
 requirements.txt |  4 +++-
 3 files changed, 11 insertions(+), 5 deletions(-)

diff --git a/Readme.md b/Readme.md
index b9480f4..2f41345 100644
--- a/Readme.md
+++ b/Readme.md
@@ -1,4 +1,4 @@
-# Face parts detection with YOLOv8 👃
+# Face parts detection with YOLOv8 🎯
 
 ## Introduction
 
diff --git a/prepare_data.py b/prepare_data.py
index 88f96cb..aa13765 100644
--- a/prepare_data.py
+++ b/prepare_data.py
@@ -9,6 +9,7 @@
 from utils import *
 import shutil
 import glob
+from unidecode import unidecode
 
 import pandas as pd
 import os
@@ -159,16 +160,19 @@ def process_names(names, split, path_data, path_dest, skip):
         path_pexels_annotations = os.path.join(path_pexels_dataset, s, "annotations", "obj_train_data")
         path_pexels_images = os.path.join(path_pexels_dataset, s, "images")
         pexels_labels = os.listdir(path_pexels_annotations)
-        pexels_names.extend([os.path.splitext(l)[0] for l in pexels_labels])
 
         for l in pexels_labels:
+            # Cleaning the names to avoid files with weird accents and characters
+            file_name_cleaned = unidecode(os.path.splitext(l)[0])
+            pexels_names.append(file_name_cleaned)
+
             img_name = os.path.splitext(l)[0] + ".jpg"
             img_source = os.path.join(path_pexels_images, img_name)
-            img_dest = os.path.join(path_processed_images, img_name)
+            img_dest = os.path.join(path_processed_images, file_name_cleaned + ".jpg")
             shutil.copy(img_source, img_dest)
 
             label_source = os.path.join(path_pexels_annotations, l)
-            label_dest = os.path.join(path_processed_labels, l)
+            label_dest = os.path.join(path_processed_labels, file_name_cleaned + ".txt")
             shutil.copy(label_source, label_dest)
 
     # Separate the Helen dataset in training and validation
diff --git a/requirements.txt b/requirements.txt
index 36d30d4..c918d8d 100644
--- a/requirements.txt
+++ b/requirements.txt
@@ -57,6 +57,7 @@ setuptools==65.6.3
 six==1.16.0
 smmap==5.0.0
 stack-data==0.6.2
+supervision~=0.3.0
 tensorboard==2.11.2
 tensorboard-data-server==0.6.1
 tensorboard-plugin-wit==1.8.1
@@ -66,7 +67,8 @@ torchvision==0.14.1+cu116
 tqdm==4.64.1
 traitlets==5.8.1
 typing_extensions==4.4.0
-ultralytics==8.0.9
+unidecode~=1.3.6
+ultralytics~=8.0.53
 urllib3==1.26.14
 wcwidth==0.2.6
 Werkzeug==2.2.2