Ecosyste.ms: Awesome

An open API service indexing awesome lists of open source software.

Awesome Lists | Featured Topics | Projects

https://github.com/DmitryUlyanov/Multicore-TSNE

Parallel t-SNE implementation with Python and Torch wrappers.
https://github.com/DmitryUlyanov/Multicore-TSNE

barnes-hut-tsne multicore py-bh-tsne tsne

Last synced: 3 months ago
JSON representation

Parallel t-SNE implementation with Python and Torch wrappers.

Awesome Lists containing this project

README 

        
# Multicore t-SNE [![Build Status](https://travis-ci.org/DmitryUlyanov/Multicore-TSNE.svg?branch=master)](https://travis-ci.org/DmitryUlyanov/Multicore-TSNE)



This is a multicore modification of [Barnes-Hut t-SNE](https://github.com/lvdmaaten/bhtsne) by L. Van der Maaten with Python CFFI-based wrappers. This code also works **faster than sklearn.TSNE** on 1 core (as of version 0.18).






# What to expect



Barnes-Hut t-SNE is done in two steps.



- First step: an efficient data structure for nearest neighbours search is built and used to compute probabilities. This can be done in parallel for each point in the dataset, this is why we can expect a good speed-up by using more cores.



- Second step: the embedding is optimized using gradient descent. This part is essentially consecutive so we can only optimize within iteration. In fact some parts can be parallelized effectively, but not all of them a parallelized for now. That is why the second step speed-up will not be as significant as first step sepeed-up but there is still room for improvement.



So when can you benefit from parallelization? It is almost true, that the second step computation time is constant of `D` and depends mostly on `N`. The first part's time depends on `D` a lot, so for small `D` `time(Step 1) << time(Step 2)`, for large `D` `time(Step 1) >> time(Step 2)`. As we are only good at parallelizing step 1 we will benefit most when `D` is large enough (MNIST's `D = 784` is large, `D = 10` even for `N=1000000` is not so much). I wrote multicore modification originally for [Springleaf competition](https://www.kaggle.com/c/springleaf-marketing-response), where my data table was about `300000 x 3000` and only several days left till the end of the competition so any speed-up was handy.



# Benchmark



### 1 core



Interestingly, this code beats other implementations. We compare to `sklearn` (Barnes-Hut of course), L. Van der Maaten's [bhtsne](https://github.com/lvdmaaten/bhtsne), [py_bh_tsne repo](https://github.com/danielfrg/tsne) (cython wrapper for bhtsne with QuadTree). `perplexity = 30, theta=0.5` for every run. In fact [py_bh_tsne repo](https://github.com/danielfrg/tsne) works at the same speed as this code when using more optimization flags for the compiler.



This is a benchmark for `70000x784` MNIST data:



| Method                       | Step 1 (sec)   | Step 2 (sec)  |

| ---------------------------- |:---------------:| --------------:|

| MulticoreTSNE(n_jobs=1)      | **912**         | **350**        |

| bhtsne                       | 4257            | 1233           |

| py_bh_tsne                   | 1232            | 367            |

| sklearn(0.18)                | ~5400           | ~20920         |



I did my best to find what is wrong with sklearn numbers, but it is the best benchmark I could do (you can find the test script in `MulticoreTSNE/examples` folder).



### Multicore



This table shows a relative to 1 core speed-up when using `n` cores.



| n_jobs        | Step 1    | Step 2   |

| ------------- |:---------:| --------:|

| 1             | 1x        | 1x       |

| 2             | 1.54x     | 1.05x    |

| 4             | 2.6x      | 1.2x     |

| 8             | 5.6x      | 1.65x    |



# How to use



### Install



#### Directly from pypi

`pip install MulticoreTSNE`



#### From source



Make sure `cmake` is installed on your system, and you will also need a sensible C++ compiler, such as `gcc` or `llvm-clang`. On macOS, you can get both via [homebrew](https://brew.sh/).



To install the package, please do:

```bash

git clone https://github.com/DmitryUlyanov/Multicore-TSNE.git

cd Multicore-TSNE/

pip install .

```



Tested with python >= 3.6 (conda).



### Run



You can use it as a near drop-in replacement for [sklearn.manifold.TSNE](http://scikit-learn.org/stable/modules/generated/sklearn.manifold.TSNE.html).



```python

from MulticoreTSNE import MulticoreTSNE as TSNE



tsne = TSNE(n_jobs=4)

Y = tsne.fit_transform(X)

```



Please refer to [sklearn TSNE manual](http://scikit-learn.org/stable/modules/generated/sklearn.manifold.TSNE.html) for parameters explanation.



This implementation `n_components=2`, which is the most common case (use [Barnes-Hut t-SNE](https://github.com/lvdmaaten/bhtsne) or sklearn otherwise). Also note that some parameters are there just for the sake of compatibility with sklearn and are otherwise ignored. See `MulticoreTSNE` class docstring for more info.



#### MNIST example

```python

from sklearn.datasets import fetch_openml

from MulticoreTSNE import MulticoreTSNE as TSNE

from matplotlib import pyplot as plt



X, _ = fetch_openml(

  "mnist_784", version=1, return_X_y=True, as_frame=False, parser="pandas"

)

embeddings = TSNE(n_jobs=4).fit_transform(X)

vis_x = embeddings[:, 0]

vis_y = embeddings[:, 1]

plt.scatter(vis_x, vis_y, c=digits.target, cmap=plt.cm.get_cmap("jet", 10), marker='.')

plt.colorbar(ticks=range(10))

plt.clim(-0.5, 9.5)

plt.show()

```



### Test



You can test it on MNIST dataset with the following command:



```bash

python MulticoreTSNE/examples/test.py --n_jobs 

```



#### Note on jupyter use

To make the computation log visible in jupyter please install `wurlitzer` (`pip install wurlitzer`) and execute this line in any cell beforehand:

```

%load_ext wurlitzer

```

Memory leakages are possible if you interrupt the process. Should be OK if you let it run until the end.



# License



Inherited from [original repo's license](https://github.com/lvdmaaten/bhtsne).



# Future work



- Allow other types than double

- Improve step 2 performance (possible)



# Citation



Please cite this repository if it was useful for your research:



```

@misc{Ulyanov2016,

  author = {Ulyanov, Dmitry},

  title = {Multicore-TSNE},

  year = {2016},

  publisher = {GitHub},

  journal = {GitHub repository},

  howpublished = {\url{https://github.com/DmitryUlyanov/Multicore-TSNE}},

}

```



Of course, do not forget to cite [L. Van der Maaten's paper](http://lvdmaaten.github.io/publications/papers/JMLR_2014.pdf)