evoasm
所属分类：人工智能/神经网络/深度学习
开发工具：kotlin
文件大小：43KB
下载次数：0
上传日期：2019-10-23 17:21:49
上 传 者：
sh-1993
说明： 一种AIMGP（通过遗传编程自动归纳机器代码）引擎
(An AIMGP (Automatic Induction of Machine code by Genetic Programming) engine)
文件列表:
.idea (0, 2019-10-24)
.idea\kotlinc.xml (286, 2019-10-24)
.idea\libraries (0, 2019-10-24)
.idea\libraries\KotlinJavaRuntime.xml (906, 2019-10-24)
.idea\libraries\KotlinJavaRuntime__2_.xml (910, 2019-10-24)
.idea\libraries\org_junit_jupiter_junit_jupiter_params_5_3_2.xml (859, 2019-10-24)
.idea\misc.xml (451, 2019-10-24)
.idea\modules.xml (925, 2019-10-24)
.idea\uiDesigner.xml (8792, 2019-10-24)
.idea\vcs.xml (180, 2019-10-24)
evoasm-gen (0, 2019-10-24)
evoasm-gen\evoasm-gen.iml (502, 2019-10-24)
evoasm-gen\src (0, 2019-10-24)
evoasm-gen\src\evoasm (0, 2019-10-24)
evoasm-gen\src\evoasm\gen (0, 2019-10-24)
evoasm-gen\src\evoasm\gen\x64 (0, 2019-10-24)
evoasm-gen\src\evoasm\gen\x64\Generator.kt (1897, 2019-10-24)
evoasm.iml (2209, 2019-10-24)
src (0, 2019-10-24)
src\evoasm (0, 2019-10-24)
src\evoasm\Ext.kt (179, 2019-10-24)
src\evoasm\FastRandom.kt (650, 2019-10-24)
src\evoasm\x64 (0, 2019-10-24)
src\evoasm\x64\AbstractPopulation.kt (13745, 2019-10-24)
src\evoasm\x64\DoublePopulation.kt (560, 2019-10-24)
src\evoasm\x64\DoubleSampleSet.kt (825, 2019-10-24)
src\evoasm\x64\DoubleVectorPopulation.kt (1532, 2019-10-24)
src\evoasm\x64\DoubleVectorSampleSet.kt (1198, 2019-10-24)
src\evoasm\x64\InstructionFinder.kt (2727, 2019-10-24)
src\evoasm\x64\InstructionGroup.kt (53597, 2019-10-24)
src\evoasm\x64\Instructions.kt (344, 2019-10-24)
src\evoasm\x64\Interpreter.kt (42776, 2019-10-24)
src\evoasm\x64\InterpreterOptions.kt (4613, 2019-10-24)
src\evoasm\x64\IntronEliminator.kt (4824, 2019-10-24)
src\evoasm\x64\Populations.kt (3078, 2019-10-24)
src\evoasm\x64\ProgramSet.kt (8699, 2019-10-24)
src\evoasm\x64\ProgramSetInput.kt (14340, 2019-10-24)
... ...
# Evoasm
## Description
*Evoasm* is an AIMGP (*Automatic Induction of Machine code by Genetic Programming*) engine.
You give it a set of examples, that is, several input/output pairs, that describe a program's behavior.
It will then try to come up with a short program (in the form of machine code) that follows your specification,
by means of genetic programming.
*Evoasm* contains a JIT that executes the generated machine code on the fly.
Currently, the only supported architecture is **x86-*****.
***NOTE**: Evoasm is currently WIP and unfinished. However, you
should be able to import it into IDEA and run the examples.*
## Interpreter
*Evoasm* just-in-time compiles a fast interpreter at runtime. The interpreter instructions
map one-to-one to *x86-**** instructions. By using an interpreter, *Evoasm* avoids the need
to re-JIT program candidates.
## Features
* Fast direct-threaded interpreter
* [x86-***](https://github.com/evoasm/kasm) up to AVX2 (no FPU)
* Parallel evaluation using OpenMP
## Examples
### Float Values
```kotlin
val options = PopulationOptions(
128_000,
4,
123456,
12,
InterpreterOptions(instructions = InstructionGroup.ARITHMETIC_SS_AVX_XMM_INSTRUCTIONS.instructions.filterNot { it == VroundssXmmXmmXmmm32Imm8 },
moveInstructions = listOf(),
compressOpcodes = false,
forceMultithreading = false,
threadCount = 5,
unsafe = false),
0.01f,
demeSize = 10,
majorGenerationFrequency = 1,
maxOffspringRatio = 0.05
)
// sqrt(x**3 + 2*x)
val sampleSet = FloatSampleSet(1,
0.0f, 0.0f,
0.5f, 1.06066017177***212f,
1.0f, 1.7320508075688772f,
1.5f, 2.5248762345905194f,
2.0f, 3.4***1016151377544f,
2.5f, 4.541475531146237f,
3.0f, 5.744562***6538029f,
3.5f, 7.0622234459127675f,
4.0f, 8.48528137423857f,
4.5f, 10.00624804809475f,
5.0f, 11.61895003862225f
)
val population = FloatPopulation(sampleSet, options)
population.run()
```
Result will be:
```
vfmsub231ss xmm0, xmm0, xmm0
vfmadd231ss xmm2, xmm1, xmm0
vaddss xmm2, xmm1, xmm2
vfmadd132ss xmm1, xmm2, xmm1
vminss xmm2, xmm2, xmm1
vfnmsub132ss xmm2, xmm0, xmm1
vfnmsub213ss xmm2, xmm2, xmm1
vsqrtss xmm0, xmm2, xmm1
```
### Double Vector Values
```kotlin
val options = PopulationOptions(
2000,
4,
Random.nextLong(),//1234567,
1,
InterpreterOptions(instructions = InstructionGroup.ARITHMETIC_PD_AVX_YMM_INSTRUCTIONS.instructions,
moveInstructions = listOf(),
unsafe = true),
0.01f,
demeSize = 1,
majorGenerationFrequency = 1
)
// simple vector addition
val sampleSet = DoubleVectorSampleSet(2, VectorSize.BITS_256,
0.0, 0.0, 0.0, 0.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0,
1.0, 1.0, 1.0, 1.0, 2.0, 2.0, 2.0, 2.0, 3.0, 3.0, 3.0, 3.0,
2.0, 2.0, 2.0, 2.0, 2.0, 2.0, 2.0, 2.0, 4.0, 4.0, 4.0, 4.0)
val population = DoubleVectorPopulation(sampleSet, options)
population.run()
```
Result will be (program size was 1):
```kotlin
vaddpd ymm0, ymm2, ymm1
```
### Instruction Finder
```kotlin
val finder = InstructionFinder(1,
0b0111, 3,
0b1, 1,
0b1111, 4,
0b11111, 5)
finder.find()
val finder2 = InstructionFinder(2,
2, 2, 4,
16, 4, 20,
100, 10, 110,
144, 12, 156)
finder2.find()
```
Result will be:
```
FOUND 3 instructions:
kasm.x***.PopcntR16Rm16@6bd61f***
kasm.x***.PopcntR16Rm16@6bd61f***
kasm.x***.PopcntR16Rm16@6bd61f***
FOUND 20 instructions:
kasm.x***.AdcRm8R8@3059cbc
kasm.x***.AdcR8Rm8@24fcf36f
kasm.x***.AddRm8R8@ea6147e
kasm.x***.AddR8Rm8@4d02f94e
kasm.x***.AdcRm16R16@7e5afaa6
kasm.x***.AdcR16Rm16@63a12c68
kasm.x***.AddRm16R16@581ac8a8
kasm.x***.AddR16Rm16@6d4e5011
kasm.x***.AdcRm32R32@4d465b11
kasm.x***.AdcR32Rm32@53fdffa1
kasm.x***.AdcxR32Rm32@5562c41e
kasm.x***.AddRm32R32@3232a28a
kasm.x***.AddR32Rm32@73e22a3d
kasm.x***.AdoxR32Rm32@47faa49c
kasm.x***.AdcRm***R***@66d3eec0
kasm.x***.AdcR***Rm***@1e04fa0a
kasm.x***.AdcxR***Rm***@1af2d44a
kasm.x***.AddRm***R***@543588e6
kasm.x***.AddR***Rm***@f5acb9d
kasm.x***.AdoxR***Rm***@4fb3ee4e
```
## Installation
First, install [kasm](https://github.com/evoasm/kasm).
Then import the project into IDEA.
Note that *kasm* and *evoasm* repositories should reside inside the same directory.
## License
[MPL-2.0][license]
近期下载者：
相关文件：
收藏者：