tangram_tree 0.7.0

Tangram makes it easy for programmers to train, deploy, and monitor machine learning models.
Documentation 
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

# Tangram Tree Benchmarks

To run the benchmarks:

1. Install the nix package manager with flake support: https://nixos.wiki/wiki/Flakes.

2. Clone the tangram repo and `cd` into it: `git clone git@github.com:tangramdotdev/tangram && cd tangram`.

3. Download the datasets:

```
mkdir data
curl -sSL https://datasets.tangram.dev/allstate_train.csv > data/allstate_train.csv
curl -sSL https://datasets.tangram.dev/allstate_test.csv > data/allstate_test.csv
curl -sSL https://datasets.tangram.dev/flights_train.csv > data/flights_train.csv
curl -sSL https://datasets.tangram.dev/flights_test.csv > data/flights_test.csv
curl -sSL https://datasets.tangram.dev/higgs_train.csv > data/higgs_train.csv
curl -sSL https://datasets.tangram.dev/higgs_test.csv > data/higgs_test.csv
```

4. Run `nix develop` to enter the dev shell.

5. Run the benchmarks:

```
cargo run --release --bin tangram_tree_benchmarks -- \
  --libraries tangram lightgbm xgboost \
  --datasets allstate flights higgs
```