1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566
//! Series represents a single column within a dataframe and wraps many `Array` like
//! functionality.
//!
//! For methods implemented for a [`Series`], please check out the trait [`SeriesTrait`]
//!
//! ## Example use:
//!
//! ```
//! use blackjack::prelude::*;
//!
//! let mut series = Series::arange(0, 5);
//!
//! // Index and change elements
//! series[0] = 1;
//! series[1] = 0;
//!
//! assert_eq!(series[0], 1);
//! assert_eq!(series.sum(), 10);
//! assert_eq!(series.len(), 5);
//! ```
use std::ops::{Range, Index, IndexMut};
use std::iter::{FromIterator, Sum};
use std::convert::From;
use std::fmt;
use std::str::FromStr;
use num::*;
use stats;
pub mod overloaders;
use prelude::*;
/// Series struct for containing underlying Array and other meta data.
#[derive(Debug, Clone, PartialEq, Deserialize, Serialize)]
pub struct Series<T>
where
T: BlackJackData
{
/// Name of the series, if added to a dataframe without a name, it will be assigned
/// a default name equalling the cound of columns in the dataframe.
pub name: Option<String>,
/// The underlying values of the Series
pub values: Vec<T>,
dtype: DType
}
/// Constructor methods for `Series<T>`
impl<T> Series<T>
where
T: BlackJackData
{
/// Create a new Series struct from an integer range with one step increments.
///
/// ## Example
/// ```
/// use blackjack::prelude::*;
///
/// let series: Series<i32> = Series::arange(0, 10);
/// ```
pub fn arange(start: T, stop: T) -> Self
where
T: Integer + BlackJackData + ToPrimitive,
Range<T>: Iterator,
Vec<T>: FromIterator<<Range<T> as Iterator>::Item>
{
let dtype = start.dtype();
let values: Vec<T> = (start..stop).collect();
Series {
name: None,
dtype,
values
}
}
/// Convert the series into another [`DType`] (creates a new series)
///
/// ## Example
/// ```
/// use blackjack::prelude::*;
///
/// let series: Series<i32> = Series::arange(0, 10);
/// assert_eq!(series[0].dtype(), DType::I32);
/// let new_series = series.astype::<f64>().unwrap();
/// assert_eq!(new_series[0].dtype(), DType::F64);
/// ```
pub fn astype<A>(&self) -> Result<Series<A>, &'static str>
where A: BlackJackData + FromStr
{
let values = self.values
.clone()
.into_iter()
.map(|v| v.to_string())
.map(|v| v.parse::<A>().map_err(|_| "Cannot cast into type"))
.collect::<Result<Vec<A>, _>>()?;
let series = Series {
name: self.name.clone(),
dtype: values[0].dtype(),
values
};
Ok(series)
}
/// Convert this series into another [`DType`] (consumes current series)
///
/// ## Example
/// ```
/// use blackjack::prelude::*;
///
/// let series: Series<i32> = Series::arange(0, 10);
/// assert_eq!(series[0].dtype(), DType::I32);
/// let new_series = series.into_type::<f64>().unwrap();
/// assert_eq!(new_series[0].dtype(), DType::F64);
/// ```
pub fn into_type<A>(self) -> Result<Series<A>, &'static str>
where A: BlackJackData + FromStr
{
let values = self.values
.into_iter()
.map(|v| v.to_string())
.map(|v| v.parse::<A>().map_err(|_| "Cannot cast into type"))
.collect::<Result<Vec<A>, _>>()?;
let series = Series {
name: self.name.clone(),
dtype: values[0].dtype(),
values
};
Ok(series)
}
/// Get a series of the unique elements held in this series
///
/// ## Example
///
/// ```
/// use blackjack::prelude::*;
///
/// let series: Series<i32> = Series::from_vec(vec![1, 2, 1, 0, 1, 0, 1, 1]);
/// let unique: Series<i32> = series.unique();
/// assert_eq!(unique, Series::from_vec(vec![0, 1, 2]));
/// ```
pub fn unique(&self) -> Series<T>
where T: PartialOrd + Copy
{
// Cannot use `HashSet` as f32 & f64 don't implement Hash
let mut unique: Vec<T> = vec![];
let mut values = self.values.clone();
values.sort_by(|a, b| a.partial_cmp(b).unwrap());
for val in values
{
if unique.len() > 0 {
if val == unique[unique.len() - 1] {
continue
} else {
unique.push(val)
}
} else {
unique.push(val)
}
}
Series::from_vec(unique)
}
/// Create a new Series struct from a vector, where T is supported by [`BlackJackData`].
///
/// ## Example
/// ```
/// use blackjack::prelude::*;
///
/// let series: Series<i32> = Series::from_vec(vec![1, 2, 3]);
/// ```
pub fn from_vec(vec: Vec<T>) -> Self
{
let dtype = vec[0].dtype(); // TODO: Do something better.
Series {
name: None,
dtype,
values: vec
}
}
/// Convert the series to a [`Vec`]
/// **Type Annotations required**
/// Will coerce elements into the desired [`DType`] primitive, just as
/// [`SeriesTrait::astype()`].
///
/// ## Example
/// ```
/// use blackjack::prelude::*;
///
/// let series = Series::from_vec(vec![1_f64, 2_f64, 3_f64]);
///
/// assert_eq!(
/// series.clone().into_vec(),
/// vec![1_f64, 2_f64, 3_f64]
/// );
/// ```
pub fn into_vec(self) -> Vec<T> {
self.values
}
/// Set the name of a series
pub fn set_name(&mut self, name: &str) -> () {
self.name = Some(name.to_string());
}
/// Get the name of the series; Series may not be assigned a string,
/// so an `Option` is returned.
///
/// ## Example
/// ```
/// use blackjack::prelude::*;
///
/// let mut series = Series::from_vec(vec![1, 2, 3]);
/// series.set_name("my-series");
///
/// assert_eq!(series.name(), Some("my-series".to_string()));
/// ```
pub fn name(&self) -> Option<String> {
match self.name {
Some(ref name) => Some(name.clone()),
None => None
}
}
/// Finds the returns a [`Series`] containing the mode(s) of the current
/// [`Series`]
pub fn mode(&self) -> Result<Self, &'static str>
where T: BlackJackData + PartialOrd + Copy + ToPrimitive
{
if self.len() == 0 {
return Err("Cannot compute mode of an empty series!")
}
let modes = stats::modes(self.values.iter().map(|v| *v));
let modes = Series::from_vec(modes);
Ok(modes)
}
/// Calculate the variance of the series
/// **NOTE** that whatever type is determined is what the values are cast to
/// during calculation of the variance.
///
/// ie. `series.var::<i32>()` will cast each element into `i32` as input
/// for calculating the variance, and yield a `i32` value. If you want all
/// values to be calculated as `f64` then specify that in the type annotation.
pub fn var(&self) -> Result<f64, &'static str>
where
T: BlackJackData + ToPrimitive + Copy
{
if self.len() == 0 {
return Err("Cannot compute variance of an empty series!");
}
let var = stats::variance(self.values.iter().map(|v| *v));
Ok(var)
}
/// Calculate the standard deviation of the series
///
/// ## Example
/// ```
/// use blackjack::prelude::*;
///
/// let series = Series::arange(0, 10).astype::<f32>().unwrap();
///
/// let std = series.std().unwrap(); // Ok(2.8722...)
/// assert!(std > 2.87);
/// assert!(std < 2.88);
/// ```
pub fn std(&self) -> Result<f64, &'static str>
where T: BlackJackData + ToPrimitive + Copy
{
if self.len() == 0 {
return Err("Cannot compute standard deviation of an empty series!")
}
let std = stats::stddev(self.values.iter().map(|v| *v));
Ok(std)
}
/// Sum a given series, yielding the same type as the elements stored in the
/// series.
pub fn sum(&self) -> T
where
T: Num + Copy + Sum
{
self.values
.iter()
.map(|v| *v)
.sum()
}
/// Average / Mean of a given series - Requires specifying desired float
/// return annotation
///
/// ## Example:
/// ```
/// use blackjack::prelude::*;
///
/// let series = Series::arange(0, 5);
/// let mean = series.mean();
///
/// match mean {
/// Ok(result) => {
/// println!("Result is: {}", &result);
/// assert_eq!(result, 2.0);
/// },
/// Err(err) => {
/// panic!("Was unable to compute mean, error: {}", err);
/// }
/// }
/// ```
pub fn mean<'a>(&'a self) -> Result<f64, &'static str>
where
T: ToPrimitive + Copy + Sum<&'a T> + Num + Sum
{
let total = self.sum().to_f64().unwrap();
let count = self.len() as f64;
Ok(total / count)
}
/// Calculate the quantile of the series
///
/// ## Example:
/// ```
/// use blackjack::prelude::*;
///
/// let series = Series::arange(0, 100).astype::<f32>().unwrap();
/// let qtl = series.quantile(0.5).unwrap(); // `49.5_f32`
///
/// assert!(qtl < 49.51);
/// assert!(qtl > 49.49);
/// ```
pub fn quantile(&self, quantile: f64) -> Result<f64, &'static str>
where
T: ToPrimitive + BlackJackData
{
use rgsl::statistics::quantile_from_sorted_data;
use std::cmp::Ordering;
let mut vec = self
.clone()
.into_vec()
.into_iter()
.map(|v| v.to_f64().unwrap())
.collect::<Vec<f64>>();
vec.sort_by(|a, b| a.partial_cmp(b).unwrap_or(Ordering::Equal));
let qtl = quantile_from_sorted_data(&vec[..], 1, vec.len(), quantile);
Ok(qtl)
}
/// Calculate the median of a series
pub fn median<'a>(&'a self) -> Result<f64, &'static str>
where T: ToPrimitive + Copy + PartialOrd
{
if self.len() == 0 {
return Err("Cannot calculate median of an empty series.")
}
let med_opt = stats::median(self.values.iter().map(|v| v.to_f64().unwrap()));
match med_opt {
Some(med) => Ok(med),
None => Err(r#"Unable to calculate median, please create an issue!
as this wasn't expected to ever happen on a non-empty
series. :("#)
}
}
/// Find the minimum of the series. If several elements are equally minimum,
/// the first element is returned. If it's empty, an Error will be returned.
///
/// ## Example
/// ```
/// use blackjack::prelude::*;
///
/// let series: Series<i32> = Series::arange(10, 100);
///
/// assert_eq!(series.min(), Ok(10));
/// ```
pub fn min(&self) -> Result<T, &'static str>
where
T: Num + Clone + Ord + BlackJackData
{
let min = self.values.iter().min();
match min {
Some(m) => Ok(m.clone()),
None => Err("Unable to find minimum of values, perhaps values is empty?")
}
}
/// Exibits the same behavior and usage of [`SeriesTrait::min`], only
/// yielding the [`Result`] of a maximum.
pub fn max(&self) -> Result<T, &'static str>
where
T: Num + Clone + Ord
{
let max = self.values.iter().max();
match max {
Some(m) => Ok(m.clone()),
None => Err("Unable to find maximum of values, perhaps values is empty?")
}
}
/// Determine the length of the Series
pub fn len(&self) -> usize { self.values.len() }
/// Determine if series is empty.
pub fn is_empty(&self) -> bool { self.len() == 0 }
/// Get the dtype, returns `None` if series dtype is unknown.
/// in such a case, calling `.astype()` to coerce all types to a single
/// type is needed.
pub fn dtype(&self) -> DType {
self.dtype.clone()
}
/// Append a [`BlackJackData`] element to the Series
///
/// ## Example
/// ```
/// use blackjack::prelude::*;
///
/// let mut series = Series::from_vec(vec![0, 1, 2]);
/// assert_eq!(series.len(), 3);
///
/// series.append(3);
/// assert_eq!(series.len(), 4);
/// ```
pub fn append<V: Into<T>>(&mut self, val: V) -> () {
let v = val.into();
self.values.push(v);
}
/// As boxed pointer, recoverable by `Box::from_raw(ptr)` or
/// `SeriesTrait::from_raw(*mut Self)`
pub fn into_raw(self) -> *mut Self {
Box::into_raw(Box::new(self))
}
/// Create from raw pointer
pub fn from_raw(ptr: *mut Self) -> Self {
unsafe { *Box::from_raw(ptr) }
}
/// Group by method for grouping elements in a [`Series`]
/// by key.
///
/// ## Example
///
/// ```
/// use blackjack::prelude::*;
///
/// let series = Series::from_vec(vec![1, 2, 3, 1, 2, 3]);
/// let keys = Series::from_vec(vec![4, 5, 6, 4, 5, 6]);
///
/// let grouped: Series<i32> = series.groupby(keys).sum();
/// assert_eq!(grouped.len(), 3);
///
/// let mut vals = grouped.into_vec();
/// vals.sort();
/// assert_eq!(vals, vec![2, 4, 6]);
/// ```
pub fn groupby(&self, keys: Series<T>) -> SeriesGroupBy<T>
where T: ToPrimitive
{
/* TODO: Revisit this to avoid the clones. Needs to keep the groups
in order based on key order; match pandas. ie:
>>> pd.Series([1, 2, 3, 1, 2, 3]).groupby([4, 5, 6, 4, 5, 6]).sum()
4 2
5 4
6 6
dtype: int64
*/
use indexmap::IndexMap;
let values = self.values.clone();
let mut map: IndexMap<String, Vec<T>> = IndexMap::new();
// Group values by their keys
for (k, v) in keys.values.into_iter().zip(values.into_iter()) {
let key = k.to_string();
let mr = map.entry(key).or_insert(vec![]);
mr.push(v);
}
// Create new series from the previous mapping.
let groups = map
.iter()
.map(|(name, values)| {
let mut series = Series::from_vec(values.clone());
series.set_name(name.as_str());
series
})
.collect();
SeriesGroupBy::new(groups)
}
}
// Support ref indexing
impl<T> Index<usize> for Series<T>
where T: BlackJackData
{
type Output = T;
fn index(&self, idx: usize) -> &T {
&self.values[idx]
}
}
// Support ref mut indexing
impl<T: BlackJackData> IndexMut<usize> for Series<T> {
fn index_mut(&mut self, idx: usize) -> &mut T {
&mut self.values[idx]
}
}
// Support Display for Series
impl<T> fmt::Display for Series<T>
where
T: BlackJackData,
String: From<T>
{
fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
use prettytable::{Table, Row, Cell};
let mut table = Table::new();
// Title (column name)
table.add_row(
Row::new(
vec![
Cell::new(&self.name().unwrap_or("<NA>".to_string()))
]
)
);
// Build remaining values.
// TODO: Limit how many are actually printed.
let _ = self.values
.iter()
.map(|v| {
let v: String = v.clone().into();
table.add_row(
Row::new(vec![
Cell::new(&format!("{}", v))
])
);
})
.collect::<Vec<()>>();
write!(f, "{}\n", table)
}
}