-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathindex.fsx
167 lines (120 loc) · 7.21 KB
/
index.fsx
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
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
#r "nuget: DiffSharp-lite,1.0.7"
(**
[](https://colab.research.google.com/github/DiffSharp/diffsharp.github.io/blob/master/index.ipynb) 
[](https://mybinder.org/v2/gh/diffsharp/diffsharp.github.io/master?filepath=index.ipynb) 
[](index.fsx) 
[](index.ipynb)
# DiffSharp: Differentiable Tensor Programming Made Simple
DiffSharp is a tensor library with support for [differentiable programming](https://en.wikipedia.org/wiki/Differentiable_programming).
It is designed for use in machine learning, probabilistic programming, optimization and other domains.
*)
<button class="button" style="vertical-align:middle" onclick="window.location.href='https://diffsharp.github.io/install.html'"><span>Install »</span></button>
(**
## Key Features
🗹 Nested and mixed-mode differentiation
🗹 Common optimizers, model elements, differentiable probability distributions
🗹 F# for robust functional programming
🗹 PyTorch familiar naming and idioms, efficient LibTorch CUDA/C++ tensors with GPU support
🗹 Linux, macOS, Windows supported
🗹 Use interactive notebooks in Jupyter and Visual Studio Code
🗹 100% open source
## Differentiable Programming
DiffSharp provides world-leading automatic differentiation capabilities for tensor code, including composable gradients, Hessians, Jacobians, directional derivatives, and matrix-free Hessian- and Jacobian-vector products over arbitrary user code. This goes beyond conventional tensor libraries such as PyTorch and TensorFlow, allowing the use of nested forward and reverse differentiation up to any level.
With DiffSharp, you can compute higher-order derivatives efficiently and differentiate functions that are internally making use of differentiation and gradient-based optimization.
*)
</br>
<img src="img/anim-intro-2.gif" width="85%" />
(**
## Practical, Familiar and Efficient
DiffSharp comes with a [LibTorch](https://pytorch.org/cppdocs/) backend, using the same C++ and CUDA implementations for tensor computations that power [PyTorch](https://pytorch.org/). On top of these raw tensors (LibTorch's ATen, excluding autograd), DiffSharp implements its own computation graph and differentiation capabilities. It is tested on Linux, macOS, and Windows, and it supports CUDA and GPUs.
The DiffSharp API is designed to be similar to [the PyTorch Python API](https://pytorch.org/docs/stable/index.html) through very similar naming and idioms, and where elements have similar names the PyTorch documentation can generally be used as a guide.
DiffSharp uses [the incredible F# programming language](https://dot.net/fsharp) for tensor programming. F# code is generally faster and more robust than equivalent Python code, while still being succinct and compact like Python, making it an ideal modern AI and machine learning implementation language. This allows fluent and productive code for tensor programming.
*)
</br>
<iframe width="85%" src="https://www.youtube.com/embed/_QnbV6CAWXc" title="YouTube video player" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture" allowfullscreen></iframe>
(**
## Interactive Notebooks
All documentation pages in this website are interactive notebooks which you can execute directly in your browser without installing anything in your local machine.
Using the buttons [](https://colab.research.google.com/github/DiffSharp/diffsharp.github.io/blob/master/index.ipynb) [](https://mybinder.org/v2/gh/diffsharp/diffsharp.github.io/master?filepath=index.ipynb) on the top of each page, you can execute the page as an interactive notebook running on cloud servers provided by [Google Colab](https://colab.research.google.com/) and [Binder](https://mybinder.org/).
Using the buttons [](index.fsx)
[](index.ipynb) you can also download a page as a script or an interactive notebook, which you can execute locally in [Jupyter](https://jupyter.org/) or [Visual Studio Code](https://code.visualstudio.com/) using [dotnet interactive](https://github.com/dotnet/interactive).
## Example
Define and add two tensors:
*)
open DiffSharp
let t1 = dsharp.tensor [ 0.0 ..0.2.. 1.0 ] // Gives [0., 0.2, 0.4, 0.6, 0.8, 1.]
let t2 = dsharp.tensor [ 1, 2, 3, 4, 5, 6 ]
t1 + t2(* output:
tensor([1.0000, 2.2000, 3.4000, 4.6000, 5.8000, 7.0000])*)
(**
Compute a convolution:
*)
let t3 = dsharp.tensor [[[[0.0 .. 10.0]]]]
let t4 = dsharp.tensor [[[[0.0 ..0.1.. 1.0]]]]
t3.conv2d(t4)(* output:
tensor([[[[38.5000]]]])*)
(**
Take the gradient of a vector-to-scalar function:
*)
let f (x: Tensor) = x.exp().sum()
dsharp.grad f (dsharp.tensor([1.8, 2.5]))(* output:
tensor([ 6.0496, 12.1825])*)
(**
Compute a nested derivative (checking for [perturbation confusion](https://doi.org/10.1007/s10990-008-9037-1)):
*)
let x0 = dsharp.tensor(1.)
let y0 = dsharp.tensor(2.)
dsharp.diff (fun x -> x * dsharp.diff (fun y -> x * y) y0) x0(* output:
tensor(2.)*)
(**
Define a model and optimize it:
*)
open DiffSharp.Data
open DiffSharp.Model
open DiffSharp.Compose
open DiffSharp.Util
open DiffSharp.Optim
let epochs = 2
let batchSize = 32
let numBatches = 5
let trainSet = MNIST("../data", train=true, transform=id)
let trainLoader = trainSet.loader(batchSize=batchSize, shuffle=true)
let validSet = MNIST("../data", train=false, transform=id)
let validLoader = validSet.loader(batchSize=batchSize, shuffle=false)
let encoder =
Conv2d(1, 32, 4, 2)
--> dsharp.relu
--> Conv2d(32, 64, 4, 2)
--> dsharp.relu
--> Conv2d(64, 128, 4, 2)
--> dsharp.flatten(1)
let decoder =
dsharp.unflatten(1, [128;1;1])
--> ConvTranspose2d(128, 64, 4, 2)
--> dsharp.relu
--> ConvTranspose2d(64, 32, 4, 3)
--> dsharp.relu
--> ConvTranspose2d(32, 1, 4, 2)
--> dsharp.sigmoid
let model = VAE([1;28;28], 64, encoder, decoder)
let lr = dsharp.tensor(0.001)
let optimizer = Adam(model, lr=lr)
for epoch = 1 to epochs do
let batches = trainLoader.epoch(numBatches)
for i, x, _ in batches do
model.reverseDiff()
let l = model.loss(x)
l.reverse()
optimizer.step()
print $"Epoch: {epoch} minibatch: {i} loss: {l}"
let validLoss =
validLoader.epoch()
|> Seq.sumBy (fun (_, x, _) -> model.loss(x, normalize=false))
print $"Validation loss: {validLoss/validSet.length}"
(**
Numerous other model definition, differentiation, and training patterns are supported. See the tutorials in the left-hand menu and [examples](https://github.com/DiffSharp/DiffSharp/tree/dev/examples) on GitHub.
## More Information
DiffSharp is developed by [Atılım Güneş Baydin](http://www.robots.ox.ac.uk/~gunes/), [Don Syme](https://www.microsoft.com/en-us/research/people/dsyme/)
and other contributors, having started as a project supervised by the automatic differentiation wizards [Barak Pearlmutter](https://scholar.google.com/citations?user=AxFrw0sAAAAJ&hl=en) and [Jeffrey Siskind](https://scholar.google.com/citations?user=CgSBtPYAAAAJ&hl=en).
Please join us [on GitHub](https://github.com/DiffSharp/DiffSharp)!
*)