Data from "Datamodels: Predicting Predictions with Training Data"

Overview

Data from "Datamodels: Predicting Predictions with Training Data"

Here we provide the data used in the paper "Datamodels: Predicting Predictions with Training Data" (arXiv, Blog).

Note that all of the data below is stored on Amazon S3 using the “requester pays” option to avoid a blowup in our data transfer costs (we put estimated AWS costs below)---if you are on a budget and do not mind waiting a bit longer, please contact us at [email protected] and we can try to arrange a free (but slower) transfer.

Citation

To cite this data, please use the following BibTeX entry:

@inproceedings{ilyas2022datamodels,
  title = {Datamodels: Predicting Predictions from Training Data},
  author = {Andrew Ilyas and Sung Min Park and Logan Engstrom and Guillaume Leclerc and Aleksander Madry},
  booktitle = {ArXiv preprint arXiv:2202.00622},
  year = {2022}
}

Overview

We provide the data used in our paper to analyze two image classification datasets: CIFAR-10 and (a modified version of) FMoW.

For each dataset, the data consists of two parts:

  1. Training data for datamodeling, which consists of:
    • Training subsets or "training masks", which are the independent variables of the regression tasks; and
    • Model outputs (correct-class margins and logits), which are the dependent variables of the regression tasks.
  2. Datamodels estimated from this data using LASSO.

For each dataset, there are multiple versions of the data depending on the choice of the hyperparameter α, the subsampling fraction (this is the random fraction of training examples on which each model is trained; see Section 2 of our paper for more information).

Following table shows the number of models we trained and used for estimating datamodels (also see Table 1 in paper):

Subsampling α (%) CIFAR-10 FMoW
10 1,500,000 N/A
20 750,000 375,000
50 300,000 150,000
75 600,000 300,000

Training data

For each dataset and $\alpha$, we provide the following data:

# M is the number of models trained
/{DATASET}/data/train_masks_{PCT}pct.npy  # [M x N_train] boolean
/{DATASET}/data/test_margins_{PCT}pct.npy # [M x N_test] np.float16
/{DATASET}/data/test_margins_{PCT}pct.npy # [M x N_train] np.float16

(The files live in the Amazon S3 bucket madrylab-datamodels; we provide instructions for acces in the next section.)

Each row of the above matrices corresponds to one instance of model trained; each column corresponds to a training or test example. CIFAR-10 examples are organized in the default order; for FMoW, see here. For example, a train mask for CIFAR-10 has the shape [M x 50,000].

For CIFAR-10, we also provide the full logits for all ten classes:

/cifar/data/train_logits_{PCT}pct.npy  # [M x N_test x 10] np.float16
/cifar/data/test_logits_{PCT}pct.npy   # [M x N_test x 10] np.float16

Note that you can also compute the margins from these logits.

We include an addtional 10,000 models for each setting that we used for evaluation; the total number of models in each matrix is M as indicated in the above table plus 10,000.

Datamodels

All estimated datamodels for each split (train or test) are provided as a dictionary in a .pt file (load with torch.load):

/{DATASET}/datamodels/train_{PCT}pct.pt
/{DATASET}/datamodels/test_{PCT}pct.pt

Each dictionary contains:

  • weight: matrix of shape N_train x N, where N is either N_train or N_test depending on the group of target examples
  • bias: vector of length N, corresponding to biases for each datamodel
  • lam: vector of length N, regularization λ chosen by CV for each datamodel

Downloading

We make all of our data available via Amazon S3. Total sizes of the training data files are as follows:

Dataset, α (%) masks, margins (GB) logits (GB)
CIFAR-10, 10 245 1688
CIFAR-10, 20 123 849
CIFAR-10, 50 49 346
CIFAR-10, 75 98 682
FMoW, 20 25.4 -
FMoW, 50 10.6 -
FMoW, 75 21.2 -

Total sizes of datamodels data (the model weights) are 16.9 GB for CIFAR-10 and 0.75 GB for FMoW.

API

You can download them using the Amazon S3 CLI interface with the requester pays option as follows (replacing the fields {...} as appropriate):

aws s3api get-object --bucket madrylab-datamodels \
                     --key {DATASET}/data/{SPLIT}_{DATA_TYPE}_{PCT}.npy \
                     --request-payer requester \
                     [OUT_FILE]

For example, to retrieve the test set margins for CIFAR-10 models trained on 50% subsets, use:

aws s3api get-object --bucket madrylab-datamodels \
                     --key cifar/data/test_margins_50pct.npy \
                     --request-payer requester \
                     test_margins_50pct.npy

Pricing

The total data transfer fee (from AWS to internet) for all of the data is around $374 (= 4155 GB x 0.09 USD per GB).

If you only download everything except for the logits (which is sufficient to reproduce all of our analysis), the fee is around $53.

Loading data

The data matrices are in numpy array format (.npy). As some of these are quite large, you can read small segments without reading the entire file into memory by additionally specifying the mmap_mode argument in np.load:

X = np.load('train_masks_10pct.npy', mmap_mode='r')
Y = np.load('test_margins_10pct.npy', mmap_mode='r')
...
# Use segments, e.g, X[:100], as appropriate
# Run regress(X, Y[:]) using choice of estimation algorithm.

FMoW data

We use a customized version of the FMoW dataset from WILDS (derived from this original dataset) that restricts the year of the training set to 2012. Our code is adapted from here.

To use the dataset, first download WILDS using:

pip install wilds

(see here for more detailed instructions).

In our paper, we only use the in-distribution training and test splits in our analysis (the original version from WILDS also has out-of-distribution as well as validation splits). Our dataset splits can be constructed as follows and used like a PyTorch dataset:

from fmow import FMoWDataset

ds = FMoWDataset(root_dir='/mnt/nfs/datasets/wilds/',
                     split_scheme='time_after_2016')

transform_steps = [
    transforms.ToTensor(),
    transforms.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225])
    ]
transform = transforms.Compose(transform_steps)

ds_train = ds.get_subset('train', transform=transform)
ds_test = ds.get_subset('id_test', transform=transform)

The columns of matrix data described above is ordered according to the default ordering of examples given by the above constructors.

Owner
Madry Lab
Towards a Principled Science of Deep Learning
Madry Lab
A handy tool for common machine learning models' hyper-parameter tuning.

Common machine learning models' hyperparameter tuning This repo is for a collection of hyper-parameter tuning for "common" machine learning models, in

Kevin Hu 2 Jan 27, 2022
Predicting Keystrokes using an Audio Side-Channel Attack and Machine Learning

Predicting Keystrokes using an Audio Side-Channel Attack and Machine Learning My

3 Apr 10, 2022
Forecast dynamically at scale with this unique package. pip install scalecast

🌄 Scalecast: Dynamic Forecasting at Scale About This package uses a scaleable forecasting approach in Python with common scikit-learn and statsmodels

Michael Keith 158 Jan 03, 2023
UpliftML: A Python Package for Scalable Uplift Modeling

UpliftML is a Python package for scalable unconstrained and constrained uplift modeling from experimental data. To accommodate working with big data, the package uses PySpark and H2O models as base l

Booking.com 254 Dec 31, 2022
[DEPRECATED] Tensorflow wrapper for DataFrames on Apache Spark

TensorFrames (Deprecated) Note: TensorFrames is deprecated. You can use pandas UDF instead. Experimental TensorFlow binding for Scala and Apache Spark

Databricks 757 Dec 31, 2022
Implementations of Machine Learning models, Regularizers, Optimizers and different Cost functions.

Linear Models Implementations of LinearRegression, LassoRegression and RidgeRegression with appropriate Regularizers and Optimizers. Linear Regression

Keivan Ipchi Hagh 1 Nov 22, 2021
XGBoost + Optuna

AutoXGB XGBoost + Optuna: no brainer auto train xgboost directly from CSV files auto tune xgboost using optuna auto serve best xgboot model using fast

abhishek thakur 517 Dec 31, 2022
PySpark + Scikit-learn = Sparkit-learn

Sparkit-learn PySpark + Scikit-learn = Sparkit-learn GitHub: https://github.com/lensacom/sparkit-learn About Sparkit-learn aims to provide scikit-lear

Lensa 1.1k Jan 04, 2023
Distributed Tensorflow, Keras and PyTorch on Apache Spark/Flink & Ray

A unified Data Analytics and AI platform for distributed TensorFlow, Keras and PyTorch on Apache Spark/Flink & Ray What is Analytics Zoo? Analytics Zo

2.5k Dec 28, 2022
Traingenerator 🧙 A web app to generate template code for machine learning ✨

Traingenerator 🧙 A web app to generate template code for machine learning ✨ 🎉 Traingenerator is now live! 🎉

Johannes Rieke 1.2k Jan 07, 2023
pure-predict: Machine learning prediction in pure Python

pure-predict speeds up and slims down machine learning prediction applications. It is a foundational tool for serverless inference or small batch prediction with popular machine learning frameworks l

Ibotta 84 Dec 29, 2022
Predict the demand for electricity (R) - FRENCH

06.demand-electricity Predict the demand for electricity (R) - FRENCH Prédisez la demande en électricité Prérequis Pour effectuer ce projet, vous devr

1 Feb 13, 2022
Anytime Learning At Macroscale

On Anytime Learning At Macroscale Learning from sequential data dumps (key) Requirements Python 3.7 Pytorch 1.9.0 Hydra 1.1.0 (pip install hydra-core

Meta Research 8 Mar 29, 2022
Predict the output which should give a fair idea about the chances of admission for a student for a particular university

Predict the output which should give a fair idea about the chances of admission for a student for a particular university.

ArvindSandhu 1 Jan 11, 2022
决策树分类与回归模型的实现和可视化

DecisionTree 决策树分类与回归模型,以及可视化 DecisionTree ID3 C4.5 CART 分类 回归 决策树绘制 分类树 回归树 调参 剪枝 ID3 ID3决策树是最朴素的决策树分类器: 无剪枝 只支持离散属性 采用信息增益准则 在data.py中,我们记录了一个小的西瓜数据

Welt Xing 10 Oct 22, 2022
A Python implementation of GRAIL, a generic framework to learn compact time series representations.

GRAIL A Python implementation of GRAIL, a generic framework to learn compact time series representations. Requirements Python 3.6+ numpy scipy tslearn

3 Nov 24, 2021
Time-series momentum for momentum investing strategy

Time-series-momentum Time-series momentum strategy. You can use the data_analysis.py file to find out the best trigger and window for a given asset an

Victor Caldeira 3 Jun 18, 2022
Mosec is a high-performance and flexible model serving framework for building ML model-enabled backend and microservices

Mosec is a high-performance and flexible model serving framework for building ML model-enabled backend and microservices. It bridges the gap between any machine learning models you just trained and t

164 Jan 04, 2023
Home repository for the Regularized Greedy Forest (RGF) library. It includes original implementation from the paper and multithreaded one written in C++, along with various language-specific wrappers.

Regularized Greedy Forest Regularized Greedy Forest (RGF) is a tree ensemble machine learning method described in this paper. RGF can deliver better r

RGF-team 363 Dec 14, 2022
A Python step-by-step primer for Machine Learning and Optimization

early-ML Presentation General Machine Learning tutorials A Python step-by-step primer for Machine Learning and Optimization This github repository gat

Dimitri Bettebghor 8 Dec 01, 2022