update README

icoxfog417 · Apr 16, 2017 · 8d32c71 · 8d32c71
1 parent 200d8c0
commit 8d32c71
Show file tree

Hide file tree

Showing 4 changed files with 120 additions and 111 deletions.
diff --git a/README.md b/README.md
@@ -30,6 +30,10 @@ The model is ported from [ai-duet](https://github.com/googlecreativelab/aiexperi
 
 ![gui.PNG](./docs/gui.PNG)
 
+### Additional Usage
+
+* [Train Your own Model](https://github.com/icoxfog417/magenta_session/tree/master/scripts)
+* [Session with your own model](https://github.com/icoxfog417/magenta_session/tree/master/server)
 
 ## Install 
 
@@ -77,119 +81,8 @@ $ docker run -it --rm -p 80:8080 magenta-session
 
 You can now play with `magenta_session` at `http://<docker-server-ipaddress>/`.
 
-
 Session Now and Enjoy Music!
 
-## Train your own model
-
-You can create your own model by your MIDI files!
-The procedure is almost same as [MelodyRNN](https://github.com/tensorflow/magenta/tree/master/magenta/models/melody_rnn). So please refer it as you need.
-
-### 1. Prepare the MIDI data
-
-Please gather your favorite MIDI files and store it to `data/raw`. You can find the MIDI files from following sites.
-
-* [midiworld.com](http://www.midiworld.com/files/142/)
-* [FreeMIDI.org](https://freemidi.org/)
-* [The Lakh MIDI Dataset v0.1](http://colinraffel.com/projects/lmd/)
-
-Game Music
-
-* [Video Game Music Archive](http://www.vgmusic.com/)
-* [THE MIDI SHRINE](http://www.midishrine.com/)
-
-### 2. Create the NoteSequence from MIDI files.
-
-Run the following command to convert the MIDI files to `NoteSequence`.
-
-```
-python scripts/data/create_note_sequences.py
-```
-
-Then, you can find `notesequences.tfrecord` in the `data/interim` folder.
-
-### 3. Convert the NoteSequence to Dataset for Model
-
-We mainly use [`MelodyRNN`](https://github.com/tensorflow/magenta/tree/master/magenta/models/melody_rnn#melody-rnn), So convert the Notesequence by its dataset script.
-
-You have to specify what kinds of model do you use by `--config` argument.
-
-* basic_rnn
-* lookback_rnn
-* attention_rnn
-
-```
-python scripts/data/convert_to_melody_dataset.py  --config attention_rnn
-```
-
-At the same time, dataset is splited to training and evaluation. You can specify its rate by `--eval_ratio` (default is `0.1`).
-
-### 4. Training the Model
-
-If dataset is prepared, you can begin the training!  
-Run the `train_model.py` and specify the model parameters like following.
-
-```
-python scripts/models/train_model.py --config attention_rnn --hparams="{'batch_size':64,'rnn_layer_sizes':[64,64]}" --num_training_steps=20000
-```
-
-([parameter is almost same as original](https://github.com/tensorflow/magenta/tree/master/magenta/models/melody_rnn#train-and-evaluate-the-model))
-
-You can watch the training state by TensorBoard.  
-Run the evaluation script...
-
-```
-python scripts/models/train_model.py --config attention_rnn --hparams="{'batch_size':64,'rnn_layer_sizes':[64,64]}" --num_training_steps=20000 --eval
-```
-
-Then invoke the TensorBoard and access [`http://localhost:6006`](http://localhost:6006).
-
-```
-tensorboard --logdir=models/logdir
-```
-
-![training.PNG](./docs/training.PNG)
-
-### 5. Create the Model
-
-After the training, then create your own model file.
-
-```
-python scripts/models/create_bundle.py --bundle_file my_model
-```
-
-Then, your model is stored in `models/` directory!
-
-### 6. Generate the MIDI files
-
-Now, let's try to generate the MIDI file! You can do it by below script.
-
-```
-python scripts/models/generate_midi.py --bundle_file=my_model --num_outputs=10 --num_steps=128 --primer_melody="[60]"
-```
-
-([parameter is almost same as original](https://github.com/tensorflow/magenta/tree/master/magenta/models/melody_rnn#generate-a-melody))
-
-If it is succeeded, MIDI files will be stored at `data/generated` directory. Sounds Good? Enjoy!
-
-
-### 7. Session with Your Model
-
-I think you want to session with your own Musical Model! 
-If so, set the `MAGENTA_MODEL` environmental variable.
-
-```
-export MAGENTA_MODEL=my_model
-```
-
-Then startup **magenta session server**!
-
-```
-python server/server.py
-```
-
-You can play the Call & Response with your model!
-
 ## Dependencies
 
 Python

diff --git a/docs/process_overview.png b/docs/process_overview.png
diff --git a/scripts/README.md b/scripts/README.md
@@ -0,0 +1,100 @@
+# Train your own model
+
+You can create your own model by your MIDI files!
+The procedure is almost same as [MelodyRNN](https://github.com/tensorflow/magenta/tree/master/magenta/models/melody_rnn). So please refer it as you need.
+
+![process_overview.png](../docs/process_overview.png)
+
+1. Prepare the MIDI data
+2. Create the NoteSequence from MIDI files
+3. Convert the NoteSequence to Dataset for Model
+4. Training the Model
+5. Create the Model
+6. Generate the MIDI files
+
+## 1. Prepare the MIDI data
+
+Please gather your favorite MIDI files and store it to `data/raw`. You can find the MIDI files from following sites.
+
+* [midiworld.com](http://www.midiworld.com/files/142/)
+* [FreeMIDI.org](https://freemidi.org/)
+* [The Lakh MIDI Dataset v0.1](http://colinraffel.com/projects/lmd/)
+
+Game Music
+
+* [Video Game Music Archive](http://www.vgmusic.com/)
+* [THE MIDI SHRINE](http://www.midishrine.com/)
+
+## 2. Create the NoteSequence from MIDI files
+
+Run the following command to convert the MIDI files to `NoteSequence`.
+
+```
+python scripts/data/create_note_sequences.py
+```
+
+Then, you can find `notesequences.tfrecord` in the `data/interim` folder.
+
+## 3. Convert the NoteSequence to Dataset for Model
+
+We mainly use [`MelodyRNN`](https://github.com/tensorflow/magenta/tree/master/magenta/models/melody_rnn#melody-rnn), So convert the Notesequence by its dataset script.
+
+You have to specify what kinds of model do you use by `--config` argument.
+
+* basic_rnn
+* lookback_rnn
+* attention_rnn
+
+```
+python scripts/data/convert_to_melody_dataset.py  --config attention_rnn
+```
+
+At the same time, dataset is splited to training and evaluation. You can specify its rate by `--eval_ratio` (default is `0.1`).
+
+## 4. Training the Model
+
+If dataset is prepared, you can begin the training!  
+Run the `train_model.py` and specify the model parameters like following.
+
+```
+python scripts/models/train_model.py --config attention_rnn --hparams="{'batch_size':64,'rnn_layer_sizes':[64,64]}" --num_training_steps=20000
+```
+
+([parameter is almost same as original](https://github.com/tensorflow/magenta/tree/master/magenta/models/melody_rnn#train-and-evaluate-the-model))
+
+You can watch the training state by TensorBoard.  
+Run the evaluation script...
+
+```
+python scripts/models/train_model.py --config attention_rnn --hparams="{'batch_size':64,'rnn_layer_sizes':[64,64]}" --num_training_steps=20000 --eval
+```
+
+Then invoke the TensorBoard and access [`http://localhost:6006`](http://localhost:6006).
+
+```
+tensorboard --logdir=models/logdir
+```
+
+![training.PNG](./docs/training.PNG)
+
+## 5. Create the Model
+
+After the training, then create your own model file.
+
+```
+python scripts/models/create_bundle.py --bundle_file my_model
+```
+
+Then, your model is stored in `models/` directory!
+
+## 6. Generate the MIDI files
+
+Now, let's try to generate the MIDI file! You can do it by below script.
+
+```
+python scripts/models/generate_midi.py --bundle_file=my_model --num_outputs=10 --num_steps=128 --primer_melody="[60]"
+```
+
+([parameter is almost same as original](https://github.com/tensorflow/magenta/tree/master/magenta/models/melody_rnn#generate-a-melody))
+
+If it is succeeded, MIDI files will be stored at `data/generated` directory. Sounds Good? Enjoy!
diff --git a/server/README.md b/server/README.md
@@ -0,0 +1,16 @@
+# Session with Your own Model
+
+I think you want to session with your own Musical Model! 
+If so, set the `MAGENTA_MODEL` environmental variable.
+
+```
+export MAGENTA_MODEL=my_model
+```
+
+Then startup **magenta session server**!
+
+```
+python server/server.py
+```
+
+You can play the Call & Response with your model!