Merge remote-tracking branch 'upstream/master'

.github/ISSUE_TEMPLATE/bug-report.md

@@ -0,0 +1,31 @@
+---
+name: Bug report
+about: Use this template if you believe you've encountered a bug.
+title: ''
+labels: bug
+assignees: K4YT3X
+
+---
+
+## Environment Information
+
+|Module|Version|
+|-|-|
+|`video2x`||
+|`ffmpeg`||
+|`waifu2x-caffe`||
+|`waifu2x-converter-cpp`||
+
+## Symptom
+
+Briefly describe what's going on.
+
+## Screenshots or Error Messages
+
+Please paste screenshots or error messages here.
+Include as much details as you can, as this greatly helps debugging the problem.
+
+Please paste your error message in the code block down below:
+
+```
+```

.github/ISSUE_TEMPLATE/feature_request.md

@@ -0,0 +1,16 @@
+---
+name: Feature request
+about: Request a new feature to be added
+title: ''
+labels: enhancement
+assignees: K4YT3X
+
+---
+
+## Description
+
+Briefly describe the feature you want to be added into `video2x`.
+
+## Sources
+
+Paste links to descriptions of related documentations, websites and etc. here.

README.md

@@ -2,41 +2,40 @@
 
 ### Official Discussion Group (Telegram): https://t.me/video2x
 
+## Download Builds (beta)
+
+You can go to the [releases page](https://github.com/k4yt3x/video2x/releases) to download the latest builds of `Video2X`. The exe files will require no Python or Python module installation.
+
+The **`full`** package provides all packages that will possibly be needed by `Video2X`, including `FFmpeg`, `waifu2x-caffe` and `waifu2x-converter-cpp`. The config file (`video2x.json`) is also already configured for the environment. All you need to do is just to launch `video2x.exe`.
+
+The **`light`** package provides only the most basic functions of `Video2X`. Only `video2x.exe`, `video2x_setup.exe` and `video2x.json` are included. To setup `FFmpeg`, `waifu2x-caffe` or `waifu2x-converter-cpp` automatically, simply launch `video2x_setup.exe`.
+
 ## Prerequisites
 
-Component names that are **bolded** are mandatory.
+Component names that are **bolded** can be automatically downloaded and configured with the `video2x_setup.py` script.
-
-Component names that are *italicized* can be automatically downloaded and configured with the `video2x_setup.py` script.
 
 1. Operating System: Windows
-1. AMD GPU / Nvidia GPU
+2. AMD GPU / Nvidia GPU
-1. AMD GPU driver / Nvidia GPU driver / Nvidia CUDNN
+3. AMD GPU driver / Nvidia GPU driver / Nvidia CUDNN
-1. [***FFMPEG***](https://ffmpeg.zeranoe.com/builds/)
+4. [**FFmpeg**](https://ffmpeg.zeranoe.com/builds/)
-1. [***waifu2x-caffe***](https://github.com/lltcggie/waifu2x-caffe/releases) / [**waifu2x-converter-cpp**](https://github.com/DeadSix27/waifu2x-converter-cpp/releases)
+5. [**waifu2x-caffe**](https://github.com/lltcggie/waifu2x-caffe/releases) / [**waifu2x-converter-cpp**](https://github.com/DeadSix27/waifu2x-converter-cpp/releases)
 
 ## Recent Changes
 
-### 2.6.3 (March 24, 2019)
+### 2.7.1 (April 18, 2019)
 
-- Added image cleaner by @BrianPetkovsek which removes upscaled frames.
+- Fixed video2x custom temp folder bug found by @cr08 .
-- Fixed some PEP8 issues.
-- Exceptions in waifu2x are now caught, and script will now stop on waifu2x error instead of keep going on to FFMPEG.
 
-### 2.6.2 (March 19, 2019)
+### 2.7.0 (March 30, 2019)
 
-- Removed `--model_dir` verification due to the rapidly evolving number of models added.
+- Added support for different extracted image formats.
-- Fixed model specifying bug. Users should now specify model using `--model_dir [path to folder containing model JSON files]`.
+- Redesigned FFmpeg wrapper, FFmpeg settings are now customizable in the `video2x.json` config file.
-- Enhanced command execution method.
+- Other minor enhancements and adjustments (e.g. argument -> method variable)
 
-### 2.6.1 (March 12, 2019)
+### Setup Script 1.2.0 (March 26, 2019)
 
-- Added `-b, --batch` option which selects applies all default values for questions automatically.
+- `video2x_setup.py` script can now automatically download and configure `waifu2x-converter-cpp`.
-- **This new version will now require `avalon_framework>=1.6.3`**. Please run `pip install -U avalon_framework` to update the existing framework.
+- replaced old progress indicator with progress bar.
-
-### 2.6.0 (March 9, 2019)
-
-- Complete redesign of configuration file format. The configuration file is now much more flexible and easy to look at.
-- Various modifications done to the rest of the program to adapt to the changes made in the configuration file. This eliminated some problems existed in the previous version.
 
 ## Description
 
@@ -60,17 +59,21 @@ Clip is from trailer of animated movie "千と千尋の神隠し". Copyright bel
 
 ## Documentations
 
-### [Video2X Wiki](https://github.com/K4YT3X/video2x/wiki)
+### [Video2X Wiki](https://github.com/k4yt3x/video2x/wiki)
 
-You can find all detailed user-facing and developer-facing documentations in the [Video2X Wiki](https://github.com/K4YT3X/video2x/wiki). It covers everything from step-by-step instructions for beginners, to the code structure of this program for advanced users and developers. If this README page doesn't answer all your questions, the wiki page is where you should head to.
+You can find all detailed user-facing and developer-facing documentations in the [Video2X Wiki](https://github.com/k4yt3x/video2x/wiki). It covers everything from step-by-step instructions for beginners, to the code structure of this program for advanced users and developers. If this README page doesn't answer all your questions, the wiki page is where you should head to.
 
-### [Step-By-Step Tutorial](https://github.com/K4YT3X/video2x/wiki/Step-By-Step-Tutorial) (Nvidia GPUs)
+### [Step-By-Step Tutorial](https://github.com/k4yt3x/video2x/wiki/Step-By-Step-Tutorial)
 
-For those who want a detailed walk-through of how to use `Video2X`, you can head to the [Step-By-Step Tutorial](https://github.com/K4YT3X/video2x/wiki/Step-By-Step-Tutorial) wiki page. It includes almost every step you need to perform in order to enlarge your first video. This tutorial currently only includes instructions for Nvidia GPUs, since AMD GPUs (OpenCL) requires installation of `waifu2x-converter-cpp` which cannot be installed automatically with Python at the moment due to its 7z compression format.
+For those who want a detailed walk-through of how to use `Video2X`, you can head to the [Step-By-Step Tutorial](https://github.com/k4yt3x/video2x/wiki/Step-By-Step-Tutorial) wiki page. It includes almost every step you need to perform in order to enlarge your first video.
 
-### [Waifu2X Drivers](https://github.com/K4YT3X/video2x/wiki/Waifu2X-Drivers)
+### [Waifu2X Drivers](https://github.com/k4yt3x/video2x/wiki/Waifu2X-Drivers)
 
-Go to the [Waifu2X Drivers](https://github.com/K4YT3X/video2x/wiki/Waifu2X-Drivers) wiki page if you want to see a detailed description on the different types of `waifu2x` drivers implemented by `Video2X`. This wiki page contains detailed difference between different drivers, and how to download and set each of them up for `Video2X`.
+Go to the [Waifu2X Drivers](https://github.com/k4yt3x/video2x/wiki/Waifu2X-Drivers) wiki page if you want to see a detailed description on the different types of `waifu2x` drivers implemented by `Video2X`. This wiki page contains detailed difference between different drivers, and how to download and set each of them up for `Video2X`.
+
+### [Q&A](https://github.com/k4yt3x/video2x/wiki/Q&A)
+
+If you have any questions, first try visiting our [Q&A](https://github.com/k4yt3x/video2x/wiki/Q&A) page to see if your question is answered there. If not, open an issue and we will respond to your questions ASAP.
 
 ---
 
@@ -80,7 +83,7 @@ Go to the [Waifu2X Drivers](https://github.com/K4YT3X/video2x/wiki/Waifu2X-Drive
 
 - **Python 3**  
 Download: https://www.python.org/downloads/windows/
-- **FFMPEG Windows Build**  
+- **FFmpeg Windows Build**  
 Download: https://ffmpeg.org/download.html  
 - **waifu2x-caffe** (for Nvidia CUDA/CUDNN)  
 Download: https://github.com/lltcggie/waifu2x-caffe/releases
@@ -91,27 +94,25 @@ Download: https://github.com/DeadSix27/waifu2x-converter-cpp/releases
 
 First, clone the video2x repository.
 
-```bash
+```shell
-$ git clone https://github.com/K4YT3X/video2x.git
+git clone https://github.com/k4yt3x/video2x.git
-$ cd video2x/bin
+cd video2x/bin
 ```
 
-Then you may run the `video2x_setup.py` script to install and configure the depencies automatically. This script is designed and tested on Windows 10.
+Then you may run the `video2x_setup.py` script to install and configure the dependencies automatically. This script is designed and tested on Windows 10.
 
-This script will install `ffmpeg`, `waifu2x-caffe` to `%LOCALAPPDATA%\\video2x` and all python libraries.
+This script will install the newest version of `ffmpeg`, either or both of `waifu2x-caffe` and `waifu2x-converter-cpp` to `%LOCALAPPDATA%\\video2x` and all required python libraries.
 
-**`waifu2x-converter-cpp` cannot be installed automatically with this script.** Please follow the [`waifu2x-converter-cpp` installation instructions](https://github.com/K4YT3X/video2x/wiki/Waifu2X-Drivers#waifu2x-converter-cpp) to install it.
+```shell
-
+python video2x_setup.py
-```bash
-$ python video2x_setup.py
 ```
 
 Alternatively, you can also install the dependencies manually. Please refer to the prerequisites section to see what's needed.
 
 Then you'll need to install python dependencies before start using video2x. Install simply by executing the following command.
 
-```bash
+```shell
-$ pip install -r requirements.txt
+pip install -r requirements.txt
 ```
 
 **Note that all command line arguments/options overwrite configuration file settings.**
@@ -131,34 +132,32 @@ Clip is from anime "さくら荘のペットな彼女". Copyright belongs to "
 
 Enlarge the video to 1920x1080 using CUDA. You may also use the `-r/--ratio` option.
 
-```bash
+```shell
-$ python video2x.py -i sample_input.mp4 -o sample_output.mp4 -m gpu --width=1920 --height=1080
+python video2x.py -i sample_input.mp4 -o sample_output.mp4 -m gpu --width=1920 --height=1080
 ```
 
 ### Nvidia CNDNN
 
 Enlarge the video to 1920x1080 using CUDNN. You may also use the `-r/--ratio` option.
 
-```bash
+```shell
-$ python video2x.py -i sample_input.mp4 -o sample_output.mp4 -m cudnn --width=1920 --height=1080
+python video2x.py -i sample_input.mp4 -o sample_output.mp4 -m cudnn --width=1920 --height=1080
 ```
 
 ### AMD or Nvidia (waifu2x-converter-cpp OpenCL)
 
 Enlarge the video by 2 times using OpenCL. Note that `waifu2x-converter-cpp` doesn't support width and height. You'll also have to explicitly specify that the driver to be used is `waifu2x_converter`.
 
-```bash
+```shell
-$ python video2x.py -i sample_input.mp4 -o sample_output.mp4 -m gpu -r 2 -d waifu2x_converter
+python video2x.py -i sample_input.mp4 -o sample_output.mp4 -m gpu -r 2 -d waifu2x_converter
 ```
 
-Corresponding sample configuration is shown below. Note that the `waifu2x_path` is different from the configuration for `waifu2x-caffe`. Instead of the binary path, folder containing extracted `waifu2x-converter-cpp.exe` should be specified.
-
 ### CPU
 
 Enlarge the video to 1920x1080 using the CPU. You may also use the `-r/--ratio` option. This is potentially much slower than using a GPU. The configuration file for this method is similar to the previous methods.
 
-```bash
+```shell
-$ python video2x.py -i sample_input.mp4 -o sample_output.mp4 -m cpu --width=1920 --height=1080
+python video2x.py -i sample_input.mp4 -o sample_output.mp4 -m cpu --width=1920 --height=1080
 ```
 
 ---
@@ -187,7 +186,7 @@ $ python video2x.py -i sample_input.mp4 -o sample_output.mp4 -m cpu --width=1920
     Waifu2x driver (default: waifu2x_caffe)
 
 ### -y MODEL_DIR, --model_dir MODEL_DIR
-    Folder containing model JSON files (default: None)
+    Folder containing model JSON files
 
 ### -t THREADS, --threads THREADS
     Number of threads to use for upscaling (default: 5)
@@ -196,18 +195,18 @@ $ python video2x.py -i sample_input.mp4 -o sample_output.mp4 -m cpu --width=1920
     Video2X config file location (default: video2x\bin\video2x.json)
 
 ### -b, --batch
-    Enable batch mode (select all default values to questions) (default: False)
+    Enable batch mode (select all default values to questions)
 
 ## Scaling Options
 
 ### --width WIDTH
-    Output video width (default: False)
+    Output video width
 
 ### --height HEIGHT
-    Output video height (default: False)
+    Output video height
 
 ### -r RATIO, --ratio RATIO
-    Scaling ratio (default: False)
+    Scaling ratio
 
 ---
 
@@ -224,10 +223,16 @@ https://www.gnu.org/licenses/gpl-3.0.txt
 
 This project relies on the following software and projects.
 
-- [FFMPEG]('https://www.ffmpeg.org/')
+- [FFmpeg]('https://www.ffmpeg.org/')
 - [waifu2x-caffe](https://github.com/lltcggie/waifu2x-caffe)
 - [waifu2x-converter-cpp](https://github.com/DeadSix27/waifu2x-converter-cpp)
 
+## Special Thanks
+
+Appreciations given to the following contributors:
+
+- @BrianPetkovsek
+
 ## Related Resources
 
 - [Dandere2x](https://github.com/CardinalPanda/dandere2x): `Dandere2x` is a lossy video upscaler also built around `waifu2x`, but with video compression techniques to shorten the time needed to process a video.

bin/ffmpeg.py

@@ -4,7 +4,7 @@
 Name: FFMPEG Class
 Author: K4YT3X
 Date Created: Feb 24, 2018
-Last Modified: March 19, 2019
+Last Modified: June 5, 2019
 
 Description: This class handles all FFMPEG related
 operations.
@@ -22,22 +22,18 @@ class Ffmpeg:
     and inserting audio tracks to videos.
     """
 
-    def __init__(self, ffmpeg_settings):
+    def __init__(self, ffmpeg_settings, image_format):
         self.ffmpeg_settings = ffmpeg_settings
-        self._parse_settings()
 
-    def _parse_settings(self):
-        """ Parse ffmpeg settings
-        """
         self.ffmpeg_path = self.ffmpeg_settings['ffmpeg_path']
         # add a forward slash to directory if not present
         # otherwise there will be a format error
         if self.ffmpeg_path[-1] != '/' and self.ffmpeg_path[-1] != '\\':
-            self.ffmpeg_path = '{}\\'.format(self.ffmpeg_path)
+            self.ffmpeg_path = f'{self.ffmpeg_path}\\'
 
-        self.ffmpeg_binary = '{}ffmpeg.exe'.format(self.ffmpeg_path)
+        self.ffmpeg_binary = f'{self.ffmpeg_path}ffmpeg.exe'
-        self.ffmpeg_hwaccel = self.ffmpeg_settings['ffmpeg_hwaccel']
+        self.ffmpeg_probe_binary = f'{self.ffmpeg_path}ffprobe.exe'
-        self.extra_arguments = self.ffmpeg_settings['extra_arguments']
+        self.image_format = image_format
 
     def get_video_info(self, input_video):
         """ Gets input video information
@@ -51,8 +47,11 @@ class Ffmpeg:
         Returns:
             dictionary -- JSON text of input video information
         """
+
+        # this execution command needs to be hard-coded
+        # since video2x only strictly recignizes this one format
         execute = [
-            '{}ffprobe.exe'.format(self.ffmpeg_path),
+            self.ffmpeg_probe_binary,
             '-v',
             'quiet',
             '-print_format',
@@ -60,10 +59,10 @@ class Ffmpeg:
             '-show_format',
             '-show_streams',
             '-i',
-            '{}'.format(input_video)
+            input_video
         ]
 
-        Avalon.debug_info('Executing: {}'.format(' '.join(execute)))
+        Avalon.debug_info(f'Executing: {" ".join(execute)}')
         json_str = subprocess.run(execute, check=True, stdout=subprocess.PIPE).stdout
         return json.loads(json_str.decode('utf-8'))
 
@@ -75,19 +74,26 @@ class Ffmpeg:
 
         Arguments:
             input_video {string} -- input video path
-            extracted_frames {string} -- video output folder
+            extracted_frames {string} -- video output directory
         """
         execute = [
-            self.ffmpeg_binary,
+            self.ffmpeg_binary
-            '-i',
-            '{}'.format(input_video),
-            '{}\\extracted_%0d.png'.format(extracted_frames),
-            '-y'
         ]
-        execute += self.extra_arguments
 
-        Avalon.debug_info('Executing: {}'.format(' '.join(execute)))
+        execute.extend([
-        subprocess.run(execute, shell=True, check=True)
+            '-i',
+            input_video
+        ])
+
+        execute.extend(self._read_configuration(phase='video_to_frames', section='output_options'))
+
+        execute.extend([
+            f'{extracted_frames}\\extracted_%0d.{self.image_format}'
+        ])
+
+        execute.extend(self._read_configuration(phase='video_to_frames'))
+
+        self._execute(execute)
 
     def convert_video(self, framerate, resolution, upscaled_frames):
         """Converts images into videos
@@ -98,31 +104,37 @@ class Ffmpeg:
         Arguments:
             framerate {float} -- target video framerate
             resolution {string} -- target video resolution
-            upscaled_frames {string} -- source images folder
+            upscaled_frames {string} -- source images directory
         """
         execute = [
             self.ffmpeg_binary,
             '-r',
             str(framerate),
-            '-f',
-            'image2',
             '-s',
-            resolution,
+            resolution
-            '-i',
-            '{}\\extracted_%d.png'.format(upscaled_frames),
-            '-vcodec',
-            'libx264',
-            '-crf',
-            '25',
-            '-pix_fmt',
-            'yuv420p',
-            '{}\\no_audio.mp4'.format(upscaled_frames),
-            '-y'
         ]
-        execute += self.extra_arguments
 
-        Avalon.debug_info('Executing: {}'.format(' '.join(execute)))
+        # read FFmpeg input options
-        subprocess.run(execute, shell=True, check=True)
+        execute.extend(self._read_configuration(phase='frames_to_video', section='input_options'))
+
+        # append input frames path into command
+        execute.extend([
+            '-i',
+            f'{upscaled_frames}\\extracted_%d.{self.image_format}'
+        ])
+
+        # read FFmpeg output options
+        execute.extend(self._read_configuration(phase='frames_to_video', section='output_options'))
+
+        # read other options
+        execute.extend(self._read_configuration(phase='frames_to_video'))
+
+        # specify output file location
+        execute.extend([
+            f'{upscaled_frames}\\no_audio.mp4'
+        ])
+
+        self._execute(execute)
 
     def migrate_audio_tracks_subtitles(self, input_video, output_video, upscaled_frames):
         """ Migrates audio tracks and subtitles from input video to output video
@@ -135,21 +147,71 @@ class Ffmpeg:
         execute = [
             self.ffmpeg_binary,
             '-i',
-            '{}\\no_audio.mp4'.format(upscaled_frames),
+            f'{upscaled_frames}\\no_audio.mp4',
             '-i',
-            '{}'.format(input_video),
+            input_video
-            '-map',
-            '0:v:0?',
-            '-map',
-            '1?',
-            '-c',
-            'copy',
-            '-map',
-            '-1:v?',
-            '{}'.format(output_video),
-            '-y'
         ]
-        execute += self.extra_arguments
 
-        Avalon.debug_info('Executing: {}'.format(' '.join(execute)))
+        execute.extend(self._read_configuration(phase='frames_to_video', section='output_options'))
-        subprocess.run(execute, shell=True, check=True)
+
+        execute.extend([
+            output_video
+        ])
+
+        execute.extend(self._read_configuration(phase='migrating_tracks'))
+
+        self._execute(execute)
+
+    def _read_configuration(self, phase, section=None):
+        """ read configuration from JSON
+
+        Read the configurations (arguments) from the JSON
+        configuration file and append them to the end of the
+        FFmpeg command.
+
+        Arguments:
+            execute {list} -- list of arguments to be executed
+            phase {str} -- phase of operation
+        """
+
+        configuration = []
+
+        # if section is specified, read configurations or keys
+        # from only that section
+        if section:
+            source = self.ffmpeg_settings[phase][section].keys()
+        else:
+            source = self.ffmpeg_settings[phase].keys()
+
+        for key in source:
+
+            if section:
+                value = self.ffmpeg_settings[phase][section][key]
+            else:
+                value = self.ffmpeg_settings[phase][key]
+
+            # null or None means that leave this option out (keep default)
+            if value is None or value is False or isinstance(value, list) or isinstance(value, dict):
+                continue
+            else:
+                configuration.append(key)
+
+                # true means key is an option
+                if value is True:
+                    continue
+
+                configuration.append(str(value))
+
+        return configuration
+
+    def _execute(self, execute):
+        """ execute command
+
+        Arguments:
+            execute {list} -- list of arguments to be executed
+
+        Returns:
+            int -- execution return code
+        """
+        Avalon.debug_info(f'Executing: {execute}')
+        return subprocess.run(execute, shell=True, check=True).returncode

bin/image_cleaner.py

@@ -5,7 +5,7 @@ Name: Video2X Image Cleaner
 Author: BrianPetkovsek
 Author: K4YT3X
 Date Created: March 24, 2019
-Last Modified: March 24, 2019
+Last Modified: April 28, 2019
 
 Description: This class is to remove the extracted frames
 that have already been upscaled.
@@ -27,11 +27,11 @@ class ImageCleaner(threading.Thread):
         threading.Thread
     """
 
-    def __init__(self, input_folder, output_folder, num_threads):
+    def __init__(self, input_directory, output_directory, threads):
         threading.Thread.__init__(self)
-        self.input_folder = input_folder
+        self.input_directory = input_directory
-        self.output_folder = output_folder
+        self.output_directory = output_directory
-        self.num_threads = num_threads
+        self.threads = threads
         self.running = False
 
     def run(self):
@@ -53,23 +53,23 @@ class ImageCleaner(threading.Thread):
         """ remove frames that have already been upscaled
 
         This method compares the files in the extracted frames
-        folder with the upscaled frames folder, and removes
+        directory with the upscaled frames directory, and removes
         the frames that has already been upscaled.
         """
 
         # list all images in the extracted frames
-        output_frames = [f for f in os.listdir(self.output_folder) if os.path.isfile(os.path.join(self.output_folder, f))]
+        output_frames = [f for f in os.listdir(self.output_directory) if os.path.isfile(os.path.join(self.output_directory, f))]
 
         # compare and remove frames downscaled images that finished being upscaled
-        # within each thread's  extracted frames folder
+        # within each thread's  extracted frames directory
-        for i in range(self.num_threads):
+        for i in range(self.threads):
-            dir_path = os.path.join(self.input_folder, str(i))
+            dir_path = os.path.join(self.input_directory, str(i))
 
-            # for each file within all the folders
+            # for each file within all the directories
             for f in os.listdir(dir_path):
                 file_path = os.path.join(dir_path, f)
 
-                # if file also exists in the output folder, then the file
+                # if file also exists in the output directory, then the file
                 # has already been processed, thus not needed anymore
                 if os.path.isfile(file_path) and f in output_frames:
                     os.remove(file_path)

bin/upscaler.py

@@ -4,7 +4,7 @@
 Name: Video2X Upscaler
 Author: K4YT3X
 Date Created: December 10, 2018
-Last Modified: March 24, 2019
+Last Modified: April 28, 2019
 
 Licensed under the GNU General Public License Version 3 (GNU GPL v3),
     available at: https://www.gnu.org/licenses/gpl-3.0.txt
@@ -13,10 +13,10 @@ Licensed under the GNU General Public License Version 3 (GNU GPL v3),
 """
 
 from avalon_framework import Avalon
-from image_cleaner import ImageCleaner
 from exceptions import *
 from ffmpeg import Ffmpeg
 from fractions import Fraction
+from image_cleaner import ImageCleaner
 from tqdm import tqdm
 from waifu2x_caffe import Waifu2xCaffe
 from waifu2x_converter import Waifu2xConverter
@@ -30,52 +30,49 @@ import time
 
 class Upscaler:
     """ An instance of this class is a upscaler that will
-    upscale all images in the given folder.
+    upscale all images in the given directory.
 
     Raises:
         Exception -- all exceptions
         ArgumentError -- if argument is not valid
     """
 
-    def __init__(self, input_video, output_video, method, waifu2x_settings, ffmpeg_settings, waifu2x_driver='waifu2x_caffe', scale_width=False, scale_height=False, scale_ratio=False, model_dir=None, threads=5, video2x_cache_folder='{}\\video2x'.format(tempfile.gettempdir()), preserve_frames=False):
+    def __init__(self, input_video, output_video, method, waifu2x_settings, ffmpeg_settings):
         # mandatory arguments
         self.input_video = input_video
         self.output_video = output_video
         self.method = method
         self.waifu2x_settings = waifu2x_settings
         self.ffmpeg_settings = ffmpeg_settings
-        self.waifu2x_driver = waifu2x_driver
-
-        # check sanity of waifu2x_driver option
-        if waifu2x_driver != 'waifu2x_caffe' and waifu2x_driver != 'waifu2x_converter':
-            raise Exception('Unrecognized waifu2x driver: {}'.format(waifu2x_driver))
 
         # optional arguments
-        self.scale_width = scale_width
+        self.waifu2x_driver = 'waifu2x_caffe'
-        self.scale_height = scale_height
+        self.scale_width = None
-        self.scale_ratio = scale_ratio
+        self.scale_height = None
-        self.model_dir = model_dir
+        self.scale_ratio = None
-        self.threads = threads
+        self.model_dir = None
+        self.threads = 5
+        self.video2x_cache_directory = f'{tempfile.gettempdir()}\\video2x'
+        self.image_format = 'png'
+        self.preserve_frames = False
 
-        # create temporary folder/directories
+    def create_temp_directories(self):
-        self.video2x_cache_folder = video2x_cache_folder
+        """create temporary directory
-        self.extracted_frames = tempfile.mkdtemp(dir=self.video2x_cache_folder)
+        """
-        Avalon.debug_info('Extracted frames are being saved to: {}'.format(self.extracted_frames))
+        self.extracted_frames = tempfile.mkdtemp(dir=self.video2x_cache_directory)
+        Avalon.debug_info(f'Extracted frames are being saved to: {self.extracted_frames}')
+        self.upscaled_frames = tempfile.mkdtemp(dir=self.video2x_cache_directory)
+        Avalon.debug_info(f'Upscaled frames are being saved to: {self.upscaled_frames}')
 
-        self.upscaled_frames = tempfile.mkdtemp(dir=self.video2x_cache_folder)
+    def cleanup_temp_directories(self):
-        Avalon.debug_info('Upscaled frames are being saved to: {}'.format(self.upscaled_frames))
+        """delete temp directories when done
-
+        """
-        self.preserve_frames = preserve_frames
-
-    def cleanup(self):
-        # delete temp directories when done
-        # avalon framework cannot be used if python is shutting down
-        # therefore, plain print is used
         if not self.preserve_frames:
-
             for directory in [self.extracted_frames, self.upscaled_frames]:
                 try:
-                    print('Cleaning up cache directory: {}'.format())
+                    # avalon framework cannot be used if python is shutting down
+                    # therefore, plain print is used
+                    print(f'Cleaning up cache directory: {directory}')
                     shutil.rmtree(directory)
                 except (OSError, FileNotFoundError):
                     pass
@@ -92,18 +89,18 @@ class Upscaler:
         elif not self.method:
             raise ArgumentError('You need to specify the enlarging processing unit')
 
-    def _progress_bar(self, extracted_frames_folders):
+    def _progress_bar(self, extracted_frames_directories):
         """ This method prints a progress bar
 
         This method prints a progress bar by keeping track
-        of the amount of frames in the input directory/folder
+        of the amount of frames in the input directory
-        and the output directory/folder. This is originally
+        and the output directory. This is originally
         suggested by @ArmandBernard.
         """
         # get number of extracted frames
         total_frames = 0
-        for folder in extracted_frames_folders:
+        for directory in extracted_frames_directories:
-            total_frames += len([f for f in os.listdir(folder) if f[-4:] == '.png'])
+            total_frames += len([f for f in os.listdir(directory) if f[-4:] == f'.{self.image_format}'])
 
         with tqdm(total=total_frames, ascii=True, desc='Upscaling Progress') as progress_bar:
 
@@ -114,7 +111,7 @@ class Upscaler:
             while not self.progress_bar_exit_signal:
 
                 try:
-                    total_frames_upscaled = len([f for f in os.listdir(self.upscaled_frames) if f[-4:] == '.png'])
+                    total_frames_upscaled = len([f for f in os.listdir(self.upscaled_frames) if f[-4:] == f'.{self.image_format}'])
                     delta = total_frames_upscaled - previous_cycle_frames
                     previous_cycle_frames = total_frames_upscaled
 
@@ -142,6 +139,10 @@ class Upscaler:
         # progress bar thread exit signal
         self.progress_bar_exit_signal = False
 
+        # create a container for exceptions in threads
+        # if this thread is not empty, then an exception has occured
+        self.upscaler_exceptions = []
+
         # it's easier to do multi-threading with waifu2x_converter
         # the number of threads can be passed directly to waifu2x_converter
         if self.waifu2x_driver == 'waifu2x_converter':
@@ -149,10 +150,10 @@ class Upscaler:
             progress_bar = threading.Thread(target=self._progress_bar, args=([self.extracted_frames],))
             progress_bar.start()
 
-            w2.upscale(self.extracted_frames, self.upscaled_frames, self.scale_ratio, self.threads)
+            w2.upscale(self.extracted_frames, self.upscaled_frames, self.scale_ratio, self.threads, self.image_format, self.upscaler_exceptions)
             for image in [f for f in os.listdir(self.upscaled_frames) if os.path.isfile(os.path.join(self.upscaled_frames, f))]:
-                renamed = re.sub('_\[.*-.*\]\[x(\d+(\.\d+)?)\]\.png', '.png', image)
+                renamed = re.sub(f'_\[.*-.*\]\[x(\d+(\.\d+)?)\]\.{self.image_format}', f'.{self.image_format}', image)
-                shutil.move('{}\\{}'.format(self.upscaled_frames, image), '{}\\{}'.format(self.upscaled_frames, renamed))
+                shutil.move(f'{self.upscaled_frames}\\{image}', f'{self.upscaled_frames}\\{renamed}')
 
             self.progress_bar_exit_signal = True
             progress_bar.join()
@@ -161,10 +162,6 @@ class Upscaler:
         # create a container for all upscaler threads
         upscaler_threads = []
 
-        # create a container for exceptions in threads
-        # if this thread is not empty, then an exception has occured
-        self.threads_exceptions = []
-
         # list all images in the extracted frames
         frames = [os.path.join(self.extracted_frames, f) for f in os.listdir(self.extracted_frames) if os.path.isfile(os.path.join(self.extracted_frames, f))]
 
@@ -173,25 +170,25 @@ class Upscaler:
         if len(frames) < self.threads:
             self.threads = len(frames)
 
-        # create a folder for each thread and append folder
+        # create a directory for each thread and append directory
         # name into a list
 
         thread_pool = []
-        thread_folders = []
+        thread_directories = []
         for thread_id in range(self.threads):
-            thread_folder = '{}\\{}'.format(self.extracted_frames, str(thread_id))
+            thread_directory = f'{self.extracted_frames}\\{str(thread_id)}'
-            thread_folders.append(thread_folder)
+            thread_directories.append(thread_directory)
 
-            # delete old folders and create new folders
+            # delete old directories and create new directories
-            if os.path.isdir(thread_folder):
+            if os.path.isdir(thread_directory):
-                shutil.rmtree(thread_folder)
+                shutil.rmtree(thread_directory)
-            os.mkdir(thread_folder)
+            os.mkdir(thread_directory)
 
-            # append folder path into list
+            # append directory path into list
-            thread_pool.append((thread_folder, thread_id))
+            thread_pool.append((thread_directory, thread_id))
 
-        # evenly distribute images into each folder
+        # evenly distribute images into each directory
-        # until there is none left in the folder
+        # until there is none left in the directory
         for image in frames:
             # move image
             shutil.move(image, thread_pool[0][0])
@@ -202,16 +199,16 @@ class Upscaler:
         for thread_info in thread_pool:
             # create thread
             if self.scale_ratio:
-                thread = threading.Thread(target=w2.upscale, args=(thread_info[0], self.upscaled_frames, self.scale_ratio, False, False, self.threads_exceptions))
+                thread = threading.Thread(target=w2.upscale, args=(thread_info[0], self.upscaled_frames, self.scale_ratio, False, False, self.image_format, self.upscaler_exceptions))
             else:
-                thread = threading.Thread(target=w2.upscale, args=(thread_info[0], self.upscaled_frames, False, self.scale_width, self.scale_height, self.threads_exceptions))
+                thread = threading.Thread(target=w2.upscale, args=(thread_info[0], self.upscaled_frames, False, self.scale_width, self.scale_height, self.image_format, self.upscaler_exceptions))
             thread.name = thread_info[1]
 
             # add threads into the pool
             upscaler_threads.append(thread)
 
         # start progress bar in a different thread
-        progress_bar = threading.Thread(target=self._progress_bar, args=(thread_folders,))
+        progress_bar = threading.Thread(target=self._progress_bar, args=(thread_directories,))
         progress_bar.start()
 
         # create the clearer and start it
@@ -233,8 +230,8 @@ class Upscaler:
 
         self.progress_bar_exit_signal = True
 
-        if len(self.threads_exceptions) != 0:
+        if len(self.upscaler_exceptions) != 0:
-            raise(self.threads_exceptions[0])
+            raise(self.upscaler_exceptions[0])
 
     def run(self):
         """Main controller for Video2X
@@ -252,7 +249,7 @@ class Upscaler:
         self.output_video = os.path.abspath(self.output_video)
 
         # initialize objects for ffmpeg and waifu2x-caffe
-        fm = Ffmpeg(self.ffmpeg_settings)
+        fm = Ffmpeg(self.ffmpeg_settings, self.image_format)
 
         # initialize waifu2x driver
         if self.waifu2x_driver == 'waifu2x_caffe':
@@ -260,7 +257,7 @@ class Upscaler:
         elif self.waifu2x_driver == 'waifu2x_converter':
             w2 = Waifu2xConverter(self.waifu2x_settings, self.model_dir)
         else:
-            raise Exception('Unrecognized waifu2x driver: {}'.format(self.waifu2x_driver))
+            raise Exception(f'Unrecognized waifu2x driver: {self.waifu2x_driver}')
 
         # extract frames from video
         fm.extract_frames(self.input_video, self.extracted_frames)
@@ -284,14 +281,14 @@ class Upscaler:
 
         # get average frame rate of video stream
         framerate = float(Fraction(video_info['streams'][video_stream_index]['avg_frame_rate']))
-        Avalon.info('Framerate: {}'.format(framerate))
+        Avalon.info(f'Framerate: {framerate}')
 
         # width/height will be coded width/height x upscale factor
         if self.scale_ratio:
-            coded_width = video_info['streams'][video_stream_index]['coded_width']
+            original_width = video_info['streams'][video_stream_index]['width']
-            coded_height = video_info['streams'][video_stream_index]['coded_height']
+            original_height = video_info['streams'][video_stream_index]['height']
-            self.scale_width = self.scale_ratio * coded_width
+            self.scale_width = int(self.scale_ratio * original_width)
-            self.scale_height = self.scale_ratio * coded_height
+            self.scale_height = int(self.scale_ratio * original_height)
 
         # upscale images one by one using waifu2x
         Avalon.info('Starting to upscale extracted images')
@@ -302,7 +299,7 @@ class Upscaler:
         Avalon.info('Converting extracted frames into video')
 
         # use user defined output size
-        fm.convert_video(framerate, '{}x{}'.format(self.scale_width, self.scale_height), self.upscaled_frames)
+        fm.convert_video(framerate, f'{self.scale_width}x{self.scale_height}', self.upscaled_frames)
         Avalon.info('Conversion completed')
 
         # migrate audio tracks and subtitles

bin/video2x.json

@@ -1,6 +1,8 @@
 {
   "waifu2x_caffe": {
     "waifu2x_caffe_path": "C:\\Users\\K4YT3X\\AppData\\Local\\video2x\\waifu2x-caffe\\waifu2x-caffe-cui.exe",
+    "input_extention_list": null,
+    "output_extention": null,
     "mode": "noise_scale",
     "scale_ratio": null,
     "scale_width": null,
@@ -19,31 +21,61 @@
     "crop_w": null,
     "crop_h": null
   },
-
   "waifu2x_converter": {
     "waifu2x_converter_path": "C:\\Users\\K4YT3X\\AppData\\Local\\video2x\\waifu2x-converter-cpp",
-    "block_size": null,
+    "output-format": null,
+    "png-compression": null,
+    "image-quality": null,
+    "block-size": null,
     "disable-gpu": null,
     "force-OpenCL": null,
     "processor": null,
     "jobs": null,
-    "model_dir": null,
+    "model-dir": null,
-    "scale_ratio": null,
+    "scale-ratio": null,
-    "noise_level": 3,
+    "noise-level": 3,
-    "mode": "noise_scale",
+    "mode": "noise-scale",
-    "quiet": true,
+    "silent": true,
     "output": null,
     "input": null
   },
-
+  "ffmpeg": {
-  "ffmpeg":{
+    "ffmpeg_path": "C:\\Users\\K4YT3X\\AppData\\Local\\video2x\\ffmpeg-latest-win64-static\\bin",
-    "ffmpeg_path": "C:\\Users\\K4YT3X\\AppData\\Local\\video2x\\ffmpeg-4.1-win64-static\\bin",
+    "video_to_frames": {
-    "ffmpeg_hwaccel": "auto",
+        "output_options":{
-    "extra_arguments": []
+            "-qscale:v": null
+        },
+        "-hwaccel": "auto",
+        "-y": true
+    },
+    "frames_to_video": {
+        "input_options":{
+            "-qscale:v": null,
+            "-qscale:a": null,
+            "-f": "image2"
+        },
+        "output_options":{
+            "-vcodec": "libx264",
+            "-crf": 17,
+            "-b:v": null,
+            "-pix_fmt": "yuv420p"
+        },
+        "-y": true
+    },
+    "migrating_tracks": {
+        "output_options":{
+            "-map": "0:v:0?",
+            "-map": "1?",
+            "-c": "copy",
+            "-map": "-1:v?",
+            "-pix_fmt": "yuv420p"
+        },
+        "-y": true
+    }
   },
-
   "video2x": {
-    "video2x_cache_folder": false,
+    "video2x_cache_directory": null,
+    "image_format": "png",
     "preserve_frames": false
   }
 }

bin/video2x.py

@@ -13,7 +13,7 @@ __      __  _       _                  ___   __   __
 Name: Video2X Controller
 Author: K4YT3X
 Date Created: Feb 24, 2018
-Last Modified: March 24, 2019
+Last Modified: April 28, 2019
 
 Licensed under the GNU General Public License Version 3 (GNU GPL v3),
     available at: https://www.gnu.org/licenses/gpl-3.0.txt
@@ -45,12 +45,14 @@ import GPUtil
 import json
 import os
 import psutil
+import re
 import shutil
+import sys
 import tempfile
 import time
 import traceback
 
-VERSION = '2.6.3'
+VERSION = '2.7.1'
 
 # each thread might take up to 2.5 GB during initialization.
 # (system memory, not to be confused with GPU memory)
@@ -76,16 +78,16 @@ def process_arguments():
     upscaler_options = parser.add_argument_group('Upscaler Options')
     upscaler_options.add_argument('-m', '--method', help='Upscaling method', action='store', default='gpu', choices=['cpu', 'gpu', 'cudnn'], required=True)
     upscaler_options.add_argument('-d', '--driver', help='Waifu2x driver', action='store', default='waifu2x_caffe', choices=['waifu2x_caffe', 'waifu2x_converter'])
-    upscaler_options.add_argument('-y', '--model_dir', help='Folder containing model JSON files', action='store', default=None)
+    upscaler_options.add_argument('-y', '--model_dir', help='directory containing model JSON files', action='store')
-    upscaler_options.add_argument('-t', '--threads', help='Number of threads to use for upscaling', action='store', type=int, default=5)
+    upscaler_options.add_argument('-t', '--threads', help='Number of threads to use for upscaling', action='store', type=int, default=1)
-    upscaler_options.add_argument('-c', '--config', help='Video2X config file location', action='store', default='{}\\video2x.json'.format(os.path.dirname(os.path.abspath(__file__))))
+    upscaler_options.add_argument('-c', '--config', help='Video2X config file location', action='store', default=f'{os.path.dirname(os.path.abspath(sys.argv[0]))}\\video2x.json')
-    upscaler_options.add_argument('-b', '--batch', help='Enable batch mode (select all default values to questions)', action='store_true', default=False)
+    upscaler_options.add_argument('-b', '--batch', help='Enable batch mode (select all default values to questions)', action='store_true')
 
     # scaling options
     scaling_options = parser.add_argument_group('Scaling Options')
-    scaling_options.add_argument('--width', help='Output video width', action='store', type=int, default=False)
+    scaling_options.add_argument('--width', help='Output video width', action='store', type=int)
-    scaling_options.add_argument('--height', help='Output video height', action='store', type=int, default=False)
+    scaling_options.add_argument('--height', help='Output video height', action='store', type=int)
-    scaling_options.add_argument('-r', '--ratio', help='Scaling ratio', action='store', type=int, default=False)
+    scaling_options.add_argument('-r', '--ratio', help='Scaling ratio', action='store', type=float)
 
     # parse arguments
     return parser.parse_args()
@@ -99,8 +101,8 @@ def print_logo():
     print('   \\  /    | | | (_| | |  __/ | (_) |  / /_   / . \\')
     print('    \\/     |_|  \\__,_|  \\___|  \\___/  |____| /_/ \\_\\')
     print('\n               Video2X Video Enlarger')
-    spaces = ((44 - len("Version {}".format(VERSION))) // 2) * " "
+    spaces = ((44 - len(f'Version {VERSION}')) // 2) * ' '
-    print('{}\n{}    Version {}\n{}'.format(Avalon.FM.BD, spaces, VERSION, Avalon.FM.RST))
+    print(f'{Avalon.FM.BD}\n{spaces}    Version {VERSION}\n{Avalon.FM.RST}')
 
 
 def check_memory():
@@ -139,17 +141,17 @@ def check_memory():
 
         # if user doesn't even have enough memory to run even one thread
         if memory_available < mem_per_thread:
-            Avalon.warning('You might have insufficient amount of {} memory available to run this program ({} GB)'.format(memory_type, memory_available))
+            Avalon.warning(f'You might have insufficient amount of {memory_type} memory available to run this program ({memory_available} GB)')
             Avalon.warning('Proceed with caution')
             if args.threads > 1:
                 if Avalon.ask('Reduce number of threads to avoid crashing?', default=True, batch=args.batch):
                     args.threads = 1
         # if memory available is less than needed, warn the user
         elif memory_available < (mem_per_thread * args.threads):
-            Avalon.warning('Each waifu2x-caffe thread will require up to 2.5 GB of system memory')
+            Avalon.warning(f'Each waifu2x-caffe thread will require up to {SYS_MEM_PER_THREAD} GB of system memory')
-            Avalon.warning('You demanded {} threads to be created, but you only have {} GB {} memory available'.format(args.threads, round(memory_available, 4), memory_type))
+            Avalon.warning(f'You demanded {args.threads} threads to be created, but you only have {round(memory_available, 4)} GB {memory_type} memory available')
-            Avalon.warning('{} GB of {} memory is recommended for {} threads'.format(mem_per_thread * args.threads, memory_type, args.threads))
+            Avalon.warning(f'{mem_per_thread * args.threads} GB of {memory_type} memory is recommended for {args.threads} threads')
-            Avalon.warning('With your current amount of {} memory available, {} threads is recommended'.format(memory_type, int(memory_available // mem_per_thread)))
+            Avalon.warning(f'With your current amount of {memory_type} memory available, {int(memory_available // mem_per_thread)} threads is recommended')
 
             # ask the user if he / she wants to change to the recommended
             # number of threads
@@ -169,12 +171,46 @@ def read_config(config_file):
         return config
 
 
+def absolutify_paths(config):
+    """ Check to see if paths to binaries are absolute
+
+    This function checks if paths to binary files are absolute.
+    If not, then absolutify the path.
+
+    Arguments:
+        config {dict} -- configuration file dictionary
+
+    Returns:
+        dict -- configuration file dictionary
+    """
+    current_directory = os.path.dirname(os.path.abspath(sys.argv[0]))
+
+    # check waifu2x-caffe path
+    if not re.match('^[a-z]:', config['waifu2x_caffe']['waifu2x_caffe_path'], re.IGNORECASE):
+        config['waifu2x_caffe']['waifu2x_caffe_path'] = f'{current_directory}\\{config["waifu2x_caffe"]["waifu2x_caffe_path"]}'
+
+    # check waifu2x-converter-cpp path
+    if not re.match('^[a-z]:', config['waifu2x_converter']['waifu2x_converter_path'], re.IGNORECASE):
+        config['waifu2x_converter']['waifu2x_converter_path'] = f'{current_directory}\\{config["waifu2x_converter"]["waifu2x_converter_path"]}'
+
+    # check ffmpeg path
+    if not re.match('^[a-z]:', config['ffmpeg']['ffmpeg_path'], re.IGNORECASE):
+        config['ffmpeg']['ffmpeg_path'] = f'{current_directory}\\{config["ffmpeg"]["ffmpeg_path"]}'
+
+    # check video2x cache path
+    if config['video2x']['video2x_cache_directory']:
+        if not re.match('^[a-z]:', config['video2x']['video2x_cache_directory'], re.IGNORECASE):
+            config['video2x']['video2x_cache_directory'] = f'{current_directory}\\{config["video2x"]["video2x_cache_directory"]}'
+
+    return config
+
+
 # /////////////////// Execution /////////////////// #
 
 # this is not a library
 if __name__ != '__main__':
     Avalon.error('This file cannot be imported')
-    raise ImportError('{} cannot be imported'.format(__file__))
+    raise ImportError(f'{__file__} cannot be imported')
 
 print_logo()
 
@@ -197,6 +233,7 @@ check_memory()
 
 # read configurations from JSON
 config = read_config(args.config)
+config = absolutify_paths(config)
 
 # load waifu2x configuration
 if args.driver == 'waifu2x_caffe':
@@ -219,25 +256,26 @@ elif args.driver == 'waifu2x_converter':
 ffmpeg_settings = config['ffmpeg']
 
 # load video2x settings
-video2x_cache_folder = config['video2x']['video2x_cache_folder']
+video2x_cache_directory = config['video2x']['video2x_cache_directory']
+image_format = config['video2x']['image_format'].lower()
 preserve_frames = config['video2x']['preserve_frames']
 
 # create temp directories if they don't exist
-if not video2x_cache_folder:
+if not video2x_cache_directory:
-    video2x_cache_folder = '{}\\video2x'.format(tempfile.gettempdir())
+    video2x_cache_directory = f'{tempfile.gettempdir()}\\video2x'
 
-if video2x_cache_folder and not os.path.isdir(video2x_cache_folder):
+if video2x_cache_directory and not os.path.isdir(video2x_cache_directory):
-    if not os.path.isfile(video2x_cache_folder) and not os.path.islink(video2x_cache_folder):
+    if not os.path.isfile(video2x_cache_directory) and not os.path.islink(video2x_cache_directory):
-        Avalon.warning('Specified cache folder/directory {} does not exist'.format(video2x_cache_folder))
+        Avalon.warning(f'Specified cache directory {video2x_cache_directory} does not exist')
-        if Avalon.ask('Create folder/directory?', default=True, batch=args.batch):
+        if Avalon.ask('Create directory?', default=True, batch=args.batch):
-            if os.mkdir(video2x_cache_folder) is None:
+            if os.mkdir(video2x_cache_directory) is None:
-                Avalon.info('{} created'.format(video2x_cache_folder))
+                Avalon.info(f'{video2x_cache_directory} created')
             else:
-                Avalon.error('Unable to create {}'.format(video2x_cache_folder))
+                Avalon.error(f'Unable to create {video2x_cache_directory}')
                 Avalon.error('Aborting...')
                 exit(1)
     else:
-        Avalon.error('Specified cache folder/directory is a file/link')
+        Avalon.error('Specified cache directory is a file/link')
         Avalon.error('Unable to continue, exiting...')
         exit(1)
 
@@ -247,33 +285,74 @@ try:
     # start timer
     begin_time = time.time()
 
+    # if input specified is a single file
     if os.path.isfile(args.input):
         """ Upscale single video file """
-        Avalon.info('Upscaling single video file: {}'.format(args.input))
+        Avalon.info(f'Upscaling single video file: {args.input}')
-        upscaler = Upscaler(input_video=args.input, output_video=args.output, method=args.method, waifu2x_settings=waifu2x_settings, ffmpeg_settings=ffmpeg_settings, waifu2x_driver=args.driver, scale_width=args.width, scale_height=args.height, scale_ratio=args.ratio, model_dir=args.model_dir, threads=args.threads, video2x_cache_folder=video2x_cache_folder)
-        upscaler.run()
-        upscaler.cleanup()
-    elif os.path.isdir(args.input):
-        """ Upscale videos in a folder/directory """
-        Avalon.info('Upscaling videos in folder/directory: {}'.format(args.input))
-        for input_video in [f for f in os.listdir(args.input) if os.path.isfile(os.path.join(args.input, f))]:
-            output_video = '{}\\{}'.format(args.output, input_video)
-            upscaler = Upscaler(input_video=os.path.join(args.input, input_video), output_video=output_video, method=args.method, waifu2x_settings=waifu2x_settings, ffmpeg_settings=ffmpeg_settings, waifu2x_driver=args.driver, scale_width=args.width, scale_height=args.height, scale_ratio=args.ratio, model_dir=args.model_dir, threads=args.threads, video2x_cache_folder=video2x_cache_folder)
-            upscaler.run()
-            upscaler.cleanup()
-    else:
-        Avalon.error('Input path is neither a file nor a folder/directory')
-        raise FileNotFoundError('{} is neither file nor folder/directory'.format(args.input))
 
-    Avalon.info('Program completed, taking {} seconds'.format(round((time.time() - begin_time), 5)))
+        # check for input output format mismatch
+        if os.path.isdir(args.output):
+            Avalon.error('Input and output path type mismatch')
+            Avalon.error('Input is single file but output is directory')
+            raise Exception('input output path type mismatch')
+        if not re.search('.*\..*$', args.output):
+            Avalon.error('No suffix found in output file path')
+            Avalon.error('Suffix must be specified for FFmpeg')
+            raise Exception('No suffix specified')
+
+        upscaler = Upscaler(input_video=args.input, output_video=args.output, method=args.method, waifu2x_settings=waifu2x_settings, ffmpeg_settings=ffmpeg_settings)
+
+        # set optional options
+        upscaler.waifu2x_driver = args.driver
+        upscaler.scale_width = args.width
+        upscaler.scale_height = args.height
+        upscaler.scale_ratio = args.ratio
+        upscaler.model_dir = args.model_dir
+        upscaler.threads = args.threads
+        upscaler.video2x_cache_directory = video2x_cache_directory
+        upscaler.image_format = image_format
+        upscaler.preserve_frames = preserve_frames
+
+        # run upscaler
+        upscaler.create_temp_directories()
+        upscaler.run()
+        upscaler.cleanup_temp_directories()
+
+    # if input specified is a directory
+    elif os.path.isdir(args.input):
+        """ Upscale videos in a directory """
+        Avalon.info(f'Upscaling videos in directory: {args.input}')
+        for input_video in [f for f in os.listdir(args.input) if os.path.isfile(os.path.join(args.input, f))]:
+            output_video = f'{args.output}\\{input_video}'
+            upscaler = Upscaler(input_video=os.path.join(args.input, input_video), output_video=output_video, method=args.method, waifu2x_settings=waifu2x_settings, ffmpeg_settings=ffmpeg_settings)
+
+            # set optional options
+            upscaler.waifu2x_driver = args.driver
+            upscaler.scale_width = args.width
+            upscaler.scale_height = args.height
+            upscaler.scale_ratio = args.ratio
+            upscaler.model_dir = args.model_dir
+            upscaler.threads = args.threads
+            upscaler.video2x_cache_directory = video2x_cache_directory
+            upscaler.image_format = image_format
+            upscaler.preserve_frames = preserve_frames
+
+            # run upscaler
+            upscaler.create_temp_directories()
+            upscaler.run()
+            upscaler.cleanup_temp_directories()
+    else:
+        Avalon.error('Input path is neither a file nor a directory')
+        raise FileNotFoundError(f'{args.input} is neither file nor directory')
+
+    Avalon.info(f'Program completed, taking {round((time.time() - begin_time), 5)} seconds')
 except Exception:
     Avalon.error('An exception has occurred')
     traceback.print_exc()
-    Avalon.warning('If you experience error \"cudaSuccess out of memory\", try reducing number of threads you\'re using')
 finally:
-    # remove Video2X Cache folder
+    # remove Video2X Cache directory
     try:
         if not preserve_frames:
-            shutil.rmtree(video2x_cache_folder)
+            shutil.rmtree(video2x_cache_directory)
     except FileNotFoundError:
         pass

bin/video2x_setup.py

@@ -4,7 +4,7 @@
 Name: Video2X Setup Script
 Author: K4YT3X
 Date Created: November 28, 2018
-Last Modified: March 9, 2019
+Last Modified: April 28, 2019
 
 Licensed under the GNU General Public License Version 3 (GNU GPL v3),
     available at: https://www.gnu.org/licenses/gpl-3.0.txt
@@ -19,9 +19,11 @@ Installation Details:
 - waifu2x-caffe: %LOCALAPPDATA%\\video2x\\waifu2x-caffe
 
 """
+import argparse
 import json
 import os
 import subprocess
+import sys
 import tempfile
 import traceback
 import zipfile
@@ -31,7 +33,20 @@ import zipfile
 # later in the script.
 # import requests
 
-VERSION = '1.1.0'
+VERSION = '1.2.1'
+
+
+def process_arguments():
+    """Processes CLI arguments
+    """
+    parser = argparse.ArgumentParser(formatter_class=argparse.ArgumentDefaultsHelpFormatter)
+
+    # video options
+    general_options = parser.add_argument_group('General Options')
+    general_options.add_argument('-d', '--driver', help='driver to download and configure', action='store', choices=['all', 'waifu2x_caffe', 'waifu2x_converter'], default='all')
+
+    # parse arguments
+    return parser.parse_args()
 
 
 class Video2xSetup:
@@ -41,19 +56,26 @@ class Video2xSetup:
     script. All files will be installed under %LOCALAPPDATA%\\video2x.
     """
 
-    def __init__(self):
+    def __init__(self, driver, download_python_modules):
+        self.driver = driver
+        self.download_python_modules = download_python_modules
         self.trash = []
 
     def run(self):
-
+        if self.download_python_modules:
-        print('\nInstalling Python libraries')
+            print('\nInstalling Python libraries')
-        self._install_python_requirements()
+            self._install_python_requirements()
 
         print('\nInstalling FFMPEG')
         self._install_ffmpeg()
 
-        print('\nInstalling waifu2x-caffe')
+        if self.driver == 'all':
-        self._install_waifu2x_caffe()
+            self._install_waifu2x_caffe()
+            self._install_waifu2x_converter_cpp()
+        elif self.driver == 'waifu2x_caffe':
+            self._install_waifu2x_caffe()
+        elif self.driver == 'waifu2x_converter':
+            self._install_waifu2x_converter_cpp()
 
         print('\nGenerating Video2X configuration file')
         self._generate_config()
@@ -66,7 +88,7 @@ class Video2xSetup:
         """
         with open('requirements.txt', 'r') as req:
             for line in req:
-                package = line.split(' ')[0]
+                package = line.split('==')[0]
                 pip_install(package)
 
     def _cleanup(self):
@@ -74,7 +96,7 @@ class Video2xSetup:
         """
         for file in self.trash:
             try:
-                print('Deleting: {}'.format(file))
+                print(f'Deleting: {file}')
                 os.remove(file)
             except FileNotFoundError:
                 pass
@@ -82,17 +104,18 @@ class Video2xSetup:
     def _install_ffmpeg(self):
         """ Install FFMPEG
         """
-        latest_release = 'https://ffmpeg.zeranoe.com/builds/win64/static/ffmpeg-4.1-win64-static.zip'
+        latest_release = 'https://ffmpeg.zeranoe.com/builds/win64/static/ffmpeg-latest-win64-static.zip'
 
         ffmpeg_zip = download(latest_release, tempfile.gettempdir())
         self.trash.append(ffmpeg_zip)
 
         with zipfile.ZipFile(ffmpeg_zip) as zipf:
-            zipf.extractall('{}\\video2x'.format(os.getenv('localappdata')))
+            zipf.extractall(f'{os.getenv("localappdata")}\\video2x')
 
     def _install_waifu2x_caffe(self):
         """ Install waifu2x_caffe
         """
+        print('\nInstalling waifu2x-caffe')
         import requests
 
         # Get latest release of waifu2x-caffe via GitHub API
@@ -104,7 +127,25 @@ class Video2xSetup:
                 self.trash.append(waifu2x_caffe_zip)
 
         with zipfile.ZipFile(waifu2x_caffe_zip) as zipf:
-            zipf.extractall('{}\\video2x'.format(os.getenv('localappdata')))
+            zipf.extractall(f'{os.getenv("localappdata")}\\video2x')
+
+    def _install_waifu2x_converter_cpp(self):
+        """ Install waifu2x_caffe
+        """
+        print('\nInstalling waifu2x-converter-cpp')
+        import re
+        import requests
+
+        # Get latest release of waifu2x-caffe via GitHub API
+        latest_release = json.loads(requests.get('https://api.github.com/repos/DeadSix27/waifu2x-converter-cpp/releases/latest').content)
+
+        for a in latest_release['assets']:
+            if re.search(r'waifu2x-DeadSix27-win64_v[0-9]*\.zip', a['browser_download_url']):
+                waifu2x_converter_cpp_zip = download(a['browser_download_url'], tempfile.gettempdir())
+                self.trash.append(waifu2x_converter_cpp_zip)
+
+        with zipfile.ZipFile(waifu2x_converter_cpp_zip) as zipf:
+            zipf.extractall(f'{os.getenv("localappdata")}\\video2x\\waifu2x-converter-cpp')
 
     def _generate_config(self):
         """ Generate video2x config
@@ -114,38 +155,48 @@ class Video2xSetup:
             template_dict = json.load(template)
             template.close()
 
-        template_dict['waifu2x_caffe']['waifu2x_caffe_path'] = '{}\\video2x\\waifu2x-caffe\\waifu2x-caffe-cui.exe'.format(os.getenv('localappdata'))
+        local_app_data = os.getenv('localappdata')
-        template_dict['ffmpeg']['ffmpeg_path'] = '{}\\video2x\\ffmpeg-4.1-win64-static\\bin'.format(os.getenv('localappdata'))
+
-        template_dict['ffmpeg']['ffmpeg_hwaccel'] = 'auto'
+        # configure only the specified drivers
-        template_dict['ffmpeg']['extra_arguments'] = []
+        if self.driver == 'all':
-        template_dict['video2x']['video2x_cache_folder'] = False
+            template_dict['waifu2x_caffe']['waifu2x_caffe_path'] = f'{local_app_data}\\video2x\\waifu2x-caffe\\waifu2x-caffe-cui.exe'
+            template_dict['waifu2x_converter']['waifu2x_converter_path'] = f'{local_app_data}\\video2x\\waifu2x-converter-cpp'
+        elif self.driver == 'waifu2x_caffe':
+            template_dict['waifu2x_caffe']['waifu2x_caffe_path'] = f'{local_app_data}\\video2x\\waifu2x-caffe\\waifu2x-caffe-cui.exe'
+        elif self.driver == 'waifu2x_converter':
+            template_dict['waifu2x_converter']['waifu2x_converter_path'] = f'{local_app_data}\\video2x\\waifu2x-converter-cpp'
+
+        template_dict['ffmpeg']['ffmpeg_path'] = f'{local_app_data}\\video2x\\ffmpeg-latest-win64-static\\bin'
+        template_dict['video2x']['video2x_cache_directory'] = None
         template_dict['video2x']['preserve_frames'] = False
 
         # Write configuration into file
         with open('video2x.json', 'w') as config:
-            json.dump(template_dict, config, indent=2)
+            json.dump(template_dict, config, indent=4)
             config.close()
 
 
 def download(url, save_path, chunk_size=4096):
     """ Download file to local with requests library
     """
+    from tqdm import tqdm
     import requests
 
-    output_file = '{}\\{}'.format(save_path, url.split('/')[-1])
+    output_file = f'{save_path}\\{url.split("/")[-1]}'
-    print('Downloading: {}'.format(url))
+    print(f'Downloading: {url}')
-    print('Chunk size: {}'.format(chunk_size))
+    print(f'Chunk size: {chunk_size}')
-    print('Saving to: {}'.format(output_file))
+    print(f'Saving to: {output_file}')
 
     stream = requests.get(url, stream=True)
+    total_size = int(stream.headers['content-length'])
 
     # Write content into file
     with open(output_file, 'wb') as output:
-        for chunk in stream.iter_content(chunk_size=chunk_size):
+        with tqdm(total=total_size, ascii=True) as progress_bar:
-            if chunk:
+            for chunk in stream.iter_content(chunk_size=chunk_size):
-                print('!', end='')
+                if chunk:
-                output.write(chunk)
+                    output.write(chunk)
-    print()
+                    progress_bar.update(len(chunk))
 
     return output_file
 
@@ -161,11 +212,20 @@ def pip_install(package):
 
 if __name__ == "__main__":
     try:
-        print('Video2x Setup Script')
+        args = process_arguments()
-        print('Version: {}'.format(VERSION))
+        print('Video2X Setup Script')
-        setup = Video2xSetup()
+        print(f'Version: {VERSION}')
+
+        # do not install pip modules if script
+        # is packaged in exe format
+        download_python_modules = True
+        if sys.argv[0].endswith('.exe'):
+            print('\nScript is packaged as exe, skipping pip module download')
+            download_python_modules = False
+
+        setup = Video2xSetup(args.driver, download_python_modules)
         setup.run()
-        print('\n Script finished successfully')
+        print('\nScript finished successfully')
     except Exception:
         traceback.print_exc()
         print('An error has occurred')

bin/waifu2x_caffe.py

@@ -4,7 +4,7 @@
 Name: Waifu2x Caffe Driver
 Author: K4YT3X
 Date Created: Feb 24, 2018
-Last Modified: March 24, 2019
+Last Modified: April 28, 2019
 
 Description: This class is a high-level wrapper
 for waifu2x-caffe.
@@ -33,20 +33,20 @@ class Waifu2xCaffe:
         self.model_dir = model_dir
         self.print_lock = threading.Lock()
 
-    def upscale(self, input_folder, output_folder, scale_ratio, scale_width, scale_height, threads_exceptions):
+    def upscale(self, input_directory, output_directory, scale_ratio, scale_width, scale_height, image_format, upscaler_exceptions):
         """This is the core function for WAIFU2X class
 
         Arguments:
-            input_folder {string} -- source folder path
+            input_directory {string} -- source directory path
-            output_folder {string} -- output folder path
+            output_directory {string} -- output directory path
             width {int} -- output video width
             height {int} -- output video height
         """
 
         try:
             # overwrite config file settings
-            self.waifu2x_settings['input_path'] = input_folder
+            self.waifu2x_settings['input_path'] = input_directory
-            self.waifu2x_settings['output_path'] = output_folder
+            self.waifu2x_settings['output_path'] = output_directory
 
             if scale_ratio:
                 self.waifu2x_settings['scale_ratio'] = scale_ratio
@@ -54,9 +54,11 @@ class Waifu2xCaffe:
                 self.waifu2x_settings['scale_width'] = scale_width
                 self.waifu2x_settings['scale_height'] = scale_height
 
+            self.waifu2x_settings['output_extention'] = image_format
+
             # print thread start message
             self.print_lock.acquire()
-            Avalon.debug_info('[upscaler] Thread {} started'.format(threading.current_thread().name))
+            Avalon.debug_info(f'[upscaler] Thread {threading.current_thread().name} started')
             self.print_lock.release()
 
             # list to be executed
@@ -72,20 +74,20 @@ class Waifu2xCaffe:
                     continue
                 else:
                     if len(key) == 1:
-                        execute.append('-{}'.format(key))
+                        execute.append(f'-{key}')
                     else:
-                        execute.append('--{}'.format(key))
+                        execute.append(f'--{key}')
                     execute.append(str(value))
 
-            Avalon.debug_info('Executing: {}'.format(execute))
+            Avalon.debug_info(f'Executing: {execute}')
             completed_command = subprocess.run(execute, check=True)
 
             # print thread exiting message
             self.print_lock.acquire()
-            Avalon.debug_info('[upscaler] Thread {} exiting'.format(threading.current_thread().name))
+            Avalon.debug_info(f'[upscaler] Thread {threading.current_thread().name} exiting')
             self.print_lock.release()
 
             # return command execution return code
             return completed_command.returncode
         except Exception as e:
-            threads_exceptions.append(e)
+            upscaler_exceptions.append(e)

bin/waifu2x_converter.py

@@ -4,7 +4,7 @@
 Name: Waifu2x Converter CPP Driver
 Author: K4YT3X
 Date Created: February 8, 2019
-Last Modified: March 24, 2019
+Last Modified: April 28, 2019
 
 Description: This class is a high-level wrapper
 for waifu2x-converter-cpp.
@@ -28,39 +28,42 @@ class Waifu2xConverter:
         self.waifu2x_settings['model_dir'] = model_dir
         self.print_lock = threading.Lock()
 
-    def upscale(self, input_folder, output_folder, scale_ratio, jobs, threads_exceptions):
+    def upscale(self, input_directory, output_directory, scale_ratio, jobs, image_format, upscaler_exceptions):
         """ Waifu2x Converter Driver Upscaler
         This method executes the upscaling of extracted frames.
 
         Arguments:
-            input_folder {string} -- source folder path
+            input_directory {string} -- source directory path
-            output_folder {string} -- output folder path
+            output_directory {string} -- output directory path
             scale_ratio {int} -- frames' scale ratio
             threads {int} -- number of threads
         """
 
         try:
             # overwrite config file settings
-            self.waifu2x_settings['input'] = input_folder
+            self.waifu2x_settings['input'] = input_directory
-            self.waifu2x_settings['output'] = output_folder
+            self.waifu2x_settings['output'] = output_directory
 
             # temporary fix for https://github.com/DeadSix27/waifu2x-converter-cpp/issues/109
-            self.waifu2x_settings['i'] = input_folder
+            """
-            self.waifu2x_settings['o'] = output_folder
+            self.waifu2x_settings['i'] = input_directory
+            self.waifu2x_settings['o'] = output_directory
             self.waifu2x_settings['input'] = None
             self.waifu2x_settings['output'] = None
+            """
 
-            self.waifu2x_settings['scale_ratio'] = scale_ratio
+            self.waifu2x_settings['scale-ratio'] = scale_ratio
             self.waifu2x_settings['jobs'] = jobs
+            self.waifu2x_settings['output-format'] = image_format
 
             # models_rgb must be specified manually for waifu2x-converter-cpp
             # if it's not specified in the arguments, create automatically
-            if self.waifu2x_settings['model_dir'] is None:
+            if self.waifu2x_settings['model-dir'] is None:
-                self.waifu2x_settings['model_dir'] = '{}\\models_rgb'.format(self.waifu2x_settings['waifu2x_converter_path'])
+                self.waifu2x_settings['model-dir'] = f'{self.waifu2x_settings["waifu2x_converter_path"]}\\models_rgb'
 
             # print thread start message
             self.print_lock.acquire()
-            Avalon.debug_info('[upscaler] Thread {} started'.format(threading.current_thread().name))
+            Avalon.debug_info(f'[upscaler] Thread {threading.current_thread().name} started')
             self.print_lock.release()
 
             # list to be executed
@@ -72,16 +75,16 @@ class Waifu2xConverter:
 
                 # the key doesn't need to be passed in this case
                 if key == 'waifu2x_converter_path':
-                    execute.append('{}\\waifu2x-converter-cpp.exe'.format(str(value)))
+                    execute.append(f'{str(value)}\\waifu2x-converter-cpp.exe')
 
                 # null or None means that leave this option out (keep default)
                 elif value is None or value is False:
                     continue
                 else:
                     if len(key) == 1:
-                        execute.append('-{}'.format(key))
+                        execute.append(f'-{key}')
                     else:
-                        execute.append('--{}'.format(key))
+                        execute.append(f'--{key}')
 
                     # true means key is an option
                     if value is True:
@@ -89,8 +92,8 @@ class Waifu2xConverter:
 
                     execute.append(str(value))
 
-            Avalon.debug_info('Executing: {}'.format(execute))
+            Avalon.debug_info(f'Executing: {execute}')
             return subprocess.run(execute, check=True).returncode
 
         except Exception as e:
-            threads_exceptions.append(e)
+            upscaler_exceptions.append(e)