Reformat readme with one line per paragraphs

Chapter01/README.md

@@ -4,14 +4,11 @@ This chapter will show you the basics of generative music and what already exist
 
 ## Code
 
-This chapter doesn't have any code, since it serves as an introduction to the
-book and as an installation procedure. You can check out the
-[Jupyter Notebook](notebook.ipynb) to test your installation.
+This chapter doesn't have any code, since it serves as an introduction to the book and as an installation procedure. You can check out the [Jupyter Notebook](notebook.ipynb) to test your installation.
 
 ## Installing Magenta
 
-Installing Magenta using Conda is easier. First
-[install Miniconda](https://conda.io/en/latest/miniconda.html), then:
+Installing Magenta using Conda is easier. First [install Miniconda](https://conda.io/en/latest/miniconda.html), then:
 
 ```bash
 # Create a new environment for Magenta with Python 3.6.x as interpreter
@@ -31,16 +28,11 @@ pip install magenta visual_midi
 python -c "import magenta; print(magenta.__version__)"
 ```
 
-## Installing Magenta GPU
-
-TODO compatibility version grid
-
 ## Utility Scripts
 
 ### [Visual MIDI](https://github.com/dubreuia/visual_midi)
 
-Transforms a MIDI file to a plot, most of the diagram in this book has been done
- with this tool.
+Transforms a MIDI file to a plot, most of the diagram in this book has been done with this tool.
 
 ### `wav2plot.py`
 

Chapter02/README.md

@@ -6,8 +6,7 @@ This chapter will show you what many consider the foundation of music—percussi
 
 ### [Example 1](chapter_02_example_01.py) or [notebook](notebook.ipynb)
 
-This example shows a basic Drums RNN generation with a hard coded primer. For
-the Python script, while in the Magenta environment (`conda activate magenta`):
+This example shows a basic Drums RNN generation with a hard coded primer. For the Python script, while in the Magenta environment (`conda activate magenta`):
 
 ```bash
 # Runs a Drums RNN generation and shows the resulting plot in the browser
@@ -22,9 +21,7 @@ jupyter notebook notebook.ipynb
 
 ### [Example 2](chapter_09_example_02.py)
 
-This example shows a basic Drums RNN generation with synthesizer playback. For
-the Python script, you'll need to first install and configure a software
-synthesizer, then while in the Magenta environment (`conda activate magenta`):
+This example shows a basic Drums RNN generation with synthesizer playback. For the Python script, you'll need to first install and configure a software synthesizer, then while in the Magenta environment (`conda activate magenta`):
 
 ```bash
 # Runs a Drums RNN generation, shosw the resulting plot in the browser, and
@@ -34,10 +31,7 @@ python chapter_02_example_02.py
 
 ### [Example 3](chapter_02_example_03.py)
 
-This example shows a basic Drums RNN generation with a looping synthesizer
-playback. For the Python script, you'll need to first install and configure a
-software synthesizer, then while in the Magenta environment
-(`conda activate magenta`):
+This example shows a basic Drums RNN generation with a looping synthesizer playback. For the Python script, you'll need to first install and configure a software synthesizer, then while in the Magenta environment (`conda activate magenta`):
 
 ```bash
 # Runs a Drums RNN generation, shosw the resulting plot in the browser, and

Chapter03/README.md

@@ -6,9 +6,7 @@ This chapter will show the importance of Long Short-Term Memory (LSTM) networks
 
 ### [Example 1](chapter_03_example_01.py) or [notebook](notebook.ipynb)
 
-This example shows a melody (monophonic) generation using the Melody RNN model
-and 3 configurations: basic, lookback and attention. For
-the Python script, while in the Magenta environment (`conda activate magenta`):
+This example shows a melody (monophonic) generation using the Melody RNN model and 3 configurations: basic, lookback and attention. For the Python script, while in the Magenta environment (`conda activate magenta`):
 
 ```bash
 # Runs 3 melody rnn generation using the basic, lookback and attention config
@@ -23,9 +21,7 @@ jupyter notebook notebook.ipynb
 
 ### [Example 2](chapter_03_example_02.py)
 
-This example shows polyphonic generations with the Polyphony RNN model. For
-the Python script, while in the Magenta environment
-(`conda activate magenta`):
+This example shows polyphonic generations with the Polyphony RNN model. For the Python script, while in the Magenta environment (`conda activate magenta`):
 
 ```bash
 # Runs 4 polyphonic generations using polyphony rnn and
@@ -35,9 +31,7 @@ python chapter_03_example_02.py
 
 ### [Example 3](chapter_03_example_03.py)
 
-This example shows polyphonic generations with the Performance RNN model. For
-the Python script, while in the Magenta environment
-(`conda activate magenta`):
+This example shows polyphonic generations with the Performance RNN model. For the Python script, while in the Magenta environment (`conda activate magenta`):
 
 ```bash
 # Runs 3 polyphonic generations using performance rnn and 

Chapter04/README.md

@@ -6,9 +6,7 @@ This chapter will show the importance of continuous latent space of Variational
 
 ### [Example 1](chapter_04_example_01.py) or [notebook](notebook.ipynb)
 
-This example shows how to sample, interpolate and humanize a drums sequence
-using MusicVAE and various configurations. For the Python script, 
-while in the Magenta environment (`conda activate magenta`):
+This example shows how to sample, interpolate and humanize a drums sequence using MusicVAE and various configurations. For the Python script, while in the Magenta environment (`conda activate magenta`):
 
 ```bash
 # Runs the example, the output files (plot, midi) will be in the "output" folder
@@ -23,9 +21,7 @@ jupyter notebook notebook.ipynb
 
 ### [Example 2](chapter_04_example_02.py)
 
-This example shows how to sample and interpolate a melody sequence
-using MusicVAE and various configurations. For the Python script, 
-while in the Magenta environment (`conda activate magenta`):
+This example shows how to sample and interpolate a melody sequence using MusicVAE and various configurations. For the Python script, while in the Magenta environment (`conda activate magenta`):
 
 ```bash
 # Runs the example, the output files (plot, midi) will be in the "output" folder
@@ -34,10 +30,7 @@ python chapter_04_example_02.py
 
 ### [Example 3](chapter_04_example_03.py)
 
-This example shows how to sample a trio (drums, melody, bass) sequence
-using MusicVAE and various configurations. For the Python script,
-while in the Magenta environment
-(`conda activate magenta`):
+This example shows how to sample a trio (drums, melody, bass) sequence using MusicVAE and various configurations. For the Python script, while in the Magenta environment (`conda activate magenta`):
 
 ```bash
 # Runs the example, the output files (plot, midi) will be in the "output" folder

Chapter05/README.md

@@ -4,10 +4,7 @@ This chapter will show audio generation. We'll first provide an overview of Wave
 
 ## Utils
 
-There are some audio utilities in the [audio_utils.py](./audio_utils.py) file,
-useful for saving and loading the encodings (`save_encoding` and 
-`load_encodings`), time stretching (`timestretch`) them and saving
-spectrogram plots (`save_spectrogram_plot` and `save_rainbowgram_plot`).
+There are some audio utilities in the [audio_utils.py](./audio_utils.py) file, useful for saving and loading the encodings (`save_encoding` and `load_encodings`), time stretching (`timestretch`) them and saving spectrogram plots (`save_spectrogram_plot` and `save_rainbowgram_plot`).
 
 ## Sounds and MIDI
 

Chapter06/README.md

@@ -4,15 +4,9 @@ This chapter will show how training our own models is crucial since it allows us
 
 ## Utils
 
-There are some utilities for processing the Lakh MIDI Dataset (LMD) 
-in the [lakh_utils.py](./lakh_utils.py) file and utilities for multiprocessing
-in the [multiprocessing_utils.py](./multiprocessing_utils.py) file with example
-usage.
+There are some utilities for processing the Lakh MIDI Dataset (LMD) in the [lakh_utils.py](./lakh_utils.py) file and utilities for multiprocessing in the [multiprocessing_utils.py](./multiprocessing_utils.py) file with example usage.
 
-There is a custom pipeline example for the Melody RNN model in the 
-[melody_rnn_pipeline_example.py](./melody_rnn_pipeline_example.py) file. Change
-directory to the folder containing the Tensorflow records of NoteSequence and 
-call the pipeline using:
+There is a custom pipeline example for the Melody RNN model in the [melody_rnn_pipeline_example.py](./melody_rnn_pipeline_example.py) file. Change directory to the folder containing the Tensorflow records of NoteSequence and call the pipeline using:
 
 ```bash
 python /path/to/the/pipeline/melody_rnn_pipeline_example.py --config="attention_rnn" --input="notesequences.tfrecord" --output_dir="sequence_examples" --eval_ratio=0.10
@@ -38,17 +32,15 @@ python chapter_06_example_01.py --sample_size=1000 --pool_size=4 --path_dataset_
 
 ### [Example 2](chapter_06_example_02.py)
 
-Lists most common genres from the Last.fm API using the LAKHs dataset
-matched with the MSD dataset.
+Lists most common genres from the Last.fm API using the LAKHs dataset matched with the MSD dataset.
 
 ```bash
 python chapter_06_example_02.py --sample_size=1000 --pool_size=4 --path_dataset_dir=PATH_DATASET --path_match_scores_file=PATH_MATCH_SCORES --last_fm_api_key=LAST_FM_API_KEY
 ```
 
 ### [Example 3](chapter_06_example_03.py)
 
-Filter on specific tags from the Last.fm API using the LAKHs dataset
-matched with the MSD dataset.
+Filter on specific tags from the Last.fm API using the LAKHs dataset matched with the MSD dataset.
 
 ```bash
 python chapter_06_example_03.py --sample_size=1000 --pool_size=4 --path_dataset_dir=PATH_DATASET --path_match_scores_file=PATH_MATCH_SCORES --last_fm_api_key=LAST_FM_API_KEY --tags="['jazz', 'blues']"
@@ -64,19 +56,15 @@ python chapter_06_example_04.py --sample_size=1000 --pool_size=4 --path_dataset_
 
 ### [Example 5](chapter_06_example_05.py)
 
-Extract drums MIDI files. Some drum tracks are split into multiple separate
-drum instruments, in which case we try to merge them into a single instrument
-and save only 1 MIDI file.
+Extract drums MIDI files. Some drum tracks are split into multiple separate drum instruments, in which case we try to merge them into a single instrument and save only 1 MIDI file.
 
 ```bash
 python chapter_06_example_05.py --sample_size=1000 --pool_size=4 --path_dataset_dir=PATH_DATASET --path_match_scores_file=PATH_MATCH_SCORES --path_output_dir=PATH_OUTPUT
 ```
 
 ### [Example 6](chapter_06_example_06.py)
 
-Extract piano MIDI files. Some piano tracks are split into multiple separate
-piano instruments, in which case we keep them split and merge them into
-multiple MIDI files.
+Extract piano MIDI files. Some piano tracks are split into multiple separate piano instruments, in which case we keep them split and merge them into multiple MIDI files.
 
 ```bash
 python chapter_06_example_06.py --sample_size=1000 --pool_size=4 --path_dataset_dir=PATH_DATASET --path_match_scores_file=PATH_MATCH_SCORES --path_output_dir=PATH_OUTPUT

Chapter07/README.md

@@ -8,17 +8,15 @@ This chapter doesn't contain a lot of code, refer to the book for the content.
 
 ### [Example 1](chapter_07_example_01.py)
 
-Configuration for the MusicVAE model, using the MIDI bass programs. To
-launch the training:
+Configuration for the MusicVAE model, using the MIDI bass programs. To launch the training:
 
 ```bash
 python chapter_07_example_01.py --config="cat-bass_2bar_small" --run_dir="..."
 ```
 
 ### [Example 3](chapter_07_example_02.py)
 
-Tensor validator and note sequence splitter (training and evaluation datasets)
-for the MusicVAE model.
+Tensor validator and note sequence splitter (training and evaluation datasets) for the MusicVAE model.
 
 ```bash
 python chapter_07_example_02.py --config="cat-drums_2bar_small" --input="notesequences.tfrecord" --output_dir="sequence_examples"

Chapter08/README.md

@@ -6,45 +6,31 @@ This chapter will show a JavaScript implementation of Magenta that gained popula
 
 ### [Example 1](chapter_08_example_01.html)
 
-An example on how to use a converted model (from a previously trained model)
-locally.
+An example on how to use a converted model (from a previously trained model) locally.
 
 ### [Example 2](chapter_08_example_02.html)
 
-An example of a Magenta.js web page using GANSynth. Press "Sample GANSynth note"
-to sample a new note using GANSynth and play it immediately. You can layer as
-many notes as you want, each note will loop each 4 seconds.
+An example of a Magenta.js web page using GANSynth. Press "Sample GANSynth note" to sample a new note using GANSynth and play it immediately. You can layer as many notes as you want, each note will loop each 4 seconds.
 
 ### [Example 2 ES6](chapter_08_example_02_es6.html)
 
 [Example 2](#example-2) using ES6 modules.
 
 ### [Example 3](chapter_08_example_03.html)
 
-An example of a Magenta.js web page using MusicVAE. Press "Sample MusicVAE trio"
-to sample a new sequence of a trio of instruments (drum kit, bass, lead) using
-MusicVAE and play it immediately. The sequence will loop.
+An example of a Magenta.js web page using MusicVAE. Press "Sample MusicVAE trio" to sample a new sequence of a trio of instruments (drum kit, bass, lead) using MusicVAE and play it immediately. The sequence will loop.
 
 ### [Example 4](chapter_08_example_04.html)
 
-An example of a Magenta.js web page using GANSynth and MusicVAE. Press "Sample
-MusicVAE trio" to sample a new sequence of a trio of instruments (drum kit,
-bass, lead) using MusicVAE and play it immediately. The sequence will loop.
-Press "Sample GANSynth note for the lead synth" to sample a new note using
-GANSynth use it in the lead synth.
+An example of a Magenta.js web page using GANSynth and MusicVAE. Press "Sample MusicVAE trio" to sample a new sequence of a trio of instruments (drum kit, bass, lead) using MusicVAE and play it immediately. The sequence will loop. Press "Sample GANSynth note for the lead synth" to sample a new note using GANSynth use it in the lead synth.
 
 ### [Example 5](chapter_08_example_05.html)
 
-An example of a Magenta.js web page using the Web Workers API. Press "Sample
-MusicVAE trio" to sample a new sequence of a trio of instruments (drum kit,
-bass, lead) using MusicVAE in a Web Worker and play it immediately.
+An example of a Magenta.js web page using the Web Workers API. Press "Sample MusicVAE trio" to sample a new sequence of a trio of instruments (drum kit, bass, lead) using MusicVAE in a Web Worker and play it immediately.
 
 ### [Example 6](chapter_08_example_06.html)
 
-An example of a Magenta.js web page using the Web MIDI API. Press "Sample
-MusicVAE trio" to sample a new sequence of a trio of instruments (drum kit,
-bass, lead) using MusicVAE and send it to the given MIDI output. The sequence
-will loop.
+An example of a Magenta.js web page using the Web MIDI API. Press "Sample MusicVAE trio" to sample a new sequence of a trio of instruments (drum kit, bass, lead) using MusicVAE and send it to the given MIDI output. The sequence will loop.
 
 ### [Example 7](chapter_08_example_07.js)
 

Chapter09/README.md

@@ -14,37 +14,31 @@ python chapter_09_example_01.py
 
 ### [Example 2](chapter_09_example_02.py)
 
-This example shows a basic Drums RNN generation with synthesizer playback,
-using a MIDI hub to send the sequence to an external device.
+This example shows a basic Drums RNN generation with synthesizer playback, using a MIDI hub to send the sequence to an external device.
 
 ```bash
 python chapter_09_example_02.py --midi_port="midi_port_name"
 ```
 
 ### [Example 3](chapter_09_example_03.py)
 
-This example shows a basic Drums RNN generation with a
-looping synthesizer playback, using a MIDI hub to send the sequence
-to an external device.
+This example shows a basic Drums RNN generation with a looping synthesizer playback, using a MIDI hub to send the sequence to an external device.
 
 ```bash
 python chapter_09_example_03.py --midi_port="midi_port_name"
 ```
 
 ### [Example 4](chapter_09_example_04.py)
 
-This example shows how to synchronize a Magenta application with an external
-device using MIDI clock and transport messages.
+This example shows how to synchronize a Magenta application with an external device using MIDI clock and transport messages.
 
 ```bash
 python chapter_09_example_04.py --midi_port="midi_port_name"
 ```
 
 ### [Example 5](chapter_09_example_05.py)
 
-This example shows a basic Drums RNN generation with a
-looping synthesizer playback, generating a new sequence at each loop,
-using a MIDI hub to send the sequence to an external device.
+This example shows a basic Drums RNN generation with a looping synthesizer playback, generating a new sequence at each loop, using a MIDI hub to send the sequence to an external device.
 
 ```bash
 python chapter_09_example_05.py --midi_port="midi_port_name"