đơn giản

l10n-placeholder1 xử lý == 1, 2, 1) kiểm tra\\(treat <- ifelse(test\\)xử lý == 1, 2, 1) đào tạo\\(y <- ifelse(train\\)y == 1, 2, 1) kiểm tra\\(y <- ifelse(test\\)y == 1, 2, 1) train\\(ts = NULL test\\)ts = NULL ``` Tham số: - `n` = số lượng mẫu - `p` = số lượng yếu tố dự đoán - `ro` = hiệp phương sai giữa các yếu tố dự đoán - `sigma` = bội số của lỗi hạn - `beta.den` = beta bị cắt xén bởi 1/beta.den Người tạo: Leo Guelman leo.guelman@gmail.com Để sử dụng bộ dữ liệu này: ```python nhập tensorflow_datasets dưới dạng tfds ds = tfds.load('simpte' , split='train') cho ex trong ds.take(4): print(ex) ``` Xem [hướng dẫn](https://www.tensorflow.org/datasets/overview) để biết thêm thông tin về [tensorflow_datasets ](https://www.tensorflow.org/datasets). " />
  • Mô tả :

Tên đầy đủ: Mô phỏng cho Hiệu ứng Điều trị Cá nhân hóa

Được tạo bằng gói Nâng cấp của R: https://rdrr.io/cran/uplift/man/sim_pte.html

Có thể tải xuống gói tại đây: https://cran.r-project.org/src/contrib/Archive/uplift/

Bộ dữ liệu được tạo trong phiên bản R 4.1.2 với mã sau:

  library(uplift)

  set.seed(123)

  train <- sim_pte(n = 1000, p = 20, rho = 0, sigma = sqrt(2), beta.den = 4)
  test <- sim_pte(n = 2000, p = 20, rho = 0, sigma = sqrt(2), beta.den = 4)

  train$treat <- ifelse(train$treat == 1, 2, 1)
  test$treat <- ifelse(test$treat == 1, 2, 1)

  train$y <- ifelse(train$y == 1, 2, 1)
  test$y <- ifelse(test$y == 1, 2, 1)

  train$ts = NULL
  test$ts = NULL

Thông số:

  • n = số mẫu
  • p = số lượng dự đoán
  • ro = hiệp phương sai giữa các yếu tố dự đoán
  • sigma = bội số của thuật ngữ lỗi
  • beta.den = beta bị cắt xén bởi 1/beta.den

Tác giả: Leo Guelman leo.guelman@gmail.com

  • Trang chủ : https://rdrr.io/cran/uplift/man/sim_pte.html

  • Mã nguồn : tfds.datasets.simpte.Builder

  • Phiên bản :

    • 1.0.0 (mặc định): Bản phát hành đầu tiên.
  • Kích thước tải xuống : Unknown size

  • Kích thước tập dữ liệu : 1.04 MiB

  • Hướng dẫn tải xuống thủ công : Bộ dữ liệu này yêu cầu bạn tải xuống dữ liệu nguồn theo cách thủ công vào download_config.manual_dir (mặc định là ~/tensorflow_datasets/downloads/manual/ ):
    Vui lòng tải dữ liệu đào tạo: sim_pte_train.csv và dữ liệu kiểm tra: sim_pte_test.csv xuống ~/tensorflow_datasets/downloads/manual/.

  • Tự động lưu vào bộ đệm ( tài liệu ): Có

  • Chia tách :

Tách ra ví dụ
'test' 2.000
'train' 1.000
  • Cấu trúc tính năng :
FeaturesDict({
    'X1': float32,
    'X10': float32,
    'X11': float32,
    'X12': float32,
    'X13': float32,
    'X14': float32,
    'X15': float32,
    'X16': float32,
    'X17': float32,
    'X18': float32,
    'X19': float32,
    'X2': float32,
    'X20': float32,
    'X3': float32,
    'X4': float32,
    'X5': float32,
    'X6': float32,
    'X7': float32,
    'X8': float32,
    'X9': float32,
    'treat': int32,
    'y': int32,
})
  • Tài liệu tính năng :
Tính năng Lớp Hình dạng Dtype Sự miêu tả
Tính năngDict
X1 tenxơ phao32
X10 tenxơ phao32
X11 tenxơ phao32
X12 tenxơ phao32
X13 tenxơ phao32
X14 tenxơ phao32
X15 tenxơ phao32
X16 tenxơ phao32
X17 tenxơ phao32
X18 tenxơ phao32
X19 tenxơ phao32
X2 tenxơ phao32
X20 tenxơ phao32
X3 tenxơ phao32
X4 tenxơ phao32
X5 tenxơ phao32
X6 tenxơ phao32
X7 tenxơ phao32
X8 tenxơ phao32
X9 tenxơ phao32
đãi tenxơ int32
y tenxơ int32
  • Các khóa được giám sát (Xem as_supervised doc ): ({'X1': 'X1', 'X10': 'X10', 'X11': 'X11', 'X12': 'X12', 'X13': 'X13', 'X14': 'X14', 'X15': 'X15', 'X16': 'X16', 'X17': 'X17', 'X18': 'X18', 'X19': 'X19', 'X2': 'X2', 'X20': 'X20', 'X3': 'X3', 'X4': 'X4', 'X5': 'X5', 'X6': 'X6', 'X7': 'X7', 'X8': 'X8', 'X9': 'X9', 'treat': 'treat'}, 'y')

  • Hình ( tfds.show_examples ): Không được hỗ trợ.

  • Ví dụ ( tfds.as_dataframe ):

  • trích dẫn :
@misc{https://doi.org/10.48550/arxiv.1212.2995,
  doi = {10.48550/ARXIV.1212.2995},
  url = {https://arxiv.org/abs/1212.2995},
  author = {Tian, Lu and Alizadeh, Ash and Gentles, Andrew and Tibshirani, Robert},
  keywords = {Methodology (stat.ME), FOS: Computer and information sciences, FOS: Computer and information sciences},
  title = {A Simple Method for Detecting Interactions between a Treatment and a Large Number of Covariates},
  publisher = {arXiv},
  year = {2012},
  copyright = {arXiv.org perpetual, non-exclusive license}
}