不满足特征绑定 `ReadError: linfa::Float`

Question

我正在尝试将 CSV 转换为 2D 数组，以使用 Linfa 板条箱训练模型。 据我了解，该函数试图同时返回 2 个错误。但我不想创建另一种错误类型。 虽然我的 array_from_csv 方法似乎无法返回可接受的结果，但这是我的代码：

use csv::ReaderBuilder;
use linfa::prelude::*;
use linfa::Dataset;
use linfa_linear::{Result, TweedieRegressor};
use ndarray::prelude::*;
use ndarray::Axis;
use ndarray_csv::{Array2Reader, ReadError};
use std::io::Read;
// Convert CSV bytes into 2D array
fn array_from_csv<R: Read>(
    csv: R,
    has_headers: bool,
    separator: u8,
) -> Result<Array2<f64>, ReadError> {
    // parse CSV
    let mut reader = ReaderBuilder::new()
        .has_headers(has_headers)
        .delimiter(separator)
        .from_reader(csv);

    // extract ndarray
    reader.deserialize_array2_dynamic();
}

我打电话到这里：

fn get_dataset() -> Dataset<f64, Ix1> {
let data = include_bytes!("examples/data/AMZN_data.csv");
let data = array_from_csv(&data[..], true, b',').unwrap();
let targets = include_bytes!("examples/data/AMZN_targets.csv");
let targets = array_from_csv(&targets[..], true, b',')
    .unwrap()
    .column(0)
    .to_owned();

let feature_names = vec![
    "date",
    "open",
    "high",
    "low",
    "close",
    "adj_close",
    "volume",
];

Dataset::new(data, targets).with_feature_names(feature_names)
}

然后我得到：

error[E0277]: the trait bound `ReadError: linfa::Float` is not satisfied
  --> src/linear_regression.rs:15:6
   |
15 | ) -> Result<Array2<f64>, ReadError> {
   |      ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ the trait `linfa::Float` is not implemented for `ReadError`
   |
   = help: the following other types implement trait `linfa::Float`:
             f32
             f64
note: required by a bound in `LinearError`
  --> /home/theod/.cargo/registry/src/index.crates.io-6f17d22bba15001f/linfa-linear-0.7.0/src/error.rs:10:25
   |
10 | pub enum LinearError<F: Float> {
   |                         ^^^^^ required by this bound in `LinearError`

这是我的 Cargo.toml：

[package]
name = "rust-programs-library"
version = "0.1.0"
edition = "2021"

# See more keys and their definitions at https://doc.rust-lang.org/cargo/reference/manifest.html

[dependencies]
ta = "0.4.0"
csv = "1.1.0"
serde = { version = "1.0", features = ["derive"] }
serde_json = "1.0"
linfa = "0.7.0"
linfa-linear = "0.7.0"
ndarray = "0.15.6"
ndarray-csv = "0.5.2"
flate2 = "1.0.28"

[dev-dependencies]
assert_approx_eq = "1.0.0"
bencher = "0.1.5"
rand = "0.6.5"
bincode = "1.3.1"

Answer 1

报告的问题来自

use linfa_linear::Result

，它将您范围内的内置

Result

替换为

linfa_linear

提供的非标准定义。

如果删除该

use

，则会收到不同的错误，单位类型与声明的

Result

之间类型不匹配。这是由于

;

末尾有多余的

array_from_csv()

造成的。一旦删除它，

array_from_csv()

就会编译。

剩下的错误是：

26 | fn get_dataset() -> Dataset<f64, Ix1> {
   |                     ----------------- expected `DatasetBase<ArrayBase<OwnedRepr<f64>, Dim<[usize; 2]>>, ArrayBase<OwnedRepr<Dim<[usize; 1]>>, Dim<[usize; 2]>>>` because of return type
...
45 |     Dataset::new(data, targets).with_feature_names(feature_names)
   |     ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ expected `Dim<[usize; 1]>`, found `f64`
   |
   = note: expected struct `DatasetBase<_, ArrayBase<OwnedRepr<Dim<[usize; 1]>>, Dim<[usize; 2]>>>`
              found struct `DatasetBase<_, ArrayBase<OwnedRepr<f64>, Dim<[usize; 1]>>>`

...但我无能为力。