import pywt
import numpy as np
import pandas as pd
from sklearn.ensemble import RandomForestRegressor
from sklearn.metrics import mean_squared_error
import matplotlib.pyplot as pltFFT analysis
FFT Examples
data = pd.read_csv(‘<file’)
cycles = 26
weeks = cycles * 2
resolution = 7
np.random.seed(0)
t = np.linspace(0, cycles * np.pi, weeks * resolution)
waveform = np.random.normal(scale=0.5, size=len(t)) + 0.5 * np.sign(np.sin(0.67 * t))
y = np.sin(t) + 0.3*np.sin(4.71*t) + waveform# Create a Pandas DataFrame
df = pd.DataFrame({'y': y})
df['unique_id'] = 1rng = pd.date_range('04/01/2021', periods=df.shape[0], freq='D')
df['ds'] = rng
df.info()<class 'pandas.core.frame.DataFrame'>
RangeIndex: 364 entries, 0 to 363
Data columns (total 3 columns):
# Column Non-Null Count Dtype
--- ------ -------------- -----
0 y 364 non-null float64
1 unique_id 364 non-null int64
2 ds 364 non-null datetime64[ns]
dtypes: datetime64[ns](1), float64(1), int64(1)
memory usage: 8.7 KBdf_plot = df.copy()
df_plot.rename(columns={'y': 'final'}, inplace=True)
df_plot['unique_id'] = 'final'
df_plot['first'] = np.sin(t)
df_plot['second'] = 0.3*np.sin(5*t)
df_plot['noise'] = np.random.normal(scale=0.2, size=len(t))
df_plot.info()<class 'pandas.core.frame.DataFrame'>
RangeIndex: 364 entries, 0 to 363
Data columns (total 6 columns):
# Column Non-Null Count Dtype
--- ------ -------------- -----
0 final 364 non-null float64
1 unique_id 364 non-null object
2 ds 364 non-null datetime64[ns]
3 first 364 non-null float64
4 second 364 non-null float64
5 noise 364 non-null float64
dtypes: datetime64[ns](1), float64(4), object(1)
memory usage: 17.2+ KBimport altair as alt
def long_form(df_plot):
return df_plot.melt('ds', var_name='unique_id', value_name='price')
def altair_plot(df_plot):
highlight = alt.selection_point(on='mouseover', fields=['unique_id'], nearest=True)
base = alt.Chart(df_plot).encode(
x='ds:T',
y='price:Q',
color='unique_id:N'
)
points = base.mark_circle().encode(
opacity=alt.value(0)
).add_params(
highlight
).properties(
width=1000
)
lines = base.mark_line().encode(
size=alt.condition(~highlight, alt.value(1), alt.value(3))
)
return points + lines
df_plot = long_form(df_plot)
altair_plot(df_plot)def wavelet_transform(data):
transformed_data = []
for column in data.columns:
coeffs = pywt.wavedec(data[column], wavelet='db5', level=5)
print(coeffs)
transformed_data.extend(coeffs)
return transformed_datay = pd.DataFrame(df['y'])
#y['second'] = waveform
y.columnsIndex(['y'], dtype='object')X_wavelet = wavelet_transform(y)[array([ 8.56037363, 8.28015354, 8.60255108, 8.33149856, 8.73389978,
7.74524132, 10.2828375 , 2.1141299 , -1.18795588, 0.44417908,
1.19875776, 0.03718912, -0.69022103, 0.39696723, -0.92086484,
0.09806285, -0.67200673, 0.71120904, -4.05185514, -4.04591833]), array([ 0.01153199, -0.25139026, 0.08876349, 1.32828559, 0.07565752,
2.63863482, 4.79537636, -1.24767023, -5.54718816, -3.35555816,
2.36944361, 4.04646868, 0.87187567, -1.68265763, 1.33179388,
-0.40272217, 0.55708821, -1.64304018, 2.00581003, -2.03032757]), array([-0.08897953, -0.4812266 , -0.50538909, -0.10168253, -2.61150232,
2.9547261 , -2.29803229, -0.09763533, 1.47903615, -2.54419368,
3.62531102, -2.47650808, 2.5678546 , -1.85454489, -0.19285758,
1.37250574, -1.5142747 , 2.39190732, -3.19870564, 2.72915131,
-3.41889537, 1.28809091, 0.52633798, -1.98882932, 3.88974806,
-5.33502469, 0.91456006, 0.25339818, 3.80649291, -3.62508319,
2.3555905 ]), array([ 0.04489604, 1.21115553, 0.30484212, 0.57478063, -0.10751276,
-0.65589434, 1.54087004, 0.69994033, 0.84825601, -0.01101841,
0.19981331, 0.33554834, -0.29570558, 1.92260008, -0.8211238 ,
0.16218484, 1.10221247, -0.2672678 , -0.39443262, -0.74778196,
-0.16734269, -1.55505881, -0.16445345, -1.09477366, -0.2275783 ,
-0.32740859, 0.09258987, 1.32728503, -1.59283858, 0.0304181 ,
-0.04143611, -0.06468654, 0.42971042, 0.40445104, 1.32661766,
-0.68588646, 0.56466584, 0.13326441, -0.17115911, 0.37710454,
-0.13755441, 1.21737271, -0.55018683, -0.22723865, -0.66991994,
0.08144042, -0.09456927, -0.88110325, 0.89665126, -1.27621373,
2.02092779, -0.82140168, -0.28145005]), array([ 0.08743568, -0.52802978, 0.76003102, -0.87477212, -0.2137142 ,
-0.08251688, 0.22470448, -0.24951939, -0.41966809, 0.24968052,
-0.82965552, 0.11282229, 0.41979293, -1.62165521, 0.63917665,
-0.05349559, -0.81191952, 0.47499175, 0.4483338 , -0.51529357,
-0.03268174, 0.81509277, -0.72212797, 0.94142773, 0.79246556,
-0.66823467, -0.04388327, -0.35251386, -0.96690333, 0.76438364,
0.61711505, -0.57417453, -0.24123739, 0.47541282, -1.33055547,
-0.04216579, 0.49976061, -0.04719167, -1.45547177, 1.4612047 ,
-1.09233401, -0.79218819, 1.4451979 , -0.52280188, -0.51661146,
0.07177687, -0.06904092, -1.00303966, 0.54283863, -0.20346968,
-0.44556147, 1.01559998, 0.21400751, -1.02416331, 1.15926639,
-0.56999735, 0.65632908, 0.4355802 , 0.48993803, -1.39774935,
0.38826329, -0.33822322, -0.40172831, 0.53523861, -0.09876577,
-0.53922567, -0.34208739, 0.67681303, 0.09242104, -0.10293512,
0.68052398, -1.18565793, -0.6305652 , -0.08625569, -1.3032623 ,
-0.4577607 , 0.47035347, 0.05809923, -0.89829346, 0.82602796,
-1.07480685, -0.05684645, 0.80296313, 0.22276104, -0.67154841,
0.41031737, -0.54248432, -1.24758935, 0.81923313, -0.76539182,
-0.06074858, 0.97558165, -0.34115139, -0.11998139, 0.21528626,
-0.60822975, 0.00416824]), array([ 7.82169804e-02, -2.12160997e-02, -1.02848006e-01, 1.11393225e+00,
-2.69082443e-01, -1.98484826e-01, -1.71201044e-01, 1.81663044e-01,
3.49069262e-01, 4.45396733e-01, -4.88410641e-01, 1.15788613e-01,
1.24733049e+00, 4.99264605e-01, 2.69837665e-01, -3.29457397e-01,
-1.78566872e-01, 5.62829176e-01, 1.68993607e-02, -4.34187851e-01,
3.75604438e-02, 1.78940706e-01, -1.31556512e+00, 1.27625750e-01,
-9.41784192e-01, -2.90424465e-01, -2.28643513e-01, 4.58050195e-01,
-4.08033755e-01, -9.86955101e-02, -5.20036649e-02, -3.78122379e-04,
9.93916246e-01, -6.37854384e-01, -6.25300155e-01, 2.95068950e-01,
4.62356890e-01, 6.19277900e-01, -3.45611868e-02, 1.00372393e-02,
-5.87213878e-01, -2.39389518e-02, 1.44606100e+00, -6.61582418e-01,
1.63701508e-01, -8.32121441e-01, -3.99609990e-01, -7.81021937e-02,
-3.17109338e-01, -6.62768834e-01, 8.28543671e-01, 1.12792812e-01,
-1.05304001e+00, -2.86248468e-01, 7.85163687e-01, -1.14392546e-01,
-3.22477044e-01, -9.26960676e-01, 1.51314348e-01, 3.69942059e-01,
-1.91301182e-01, 4.69421895e-01, -7.35694506e-01, 3.20851631e-01,
-7.78526682e-01, 3.67410679e-02, -5.64428095e-01, -2.68241703e-01,
-2.10079402e-03, 3.51694972e-01, -1.49792951e-01, -2.97840420e-01,
4.72250268e-01, -9.50879824e-01, -5.42413251e-01, -1.15215252e-01,
-8.46279893e-01, 5.97270295e-01, 4.20475483e-01, 1.18398345e-01,
-2.68442902e-01, -5.38487858e-01, -1.44555769e-01, -7.29126413e-01,
5.81750516e-01, 1.31050332e-01, 4.13242858e-01, -1.69833616e-01,
-5.93391260e-01, 2.54453871e-01, -2.24416337e-01, 5.22312215e-01,
9.80914799e-01, -1.45058262e-02, -9.80040422e-01, -7.46969348e-01,
-7.80924496e-02, 6.39680098e-01, 1.83881590e-01, 2.19077815e-01,
-2.97418275e-01, 4.84247670e-01, -1.16957224e-02, 8.16785915e-01,
5.63532338e-03, -9.60587176e-02, 3.83645651e-01, 2.16475782e-02,
-1.36875879e-01, 5.94808227e-01, 5.22964942e-03, -2.75391075e-01,
-6.37982756e-01, 8.51258601e-03, -3.16836196e-01, -6.85667705e-01,
1.08353986e-02, -7.82748435e-01, 4.28636022e-01, 1.03055876e+00,
-7.63941608e-01, 3.20021138e-01, 6.56538023e-02, 7.50973676e-01,
-5.42819573e-01, 6.88499408e-01, -2.69482031e-01, 4.58058772e-01,
-3.65725409e-01, -7.34325562e-02, -7.43466155e-01, 6.62097224e-02,
-7.63881741e-01, 7.13215259e-02, -7.17346313e-02, 3.79695244e-01,
1.97768780e+00, -6.49823532e-01, 3.26231535e-01, -6.76495937e-01,
-1.04392496e+00, -4.81300683e-01, -4.17198698e-01, -4.93617859e-01,
-4.36926650e-01, -3.29281763e-01, 1.40802817e+00, 1.33180958e-01,
-4.08391392e-01, -2.60750946e-01, -7.00602812e-01, -5.40610047e-01,
6.22860442e-01, -2.48715379e-01, 6.80335666e-01, -2.22681683e-01,
-5.17525473e-01, 5.61675223e-01, -3.38477480e-01, 1.43883793e-01,
-6.74863552e-02, 3.17695822e-01, -7.23998209e-02, -1.88063003e-01,
-1.19267630e+00, 2.18871157e-01, 9.27703437e-01, -1.90193785e-01,
-1.21027604e+00, -3.24657030e-01, -3.35131838e-01, 3.43673865e-01,
2.20526430e-01, -2.57544687e-01, 1.10155401e-01, -1.05068944e-01,
8.65358978e-01, 2.03499300e-01, -7.50242361e-02, -3.86384134e-01,
-7.66133144e-03, 4.64230598e-01, -1.63102790e-01, -3.32783358e-01,
1.55127572e-01, 2.00635763e-01])]X_wavelet[0]array([ 1.69782638e+00, 2.86633130e+00, 2.19476886e+00, 1.17397138e+00,
2.01586836e+00, 2.70074412e+00, 1.98287045e+00, 2.53977548e+00,
1.91548731e+00, 2.07104557e+00, 1.49206799e+00, 6.24585355e-01,
3.63488119e-03, -1.23050798e+00, -2.22516777e+00, -1.45626892e+00,
-1.57360056e+00, -6.70425537e-01, 3.60216320e-01, 3.92006830e-01,
-6.67448566e-01, 1.20076474e+00, 1.34902772e+00, -8.80108851e-01,
-8.48793754e-01, 9.86961240e-01, -1.00889684e+00, -9.96228925e-01,
-7.58473590e-01, -1.09936767e+00, -4.74542101e-01, 1.07154093e+00,
6.71358965e-01, 1.41619721e+00, 1.77307686e+00, 7.40809286e-02,
5.62297202e-02, 1.48342153e-02, -7.92568112e-01, -6.72164798e-01,
-1.47818506e+00, -2.98386840e+00, -2.19059661e+00, -1.95613078e+00,
-2.01711333e+00, -7.02480981e-04, 2.58409010e+00, 1.03976676e+00,
2.38187317e+00, 2.73694743e+00, 1.94806995e+00, 2.13110002e+00,
1.41275222e+00, 6.61209183e-01, -6.26994065e-01, -3.83339616e-01,
-2.24552771e+00, -1.36530903e+00, 5.20891491e-02, -9.63966984e-01,
4.33903511e-01, 8.65625523e-01, 5.39747879e-01, 1.31753836e+00,
5.60365478e-01, -3.40131488e-01, 1.80168614e-01, 1.10758938e+00,
-6.84122197e-01, -3.34468694e-01, -9.67974799e-01, -5.68087480e-01,
5.62314033e-01, -1.94723399e-01, 5.90536838e-01, 3.50383895e+00,
1.71759551e+00, 9.24381133e-02, 8.06891772e-01, 1.74700310e-01,
-1.42390185e+00, -5.70536402e-01, -2.47344630e+00, -2.75288846e+00,
-7.27528289e-01, -2.10474614e+00, -2.16611230e+00, 6.83276964e-01,
1.45111868e+00, 2.05263083e+00, 2.10768670e+00, 1.71995702e+00,
1.28707724e+00, 1.71366022e+00, -3.45686267e-01, -5.42679676e-01,
-5.36517797e-01, -1.64763345e+00, -2.23588848e+00, -1.62481503e+00,
-1.88789349e+00, -8.67276586e-01, 1.49540135e+00, 3.48492854e-01,
8.75034108e-01, 1.48134970e+00, -3.57533461e-01, -3.50851776e-01,
1.21067431e+00, 3.34407247e-01, -7.22731370e-01, -7.83948722e-01,
-8.00539078e-03, -6.27089015e-01, 2.58212559e-01, -2.01821012e-01,
1.25658523e+00, 2.32748366e+00, 3.63168070e-01, 5.26949190e-01,
3.16572282e-01, 6.48425185e-03, -7.71003662e-01, -1.01376207e+00,
-2.32236401e+00, -2.44432130e+00, -1.75157248e+00, -2.24439864e+00,
-1.59672927e+00, 5.30976701e-01, 1.14611200e+00, 2.04313147e+00,
1.41599179e+00, 1.26529130e+00, 1.84280178e+00, 6.76185118e-01,
4.80539100e-01, 3.72875688e-01, -4.81506716e-01, -2.27614039e+00,
-1.94089445e+00, -1.49267556e+00, -2.35262541e+00, 1.51879060e-01,
4.85279258e-02, 6.93119371e-01, 7.50628470e-01, 1.58969214e+00,
-2.73848686e-01, 1.46411682e+00, 5.94325381e-01, 1.20018857e-01,
1.42775133e-01, -9.01552690e-01, -8.36577719e-01, 1.87829611e-01,
-6.83570959e-01, 9.44528652e-02, 1.85610930e+00, 1.72844555e+00,
9.30680887e-02, 1.00389764e+00, 2.83265909e-01, -8.17614709e-01,
-4.59295934e-01, -2.21982178e+00, -2.08909812e+00, -1.23495410e+00,
-1.92722736e+00, -2.09438316e+00, -1.27483495e+00, 2.47151395e-02,
8.62875878e-01, 3.14299902e+00, 1.53494389e+00, 1.35202124e+00,
1.84182300e+00, 1.45584181e+00, 1.20316826e-01, 3.96518649e-01,
-7.89934160e-01, -2.90731899e+00, -2.14249061e+00, -1.76722203e+00,
-1.01628309e+00, 1.55622881e-01])plt.plot(X_wavelet[0])
plt.plot(X_wavelet[1])
plt.plot(X_wavelet[2])
plt.plot(X_wavelet[3])
X_wavelet--------------------------------------------------------------------------- AttributeError Traceback (most recent call last) Cell In[56], line 1 ----> 1 X_wavelet.info() AttributeError: 'numpy.ndarray' object has no attribute 'info'
split_point = int(len(data) * 0.8)--------------------------------------------------------------------------- NameError Traceback (most recent call last) Cell In[54], line 1 ----> 1 split_point = int(len(data) * 0.8) NameError: name 'data' is not defined
X_train, X_data = X_wavelet[:split_point], X_wavelet[split_point:, :]
y_train, y_test = y.iloc[:split_point], y.iloc[split_point:]model = RandomForestRegressor()
model.fit(X_train, y_train)