BITCOIN.PY

import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sb

from sklearn.model_selection import train_test_split
from sklearn.preprocessing import StandardScaler
from sklearn.linear_model import LogisticRegression
from sklearn.svm import SVC
from xgboost import XGBClassifier
from sklearn import metrics

import warnings
warnings.filterwarnings('ignore')

# Load the dataset
df = pd.read_csv('bitcoin.csv')
print(df.head())
print(df.shape)
print(df.describe())
print(df.info())

# Plotting the 'close' prices
plt.figure(figsize=(15, 5))
plt.plot(df['close'])
plt.title('Bitcoin Close Price', fontsize=15)
plt.ylabel('Price in dollars')
plt.show()

# Checking for null values
print("Check if data is null:")
print(df.isnull().sum())

# Defining the features to visualize
features = ['open', 'high', 'low', 'close']

# Plotting distributions of features
plt.figure(figsize=(20, 10))
for i, col in enumerate(features):
    plt.subplot(2, 2, i+1)
    sb.distplot(df[col])
plt.show()

# Plotting distributions of features with a filtered range
plt.figure(figsize=(20, 10))
for i, col in enumerate(features):
    range_min = 0.0
    range_max = 10000.0
    filtered_data = [x for x in df[col] if range_min <= x <= range_max]
    plt.subplot(2, 2, i+1)
    sb.distplot(filtered_data)
plt.show()

# Plotting boxplots of features
plt.figure(figsize=(20, 10))
for i, col in enumerate(features):
    plt.subplot(2, 2, i+1)
    ax = sb.boxplot(df[col])
    q1, median, q3 = df[col].quantile([0.25, 0.5, 0.75])
    label_text = f"25%: {q1:.2f}   median: {median:.2f}   75%: {q3:.2f}"
    plt.text(20000, -0.35, label_text, fontsize=12)
    IQR = q3 - q1
    k = 1.5  # Adjust this value if needed
    lower_fence = q1 - k * IQR
    upper_fence = q3 + k * IQR
    label_fence = f"Fence line: {upper_fence:.2f}"
    plt.text(20000, -0.25, label_fence, fontsize=12)
plt.show()

# Splitting the date column and handling time separately
df['date'] = pd.to_datetime(df['date'])
df['year'] = df['date'].dt.year
df['month'] = df['date'].dt.month
df['day'] = df['date'].dt.day
print(df.head())

# Grouping data by year and plotting
data_grouped = df.groupby('year').mean()
plt.figure(figsize=(20, 10))
for i, col in enumerate(['open', 'high', 'low', 'close']):
    plt.subplot(2, 2, i+1)
    data_grouped[col].plot.bar()
    dataLabel = f"{col}"
    plt.text(3, 39000, dataLabel, fontsize=12)
plt.show()

# Feature engineering
df['is_quarter_end'] = np.where(df['month'] % 3 == 0, 1, 0)
print(df.head())
df['open-close'] = df['open'] - df['close']
df['low-high'] = df['low'] - df['high']
df['target'] = np.where(df['close'].shift(-1) > df['close'], 0, 1)

# Plotting target distribution
plt.figure(figsize=(10, 10))
plt.pie(df['target'].value_counts().values, labels=["Goes down", "Goes up"], autopct='%1.1f%%')
plt.show()

# Plotting heatmap of correlations
plt.figure(figsize=(10, 10))
sb.heatmap(df.corr() > 0.8, annot=True, cbar=False)
plt.show()

# Preparing features and target for model training
features = df[['open-close', 'low-high', 'is_quarter_end']]
target = df['target']

scaler = StandardScaler()
features = scaler.fit_transform(features)

X_train, X_valid, Y_train, Y_valid = train_test_split(features, target, test_size=0.1, random_state=2022)
print(X_train.shape, X_valid.shape)

# Training and evaluating models
models = [LogisticRegression(), SVC(kernel='poly', probability=True), XGBClassifier()]

for i in range(3):
    models[i].fit(X_train, Y_train)
    print(f'{models[i]}: ')
    print('Training Accuracy:', metrics.roc_auc_score(Y_train, models[i].predict_proba(X_train)[:, 1]))
    print('Validation Accuracy:', metrics.roc_auc_score(Y_valid, models[i].predict_proba(X_valid)[:, 1]))
    print()

print('\n\n0 : Goes up')
print('1 : Goes down')
metrics.plot_confusion_matrix(models[0], X_valid, Y_valid)
plt.show()