What type of star is it?¶
Credit: AITS Cainvas Community
Photo by Alex Kunchevsky for OUTLΛNE on Dribbble
Identify the type of star using its characteristics like luminosity, temperature, colour, etc.
In [1]:
import pandas as pd
import numpy as np
from sklearn.model_selection import train_test_split
from sklearn.preprocessing import StandardScaler, MinMaxScaler
from tensorflow.keras import models, optimizers, losses, layers, callbacks
from sklearn.metrics import confusion_matrix
import matplotlib.pyplot as plt
import random
Dataset¶
On Kaggle by Deepraj Baidya | Github
The dataset took 3 weeks to collect for 240 stars which are mostly collected from web. The missing data were manually calculated using equations of astrophysics.
The dataset is a CSV file with characteristics of a star like luminosity, temperature, colour, radius etc that help classify them into one of the 6 classes - Brown Dwarf, Red Dwarf, White Dwarf, Main Sequence, Supergiant, Hypergiant.
In [2]:
df = pd.read_csv('https://cainvas-static.s3.amazonaws.com/media/user_data/cainvas-admin/star.csv')
df
Out[2]:
Preprocessing¶
In [3]:
df['Star color'].value_counts()
Out[3]:
There are many shades of colours mentioned, some similar like Yellowish white and White-Yellow, and many spelling for blue-white.
We can identify 5 basic colours from the given list - blue, white, yellow, orange, red. Lets rewrite the column as 5 columns with multilabel values.
In [4]:
colours = ['Blu', 'Whit', 'Yellow', 'Orang', 'Red'] # using root word of colours as the spelling can differ while specifying shades
df[colours] = 0
for c in colours:
df.loc[df['Star color'].str.contains(c, case = False), c]=1
In [5]:
df['Spectral Class'].value_counts()
Out[5]:
In [6]:
# One hot encoding the input column
df_dummies = pd.get_dummies(df['Spectral Class'], drop_first = True, prefix = 'Spectral')
for column in df_dummies:
df[column] = df_dummies[column]
# One hot encoding the output column
y = pd.get_dummies(df['Star type'])
# Dropping the encoded columns
df = df.drop(columns = ['Spectral Class', 'Star type', 'Star color'])
Looking into the variable value ranges
In [7]:
df.describe()
Out[7]:
The standard deviation values differ across the attributes.
Splitting the dataset before standardization (mean = 0, sd = 1).
Train val test split¶
In [8]:
# Splitting into train, val and test set -- 80-10-10 split
# First, an 80-20 split
Xtrain, X_val_test, ytrain, y_val_test = train_test_split(df, y, test_size = 0.2)
# Then split the 20% into half
Xval, Xtest, yval, ytest = train_test_split(X_val_test, y_val_test, test_size = 0.5)
print("Number of samples in...")
print("Training set: ", len(Xtrain))
print("Validation set: ", len(Xval))
print("Testing set: ", len(Xtest))
Standardization¶
In [9]:
ss = StandardScaler()
Xtrain = ss.fit_transform(Xtrain)
Xval = ss.transform(Xval)
Xtest = ss.transform(Xtest)
The model¶
In [10]:
model = models.Sequential([
layers.Dense(16, activation = 'relu', input_shape = Xtrain[0].shape),
layers.Dense(8, activation = 'relu'),
layers.Dense(6, activation = 'softmax')
])
cb = callbacks.EarlyStopping(patience = 5, restore_best_weights = True)
In [11]:
model.summary()
In [12]:
model.compile(optimizer = optimizers.Adam(0.001), loss = losses.CategoricalCrossentropy(), metrics = ['accuracy'])
history = model.fit(Xtrain, ytrain, validation_data = (Xval, yval), epochs = 256, callbacks = cb)
In [13]:
model.evaluate(Xtest, ytest)
Out[13]:
In [14]:
cm = confusion_matrix(np.argmax(ytest.values, axis = 1), (np.argmax(model.predict(Xtest), axis = 1)))
cm = cm.astype('int') / cm.sum(axis=1)[:, np.newaxis]
fig = plt.figure(figsize = (10, 10))
ax = fig.add_subplot(111)
for i in range(cm.shape[1]):
for j in range(cm.shape[0]):
if cm[i,j] > 0.8:
clr = "white"
else:
clr = "black"
ax.text(j, i, format(cm[i, j], '.2f'), horizontalalignment="center", color=clr)
_ = ax.imshow(cm, cmap=plt.cm.Blues)
ax.set_xticks(range(6))
ax.set_yticks(range(6))
ax.set_xticklabels(range(6))
ax.set_yticklabels(range(6))
plt.xlabel('Predicted')
plt.ylabel('True')
plt.show()
Plotting the metrics¶
In [15]:
def plot(history, variable, variable2):
plt.plot(range(len(history[variable])), history[variable])
plt.plot(range(len(history[variable2])), history[variable2])
plt.legend([variable, variable2])
plt.title(variable)
In [16]:
plot(history.history, "loss", "val_loss")
In [17]:
plot(history.history, "accuracy", "val_accuracy")
Prediction¶
In [18]:
classes = ['Brown Dwarf', 'Red Dwarf', 'White Dwarf', 'Main Sequence', 'Supergiant', 'Hypergiant']
In [19]:
# pick random test data sample from one batch
x = random.randint(0, len(Xtest) - 1)
output = model.predict(Xtest[x].reshape(1, -1))[0]
print("Predicted: ", classes[np.argmax(output)])
print("Probability: ", output[np.argmax(output)])
print("True: ", classes[np.argmax(ytest.values[x])])
deepC¶
In [20]:
model.save('star.h5')
!deepCC star.h5
In [21]:
x = random.randint(0, len(Xtest) - 1)
print(x)
np.savetxt('sample.data', Xtest[x]) # xth sample into text file
# run exe with input
!star_deepC/star.exe sample.data
# show predicted output
nn_out = np.loadtxt('deepSea_result_1.out')
print("Model output: ", model.predict(Xtest[x].reshape(1, -1))[0])
print("deepC output:",nn_out)
print("Predicted: ", classes[np.argmax(nn_out)])
print("Probability: ", nn_out[np.argmax(nn_out)])
print("True: ", classes[np.argmax(ytest.values[x])])