amanbthakkar/log_fit.py Secret

## log_fit.py
#this is the function we want to fit over our data: a.log(x)+b
#we need to find appropriate coefficients
def func(x, p1, p2):
    return p1*np.log(x) + p2

#we are fitting log of price of BTC against the function, not actual price
ydata = np.log(df["Value"])
xdata = [x+1 for x in range(len(df))] #just use numbers for dates

extended_dates = pd.date_range(df["Date"].iloc[0], "2023-01-01")

#extract optimal coefficients using curve fit
popt, pcov = curve_fit(func, xdata, ydata, p0=(3.0, -10))

#try to get ydata from xdata and function
#popt has coefficients, pcov has covariances between them
print(popt)

#generate fitted Y data
fittedYdata = func(np.array([x+1 for x in range(len(df))]), popt[0], popt[1])#pass values to function

plt.style.use("dark_background")
fig, ax = plt.subplots()
ax.semilogy(df["Date"], df["Value"])
plt.yscale('log', subsy=[1])
ax.yaxis.set_major_formatter(matplotlib.ticker.ScalarFormatter())
ax.yaxis.set_minor_formatter(matplotlib.ticker.ScalarFormatter())
plt.plot(df["Date"], np.exp(fittedYdata)) #exponentiate the data

plt.title("BTC logarithmic regression")
plt.ylabel("Price in USD")
plt.ylim(bottom=0.1)
plt.show()
	#this is the function we want to fit over our data: a.log(x)+b
	#we need to find appropriate coefficients
	def func(x, p1, p2):
	return p1*np.log(x) + p2

	#we are fitting log of price of BTC against the function, not actual price
	ydata = np.log(df["Value"])
	xdata = [x+1 for x in range(len(df))] #just use numbers for dates

	extended_dates = pd.date_range(df["Date"].iloc[0], "2023-01-01")

	#extract optimal coefficients using curve fit
	popt, pcov = curve_fit(func, xdata, ydata, p0=(3.0, -10))

	#try to get ydata from xdata and function
	#popt has coefficients, pcov has covariances between them
	print(popt)

	#generate fitted Y data
	fittedYdata = func(np.array([x+1 for x in range(len(df))]), popt[0], popt[1])#pass values to function

	plt.style.use("dark_background")
	fig, ax = plt.subplots()
	ax.semilogy(df["Date"], df["Value"])
	plt.yscale('log', subsy=[1])
	ax.yaxis.set_major_formatter(matplotlib.ticker.ScalarFormatter())
	ax.yaxis.set_minor_formatter(matplotlib.ticker.ScalarFormatter())
	plt.plot(df["Date"], np.exp(fittedYdata)) #exponentiate the data

	plt.title("BTC logarithmic regression")
	plt.ylabel("Price in USD")
	plt.ylim(bottom=0.1)
	plt.show()