├── data ├── __init__.py ├── raw │ └── putYourDataHere ├── images │ └── imagesAreCreatedHere ├── prepareDownload.py └── prepareRawData.py ├── training ├── __init__.py ├── model.py ├── datapipe.py ├── train.py └── callbacks.py ├── requirements.txt ├── assets ├── model.png └── latentspace.png ├── weightsTrained └── weights_cpk.h5 ├── config.py ├── .gitignore ├── README.md └── inference └── inference.ipynb /data/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /training/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /data/raw/putYourDataHere: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /data/images/imagesAreCreatedHere: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /requirements.txt: -------------------------------------------------------------------------------- 1 | numpy 2 | matplotlib 3 | tensorflow>=2.0 4 | scipy 5 | pandas 6 | beautifulsoup4 -------------------------------------------------------------------------------- /assets/model.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/astridwalle/neural-airfoil-designer/main/assets/model.png -------------------------------------------------------------------------------- /assets/latentspace.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/astridwalle/neural-airfoil-designer/main/assets/latentspace.png -------------------------------------------------------------------------------- /weightsTrained/weights_cpk.h5: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/astridwalle/neural-airfoil-designer/main/weightsTrained/weights_cpk.h5 -------------------------------------------------------------------------------- /config.py: -------------------------------------------------------------------------------- 1 | import argparse 2 | 3 | 4 | def parse_args(): 5 | parser = argparse.ArgumentParser() 6 | 7 | parser.add_argument('--ih', default=256, type=int) 8 | parser.add_argument('--latentDim', default=2, type=int) 9 | parser.add_argument('--batchSize', default=8, type=int) 10 | parser.add_argument('--learnRate', default=5e-5, type=float) 11 | parser.add_argument('--testTrainSplit', default=5e-5, type=float) 12 | parser.add_argument('--epochs', default=3000, type=int) 13 | 14 | arg = parser.parse_args() 15 | 16 | return arg 17 | 18 | -------------------------------------------------------------------------------- /data/prepareDownload.py: -------------------------------------------------------------------------------- 1 | from bs4 import BeautifulSoup 2 | import re 3 | import urllib.request as urllib2 4 | 5 | 6 | basePath = "https://m-selig.ae.illinois.edu/ads/" 7 | 8 | html_page = urllib2.urlopen(f"{basePath}coord_database.html") 9 | soup = BeautifulSoup(html_page, 'lxml') 10 | 11 | pattern = re.compile('\.dat', re.IGNORECASE) 12 | 13 | ind = 1 14 | links = [] 15 | 16 | for link in soup.find_all("a", attrs={'href': pattern}): 17 | links.append(link.get('href')) 18 | 19 | urllib2.urlretrieve(basePath+link.get('href'), "./data/raw/"+link.get('href').rsplit('/',1)[-1]) 20 | 21 | print(link) 22 | 23 | 24 | -------------------------------------------------------------------------------- /.gitignore: -------------------------------------------------------------------------------- 1 | # Byte-compiled / optimized / DLL files 2 | __pycache__/ 3 | *.py[cod] 4 | *$py.class 5 | 6 | # C extensions 7 | *.so 8 | 9 | # Distribution / packaging 10 | .Python 11 | build/ 12 | develop-eggs/ 13 | dist/ 14 | downloads/ 15 | eggs/ 16 | .eggs/ 17 | lib/ 18 | lib64/ 19 | parts/ 20 | sdist/ 21 | var/ 22 | wheels/ 23 | *.egg-info/ 24 | .installed.cfg 25 | *.egg 26 | MANIFEST 27 | 28 | # PyInstaller 29 | # Usually these files are written by a python script from a template 30 | # before PyInstaller builds the exe, so as to inject date/other infos into it. 31 | *.manifest 32 | *.spec 33 | 34 | *.dat 35 | *.json 36 | tblogs/ 37 | images/ 38 | 39 | # Installer logs 40 | pip-log.txt 41 | pip-delete-this-directory.txt 42 | 43 | # Unit test / coverage reports 44 | htmlcov/ 45 | .tox/ 46 | .coverage 47 | .coverage.* 48 | .cache 49 | nosetests.xml 50 | coverage.xml 51 | *.cover 52 | .hypothesis/ 53 | .pytest_cache/ 54 | 55 | # Translations 56 | *.mo 57 | *.pot 58 | 59 | # Django stuff: 60 | *.log 61 | local_settings.py 62 | db.sqlite3 63 | 64 | # Flask stuff: 65 | instance/ 66 | .webassets-cache 67 | 68 | # Scrapy stuff: 69 | .scrapy 70 | 71 | # Sphinx documentation 72 | docs/_build/ 73 | 74 | # PyBuilder 75 | target/ 76 | 77 | # Jupyter Notebook 78 | .ipynb_checkpoints 79 | 80 | # pyenv 81 | .python-version 82 | 83 | # celery beat schedule file 84 | celerybeat-schedule 85 | 86 | # SageMath parsed files 87 | *.sage.py 88 | 89 | # Environments 90 | .env 91 | .venv 92 | env/ 93 | venv/ 94 | ENV/ 95 | env.bak/ 96 | venv.bak/ 97 | 98 | # Spyder project settings 99 | .spyderproject 100 | .spyproject 101 | 102 | # Rope project settings 103 | .ropeproject 104 | 105 | # mkdocs documentation 106 | /site 107 | 108 | # mypy 109 | .mypy_cache/ 110 | -------------------------------------------------------------------------------- /training/model.py: -------------------------------------------------------------------------------- 1 | 2 | import tensorflow as tf 3 | from tensorflow.keras.layers import Conv1D, Lambda,Reshape, Flatten, UpSampling1D, Dense, AveragePooling1D 4 | from datetime import datetime 5 | from keras import backend as K 6 | 7 | 8 | 9 | def createVAEModel(ih,iw,latent_dim): 10 | 11 | 12 | def sampling(args): 13 | z_mean, z_log_sigma = args 14 | epsilon = K.random_normal(shape=(K.shape(z_mean)[0], latent_dim), 15 | mean=0., stddev=0.1) 16 | return z_mean + K.exp(z_log_sigma) * epsilon 17 | 18 | 19 | 20 | # ============================= 21 | # Encoder Model 22 | inputs = tf.keras.layers.Input((ih,iw), name="rgb") 23 | 24 | x = Conv1D(12,3, activation='tanh', padding='same',dilation_rate=2)(inputs) 25 | x1 = AveragePooling1D(2)(x) 26 | x2 = Conv1D(8,3, activation='tanh', padding='same',dilation_rate=2)(x1) 27 | x3 = AveragePooling1D(2)(x2) 28 | x2 = Conv1D(4,3, activation='tanh', padding='same',dilation_rate=2)(x2) 29 | x4 = AveragePooling1D(2)(x2) 30 | h = Flatten()(x4) 31 | 32 | z_mean = Dense(latent_dim)(h) 33 | z_log_sigma = Dense(latent_dim)(h) 34 | z = Lambda(sampling)([z_mean, z_log_sigma]) 35 | 36 | # ============================= 37 | # Decoder Model 38 | l = tf.keras.layers.Input(latent_dim, name="latent") 39 | d1 = Dense(128)(l) 40 | d2 = Reshape((32,4))(d1) 41 | 42 | d3 = Conv1D(4,1,strides=1, activation='tanh', padding='same')(d2) 43 | d4 = UpSampling1D(2)(d3) 44 | 45 | d40 = Conv1D(4,1,strides=1, activation='tanh', padding='same')(d4) 46 | d5 = UpSampling1D(2)(d40) 47 | 48 | d5 = Conv1D(8,1,strides=1, activation='tanh', padding='same')(d5) 49 | d6 = UpSampling1D(2)(d5) 50 | 51 | decoded = Conv1D(4,1,strides=1, activation='linear', padding='same')(d6) 52 | 53 | 54 | 55 | encoder = tf.keras.Model(inputs=[inputs], outputs=[z_mean, z_log_sigma, z], name='encoder') 56 | decoder = tf.keras.Model(inputs=[l], outputs=[decoded], name='decoder') 57 | 58 | 59 | # ============================= 60 | # VAE Model 61 | outputs = decoder(encoder(inputs)[2]) 62 | 63 | model = tf.keras.Model(inputs, outputs, name='autoencoder') 64 | 65 | # ============================= 66 | # Add VAE loss 67 | 68 | reconstruction_loss = 0.5*K.sum(K.square(inputs-outputs))/0.01 69 | 70 | kl_loss = 1 + z_log_sigma - K.square(z_mean) - K.exp(z_log_sigma) 71 | kl_loss = K.sum(kl_loss, axis=-1) 72 | kl_loss *= -0.5 73 | vae_loss = K.mean(reconstruction_loss + kl_loss) 74 | model.add_loss(vae_loss) 75 | 76 | return model, encoder, decoder 77 | -------------------------------------------------------------------------------- /training/datapipe.py: -------------------------------------------------------------------------------- 1 | import glob 2 | import tensorflow as tf 3 | import matplotlib.pyplot as plt 4 | import json 5 | import os 6 | import numpy as np 7 | 8 | 9 | 10 | 11 | 12 | 13 | class Datapipe: 14 | def __init__(self, path): 15 | 16 | 17 | self.filenames = glob.glob(path) 18 | 19 | 20 | # ================================================= 21 | @staticmethod 22 | def readJson(jsonfile): 23 | 24 | 25 | with open(jsonfile, 'r') as f1: 26 | data = json.load(f1) 27 | 28 | 29 | name = data["name"] 30 | ss = np.asarray(data["ss"], dtype=np.float32) 31 | ps = np.asarray(data["ps"], dtype=np.float32) 32 | 33 | 34 | geom = np.vstack((ss,ps)).T 35 | 36 | return name, geom 37 | 38 | # ================================================= 39 | def autoencoder(self, name, geom): 40 | return geom, geom 41 | 42 | def scaleer(self, name, geom): 43 | 44 | geom = geom*tf.constant([1.0, 5.0, 1.0, -5.0], dtype=tf.float32) + tf.constant([[-0.5, 0.0, -0.5, -0.0]], dtype=tf.float32) 45 | geom = geom*0.1 46 | 47 | return name, geom 48 | # ================================================= 49 | def create(self, split=0.1, batchSize=10): 50 | 51 | dataset = tf.data.Dataset.from_tensor_slices(self.filenames) 52 | 53 | dataset = dataset.shuffle(buffer_size=5000) 54 | dataset = dataset.repeat(1) 55 | 56 | # Load the Json 57 | dataset = dataset.map(self._loadJson) 58 | 59 | dataset = dataset.map(self.scaleer) 60 | dataset = dataset.map(self.autoencoder) 61 | 62 | 63 | train_size = int((1-split) * len(self.filenames)) 64 | 65 | 66 | train_dataset = dataset.take(train_size) 67 | test_dataset = dataset.skip(train_size) 68 | 69 | 70 | train_dataset = train_dataset.batch(batchSize) 71 | test_dataset = test_dataset.batch(batchSize) 72 | 73 | return train_dataset, test_dataset 74 | 75 | # ============================ 76 | def _loadJson(self, jsonfile): 77 | 78 | def _pyLoadJson(content): 79 | return Datapipe.readJson( 80 | jsonfile=content.numpy().decode("utf-8"), 81 | ) 82 | 83 | 84 | name, geom = tf.py_function( 85 | _pyLoadJson, [jsonfile], [tf.string, tf.float32] 86 | ) 87 | 88 | return name, geom 89 | 90 | 91 | 92 | if __name__ == "__main__": 93 | 94 | 95 | dp = Datapipe("database_processed/*.json") 96 | 97 | g, gt = dp.create(split=0.3) 98 | 99 | 100 | for (x,y) in g.take(4): 101 | print(x) 102 | print(y) 103 | 104 | 105 | plt.plot(y[0,:,0], y[0,:,1],'b-') 106 | plt.plot(y[0,:,2], y[0,:,3],'r-') 107 | 108 | plt.show() 109 | 110 | plt.imshow(y[0,...], aspect="auto") 111 | 112 | plt.show() -------------------------------------------------------------------------------- /training/train.py: -------------------------------------------------------------------------------- 1 | import glob 2 | import tensorflow as tf 3 | import matplotlib.pyplot as plt 4 | import json 5 | import os 6 | import sys 7 | import numpy as np 8 | from tensorflow.keras.layers import Dropout, BatchNormalization, Conv1D, Lambda, MaxPooling1D, Reshape, BatchNormalization, Flatten, UpSampling1D, Dense, AveragePooling1D 9 | from datetime import datetime 10 | from tensorflow.keras import regularizers 11 | from keras import backend as K 12 | 13 | sys.path.insert(0, os.path.abspath(os.path.join(os.path.dirname(__file__), '..'))) 14 | 15 | from training.datapipe import Datapipe 16 | from training.callbacks import DrawImageCallback 17 | from training.model import createVAEModel 18 | from config import parse_args 19 | 20 | 21 | 22 | def main(arg): 23 | 24 | 25 | ih, iw = arg.ih, 4 26 | learnrate = arg.learnRate 27 | batchSize = arg.batchSize 28 | 29 | 30 | dp = Datapipe("./data/processed/*.json") 31 | g, gt = dp.create(split= arg.testTrainSplit, batchSize=batchSize) 32 | 33 | latent_dim = arg.latentDim 34 | 35 | 36 | # ============================= 37 | # VAE Model 38 | 39 | model, encoder, decoder = createVAEModel(ih,iw,latent_dim) 40 | 41 | print(model.summary()) 42 | 43 | opti = tf.keras.optimizers.Adam(learnrate) 44 | 45 | model.compile(optimizer=opti) 46 | 47 | 48 | #model.load_weights("weights_cpk.h5") 49 | 50 | 51 | # ================================ 52 | now = datetime.now() 53 | #timestamp = datetime.timestamp(now) 54 | timestamp = now.strftime('%Y-%m-%d-%H%M%S') 55 | 56 | drwcb = DrawImageCallback(logdir=f"./tblogs/ae/{timestamp}",tfdataset=gt, encoder=encoder) 57 | 58 | 59 | tfbcb = tf.keras.callbacks.TensorBoard( 60 | log_dir=f"./tblogs/ae/{timestamp}", histogram_freq=0, write_graph=True, 61 | write_images=True, update_freq='batch', 62 | profile_batch=2, embeddings_freq=0, embeddings_metadata=None 63 | ) 64 | 65 | estcb = tf.keras.callbacks.EarlyStopping( 66 | monitor='val_loss', min_delta=0, patience=140, verbose=0, 67 | mode='auto', baseline=None, restore_best_weights=True 68 | ) 69 | 70 | mcpcb = tf.keras.callbacks.ModelCheckpoint( 71 | os.path.join('weights_cpk.h5'), monitor='val_loss', verbose=0, save_best_only=True, 72 | save_weights_only=True, mode='auto', save_freq='epoch', 73 | ) 74 | 75 | rlrcb = tf.keras.callbacks.ReduceLROnPlateau( 76 | monitor='val_loss', 77 | factor=0.5, 78 | patience=35, 79 | verbose=0, 80 | mode='auto', 81 | min_delta=0.0001, 82 | cooldown=0, 83 | min_lr=0, 84 | ) 85 | 86 | term = tf.keras.callbacks.TerminateOnNaN() 87 | 88 | # ================================ 89 | 90 | 91 | model.fit( 92 | g, epochs=arg.epochs, 93 | callbacks = [tfbcb, mcpcb, estcb, rlrcb, term, drwcb], 94 | validation_data=gt 95 | ) 96 | 97 | model.save_weights("weights_final.h5") 98 | 99 | if __name__ == "__main__": 100 | arg = parse_args() 101 | main(arg) -------------------------------------------------------------------------------- /training/callbacks.py: -------------------------------------------------------------------------------- 1 | import os 2 | import tensorflow as tf 3 | from matplotlib.backends.backend_agg import FigureCanvasAgg 4 | from matplotlib.figure import Figure 5 | import numpy as np 6 | 7 | 8 | 9 | 10 | 11 | 12 | class DrawImageCallback(tf.keras.callbacks.Callback): 13 | def __init__( 14 | self, 15 | logdir, 16 | tfdataset, 17 | encoder, 18 | writerName="imager", 19 | ): 20 | super(DrawImageCallback, self).__init__() 21 | 22 | self.tbcb = tf.summary.create_file_writer(os.path.join(logdir, writerName)) 23 | self.writerName = writerName 24 | self.step_number = 0 25 | self.tfdataset = tfdataset 26 | self.encoder = encoder 27 | 28 | def on_epoch_end(self, epoch, logs=None): 29 | """Draw images at the end of an epoche 30 | 31 | Args: 32 | epoch ([type]): [description] 33 | logs ([type], optional): [description]. Defaults to None. 34 | """ 35 | 36 | 37 | latent = [] 38 | for (x, ytrue) in self.tfdataset: 39 | latent.append(self.encoder.predict(x)[2]) 40 | 41 | latent = np.vstack(latent) 42 | 43 | 44 | fig = Figure(figsize=(5, 4), dpi=100) 45 | canvas = FigureCanvasAgg(fig) 46 | 47 | # Do some plotting here 48 | ax = fig.add_subplot(111) 49 | 50 | ax.scatter(latent[:,0], latent[:,1]) # latent[:,2] 51 | 52 | ax.grid(True) 53 | 54 | # Retrieve a view on the renderer buffer 55 | canvas.draw() 56 | latentImgs = np.expand_dims(np.asarray(canvas.buffer_rgba()),0) 57 | # convert to a NumPy array 58 | 59 | 60 | 61 | 62 | x, ytrue = None, None 63 | 64 | 65 | for (x, ytrue) in self.tfdataset: 66 | ypred = self.model.predict(x) 67 | 68 | break 69 | 70 | 71 | imgs = [] 72 | 73 | for i in range(x.shape[0]): 74 | # make a Figure and attach it to a canvas. 75 | fig = Figure(figsize=(5, 4), dpi=100) 76 | canvas = FigureCanvasAgg(fig) 77 | 78 | # Do some plotting here 79 | ax = fig.add_subplot(111) 80 | 81 | 82 | ax.plot(ytrue[i,:,0], ytrue[i,:,1],'b--') 83 | ax.plot(ytrue[i,:,2], ytrue[i,:,3],'r--') 84 | 85 | ax.plot(ypred[i,:,0], ypred[i,:,1],'b-') 86 | ax.plot(ypred[i,:,2], ypred[i,:,3],'r-') 87 | ax.grid(True) 88 | ax.axis([-0.5*0.1,0.5*0.1,-0.3*0.1,0.7*0.1]) 89 | # Retrieve a view on the renderer buffer 90 | canvas.draw() 91 | buf = np.asarray(canvas.buffer_rgba()) 92 | # convert to a NumPy array 93 | imgs.append(np.expand_dims(buf,0)) 94 | 95 | imgs = np.vstack(imgs) 96 | 97 | 98 | with self.tbcb.as_default(): 99 | 100 | tf.summary.image( 101 | "Images", imgs[...,:3], max_outputs=25, step=self.step_number 102 | ) 103 | tf.summary.image( 104 | "LatentSpace", latentImgs[...,:3], max_outputs=25, step=self.step_number 105 | ) 106 | 107 | 108 | self.step_number += 1 -------------------------------------------------------------------------------- /README.md: -------------------------------------------------------------------------------- 1 | # Neural Airfoil Generator 2 | 3 | Generate airfoils with the help of a varational autoencoder VAE. 4 | 5 | - Uses public database 6 | 7 | - Training of a VAE to predict shapes 8 | 9 | You can specify the following parameters: 10 | - `--ih`: Resolution of pressure and suction side (int) 11 | - `--learnRate`: learning rate (float) 12 | - `--testTrainSplit`: Test train split (float) 13 | - `--epochs`: number of training epochs (int) 14 | - `--latentDim`: latent dimensions (int) 15 | - `--batchSize`: batch size (int) 16 | 17 | 18 | ![Model](assets/model.png). 19 | ![Samples from Latentspace](assets/latentspace.png). 20 | 21 | ## Requirements 22 | 23 | Install the requirements with: 24 | 25 | `pip install -r requirements.txt` 26 | 27 | 28 | ## Get the dataset 29 | 30 | Data are taken from here: [Illinois Airfoil Database](https://m-selig.ae.illinois.edu/ads/coord_database.html). 31 | 32 | To prepare the dataset run from project root dir: 33 | 34 | python3 ./data/prepareDownload.py 35 | 36 | This script will download all .dat files and place them in ./data/raw. For more information on the download script see here: [Josh the engineer](https://www.youtube.com/watch?v=nILo18DlqAo). To prepare the raw data run 37 | 38 | python3 ./data/prepareRawData.py 39 | 40 | in the same folder (./data). This will generated a json file for each .dat file with the following structure: 41 | 42 | name: "name of the airfoil", 43 | ss: [List x coordinates, List y coordinates], 44 | ps: [List x coordinates, List y coordinates], 45 | 46 | 47 | ## Training 48 | 49 | To train the variational autoencoder run 50 | 51 | python training/train.py 52 | 53 | By default, the model is trained with a batch norm of 8, learning rate of 1e-4, 3000 epochs and saved as `weights-cpk.h5`. 54 | 55 | 56 | __________________________________________________________________________________________________ 57 | Layer (type) Output Shape Param # Connected to 58 | ================================================================================================== 59 | rgb (InputLayer) [(None, 256, 4)] 0 [] 60 | 61 | encoder (Functional) [(None, 2), 1580 ['rgb[0][0]'] 62 | (None, 2), 63 | (None, 2)] 64 | 65 | decoder (Functional) (None, 256, 4) 500 ['encoder[0][2]'] 66 | 67 | conv1d (Conv1D) (None, 256, 12) 156 ['rgb[0][0]'] 68 | 69 | average_pooling1d (AveragePool (None, 128, 12) 0 ['conv1d[0][0]'] 70 | ing1D) 71 | 72 | conv1d_1 (Conv1D) (None, 128, 8) 296 ['average_pooling1d[0][0]'] 73 | 74 | conv1d_2 (Conv1D) (None, 128, 4) 100 ['conv1d_1[0][0]'] 75 | 76 | average_pooling1d_2 (AveragePo (None, 64, 4) 0 ['conv1d_2[0][0]'] 77 | oling1D) 78 | 79 | flatten (Flatten) (None, 256) 0 ['average_pooling1d_2[0][0]'] 80 | 81 | dense_1 (Dense) (None, 2) 514 ['flatten[0][0]'] 82 | 83 | dense (Dense) (None, 2) 514 ['flatten[0][0]'] 84 | 85 | tf.math.subtract (TFOpLambda) (None, 256, 4) 0 ['rgb[0][0]', 86 | 'decoder[0][0]'] 87 | 88 | tf.__operators__.add (TFOpLamb (None, 2) 0 ['dense_1[0][0]'] 89 | da) 90 | 91 | tf.math.square_1 (TFOpLambda) (None, 2) 0 ['dense[0][0]'] 92 | 93 | tf.math.square (TFOpLambda) (None, 256, 4) 0 ['tf.math.subtract[0][0]'] 94 | 95 | tf.math.subtract_1 (TFOpLambda (None, 2) 0 ['tf.__operators__.add[0][0]', 96 | ) 'tf.math.square_1[0][0]'] 97 | 98 | tf.math.exp (TFOpLambda) (None, 2) 0 ['dense_1[0][0]'] 99 | 100 | tf.math.reduce_sum (TFOpLambda () 0 ['tf.math.square[0][0]'] 101 | ) 102 | 103 | tf.math.subtract_2 (TFOpLambda (None, 2) 0 ['tf.math.subtract_1[0][0]', 104 | ) 'tf.math.exp[0][0]'] 105 | 106 | tf.math.multiply (TFOpLambda) () 0 ['tf.math.reduce_sum[0][0]'] 107 | 108 | tf.math.reduce_sum_1 (TFOpLamb (None,) 0 ['tf.math.subtract_2[0][0]'] 109 | da) 110 | 111 | tf.math.truediv (TFOpLambda) () 0 ['tf.math.multiply[0][0]'] 112 | 113 | tf.math.multiply_1 (TFOpLambda (None,) 0 ['tf.math.reduce_sum_1[0][0]'] 114 | ) 115 | 116 | tf.__operators__.add_1 (TFOpLa (None,) 0 ['tf.math.truediv[0][0]', 117 | mbda) 'tf.math.multiply_1[0][0]'] 118 | 119 | tf.math.reduce_mean (TFOpLambd () 0 ['tf.__operators__.add_1[0][0]'] 120 | a) 121 | 122 | add_loss (AddLoss) () 0 ['tf.math.reduce_mean[0][0]'] 123 | 124 | ================================================================================================== 125 | Total params: 2,080 126 | Trainable params: 2,080 127 | Non-trainable params: 0 128 | __________________________________________________________________________________________________ 129 | 130 | 131 | ## Inference 132 | 133 | You can test your model using the file `inference.ipynb`. Use the following prompt to predict the rotation of a specific image: 134 | 135 | -------------------------------------------------------------------------------- /data/prepareRawData.py: -------------------------------------------------------------------------------- 1 | # ================ 2 | import numpy as np 3 | import scipy.interpolate as si 4 | import matplotlib.pyplot as plt 5 | import pandas as pd 6 | import glob 7 | import os 8 | import sys 9 | import json 10 | from scipy.interpolate import splprep, splev 11 | 12 | sys.path.insert(0, os.path.abspath(os.path.join(os.path.dirname(__file__), '..'))) 13 | 14 | from config import parse_args 15 | 16 | 17 | def main(arg): 18 | 19 | for ctr,d in enumerate(glob.glob("./data/raw/*.dat")): 20 | 21 | df = pd.read_table(d, sep="\s+", skipfooter=1) 22 | 23 | 24 | if len(df.columns) == 1: 25 | df = df.reset_index(level=0) 26 | 27 | print(df) 28 | 29 | if True: 30 | # df.columns = ["x", "y"] 31 | 32 | try: 33 | df = df[df.iloc[:,0].notna()] 34 | df = df[df.iloc[:,1].notna()] 35 | df = df[df.iloc[:,0].apply(lambda x: isinstance(x, (int, np.int64, float)))] 36 | df = df[df.iloc[:,1].apply(lambda x: isinstance(x, (int, np.int64, float)))] 37 | df = df[(df.iloc[:,0]<=1.0) & (df.iloc[:,1]<=1.0) & (df.iloc[:,0]>=0.0) & (df.iloc[:,1]>=-1.0)] 38 | 39 | except: 40 | print(f"Couldnt process {d}") 41 | 42 | 43 | 44 | ntepts = 0 45 | eps = 1e-8 46 | 47 | print("=========",d,"========") 48 | 49 | while ntepts == 0: 50 | 51 | df["teidx"] = np.abs(df.iloc[:,0]-1.0)< eps 52 | ntepts = df.teidx.sum() 53 | eps *= 1.1 54 | 55 | 56 | nlepts = 0 57 | eps = 1e-8 58 | 59 | while nlepts == 0: 60 | 61 | df["leidx"] = np.abs(df.iloc[:,0]) < eps 62 | 63 | nlepts = df.leidx.sum() 64 | eps *= 1.1 65 | 66 | 67 | print(ntepts, nlepts) 68 | 69 | 70 | # Split into SS and PS 71 | x = df.iloc[:,0].to_numpy() 72 | y = df.iloc[:,1].to_numpy() 73 | 74 | teidx = df.index[df['teidx']].tolist() 75 | leidx = df.index[df['leidx']].tolist() 76 | 77 | 78 | if ntepts == 2 and nlepts ==1: 79 | 80 | x1, x2 = x[teidx[0]:leidx[0]+1][::-1], x[leidx[0]:] 81 | y1, y2 = y[teidx[0]:leidx[0]+1][::-1], y[leidx[0]:] 82 | 83 | 84 | # Case 2 85 | elif ntepts == 2 and nlepts ==2: 86 | 87 | # [1, 50] [49, 98] 88 | 89 | # [60, 61] [0, 119] 90 | 91 | 92 | # Case 2a 93 | if leidx[1] - teidx[0] == 1: 94 | 95 | x1, x2 = x[leidx[0]-1:teidx[0]], x[leidx[1]-1:] 96 | y1, y2 = y[leidx[0]-1:teidx[0]], y[leidx[1]-1:] 97 | case = "2a" 98 | 99 | # Case 2b 100 | elif leidx[1] - leidx[0] == 1: 101 | 102 | x1, x2 = x[teidx[0]:leidx[0]+1][::-1], x[leidx[1]:] 103 | y1, y2 = y[teidx[0]:leidx[0]+1][::-1], y[leidx[1]:] 104 | case = "2b" 105 | 106 | 107 | 108 | 109 | # Case 3 110 | elif ntepts == 1 and nlepts ==2: 111 | 112 | print(leidx, teidx) 113 | # 3a: [57, 58] [1] 114 | # 3b: [1, 68] [67] 115 | # 3c: [1, 41] [81] , [1, 36] [67] 116 | # 117 | if leidx[1]-leidx[0] == 1: 118 | 119 | x1, x2 = x[teidx[0]:leidx[0]][::-1], x[leidx[1]:] 120 | y1, y2 = y[teidx[0]:leidx[0]][::-1], y[leidx[1]:] 121 | case = "3a" 122 | 123 | elif leidx[1]-teidx[0] == 1: 124 | 125 | x1, x2 = x[leidx[0]:teidx[0]], x[leidx[1]:] 126 | y1, y2 = y[leidx[0]:teidx[0]], y[leidx[1]:] 127 | case = "3b" 128 | 129 | else: 130 | 131 | teidx = [leidx[1]-1, teidx[0]] 132 | 133 | 134 | # Case 2a 135 | if leidx[1] - teidx[0] == 1: 136 | 137 | x1, x2 = x[leidx[0]-1:teidx[0]-1], x[leidx[1]-1:] 138 | y1, y2 = y[leidx[0]-1:teidx[0]-1], y[leidx[1]-1:] 139 | 140 | case = "2a" 141 | 142 | # Case 2b 143 | elif leidx[1] - leidx[0] == 1: 144 | 145 | x1, x2 = x[teidx[0]:leidx[0]+1][::-1], x[leidx[1]:] 146 | y1, y2 = y[teidx[0]:leidx[0]+1][::-1], y[leidx[1]:] 147 | case = "2b" 148 | 149 | 150 | 151 | legap = np.sqrt((x1[0] - x2[0])**2 + (y1[0] - y2[0])**2) 152 | tegap = np.sqrt((x1[-1] - x2[-1])**2 + (y1[-1] - y2[-1])**2) 153 | 154 | if legap < 1e-4: 155 | pass 156 | elif legap < 0.14: 157 | xmean = 0.5*(x1[0]+ x2[0]) 158 | ymean = 0.5*(y1[0]+ y2[0]) 159 | 160 | x1 = np.concatenate((np.asarray([xmean]), x1)) 161 | x2 = np.concatenate((np.asarray([xmean]), x2)) 162 | y1 = np.concatenate((np.asarray([ymean]), y1)) 163 | y2 = np.concatenate((np.asarray([ymean]), y2)) 164 | 165 | else: 166 | print("Warning:", d) 167 | continue 168 | 169 | if tegap < 1e-4: 170 | pass 171 | elif tegap < 0.14: 172 | xmean = 0.5*(x1[-1]+ x2[-1]) 173 | ymean = 0.5*(y1[-1]+ y2[-1]) 174 | 175 | x1 = np.concatenate((x1, np.asarray([xmean]))) 176 | x2 = np.concatenate((x2, np.asarray([xmean]))) 177 | y1 = np.concatenate((y1, np.asarray([ymean]))) 178 | y2 = np.concatenate((y2, np.asarray([ymean]))) 179 | 180 | else: 181 | print("Warning:", d) 182 | continue 183 | 184 | # Spline reconstruction 185 | try: 186 | tck1, _ = splprep([x1, y1], s=0,k=1) 187 | tck2, _ = splprep([x2, y2], s=0,k=1) 188 | 189 | u = np.linspace(0,1, arg.ih) 190 | 191 | xss = splev(u, tck1) 192 | xps = splev(u, tck2) 193 | 194 | 195 | except: 196 | print("Spline Error" , d) 197 | continue 198 | 199 | fp, fn = os.path.split(d) 200 | name = fn.split('.')[0] 201 | 202 | plt.plot(x1,y1,'b-') 203 | plt.plot(x2,y2,'r-') 204 | plt.plot(x1[0],y1[0],'bs') 205 | plt.plot(x2[0],y2[0],'rs') 206 | plt.plot(x1[-1],y1[-1],'bo') 207 | plt.plot(x2[-1],y2[-1],'ro') 208 | plt.plot(xss[0], xss[1], 'k--') 209 | plt.plot(xps[0], xps[1], 'k--') 210 | 211 | plt.axis("equal") 212 | plt.title(f"{d} {case}") 213 | plt.savefig(f"./data/images/{name}.png") 214 | 215 | 216 | plt.close() 217 | 218 | data = {"name": name, "ss": [xss[0].tolist(), xss[1].tolist()], "ps": [xps[0].tolist(), xps[1].tolist()]} 219 | 220 | with open(f"./data/processed/{name}.json", "w") as outfile: 221 | outfile.write(json.dumps(data, indent=4)) 222 | 223 | 224 | if __name__ == "__main__": 225 | arg = parse_args() 226 | main(arg) -------------------------------------------------------------------------------- /inference/inference.ipynb: -------------------------------------------------------------------------------- 1 | { 2 | "cells": [ 3 | { 4 | "cell_type": "code", 5 | "execution_count": 15, 6 | "metadata": {}, 7 | "outputs": [ 8 | { 9 | "name": "stdout", 10 | "output_type": "stream", 11 | "text": [ 12 | "c:\\Users\\z004hkut\\projects\\20_tda\\neural-airfoil-generator\\inference\n" 13 | ] 14 | } 15 | ], 16 | "source": [ 17 | "import glob\n", 18 | "import tensorflow as tf\n", 19 | "import matplotlib.pyplot as plt\n", 20 | "import json\n", 21 | "import os\n", 22 | "\n", 23 | "\n", 24 | "from ipywidgets import *\n", 25 | "import numpy as np\n", 26 | "import matplotlib.pyplot as plt\n", 27 | "\n", 28 | "\n", 29 | "import numpy as np\n", 30 | "from tensorflow.keras.layers import Dropout, BatchNormalization, Conv1D, Lambda, MaxPooling1D, Reshape, BatchNormalization, Flatten, UpSampling1D, Dense, AveragePooling1D\n", 31 | "from datetime import datetime\n", 32 | "from tensorflow.keras import regularizers\n", 33 | "from keras import backend as K\n", 34 | "\n", 35 | "\n", 36 | "from matplotlib.widgets import Slider, Button, RadioButtons\n", 37 | "\n", 38 | "\n", 39 | "\n", 40 | "\n", 41 | "import sys\n", 42 | "\n", 43 | "print(os.getcwd())\n", 44 | "sys.path.insert(0, '..')\n", 45 | "\n", 46 | "from training.datapipe import Datapipe\n", 47 | "from training.callbacks import DrawImageCallback\n", 48 | "from training.model import createVAEModel\n", 49 | "\n" 50 | ] 51 | }, 52 | { 53 | "cell_type": "code", 54 | "execution_count": 7, 55 | "metadata": {}, 56 | "outputs": [], 57 | "source": [ 58 | "\n", 59 | "# The parameters of the config file\n", 60 | "ih, iw = 256, 4\n", 61 | "latent_dim = 2\n", 62 | "\n", 63 | "\n", 64 | "# Load model and weights\n", 65 | "model, encoder, decoder = createVAEModel(ih,iw,latent_dim)\n", 66 | "model.load_weights(\"../weightsTrained/weights_cpk.h5\")\n" 67 | ] 68 | }, 69 | { 70 | "cell_type": "code", 71 | "execution_count": 42, 72 | "metadata": {}, 73 | "outputs": [ 74 | { 75 | "data": { 76 | "application/vnd.jupyter.widget-view+json": { 77 | "model_id": "466f04b8f0b145f5a95b0af46bf8c455", 78 | "version_major": 2, 79 | "version_minor": 0 80 | }, 81 | "text/plain": [ 82 | "interactive(children=(FloatSlider(value=0.0, description='z1', max=7.0, min=-7.0, step=0.02), FloatSlider(valu…" 83 | ] 84 | }, 85 | "metadata": {}, 86 | "output_type": "display_data" 87 | } 88 | ], 89 | "source": [ 90 | "%matplotlib inline\n", 91 | "from IPython.display import clear_output\n", 92 | "\n", 93 | "\n", 94 | "\n", 95 | "@widgets.interact(z1=(-7,7,0.02), z2=(-7,7,0.02))\n", 96 | "def foo(z1,z2):\n", 97 | " \"\"\"\n", 98 | " Print the current widget value in short sentence\n", 99 | " \"\"\"\n", 100 | " clear_output(wait=True)\n", 101 | "\n", 102 | " x = np.asarray([[z1,z2]])\n", 103 | "\n", 104 | " geoms = decoder(x)\n", 105 | "\n", 106 | " fig = plt.figure()\n", 107 | " ax = fig.add_subplot(111)\n", 108 | " line1, = ax.plot(geoms[0,:,0], geoms[0,:,1]/5, 'b-') # Returns a tuple of line objects, thus the comma\n", 109 | " line2, = ax.plot(geoms[0,:,2], geoms[0,:,3]/5, 'r-') # Returns a tuple of line objects, thus the comma\n", 110 | "\n", 111 | " plt.axis(\"equal\")\n", 112 | " #plt.axis([0.05,-0.05,-0.01,0.01])\n", 113 | "\n", 114 | " plt.grid(True)\n", 115 | " plt.xlabel('axis x')\n", 116 | " plt.ylabel('axis y')\n", 117 | " plt.show()\n", 118 | " \n" 119 | ] 120 | }, 121 | { 122 | "cell_type": "code", 123 | "execution_count": null, 124 | "metadata": {}, 125 | "outputs": [], 126 | "source": [] 127 | }, 128 | { 129 | "cell_type": "code", 130 | "execution_count": 41, 131 | "metadata": {}, 132 | "outputs": [ 133 | { 134 | "data": { 135 | "image/png": "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", 136 | "text/plain": [ 137 | "
" 138 | ] 139 | }, 140 | "metadata": {}, 141 | "output_type": "display_data" 142 | } 143 | ], 144 | "source": [ 145 | "\n", 146 | "\n", 147 | "x = np.random.normal(size=(10,2))\n", 148 | "\n", 149 | "geoms = decoder(x)\n", 150 | "\n", 151 | "w=2\n", 152 | "h=5\n", 153 | "\n", 154 | "fig, axs = plt.subplots(h, w, figsize=(10,10))\n", 155 | "\n", 156 | "\n", 157 | "for idx in range(geoms.shape[0]):\n", 158 | " j,i = idx%w, idx//w\n", 159 | " #axs[i,j].title(f\"Latent Space {np.around(x[idx,:],2)}\")\n", 160 | " axs[i,j].plot(geoms[idx,:,0], geoms[idx,:,1]/5,'b-')\n", 161 | " axs[i,j].plot(geoms[idx,:,2], -geoms[idx,:,3]/5,'r-')\n", 162 | " axs[i,j].axis(\"equal\")\n", 163 | " axs[i,j].axis('off')\n", 164 | " axs[i,j].title.set_text(f\"Latent Space {np.around(x[idx,:],2)}\")\n", 165 | "\n", 166 | "plt.show()" 167 | ] 168 | }, 169 | { 170 | "cell_type": "code", 171 | "execution_count": 28, 172 | "metadata": {}, 173 | "outputs": [ 174 | { 175 | "name": "stdout", 176 | "output_type": "stream", 177 | "text": [ 178 | "0 0\n", 179 | "0 1\n", 180 | "1 0\n", 181 | "1 1\n", 182 | "2 0\n", 183 | "2 1\n", 184 | "3 0\n", 185 | "3 1\n", 186 | "4 0\n", 187 | "4 1\n" 188 | ] 189 | }, 190 | { 191 | "data": { 192 | "image/png": "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", 193 | "text/plain": [ 194 | "
" 195 | ] 196 | }, 197 | "metadata": {}, 198 | "output_type": "display_data" 199 | } 200 | ], 201 | "source": [] 202 | }, 203 | { 204 | "cell_type": "code", 205 | "execution_count": null, 206 | "metadata": {}, 207 | "outputs": [], 208 | "source": [] 209 | } 210 | ], 211 | "metadata": { 212 | "kernelspec": { 213 | "display_name": "Python 3.8.7 ('venv': venv)", 214 | "language": "python", 215 | "name": "python3" 216 | }, 217 | "language_info": { 218 | "codemirror_mode": { 219 | "name": "ipython", 220 | "version": 3 221 | }, 222 | "file_extension": ".py", 223 | "mimetype": "text/x-python", 224 | "name": "python", 225 | "nbconvert_exporter": "python", 226 | "pygments_lexer": "ipython3", 227 | "version": "3.8.7" 228 | }, 229 | "orig_nbformat": 4, 230 | "vscode": { 231 | "interpreter": { 232 | "hash": "db564559e495a4953c5364f7c8b05ad4aa98284453e9755dc733f207a2f5474d" 233 | } 234 | } 235 | }, 236 | "nbformat": 4, 237 | "nbformat_minor": 2 238 | } 239 | --------------------------------------------------------------------------------