README.md 4.93 KB
Newer Older
Gustavo Valiente's avatar
Gustavo Valiente committed
1
# pocket-tensor
Gustavo Valiente's avatar
Readme  
Gustavo Valiente committed
2

Gustavo Valiente's avatar
Gustavo Valiente committed
3
pocket-tensor is a [Kerasify](https://github.com/moof2k/kerasify) fork designed for running trained Keras models from a C++ application on embedded devices.
Gustavo Valiente's avatar
Readme  
Gustavo Valiente committed
4 5 6 7

## Design goals

* Compatibility with sequential networks generated by Keras 2.x using Tensorflow backend.
8 9
* Multithread CPU support (no GPU support).
* Low RAM usage.
Gustavo Valiente's avatar
Readme  
Gustavo Valiente committed
10
* Easy to build and run (no external dependencies).
11
* Fast build times.
Gustavo Valiente's avatar
Readme  
Gustavo Valiente committed
12 13 14 15 16 17 18 19 20 21 22

## Improvements over Kerasify

* Thanks to the awesome [libsimdpp library](https://github.com/p12tic/libsimdpp), tensor operations have been rewritten using SIMD instructions to improve prediction performance.
* Memory (re)usage has been improved in order to reduce memory allocations.
* Apart from `float`, `double` precision tensors are supported (see `pt_tweakme.h` file).
* Tensor dimensions are rigorously validated on each layer to avoid bad models usage.
* Besides GCC and Clang, Visual Studio compiler is properly supported.

## Hardware requirements

Gustavo Valiente's avatar
Gustavo Valiente committed
23
Since there's no GPU support, by default pocket-tensor requires the following CPU SIMD instruction sets:
Gustavo Valiente's avatar
Readme  
Gustavo Valiente committed
24 25 26 27 28 29 30 31 32 33

* ARM: NEON with floating point support.
* x86: AVX.

Required SIMD instruction sets are specified in the `pt_tweakme.h` file, so they can be modified with ease.

## Software requirements

Since a copy of libsimdpp comes bundled with this library, there's no external dependencies required, so the only software requirements are a C++11-compatible compiler and CMake >= 3.4.  

Gustavo Valiente's avatar
Gustavo Valiente committed
34
pocket-tensor has been tested with these compilers: 
Gustavo Valiente's avatar
Readme  
Gustavo Valiente committed
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

* GCC 4.9.
* MSVC 2017.
* Whatever Clang comes with Apple LLVM 9.1.0.

## How to build

A CMakeLists.txt is provided with this library, so in order to use it you only need to include this file in your CMake project.  

To build and run the unit tests, you need to generate them first:

```
python make_tests.py
mkdir tests_build
cd tests_build
cmake -DPT_BUILD_TESTS=ON -DCMAKE_BUILD_TYPE=Release ../..
make
./tests/pocket-tensor-tests
```

## Usage

1) Use Keras to build (`model.compile(...)`) and train (`model.fit(...)`) your model as usual.

2) Now convert it to the Kerasify file format with `kerasify.export_model(model, 'example.model')`.

3) Finally load it in C++ (`pt::create("example.model")`) and use `model->predict(...)` to perform a prediction with your data.

The following example shows the full workflow:

```python
# make_model.py:

import numpy as np
from keras.models import Sequential
from keras.layers import Dense
from kerasify import export_model

test_x = np.random.rand(10, 10).astype('f')
test_y = np.random.rand(10).astype('f')

model = Sequential()
model.add(Dense(1, input_dim=10))

model.compile(loss='mean_squared_error', optimizer='adamax')
model.fit(test_x, test_y, epochs=1)

print model.predict(np.array([[0, 1, 2, 3, 4, 5, 6, 7, 8, 9]]))

export_model(model, 'example.model')
```

```cpp
// main.cpp:

#include <iostream>
#include "pt_model.h"
#include "pt_tensor.h"

int main()
{
    // Initialize model:
    auto model = pt::create("example.model");
Gustavo Valiente's avatar
Gustavo Valiente committed
98
    // REQUIRE(model);
Gustavo Valiente's avatar
Readme  
Gustavo Valiente committed
99 100 101 102 103 104 105 106

    // Create input tensor:
    pt::Tensor in(10);
    in.setData({0, 1, 2, 3, 4, 5, 6, 7, 8, 9});

    // Run prediction:
    pt::Tensor out;
    bool success = model->predict(std::move(in), out);
Gustavo Valiente's avatar
Gustavo Valiente committed
107
    // REQUIRE(success);
Gustavo Valiente's avatar
Readme  
Gustavo Valiente committed
108 109 110 111 112 113 114 115 116 117 118 119 120 121 122
	
    // Print output:
    std::cout << out << std::endl;
    return 0;
}
```

## Supported layer types

The most common layer types used in image recognition and sequences prediction are supported, making many popular model architectures possible:

* Convolutions: `Conv1D`, `Conv2D`, `LocallyConnected1D`.
* Sequences related: `LSTM`, `Embedding`.
* Activations: `Linear`, `ReLU`, `ELU`, `Softplus`, `Softsign`, `Tanh`, `Sigmoid`, `HardSigmoid`, `Softmax`.
* Other: `Dense`, `Flatten`, `MaxPooling2D`, `BatchNormalization`, `ELU`.
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

## Performance

The prediction time of the following models have been measured on a Raspberry Pi 3:

CNN:

```python
model = Sequential()
model.add(Conv2D(32, kernel_size=(3, 3),
                 activation='relu',
                 input_shape=input_shape))
model.add(Conv2D(64, (3, 3), activation='relu'))
model.add(MaxPooling2D(pool_size=(2, 2)))
model.add(Dropout(0.25))
model.add(Flatten())
model.add(Dense(128, activation='relu'))
model.add(Dropout(0.5))
model.add(Dense(num_classes, activation='sigmoid'))
```

| Library       | Elapsed time (us) |
| ------------- | ----------------: |
| Keras         |             23363 |
| Kerasify      |             64238 |
| frugally-deep |             29298 |
| pocket-tensor |             27329 |

LSTM:

```python
model = Sequential()
model.add(Embedding(max_features, 128))
model.add(LSTM(128, return_sequences=True, dropout=0.2, recurrent_dropout=0.2))
model.add(LSTM(128, return_sequences=False, dropout=0.2, recurrent_dropout=0.2))
model.add(Dense(1, activation='sigmoid'))
```

| Library       | Elapsed time (us) |
| ------------- | ----------------: |
| Keras         |             89344 |
| Kerasify      |             79060 |
| frugally-deep |     Not supported |
| pocket-tensor |             67115 |