juliuskittler

"""
This is an adjusted version of the following file: https://github.com/outerbounds/dsbook/blob/main/chapter-3/classifier_predict.py
"""

from metaflow import FlowSpec, step, Flow, Parameter, JSONType, project, conda_base

@project(name="dummy_example")
@conda_base(python="3.8.13", libraries={"pandas": "1.5", "scikit-learn": "1.1.2"})
class ClassifierPredictFlow(FlowSpec):

juliuskittler

"""
This is an adjusted version of the following file: https://github.com/outerbounds/dsbook/blob/main/chapter-3/classifier_train.py
"""

from metaflow import FlowSpec, step, project, conda_base

@project(name="dummy_example")
@conda_base(python="3.8.13", libraries={"pandas": "1.5", "scikit-learn": "1.1.2"})
class ClassifierTrainFlow(FlowSpec):

juliuskittler

[tool.poetry]
name = "test"
version = "0.1.0"
description = "Test"
authors = ["Test"]

[tool.poetry.dependencies]
python = "3.8.x"
pandas = "*"

juliuskittler

df = data.frame(x = c("Q1", "Q2", "Q3", "Q4", "Q1", "Q2", "Q3", "Q4"), 
                y = c(1000, 2000, 4000, 8000, 500, 1000, 2000, 4000), 
                type = c(rep("revenue", 4), rep("cost", 4)))

ggplot(data = df, aes(x = x, y = y, fill = type)) + 
  geom_col(position = "dodge", alpha = 0.8) + 
  scale_fill_manual(values = c("grey", "steelblue4")) + 
  labs(title = "Revenue and Costs by Quarter [in $1M]", y = "Revenue", x = "Quarter") + 
  theme_bw() + theme(plot.title = element_text(hjust = 0.5))

juliuskittler

df = data.frame(x = c("Q1", "Q2", "Q3", "Q4"), 
                y = c(1000, 2000, 4000, 8000))

ggplot(data = df, aes(x = x, y = y)) + 
  geom_col(fill = "steelblue4", alpha = 0.8) + 
  labs(title = "Revenue by Quarter [in $1M]", y = "Revenue", x = "Quarter") + 
  theme_bw() + theme(plot.title = element_text(hjust = 0.5))

juliuskittler

df = ggplot2::diamonds # data frame with 1 character/factor variable 

ggplot(data = df, aes(x = cut, fill = color)) + 
  geom_bar(position = "stack", alpha = 0.8) + # alternative: position = "dodge" 
  labs(title = "Number of Diamonds by Cut and Color", y = "Count", x = "Cut") + 
  theme_bw() + theme(plot.title = element_text(hjust = 0.5))

juliuskittler

df = ggplot2::diamonds # data frame with 1 character or factor variable 

ggplot(data = df, aes(x = cut)) + 
  geom_bar(fill = "steelblue4", alpha = 0.8) + 
  labs(title = "Number of Diamonds by Cut", y = "Count", x = "Cut") + 
  theme_bw() + theme(plot.title = element_text(hjust = 0.5))

juliuskittler

par(mfrow=c(5,2), mar=c(2,2,2,2))
curve(sin(x), from = -10, to = 10, main = "sin(x)")
curve(sinpi(x), from = -10, to = 10, main = "sin(pi*x)")
curve(sinh(x), from = -10, to = 10, main = "sinh(x)")
curve(asinh(x), from = -10, to = 10, main = "asinh(x)")
curve(cos(x), from = -10, to = 10, main = "cos(x)")
curve(cospi(x), from = -10, to = 10, main = "cos(pi*x)")
curve(cosh(x), from = -10, to = 10, main = "cosh(x)")
curve(acosh(x), from = -10, to = 10, main = "acosh(x)")
curve(tan(x), from = -10, to = 10, main = "tan(x)")

juliuskittler

par(mfrow=c(5,2), mar=c(2,2,2,2))
curve(1 / (1 + exp(-x)), from = -10, to = 10, main = "sigmoid(x)")
curve(10 / (1 + exp(-x)), from = -10, to = 10, main = "10*sigmoid(x)")
curve(tanh(x), from = -10, to = 10, main = "tanh(x)")
curve(exp(x) / sum(exp(x)), from = -10, to = 10, main = "softmax(x)")
curve(pmax(0, x), from = -10, to = 10, main = "rectifier(x)")
curve(log(1 + exp(x)), from = -10, to = 10, main = "softplus(x)")
curve(sign(x), from = -10, to = 10, main = "sign(x)")
curve(floor(x), from = -10, to = 10, main = "floor(x)")
curve(round(x), from = -10, to = 10, main = "round(x)")

juliuskittler

par(mfrow=c(5,2), mar=c(2,2,2,2))
curve(factorial(x), from = -10, to = 10, main = "x!")
curve(2^x, from = -10, to = 10, main = "2^x")
curve(1/sqrt(x), from = -10, to = 10, main = "1/x^(1/2)")
curve(exp(-x), from = -10, to = 10, main = "e^(-x)")
curve(log(x), from = -10, to = 10, main = "ln(x)")
curve(exp(x), from = -10, to = 10, main = "e^x")
curve(log2(x), from = -10, to = 10, main = "log2(x)")
curve(log10(x), from = -10, to = 10, main = "log10(x)")
curve(x*log(x), from = -10, to = 10, main = "x*ln(x)")