mirror of
https://github.com/wassname/keras-js.git
synced 2026-09-09 11:25:25 +08:00
implement UpSampling1D/2D/3D layers, with tests
This commit is contained in:
Leon Chen
@@ -49,6 +49,12 @@
|
||||
<script src="/test/convolutional/Convolution1D.js"></script>
|
||||
<script src="/test/convolutional/data_Convolution2D.js"></script>
|
||||
<script src="/test/convolutional/Convolution2D.js"></script>
|
||||
<script src="/test/convolutional/data_UpSampling1D.js"></script>
|
||||
<script src="/test/convolutional/UpSampling1D.js"></script>
|
||||
<script src="/test/convolutional/data_UpSampling2D.js"></script>
|
||||
<script src="/test/convolutional/UpSampling2D.js"></script>
|
||||
<script src="/test/convolutional/data_UpSampling3D.js"></script>
|
||||
<script src="/test/convolutional/UpSampling3D.js"></script>
|
||||
|
||||
<script>
|
||||
// mocha.checkLeaks();
|
||||
|
||||
@@ -51,7 +51,7 @@
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": 6,
|
||||
"execution_count": 15,
|
||||
"metadata": {
|
||||
"collapsed": false
|
||||
},
|
||||
@@ -109,7 +109,7 @@
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": 8,
|
||||
"execution_count": 16,
|
||||
"metadata": {
|
||||
"collapsed": false
|
||||
},
|
||||
@@ -163,7 +163,7 @@
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": 12,
|
||||
"execution_count": 17,
|
||||
"metadata": {
|
||||
"collapsed": false
|
||||
},
|
||||
@@ -216,12 +216,12 @@
|
||||
"cell_type": "markdown",
|
||||
"metadata": {},
|
||||
"source": [
|
||||
"**[convolutional.Convolution1D.4] 2 length 7 filters on 8x3 input, activation='tanh', border_mode='same', subsample_length=1, bias=True**"
|
||||
"**[convolutional.Convolution1D.3] 2 length 7 filters on 8x3 input, activation='tanh', border_mode='same', subsample_length=1, bias=True**"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": 14,
|
||||
"execution_count": 18,
|
||||
"metadata": {
|
||||
"collapsed": false
|
||||
},
|
||||
|
||||
@@ -0,0 +1,158 @@
|
||||
{
|
||||
"cells": [
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": 2,
|
||||
"metadata": {
|
||||
"collapsed": false
|
||||
},
|
||||
"outputs": [],
|
||||
"source": [
|
||||
"import numpy as np\n",
|
||||
"from keras.models import Model\n",
|
||||
"from keras.layers import Input\n",
|
||||
"from keras.layers.convolutional import UpSampling1D\n",
|
||||
"from keras import backend as K"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": 3,
|
||||
"metadata": {
|
||||
"collapsed": true
|
||||
},
|
||||
"outputs": [],
|
||||
"source": [
|
||||
"def format_decimal(arr, places=6):\n",
|
||||
" return [round(x * 10**places) / 10**places for x in arr]"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "markdown",
|
||||
"metadata": {},
|
||||
"source": [
|
||||
"### UpSampling1D"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "markdown",
|
||||
"metadata": {},
|
||||
"source": [
|
||||
"**[convolutional.UpSampling1D.0] length 2 upsampling on 3x5 input**"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": 5,
|
||||
"metadata": {
|
||||
"collapsed": false
|
||||
},
|
||||
"outputs": [
|
||||
{
|
||||
"name": "stdout",
|
||||
"output_type": "stream",
|
||||
"text": [
|
||||
"\n",
|
||||
"in shape: (3, 5)\n",
|
||||
"in: [0.262, 0.764609, -0.482897, -0.371755, 0.871769, 0.490033, -0.986894, -0.960468, 0.373039, 0.911356, 0.00298, 0.270652, 0.749006, 0.692235, 0.471778]\n",
|
||||
"out shape: (6, 5)\n",
|
||||
"out: [0.262, 0.764609, -0.482897, -0.371755, 0.871769, 0.262, 0.764609, -0.482897, -0.371755, 0.871769, 0.490033, -0.986894, -0.960468, 0.373039, 0.911356, 0.490033, -0.986894, -0.960468, 0.373039, 0.911356, 0.00298, 0.270652, 0.749006, 0.692235, 0.471778, 0.00298, 0.270652, 0.749006, 0.692235, 0.471778]\n"
|
||||
]
|
||||
}
|
||||
],
|
||||
"source": [
|
||||
"data_in_shape = (3, 5)\n",
|
||||
"L = UpSampling1D(length=2)\n",
|
||||
"\n",
|
||||
"layer_0 = Input(shape=data_in_shape)\n",
|
||||
"layer_1 = L(layer_0)\n",
|
||||
"model = Model(input=layer_0, output=layer_1)\n",
|
||||
"\n",
|
||||
"# set weights to random (use seed for reproducibility)\n",
|
||||
"np.random.seed(230)\n",
|
||||
"data_in = 2 * np.random.random(data_in_shape) - 1\n",
|
||||
"print('')\n",
|
||||
"print('in shape:', data_in_shape)\n",
|
||||
"print('in:', format_decimal(data_in.ravel().tolist()))\n",
|
||||
"result = model.predict(np.array([data_in]))\n",
|
||||
"print('out shape:', result[0].shape)\n",
|
||||
"print('out:', format_decimal(result[0].ravel().tolist()))"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "markdown",
|
||||
"metadata": {},
|
||||
"source": [
|
||||
"**[convolutional.UpSampling1D.0] length 3 upsampling on 4x4 input**"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": 6,
|
||||
"metadata": {
|
||||
"collapsed": false
|
||||
},
|
||||
"outputs": [
|
||||
{
|
||||
"name": "stdout",
|
||||
"output_type": "stream",
|
||||
"text": [
|
||||
"\n",
|
||||
"in shape: (4, 4)\n",
|
||||
"in: [0.562988, 0.168418, -0.14658, -0.369311, 0.653777, 0.806859, -0.922124, 0.830445, -0.878989, -0.638546, -0.855401, -0.082476, 0.416718, -0.03353, -0.949107, -0.866195]\n",
|
||||
"out shape: (12, 4)\n",
|
||||
"out: [0.562988, 0.168418, -0.14658, -0.369311, 0.562988, 0.168418, -0.14658, -0.369311, 0.562988, 0.168418, -0.14658, -0.369311, 0.653777, 0.806859, -0.922124, 0.830445, 0.653777, 0.806859, -0.922124, 0.830445, 0.653777, 0.806859, -0.922124, 0.830445, -0.878989, -0.638546, -0.855401, -0.082476, -0.878989, -0.638546, -0.855401, -0.082476, -0.878989, -0.638546, -0.855401, -0.082476, 0.416718, -0.03353, -0.949107, -0.866195, 0.416718, -0.03353, -0.949107, -0.866195, 0.416718, -0.03353, -0.949107, -0.866195]\n"
|
||||
]
|
||||
}
|
||||
],
|
||||
"source": [
|
||||
"data_in_shape = (4, 4)\n",
|
||||
"L = UpSampling1D(length=3)\n",
|
||||
"\n",
|
||||
"layer_0 = Input(shape=data_in_shape)\n",
|
||||
"layer_1 = L(layer_0)\n",
|
||||
"model = Model(input=layer_0, output=layer_1)\n",
|
||||
"\n",
|
||||
"# set weights to random (use seed for reproducibility)\n",
|
||||
"np.random.seed(231)\n",
|
||||
"data_in = 2 * np.random.random(data_in_shape) - 1\n",
|
||||
"print('')\n",
|
||||
"print('in shape:', data_in_shape)\n",
|
||||
"print('in:', format_decimal(data_in.ravel().tolist()))\n",
|
||||
"result = model.predict(np.array([data_in]))\n",
|
||||
"print('out shape:', result[0].shape)\n",
|
||||
"print('out:', format_decimal(result[0].ravel().tolist()))"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": null,
|
||||
"metadata": {
|
||||
"collapsed": true
|
||||
},
|
||||
"outputs": [],
|
||||
"source": []
|
||||
}
|
||||
],
|
||||
"metadata": {
|
||||
"kernelspec": {
|
||||
"display_name": "Python 3",
|
||||
"language": "python",
|
||||
"name": "python3"
|
||||
},
|
||||
"language_info": {
|
||||
"codemirror_mode": {
|
||||
"name": "ipython",
|
||||
"version": 3
|
||||
},
|
||||
"file_extension": ".py",
|
||||
"mimetype": "text/x-python",
|
||||
"name": "python",
|
||||
"nbconvert_exporter": "python",
|
||||
"pygments_lexer": "ipython3",
|
||||
"version": "3.5.2"
|
||||
}
|
||||
},
|
||||
"nbformat": 4,
|
||||
"nbformat_minor": 0
|
||||
}
|
||||
@@ -0,0 +1,256 @@
|
||||
{
|
||||
"cells": [
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": 1,
|
||||
"metadata": {
|
||||
"collapsed": false
|
||||
},
|
||||
"outputs": [
|
||||
{
|
||||
"name": "stderr",
|
||||
"output_type": "stream",
|
||||
"text": [
|
||||
"Using TensorFlow backend.\n"
|
||||
]
|
||||
}
|
||||
],
|
||||
"source": [
|
||||
"import numpy as np\n",
|
||||
"from keras.models import Model\n",
|
||||
"from keras.layers import Input\n",
|
||||
"from keras.layers.convolutional import UpSampling2D\n",
|
||||
"from keras import backend as K"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": 2,
|
||||
"metadata": {
|
||||
"collapsed": true
|
||||
},
|
||||
"outputs": [],
|
||||
"source": [
|
||||
"def format_decimal(arr, places=6):\n",
|
||||
" return [round(x * 10**places) / 10**places for x in arr]"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "markdown",
|
||||
"metadata": {},
|
||||
"source": [
|
||||
"### UpSampling2D"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "markdown",
|
||||
"metadata": {},
|
||||
"source": [
|
||||
"**[convolutional.UpSampling2D.0] size 2x2 upsampling on 3x3x3 input, dim_ordering='tf'**"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": 3,
|
||||
"metadata": {
|
||||
"collapsed": false
|
||||
},
|
||||
"outputs": [
|
||||
{
|
||||
"name": "stdout",
|
||||
"output_type": "stream",
|
||||
"text": [
|
||||
"\n",
|
||||
"in shape: (3, 3, 3)\n",
|
||||
"in: [-0.570441, -0.454673, -0.285321, 0.237249, 0.282682, 0.428035, 0.160547, -0.332203, 0.546391, 0.272735, 0.010827, -0.763164, -0.442696, 0.381948, -0.676994, 0.753553, -0.031788, 0.915329, -0.738844, 0.269075, 0.434091, 0.991585, -0.944288, 0.258834, 0.162138, 0.565201, -0.492094]\n",
|
||||
"out shape: (6, 6, 3)\n",
|
||||
"out: [-0.570441, -0.454673, -0.285321, -0.570441, -0.454673, -0.285321, 0.237249, 0.282682, 0.428035, 0.237249, 0.282682, 0.428035, 0.160547, -0.332203, 0.546391, 0.160547, -0.332203, 0.546391, -0.570441, -0.454673, -0.285321, -0.570441, -0.454673, -0.285321, 0.237249, 0.282682, 0.428035, 0.237249, 0.282682, 0.428035, 0.160547, -0.332203, 0.546391, 0.160547, -0.332203, 0.546391, 0.272735, 0.010827, -0.763164, 0.272735, 0.010827, -0.763164, -0.442696, 0.381948, -0.676994, -0.442696, 0.381948, -0.676994, 0.753553, -0.031788, 0.915329, 0.753553, -0.031788, 0.915329, 0.272735, 0.010827, -0.763164, 0.272735, 0.010827, -0.763164, -0.442696, 0.381948, -0.676994, -0.442696, 0.381948, -0.676994, 0.753553, -0.031788, 0.915329, 0.753553, -0.031788, 0.915329, -0.738844, 0.269075, 0.434091, -0.738844, 0.269075, 0.434091, 0.991585, -0.944288, 0.258834, 0.991585, -0.944288, 0.258834, 0.162138, 0.565201, -0.492094, 0.162138, 0.565201, -0.492094, -0.738844, 0.269075, 0.434091, -0.738844, 0.269075, 0.434091, 0.991585, -0.944288, 0.258834, 0.991585, -0.944288, 0.258834, 0.162138, 0.565201, -0.492094, 0.162138, 0.565201, -0.492094]\n"
|
||||
]
|
||||
}
|
||||
],
|
||||
"source": [
|
||||
"data_in_shape = (3, 3, 3)\n",
|
||||
"L = UpSampling2D(size=(2, 2), dim_ordering='tf')\n",
|
||||
"\n",
|
||||
"layer_0 = Input(shape=data_in_shape)\n",
|
||||
"layer_1 = L(layer_0)\n",
|
||||
"model = Model(input=layer_0, output=layer_1)\n",
|
||||
"\n",
|
||||
"# set weights to random (use seed for reproducibility)\n",
|
||||
"np.random.seed(250)\n",
|
||||
"data_in = 2 * np.random.random(data_in_shape) - 1\n",
|
||||
"print('')\n",
|
||||
"print('in shape:', data_in_shape)\n",
|
||||
"print('in:', format_decimal(data_in.ravel().tolist()))\n",
|
||||
"result = model.predict(np.array([data_in]))\n",
|
||||
"print('out shape:', result[0].shape)\n",
|
||||
"print('out:', format_decimal(result[0].ravel().tolist()))"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "markdown",
|
||||
"metadata": {},
|
||||
"source": [
|
||||
"**[convolutional.UpSampling2D.0] size 2x2 upsampling on 3x3x3 input, dim_ordering='th'**"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": 4,
|
||||
"metadata": {
|
||||
"collapsed": false
|
||||
},
|
||||
"outputs": [
|
||||
{
|
||||
"name": "stdout",
|
||||
"output_type": "stream",
|
||||
"text": [
|
||||
"\n",
|
||||
"in shape: (3, 3, 3)\n",
|
||||
"in: [-0.570441, -0.454673, -0.285321, 0.237249, 0.282682, 0.428035, 0.160547, -0.332203, 0.546391, 0.272735, 0.010827, -0.763164, -0.442696, 0.381948, -0.676994, 0.753553, -0.031788, 0.915329, -0.738844, 0.269075, 0.434091, 0.991585, -0.944288, 0.258834, 0.162138, 0.565201, -0.492094]\n",
|
||||
"out shape: (3, 6, 6)\n",
|
||||
"out: [-0.570441, -0.570441, -0.454673, -0.454673, -0.285321, -0.285321, -0.570441, -0.570441, -0.454673, -0.454673, -0.285321, -0.285321, 0.237249, 0.237249, 0.282682, 0.282682, 0.428035, 0.428035, 0.237249, 0.237249, 0.282682, 0.282682, 0.428035, 0.428035, 0.160547, 0.160547, -0.332203, -0.332203, 0.546391, 0.546391, 0.160547, 0.160547, -0.332203, -0.332203, 0.546391, 0.546391, 0.272735, 0.272735, 0.010827, 0.010827, -0.763164, -0.763164, 0.272735, 0.272735, 0.010827, 0.010827, -0.763164, -0.763164, -0.442696, -0.442696, 0.381948, 0.381948, -0.676994, -0.676994, -0.442696, -0.442696, 0.381948, 0.381948, -0.676994, -0.676994, 0.753553, 0.753553, -0.031788, -0.031788, 0.915329, 0.915329, 0.753553, 0.753553, -0.031788, -0.031788, 0.915329, 0.915329, -0.738844, -0.738844, 0.269075, 0.269075, 0.434091, 0.434091, -0.738844, -0.738844, 0.269075, 0.269075, 0.434091, 0.434091, 0.991585, 0.991585, -0.944288, -0.944288, 0.258834, 0.258834, 0.991585, 0.991585, -0.944288, -0.944288, 0.258834, 0.258834, 0.162138, 0.162138, 0.565201, 0.565201, -0.492094, -0.492094, 0.162138, 0.162138, 0.565201, 0.565201, -0.492094, -0.492094]\n"
|
||||
]
|
||||
}
|
||||
],
|
||||
"source": [
|
||||
"data_in_shape = (3, 3, 3)\n",
|
||||
"L = UpSampling2D(size=(2, 2), dim_ordering='th')\n",
|
||||
"\n",
|
||||
"layer_0 = Input(shape=data_in_shape)\n",
|
||||
"layer_1 = L(layer_0)\n",
|
||||
"model = Model(input=layer_0, output=layer_1)\n",
|
||||
"\n",
|
||||
"# set weights to random (use seed for reproducibility)\n",
|
||||
"np.random.seed(250)\n",
|
||||
"data_in = 2 * np.random.random(data_in_shape) - 1\n",
|
||||
"print('')\n",
|
||||
"print('in shape:', data_in_shape)\n",
|
||||
"print('in:', format_decimal(data_in.ravel().tolist()))\n",
|
||||
"result = model.predict(np.array([data_in]))\n",
|
||||
"print('out shape:', result[0].shape)\n",
|
||||
"print('out:', format_decimal(result[0].ravel().tolist()))"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "markdown",
|
||||
"metadata": {},
|
||||
"source": [
|
||||
"**[convolutional.UpSampling2D.2] size 3x2 upsampling on 4x2x2 input, dim_ordering='tf'**"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": 5,
|
||||
"metadata": {
|
||||
"collapsed": false
|
||||
},
|
||||
"outputs": [
|
||||
{
|
||||
"name": "stdout",
|
||||
"output_type": "stream",
|
||||
"text": [
|
||||
"\n",
|
||||
"in shape: (4, 2, 2)\n",
|
||||
"in: [0.275222, -0.793967, -0.468107, -0.841484, -0.295362, 0.78175, 0.068787, -0.261747, -0.625733, -0.042907, 0.861141, 0.85267, 0.956439, 0.717838, -0.99869, -0.963008]\n",
|
||||
"out shape: (12, 4, 2)\n",
|
||||
"out: [0.275222, -0.793967, 0.275222, -0.793967, -0.468107, -0.841484, -0.468107, -0.841484, 0.275222, -0.793967, 0.275222, -0.793967, -0.468107, -0.841484, -0.468107, -0.841484, 0.275222, -0.793967, 0.275222, -0.793967, -0.468107, -0.841484, -0.468107, -0.841484, -0.295362, 0.78175, -0.295362, 0.78175, 0.068787, -0.261747, 0.068787, -0.261747, -0.295362, 0.78175, -0.295362, 0.78175, 0.068787, -0.261747, 0.068787, -0.261747, -0.295362, 0.78175, -0.295362, 0.78175, 0.068787, -0.261747, 0.068787, -0.261747, -0.625733, -0.042907, -0.625733, -0.042907, 0.861141, 0.85267, 0.861141, 0.85267, -0.625733, -0.042907, -0.625733, -0.042907, 0.861141, 0.85267, 0.861141, 0.85267, -0.625733, -0.042907, -0.625733, -0.042907, 0.861141, 0.85267, 0.861141, 0.85267, 0.956439, 0.717838, 0.956439, 0.717838, -0.99869, -0.963008, -0.99869, -0.963008, 0.956439, 0.717838, 0.956439, 0.717838, -0.99869, -0.963008, -0.99869, -0.963008, 0.956439, 0.717838, 0.956439, 0.717838, -0.99869, -0.963008, -0.99869, -0.963008]\n"
|
||||
]
|
||||
}
|
||||
],
|
||||
"source": [
|
||||
"data_in_shape = (4, 2, 2)\n",
|
||||
"L = UpSampling2D(size=(3, 2), dim_ordering='tf')\n",
|
||||
"\n",
|
||||
"layer_0 = Input(shape=data_in_shape)\n",
|
||||
"layer_1 = L(layer_0)\n",
|
||||
"model = Model(input=layer_0, output=layer_1)\n",
|
||||
"\n",
|
||||
"# set weights to random (use seed for reproducibility)\n",
|
||||
"np.random.seed(251)\n",
|
||||
"data_in = 2 * np.random.random(data_in_shape) - 1\n",
|
||||
"print('')\n",
|
||||
"print('in shape:', data_in_shape)\n",
|
||||
"print('in:', format_decimal(data_in.ravel().tolist()))\n",
|
||||
"result = model.predict(np.array([data_in]))\n",
|
||||
"print('out shape:', result[0].shape)\n",
|
||||
"print('out:', format_decimal(result[0].ravel().tolist()))"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "markdown",
|
||||
"metadata": {},
|
||||
"source": [
|
||||
"**[convolutional.UpSampling2D.3] size 1x3 upsampling on 4x3x2 input, dim_ordering='th'**"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": 7,
|
||||
"metadata": {
|
||||
"collapsed": false
|
||||
},
|
||||
"outputs": [
|
||||
{
|
||||
"name": "stdout",
|
||||
"output_type": "stream",
|
||||
"text": [
|
||||
"\n",
|
||||
"in shape: (4, 3, 2)\n",
|
||||
"in: [-0.989173, -0.133618, -0.505338, 0.023259, 0.503982, -0.303769, -0.436321, 0.793911, 0.416102, 0.806405, -0.098342, -0.738022, -0.982676, 0.805073, 0.741244, -0.941634, -0.253526, -0.136544, -0.295772, 0.207565, -0.517246, -0.686963, -0.176235, -0.354111]\n",
|
||||
"out shape: (4, 3, 6)\n",
|
||||
"out: [-0.989173, -0.989173, -0.989173, -0.133618, -0.133618, -0.133618, -0.505338, -0.505338, -0.505338, 0.023259, 0.023259, 0.023259, 0.503982, 0.503982, 0.503982, -0.303769, -0.303769, -0.303769, -0.436321, -0.436321, -0.436321, 0.793911, 0.793911, 0.793911, 0.416102, 0.416102, 0.416102, 0.806405, 0.806405, 0.806405, -0.098342, -0.098342, -0.098342, -0.738022, -0.738022, -0.738022, -0.982676, -0.982676, -0.982676, 0.805073, 0.805073, 0.805073, 0.741244, 0.741244, 0.741244, -0.941634, -0.941634, -0.941634, -0.253526, -0.253526, -0.253526, -0.136544, -0.136544, -0.136544, -0.295772, -0.295772, -0.295772, 0.207565, 0.207565, 0.207565, -0.517246, -0.517246, -0.517246, -0.686963, -0.686963, -0.686963, -0.176235, -0.176235, -0.176235, -0.354111, -0.354111, -0.354111]\n"
|
||||
]
|
||||
}
|
||||
],
|
||||
"source": [
|
||||
"data_in_shape = (4, 3, 2)\n",
|
||||
"L = UpSampling2D(size=(1, 3), dim_ordering='th')\n",
|
||||
"\n",
|
||||
"layer_0 = Input(shape=data_in_shape)\n",
|
||||
"layer_1 = L(layer_0)\n",
|
||||
"model = Model(input=layer_0, output=layer_1)\n",
|
||||
"\n",
|
||||
"# set weights to random (use seed for reproducibility)\n",
|
||||
"np.random.seed(252)\n",
|
||||
"data_in = 2 * np.random.random(data_in_shape) - 1\n",
|
||||
"print('')\n",
|
||||
"print('in shape:', data_in_shape)\n",
|
||||
"print('in:', format_decimal(data_in.ravel().tolist()))\n",
|
||||
"result = model.predict(np.array([data_in]))\n",
|
||||
"print('out shape:', result[0].shape)\n",
|
||||
"print('out:', format_decimal(result[0].ravel().tolist()))"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": null,
|
||||
"metadata": {
|
||||
"collapsed": true
|
||||
},
|
||||
"outputs": [],
|
||||
"source": []
|
||||
}
|
||||
],
|
||||
"metadata": {
|
||||
"kernelspec": {
|
||||
"display_name": "Python 3",
|
||||
"language": "python",
|
||||
"name": "python3"
|
||||
},
|
||||
"language_info": {
|
||||
"codemirror_mode": {
|
||||
"name": "ipython",
|
||||
"version": 3
|
||||
},
|
||||
"file_extension": ".py",
|
||||
"mimetype": "text/x-python",
|
||||
"name": "python",
|
||||
"nbconvert_exporter": "python",
|
||||
"pygments_lexer": "ipython3",
|
||||
"version": "3.5.2"
|
||||
}
|
||||
},
|
||||
"nbformat": 4,
|
||||
"nbformat_minor": 0
|
||||
}
|
||||
@@ -0,0 +1,256 @@
|
||||
{
|
||||
"cells": [
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": 1,
|
||||
"metadata": {
|
||||
"collapsed": false
|
||||
},
|
||||
"outputs": [
|
||||
{
|
||||
"name": "stderr",
|
||||
"output_type": "stream",
|
||||
"text": [
|
||||
"Using TensorFlow backend.\n"
|
||||
]
|
||||
}
|
||||
],
|
||||
"source": [
|
||||
"import numpy as np\n",
|
||||
"from keras.models import Model\n",
|
||||
"from keras.layers import Input\n",
|
||||
"from keras.layers.convolutional import UpSampling3D\n",
|
||||
"from keras import backend as K"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": 2,
|
||||
"metadata": {
|
||||
"collapsed": true
|
||||
},
|
||||
"outputs": [],
|
||||
"source": [
|
||||
"def format_decimal(arr, places=6):\n",
|
||||
" return [round(x * 10**places) / 10**places for x in arr]"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "markdown",
|
||||
"metadata": {},
|
||||
"source": [
|
||||
"### UpSampling3D"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "markdown",
|
||||
"metadata": {},
|
||||
"source": [
|
||||
"**[convolutional.UpSampling3D.0] size 2x2x2 upsampling on 2x2x2x3 input, dim_ordering='tf'**"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": 6,
|
||||
"metadata": {
|
||||
"collapsed": false
|
||||
},
|
||||
"outputs": [
|
||||
{
|
||||
"name": "stdout",
|
||||
"output_type": "stream",
|
||||
"text": [
|
||||
"\n",
|
||||
"in shape: (2, 2, 2, 3)\n",
|
||||
"in: [-0.806777, -0.564841, -0.481331, 0.559626, 0.274958, -0.659222, -0.178541, 0.689453, -0.028873, 0.053859, -0.446394, -0.53406, 0.776897, -0.700858, -0.802179, -0.616515, 0.718677, 0.303042, -0.080606, -0.850593, -0.795971, 0.860487, -0.90685, 0.89858]\n",
|
||||
"out shape: (4, 4, 4, 3)\n",
|
||||
"out: [-0.806777, -0.564841, -0.481331, -0.806777, -0.564841, -0.481331, 0.559626, 0.274958, -0.659222, 0.559626, 0.274958, -0.659222, -0.806777, -0.564841, -0.481331, -0.806777, -0.564841, -0.481331, 0.559626, 0.274958, -0.659222, 0.559626, 0.274958, -0.659222, -0.178541, 0.689453, -0.028873, -0.178541, 0.689453, -0.028873, 0.053859, -0.446394, -0.53406, 0.053859, -0.446394, -0.53406, -0.178541, 0.689453, -0.028873, -0.178541, 0.689453, -0.028873, 0.053859, -0.446394, -0.53406, 0.053859, -0.446394, -0.53406, -0.806777, -0.564841, -0.481331, -0.806777, -0.564841, -0.481331, 0.559626, 0.274958, -0.659222, 0.559626, 0.274958, -0.659222, -0.806777, -0.564841, -0.481331, -0.806777, -0.564841, -0.481331, 0.559626, 0.274958, -0.659222, 0.559626, 0.274958, -0.659222, -0.178541, 0.689453, -0.028873, -0.178541, 0.689453, -0.028873, 0.053859, -0.446394, -0.53406, 0.053859, -0.446394, -0.53406, -0.178541, 0.689453, -0.028873, -0.178541, 0.689453, -0.028873, 0.053859, -0.446394, -0.53406, 0.053859, -0.446394, -0.53406, 0.776897, -0.700858, -0.802179, 0.776897, -0.700858, -0.802179, -0.616515, 0.718677, 0.303042, -0.616515, 0.718677, 0.303042, 0.776897, -0.700858, -0.802179, 0.776897, -0.700858, -0.802179, -0.616515, 0.718677, 0.303042, -0.616515, 0.718677, 0.303042, -0.080606, -0.850593, -0.795971, -0.080606, -0.850593, -0.795971, 0.860487, -0.90685, 0.89858, 0.860487, -0.90685, 0.89858, -0.080606, -0.850593, -0.795971, -0.080606, -0.850593, -0.795971, 0.860487, -0.90685, 0.89858, 0.860487, -0.90685, 0.89858, 0.776897, -0.700858, -0.802179, 0.776897, -0.700858, -0.802179, -0.616515, 0.718677, 0.303042, -0.616515, 0.718677, 0.303042, 0.776897, -0.700858, -0.802179, 0.776897, -0.700858, -0.802179, -0.616515, 0.718677, 0.303042, -0.616515, 0.718677, 0.303042, -0.080606, -0.850593, -0.795971, -0.080606, -0.850593, -0.795971, 0.860487, -0.90685, 0.89858, 0.860487, -0.90685, 0.89858, -0.080606, -0.850593, -0.795971, -0.080606, -0.850593, -0.795971, 0.860487, -0.90685, 0.89858, 0.860487, -0.90685, 0.89858]\n"
|
||||
]
|
||||
}
|
||||
],
|
||||
"source": [
|
||||
"data_in_shape = (2, 2, 2, 3)\n",
|
||||
"L = UpSampling3D(size=(2, 2, 2), dim_ordering='tf')\n",
|
||||
"\n",
|
||||
"layer_0 = Input(shape=data_in_shape)\n",
|
||||
"layer_1 = L(layer_0)\n",
|
||||
"model = Model(input=layer_0, output=layer_1)\n",
|
||||
"\n",
|
||||
"# set weights to random (use seed for reproducibility)\n",
|
||||
"np.random.seed(260)\n",
|
||||
"data_in = 2 * np.random.random(data_in_shape) - 1\n",
|
||||
"print('')\n",
|
||||
"print('in shape:', data_in_shape)\n",
|
||||
"print('in:', format_decimal(data_in.ravel().tolist()))\n",
|
||||
"result = model.predict(np.array([data_in]))\n",
|
||||
"print('out shape:', result[0].shape)\n",
|
||||
"print('out:', format_decimal(result[0].ravel().tolist()))"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "markdown",
|
||||
"metadata": {},
|
||||
"source": [
|
||||
"**[convolutional.UpSampling3D.0] size 2x2x2 upsampling on 2x2x2x3 input, dim_ordering='th'**"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": 7,
|
||||
"metadata": {
|
||||
"collapsed": false
|
||||
},
|
||||
"outputs": [
|
||||
{
|
||||
"name": "stdout",
|
||||
"output_type": "stream",
|
||||
"text": [
|
||||
"\n",
|
||||
"in shape: (2, 2, 2, 3)\n",
|
||||
"in: [-0.806777, -0.564841, -0.481331, 0.559626, 0.274958, -0.659222, -0.178541, 0.689453, -0.028873, 0.053859, -0.446394, -0.53406, 0.776897, -0.700858, -0.802179, -0.616515, 0.718677, 0.303042, -0.080606, -0.850593, -0.795971, 0.860487, -0.90685, 0.89858]\n",
|
||||
"out shape: (2, 4, 4, 6)\n",
|
||||
"out: [-0.806777, -0.806777, -0.564841, -0.564841, -0.481331, -0.481331, -0.806777, -0.806777, -0.564841, -0.564841, -0.481331, -0.481331, 0.559626, 0.559626, 0.274958, 0.274958, -0.659222, -0.659222, 0.559626, 0.559626, 0.274958, 0.274958, -0.659222, -0.659222, -0.806777, -0.806777, -0.564841, -0.564841, -0.481331, -0.481331, -0.806777, -0.806777, -0.564841, -0.564841, -0.481331, -0.481331, 0.559626, 0.559626, 0.274958, 0.274958, -0.659222, -0.659222, 0.559626, 0.559626, 0.274958, 0.274958, -0.659222, -0.659222, -0.178541, -0.178541, 0.689453, 0.689453, -0.028873, -0.028873, -0.178541, -0.178541, 0.689453, 0.689453, -0.028873, -0.028873, 0.053859, 0.053859, -0.446394, -0.446394, -0.53406, -0.53406, 0.053859, 0.053859, -0.446394, -0.446394, -0.53406, -0.53406, -0.178541, -0.178541, 0.689453, 0.689453, -0.028873, -0.028873, -0.178541, -0.178541, 0.689453, 0.689453, -0.028873, -0.028873, 0.053859, 0.053859, -0.446394, -0.446394, -0.53406, -0.53406, 0.053859, 0.053859, -0.446394, -0.446394, -0.53406, -0.53406, 0.776897, 0.776897, -0.700858, -0.700858, -0.802179, -0.802179, 0.776897, 0.776897, -0.700858, -0.700858, -0.802179, -0.802179, -0.616515, -0.616515, 0.718677, 0.718677, 0.303042, 0.303042, -0.616515, -0.616515, 0.718677, 0.718677, 0.303042, 0.303042, 0.776897, 0.776897, -0.700858, -0.700858, -0.802179, -0.802179, 0.776897, 0.776897, -0.700858, -0.700858, -0.802179, -0.802179, -0.616515, -0.616515, 0.718677, 0.718677, 0.303042, 0.303042, -0.616515, -0.616515, 0.718677, 0.718677, 0.303042, 0.303042, -0.080606, -0.080606, -0.850593, -0.850593, -0.795971, -0.795971, -0.080606, -0.080606, -0.850593, -0.850593, -0.795971, -0.795971, 0.860487, 0.860487, -0.90685, -0.90685, 0.89858, 0.89858, 0.860487, 0.860487, -0.90685, -0.90685, 0.89858, 0.89858, -0.080606, -0.080606, -0.850593, -0.850593, -0.795971, -0.795971, -0.080606, -0.080606, -0.850593, -0.850593, -0.795971, -0.795971, 0.860487, 0.860487, -0.90685, -0.90685, 0.89858, 0.89858, 0.860487, 0.860487, -0.90685, -0.90685, 0.89858, 0.89858]\n"
|
||||
]
|
||||
}
|
||||
],
|
||||
"source": [
|
||||
"data_in_shape = (2, 2, 2, 3)\n",
|
||||
"L = UpSampling3D(size=(2, 2, 2), dim_ordering='th')\n",
|
||||
"\n",
|
||||
"layer_0 = Input(shape=data_in_shape)\n",
|
||||
"layer_1 = L(layer_0)\n",
|
||||
"model = Model(input=layer_0, output=layer_1)\n",
|
||||
"\n",
|
||||
"# set weights to random (use seed for reproducibility)\n",
|
||||
"np.random.seed(260)\n",
|
||||
"data_in = 2 * np.random.random(data_in_shape) - 1\n",
|
||||
"print('')\n",
|
||||
"print('in shape:', data_in_shape)\n",
|
||||
"print('in:', format_decimal(data_in.ravel().tolist()))\n",
|
||||
"result = model.predict(np.array([data_in]))\n",
|
||||
"print('out shape:', result[0].shape)\n",
|
||||
"print('out:', format_decimal(result[0].ravel().tolist()))"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "markdown",
|
||||
"metadata": {},
|
||||
"source": [
|
||||
"**[convolutional.UpSampling2D.2] size 1x3x2 upsampling on 2x1x3x2 input, dim_ordering='tf'**"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": 8,
|
||||
"metadata": {
|
||||
"collapsed": false
|
||||
},
|
||||
"outputs": [
|
||||
{
|
||||
"name": "stdout",
|
||||
"output_type": "stream",
|
||||
"text": [
|
||||
"\n",
|
||||
"in shape: (2, 1, 3, 2)\n",
|
||||
"in: [-0.989173, -0.133618, -0.505338, 0.023259, 0.503982, -0.303769, -0.436321, 0.793911, 0.416102, 0.806405, -0.098342, -0.738022]\n",
|
||||
"out shape: (2, 3, 6, 2)\n",
|
||||
"out: [-0.989173, -0.133618, -0.989173, -0.133618, -0.505338, 0.023259, -0.505338, 0.023259, 0.503982, -0.303769, 0.503982, -0.303769, -0.989173, -0.133618, -0.989173, -0.133618, -0.505338, 0.023259, -0.505338, 0.023259, 0.503982, -0.303769, 0.503982, -0.303769, -0.989173, -0.133618, -0.989173, -0.133618, -0.505338, 0.023259, -0.505338, 0.023259, 0.503982, -0.303769, 0.503982, -0.303769, -0.436321, 0.793911, -0.436321, 0.793911, 0.416102, 0.806405, 0.416102, 0.806405, -0.098342, -0.738022, -0.098342, -0.738022, -0.436321, 0.793911, -0.436321, 0.793911, 0.416102, 0.806405, 0.416102, 0.806405, -0.098342, -0.738022, -0.098342, -0.738022, -0.436321, 0.793911, -0.436321, 0.793911, 0.416102, 0.806405, 0.416102, 0.806405, -0.098342, -0.738022, -0.098342, -0.738022]\n"
|
||||
]
|
||||
}
|
||||
],
|
||||
"source": [
|
||||
"data_in_shape = (2, 1, 3, 2)\n",
|
||||
"L = UpSampling3D(size=(1, 3, 2), dim_ordering='tf')\n",
|
||||
"\n",
|
||||
"layer_0 = Input(shape=data_in_shape)\n",
|
||||
"layer_1 = L(layer_0)\n",
|
||||
"model = Model(input=layer_0, output=layer_1)\n",
|
||||
"\n",
|
||||
"# set weights to random (use seed for reproducibility)\n",
|
||||
"np.random.seed(252)\n",
|
||||
"data_in = 2 * np.random.random(data_in_shape) - 1\n",
|
||||
"print('')\n",
|
||||
"print('in shape:', data_in_shape)\n",
|
||||
"print('in:', format_decimal(data_in.ravel().tolist()))\n",
|
||||
"result = model.predict(np.array([data_in]))\n",
|
||||
"print('out shape:', result[0].shape)\n",
|
||||
"print('out:', format_decimal(result[0].ravel().tolist()))"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "markdown",
|
||||
"metadata": {},
|
||||
"source": [
|
||||
"**[convolutional.UpSampling2D.2] size 2x1x2 upsampling on 2x1x3x3 input, dim_ordering='th'**"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": 10,
|
||||
"metadata": {
|
||||
"collapsed": false
|
||||
},
|
||||
"outputs": [
|
||||
{
|
||||
"name": "stdout",
|
||||
"output_type": "stream",
|
||||
"text": [
|
||||
"\n",
|
||||
"in shape: (2, 1, 3, 3)\n",
|
||||
"in: [-0.47588, 0.366985, 0.040173, 0.015578, -0.906159, 0.241982, -0.771299, -0.443554, -0.56404, -0.17751, 0.541277, -0.233327, 0.024369, 0.858275, 0.496191, 0.980574, -0.59522, 0.480899]\n",
|
||||
"out shape: (2, 2, 3, 6)\n",
|
||||
"out: [-0.47588, -0.47588, 0.366985, 0.366985, 0.040173, 0.040173, 0.015578, 0.015578, -0.906159, -0.906159, 0.241982, 0.241982, -0.771299, -0.771299, -0.443554, -0.443554, -0.56404, -0.56404, -0.47588, -0.47588, 0.366985, 0.366985, 0.040173, 0.040173, 0.015578, 0.015578, -0.906159, -0.906159, 0.241982, 0.241982, -0.771299, -0.771299, -0.443554, -0.443554, -0.56404, -0.56404, -0.17751, -0.17751, 0.541277, 0.541277, -0.233327, -0.233327, 0.024369, 0.024369, 0.858275, 0.858275, 0.496191, 0.496191, 0.980574, 0.980574, -0.59522, -0.59522, 0.480899, 0.480899, -0.17751, -0.17751, 0.541277, 0.541277, -0.233327, -0.233327, 0.024369, 0.024369, 0.858275, 0.858275, 0.496191, 0.496191, 0.980574, 0.980574, -0.59522, -0.59522, 0.480899, 0.480899]\n"
|
||||
]
|
||||
}
|
||||
],
|
||||
"source": [
|
||||
"data_in_shape = (2, 1, 3, 3)\n",
|
||||
"L = UpSampling3D(size=(2, 1, 2), dim_ordering='th')\n",
|
||||
"\n",
|
||||
"layer_0 = Input(shape=data_in_shape)\n",
|
||||
"layer_1 = L(layer_0)\n",
|
||||
"model = Model(input=layer_0, output=layer_1)\n",
|
||||
"\n",
|
||||
"# set weights to random (use seed for reproducibility)\n",
|
||||
"np.random.seed(253)\n",
|
||||
"data_in = 2 * np.random.random(data_in_shape) - 1\n",
|
||||
"print('')\n",
|
||||
"print('in shape:', data_in_shape)\n",
|
||||
"print('in:', format_decimal(data_in.ravel().tolist()))\n",
|
||||
"result = model.predict(np.array([data_in]))\n",
|
||||
"print('out shape:', result[0].shape)\n",
|
||||
"print('out:', format_decimal(result[0].ravel().tolist()))"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": null,
|
||||
"metadata": {
|
||||
"collapsed": true
|
||||
},
|
||||
"outputs": [],
|
||||
"source": []
|
||||
}
|
||||
],
|
||||
"metadata": {
|
||||
"kernelspec": {
|
||||
"display_name": "Python 3",
|
||||
"language": "python",
|
||||
"name": "python3"
|
||||
},
|
||||
"language_info": {
|
||||
"codemirror_mode": {
|
||||
"name": "ipython",
|
||||
"version": 3
|
||||
},
|
||||
"file_extension": ".py",
|
||||
"mimetype": "text/x-python",
|
||||
"name": "python",
|
||||
"nbconvert_exporter": "python",
|
||||
"pygments_lexer": "ipython3",
|
||||
"version": "3.5.2"
|
||||
}
|
||||
},
|
||||
"nbformat": 4,
|
||||
"nbformat_minor": 0
|
||||
}
|
||||
@@ -0,0 +1,36 @@
|
||||
import Layer from '../../engine/Layer'
|
||||
import Tensor from '../../Tensor'
|
||||
import ops from 'ndarray-ops'
|
||||
|
||||
/**
|
||||
* UpSampling1D layer class
|
||||
*/
|
||||
export default class UpSampling1D extends Layer {
|
||||
/**
|
||||
* Creates a UpSampling1D activation layer
|
||||
* @param {number} length - upsampling factor
|
||||
*/
|
||||
constructor (length = 2, attrs = {}) {
|
||||
super(attrs)
|
||||
this.length = length
|
||||
}
|
||||
|
||||
/**
|
||||
* Method for layer computational logic
|
||||
* @param {Tensor} x
|
||||
* @returns {Tensor} x
|
||||
*/
|
||||
call = x => {
|
||||
const inputShape = x.tensor.shape
|
||||
const outputShape = [inputShape[0] * this.length, inputShape[1]]
|
||||
let y = new Tensor([], outputShape)
|
||||
for (let i = 0; i < this.length; i++) {
|
||||
ops.assign(
|
||||
y.tensor.lo(i, 0).step(this.length, 1),
|
||||
x.tensor
|
||||
)
|
||||
}
|
||||
x.tensor = y.tensor
|
||||
return x
|
||||
}
|
||||
}
|
||||
@@ -0,0 +1,58 @@
|
||||
import Layer from '../../engine/Layer'
|
||||
import Tensor from '../../Tensor'
|
||||
import ops from 'ndarray-ops'
|
||||
|
||||
/**
|
||||
* UpSampling2D layer class
|
||||
*/
|
||||
export default class UpSampling2D extends Layer {
|
||||
/**
|
||||
* Creates a UpSampling2D activation layer
|
||||
* @param {number} size - upsampling factor
|
||||
*/
|
||||
constructor (size = [2, 2], attrs = {}) {
|
||||
super(attrs)
|
||||
const {
|
||||
dimOrdering = 'tf'
|
||||
} = attrs
|
||||
|
||||
this.size = size
|
||||
this.dimOrdering = dimOrdering
|
||||
}
|
||||
|
||||
/**
|
||||
* Method for layer computational logic
|
||||
* @param {Tensor} x
|
||||
* @returns {Tensor} x
|
||||
*/
|
||||
call = x => {
|
||||
// convert to tf ordering
|
||||
if (this.dimOrdering === 'th') {
|
||||
x.tensor = x.tensor.transpose(1, 2, 0)
|
||||
}
|
||||
|
||||
const inputShape = x.tensor.shape
|
||||
const outputShape = [
|
||||
inputShape[0] * this.size[0],
|
||||
inputShape[1] * this.size[1],
|
||||
inputShape[2]
|
||||
]
|
||||
let y = new Tensor([], outputShape)
|
||||
for (let i = 0; i < this.size[0]; i++) {
|
||||
for (let j = 0; j < this.size[1]; j++) {
|
||||
ops.assign(
|
||||
y.tensor.lo(i, j, 0).step(this.size[0], this.size[1], 1),
|
||||
x.tensor
|
||||
)
|
||||
}
|
||||
}
|
||||
x.tensor = y.tensor
|
||||
|
||||
// convert back to th ordering if necessary
|
||||
if (this.dimOrdering === 'th') {
|
||||
x.tensor = x.tensor.transpose(2, 0, 1)
|
||||
}
|
||||
|
||||
return x
|
||||
}
|
||||
}
|
||||
@@ -0,0 +1,61 @@
|
||||
import Layer from '../../engine/Layer'
|
||||
import Tensor from '../../Tensor'
|
||||
import ops from 'ndarray-ops'
|
||||
|
||||
/**
|
||||
* UpSampling3D layer class
|
||||
*/
|
||||
export default class UpSampling3D extends Layer {
|
||||
/**
|
||||
* Creates a UpSampling3D activation layer
|
||||
* @param {number} size - upsampling factor
|
||||
*/
|
||||
constructor (size = [2, 2, 2], attrs = {}) {
|
||||
super(attrs)
|
||||
const {
|
||||
dimOrdering = 'tf'
|
||||
} = attrs
|
||||
|
||||
this.size = size
|
||||
this.dimOrdering = dimOrdering
|
||||
}
|
||||
|
||||
/**
|
||||
* Method for layer computational logic
|
||||
* @param {Tensor} x
|
||||
* @returns {Tensor} x
|
||||
*/
|
||||
call = x => {
|
||||
// convert to tf ordering
|
||||
if (this.dimOrdering === 'th') {
|
||||
x.tensor = x.tensor.transpose(1, 2, 3, 0)
|
||||
}
|
||||
|
||||
const inputShape = x.tensor.shape
|
||||
const outputShape = [
|
||||
inputShape[0] * this.size[0],
|
||||
inputShape[1] * this.size[1],
|
||||
inputShape[2] * this.size[2],
|
||||
inputShape[3]
|
||||
]
|
||||
let y = new Tensor([], outputShape)
|
||||
for (let i = 0; i < this.size[0]; i++) {
|
||||
for (let j = 0; j < this.size[1]; j++) {
|
||||
for (let k = 0; k < this.size[2]; k++) {
|
||||
ops.assign(
|
||||
y.tensor.lo(i, j, k, 0).step(this.size[0], this.size[1], this.size[2], 1),
|
||||
x.tensor
|
||||
)
|
||||
}
|
||||
}
|
||||
}
|
||||
x.tensor = y.tensor
|
||||
|
||||
// convert back to th ordering if necessary
|
||||
if (this.dimOrdering === 'th') {
|
||||
x.tensor = x.tensor.transpose(3, 0, 1, 2)
|
||||
}
|
||||
|
||||
return x
|
||||
}
|
||||
}
|
||||
@@ -1,7 +1,13 @@
|
||||
import Convolution1D from './Convolution1D'
|
||||
import Convolution2D from './Convolution2D'
|
||||
import UpSampling1D from './UpSampling1D'
|
||||
import UpSampling2D from './UpSampling2D'
|
||||
import UpSampling3D from './UpSampling3D'
|
||||
|
||||
export {
|
||||
Convolution1D,
|
||||
Convolution2D
|
||||
Convolution2D,
|
||||
UpSampling1D,
|
||||
UpSampling2D,
|
||||
UpSampling3D
|
||||
}
|
||||
|
||||
@@ -0,0 +1,48 @@
|
||||
/* eslint-env browser, mocha */
|
||||
|
||||
describe('convolutional layer: UpSampling1D', function () {
|
||||
const assert = chai.assert
|
||||
const styles = testGlobals.styles
|
||||
const logTime = testGlobals.logTime
|
||||
const stringifyCondensed = testGlobals.stringifyCondensed
|
||||
const approxEquals = KerasJS.testUtils.approxEquals
|
||||
const layers = KerasJS.layers
|
||||
|
||||
before(function () {
|
||||
console.log('\n%cconvolutional layer: UpSampling1D', styles.h1)
|
||||
})
|
||||
|
||||
it(`[convolutional.UpSampling1D.0] length 2 upsampling on 3x5 input`, function () {
|
||||
const key = `convolutional.UpSampling1D.0`
|
||||
console.log(`\n%c[${key}] length 2 upsampling on 3x5 input`, styles.h3)
|
||||
let testLayer = new layers.UpSampling1D(2)
|
||||
let t = new KerasJS.Tensor(TEST_DATA[key].input.data, TEST_DATA[key].input.shape)
|
||||
console.log('%cin', styles.h4, stringifyCondensed(t.tensor))
|
||||
const startTime = performance.now()
|
||||
t = testLayer.call(t)
|
||||
const endTime = performance.now()
|
||||
console.log('%cout', styles.h4, stringifyCondensed(t.tensor))
|
||||
logTime(startTime, endTime)
|
||||
const dataExpected = new Float32Array(TEST_DATA[key].expected.data)
|
||||
const shapeExpected = TEST_DATA[key].expected.shape
|
||||
assert.deepEqual(t.tensor.shape, shapeExpected)
|
||||
assert.isTrue(approxEquals(t.tensor, dataExpected))
|
||||
})
|
||||
|
||||
it(`[convolutional.UpSampling1D.1] length 3 upsampling on 4x4 input`, function () {
|
||||
const key = `convolutional.UpSampling1D.1`
|
||||
console.log(`\n%c[${key}] length 3 upsampling on 4x4 input`, styles.h3)
|
||||
let testLayer = new layers.UpSampling1D(3)
|
||||
let t = new KerasJS.Tensor(TEST_DATA[key].input.data, TEST_DATA[key].input.shape)
|
||||
console.log('%cin', styles.h4, stringifyCondensed(t.tensor))
|
||||
const startTime = performance.now()
|
||||
t = testLayer.call(t)
|
||||
const endTime = performance.now()
|
||||
console.log('%cout', styles.h4, stringifyCondensed(t.tensor))
|
||||
logTime(startTime, endTime)
|
||||
const dataExpected = new Float32Array(TEST_DATA[key].expected.data)
|
||||
const shapeExpected = TEST_DATA[key].expected.shape
|
||||
assert.deepEqual(t.tensor.shape, shapeExpected)
|
||||
assert.isTrue(approxEquals(t.tensor, dataExpected))
|
||||
})
|
||||
})
|
||||
@@ -0,0 +1,82 @@
|
||||
/* eslint-env browser, mocha */
|
||||
|
||||
describe('convolutional layer: UpSampling2D', function () {
|
||||
const assert = chai.assert
|
||||
const styles = testGlobals.styles
|
||||
const logTime = testGlobals.logTime
|
||||
const stringifyCondensed = testGlobals.stringifyCondensed
|
||||
const approxEquals = KerasJS.testUtils.approxEquals
|
||||
const layers = KerasJS.layers
|
||||
|
||||
before(function () {
|
||||
console.log('\n%cconvolutional layer: UpSampling2D', styles.h1)
|
||||
})
|
||||
|
||||
it(`[convolutional.UpSampling2D.0] size 2x2 upsampling on 3x3x3 input, dimOrdering=tf`, function () {
|
||||
const key = `convolutional.UpSampling2D.0`
|
||||
console.log(`\n%c[${key}] size 2x2 upsampling on 3x3x3 input, dimOrdering=tf`, styles.h3)
|
||||
let testLayer = new layers.UpSampling2D([2, 2], { dimOrdering: 'tf' })
|
||||
let t = new KerasJS.Tensor(TEST_DATA[key].input.data, TEST_DATA[key].input.shape)
|
||||
console.log('%cin', styles.h4, stringifyCondensed(t.tensor))
|
||||
const startTime = performance.now()
|
||||
t = testLayer.call(t)
|
||||
const endTime = performance.now()
|
||||
console.log('%cout', styles.h4, stringifyCondensed(t.tensor))
|
||||
logTime(startTime, endTime)
|
||||
const dataExpected = new Float32Array(TEST_DATA[key].expected.data)
|
||||
const shapeExpected = TEST_DATA[key].expected.shape
|
||||
assert.deepEqual(t.tensor.shape, shapeExpected)
|
||||
assert.isTrue(approxEquals(t.tensor, dataExpected))
|
||||
})
|
||||
|
||||
it(`[convolutional.UpSampling2D.1] size 2x2 upsampling on 3x3x3 input, dimOrdering=th`, function () {
|
||||
const key = `convolutional.UpSampling2D.1`
|
||||
console.log(`\n%c[${key}] size 2x2 upsampling on 3x3x3 input, dimOrdering=th`, styles.h3)
|
||||
let testLayer = new layers.UpSampling2D([2, 2], { dimOrdering: 'th' })
|
||||
let t = new KerasJS.Tensor(TEST_DATA[key].input.data, TEST_DATA[key].input.shape)
|
||||
console.log('%cin', styles.h4, stringifyCondensed(t.tensor))
|
||||
const startTime = performance.now()
|
||||
t = testLayer.call(t)
|
||||
const endTime = performance.now()
|
||||
console.log('%cout', styles.h4, stringifyCondensed(t.tensor))
|
||||
logTime(startTime, endTime)
|
||||
const dataExpected = new Float32Array(TEST_DATA[key].expected.data)
|
||||
const shapeExpected = TEST_DATA[key].expected.shape
|
||||
assert.deepEqual(t.tensor.shape, shapeExpected)
|
||||
assert.isTrue(approxEquals(t.tensor, dataExpected))
|
||||
})
|
||||
|
||||
it(`[convolutional.UpSampling2D.2] size 3x2 upsampling on 4x2x2 input, dimOrdering=tf`, function () {
|
||||
const key = `convolutional.UpSampling2D.2`
|
||||
console.log(`\n%c[${key}] size 3x2 upsampling on 4x2x2 input, dimOrdering=tf`, styles.h3)
|
||||
let testLayer = new layers.UpSampling2D([3, 2], { dimOrdering: 'tf' })
|
||||
let t = new KerasJS.Tensor(TEST_DATA[key].input.data, TEST_DATA[key].input.shape)
|
||||
console.log('%cin', styles.h4, stringifyCondensed(t.tensor))
|
||||
const startTime = performance.now()
|
||||
t = testLayer.call(t)
|
||||
const endTime = performance.now()
|
||||
console.log('%cout', styles.h4, stringifyCondensed(t.tensor))
|
||||
logTime(startTime, endTime)
|
||||
const dataExpected = new Float32Array(TEST_DATA[key].expected.data)
|
||||
const shapeExpected = TEST_DATA[key].expected.shape
|
||||
assert.deepEqual(t.tensor.shape, shapeExpected)
|
||||
assert.isTrue(approxEquals(t.tensor, dataExpected))
|
||||
})
|
||||
|
||||
it(`[convolutional.UpSampling2D.3] size 1x3 upsampling on 4x3x2 input, dimOrdering=th`, function () {
|
||||
const key = `convolutional.UpSampling2D.3`
|
||||
console.log(`\n%c[${key}] size 1x3 upsampling on 4x3x2 input, dimOrdering=th`, styles.h3)
|
||||
let testLayer = new layers.UpSampling2D([1, 3], { dimOrdering: 'th' })
|
||||
let t = new KerasJS.Tensor(TEST_DATA[key].input.data, TEST_DATA[key].input.shape)
|
||||
console.log('%cin', styles.h4, stringifyCondensed(t.tensor))
|
||||
const startTime = performance.now()
|
||||
t = testLayer.call(t)
|
||||
const endTime = performance.now()
|
||||
console.log('%cout', styles.h4, stringifyCondensed(t.tensor))
|
||||
logTime(startTime, endTime)
|
||||
const dataExpected = new Float32Array(TEST_DATA[key].expected.data)
|
||||
const shapeExpected = TEST_DATA[key].expected.shape
|
||||
assert.deepEqual(t.tensor.shape, shapeExpected)
|
||||
assert.isTrue(approxEquals(t.tensor, dataExpected))
|
||||
})
|
||||
})
|
||||
@@ -0,0 +1,82 @@
|
||||
/* eslint-env browser, mocha */
|
||||
|
||||
describe('convolutional layer: UpSampling3D', function () {
|
||||
const assert = chai.assert
|
||||
const styles = testGlobals.styles
|
||||
const logTime = testGlobals.logTime
|
||||
const stringifyCondensed = testGlobals.stringifyCondensed
|
||||
const approxEquals = KerasJS.testUtils.approxEquals
|
||||
const layers = KerasJS.layers
|
||||
|
||||
before(function () {
|
||||
console.log('\n%cconvolutional layer: UpSampling3D', styles.h1)
|
||||
})
|
||||
|
||||
it(`[convolutional.UpSampling3D.0] size 2x2x2 upsampling on 2x2x2x3 input, dimOrdering=tf`, function () {
|
||||
const key = `convolutional.UpSampling3D.0`
|
||||
console.log(`\n%c[${key}] size 2x2x2 upsampling on 2x2x2x3 input, dimOrdering=tf`, styles.h3)
|
||||
let testLayer = new layers.UpSampling3D([2, 2, 2], { dimOrdering: 'tf' })
|
||||
let t = new KerasJS.Tensor(TEST_DATA[key].input.data, TEST_DATA[key].input.shape)
|
||||
console.log('%cin', styles.h4, stringifyCondensed(t.tensor))
|
||||
const startTime = performance.now()
|
||||
t = testLayer.call(t)
|
||||
const endTime = performance.now()
|
||||
console.log('%cout', styles.h4, stringifyCondensed(t.tensor))
|
||||
logTime(startTime, endTime)
|
||||
const dataExpected = new Float32Array(TEST_DATA[key].expected.data)
|
||||
const shapeExpected = TEST_DATA[key].expected.shape
|
||||
assert.deepEqual(t.tensor.shape, shapeExpected)
|
||||
assert.isTrue(approxEquals(t.tensor, dataExpected))
|
||||
})
|
||||
|
||||
it(`[convolutional.UpSampling3D.1] size 2x2x2 upsampling on 2x2x2x3 input, dimOrdering=th`, function () {
|
||||
const key = `convolutional.UpSampling3D.1`
|
||||
console.log(`\n%c[${key}] size 2x2x2 upsampling on 2x2x2x3 input, dimOrdering=th`, styles.h3)
|
||||
let testLayer = new layers.UpSampling3D([2, 2, 2], { dimOrdering: 'th' })
|
||||
let t = new KerasJS.Tensor(TEST_DATA[key].input.data, TEST_DATA[key].input.shape)
|
||||
console.log('%cin', styles.h4, stringifyCondensed(t.tensor))
|
||||
const startTime = performance.now()
|
||||
t = testLayer.call(t)
|
||||
const endTime = performance.now()
|
||||
console.log('%cout', styles.h4, stringifyCondensed(t.tensor))
|
||||
logTime(startTime, endTime)
|
||||
const dataExpected = new Float32Array(TEST_DATA[key].expected.data)
|
||||
const shapeExpected = TEST_DATA[key].expected.shape
|
||||
assert.deepEqual(t.tensor.shape, shapeExpected)
|
||||
assert.isTrue(approxEquals(t.tensor, dataExpected))
|
||||
})
|
||||
|
||||
it(`[convolutional.UpSampling3D.2] size 1x3x2 upsampling on 2x1x3x2 input, dimOrdering=tf`, function () {
|
||||
const key = `convolutional.UpSampling3D.2`
|
||||
console.log(`\n%c[${key}] size 1x3x2 upsampling on 2x1x3x2 input, dimOrdering=tf`, styles.h3)
|
||||
let testLayer = new layers.UpSampling3D([1, 3, 2], { dimOrdering: 'tf' })
|
||||
let t = new KerasJS.Tensor(TEST_DATA[key].input.data, TEST_DATA[key].input.shape)
|
||||
console.log('%cin', styles.h4, stringifyCondensed(t.tensor))
|
||||
const startTime = performance.now()
|
||||
t = testLayer.call(t)
|
||||
const endTime = performance.now()
|
||||
console.log('%cout', styles.h4, stringifyCondensed(t.tensor))
|
||||
logTime(startTime, endTime)
|
||||
const dataExpected = new Float32Array(TEST_DATA[key].expected.data)
|
||||
const shapeExpected = TEST_DATA[key].expected.shape
|
||||
assert.deepEqual(t.tensor.shape, shapeExpected)
|
||||
assert.isTrue(approxEquals(t.tensor, dataExpected))
|
||||
})
|
||||
|
||||
it(`[convolutional.UpSampling3D.3] 2x1x2 upsampling on 2x1x3x3 input, dimOrdering=th`, function () {
|
||||
const key = `convolutional.UpSampling3D.3`
|
||||
console.log(`\n%c[${key}] 2x1x2 upsampling on 2x1x3x3 input, dimOrdering=th`, styles.h3)
|
||||
let testLayer = new layers.UpSampling3D([2, 1, 2], { dimOrdering: 'th' })
|
||||
let t = new KerasJS.Tensor(TEST_DATA[key].input.data, TEST_DATA[key].input.shape)
|
||||
console.log('%cin', styles.h4, stringifyCondensed(t.tensor))
|
||||
const startTime = performance.now()
|
||||
t = testLayer.call(t)
|
||||
const endTime = performance.now()
|
||||
console.log('%cout', styles.h4, stringifyCondensed(t.tensor))
|
||||
logTime(startTime, endTime)
|
||||
const dataExpected = new Float32Array(TEST_DATA[key].expected.data)
|
||||
const shapeExpected = TEST_DATA[key].expected.shape
|
||||
assert.deepEqual(t.tensor.shape, shapeExpected)
|
||||
assert.isTrue(approxEquals(t.tensor, dataExpected))
|
||||
})
|
||||
})
|
||||
@@ -0,0 +1,30 @@
|
||||
// TEST DATA
|
||||
// Keyed by mocha test ID
|
||||
// Python code for generating test data can be found in the matching jupyter notebook in folder `notebooks/`.
|
||||
|
||||
(function () {
|
||||
var DATA = {
|
||||
'convolutional.UpSampling1D.0': {
|
||||
input: {
|
||||
data: [0.262, 0.764609, -0.482897, -0.371755, 0.871769, 0.490033, -0.986894, -0.960468, 0.373039, 0.911356, 0.00298, 0.270652, 0.749006, 0.692235, 0.471778],
|
||||
shape: [3, 5]
|
||||
},
|
||||
expected: {
|
||||
data: [0.262, 0.764609, -0.482897, -0.371755, 0.871769, 0.262, 0.764609, -0.482897, -0.371755, 0.871769, 0.490033, -0.986894, -0.960468, 0.373039, 0.911356, 0.490033, -0.986894, -0.960468, 0.373039, 0.911356, 0.00298, 0.270652, 0.749006, 0.692235, 0.471778, 0.00298, 0.270652, 0.749006, 0.692235, 0.471778],
|
||||
shape: [6, 5]
|
||||
}
|
||||
},
|
||||
'convolutional.UpSampling1D.1': {
|
||||
input: {
|
||||
data: [0.562988, 0.168418, -0.14658, -0.369311, 0.653777, 0.806859, -0.922124, 0.830445, -0.878989, -0.638546, -0.855401, -0.082476, 0.416718, -0.03353, -0.949107, -0.866195],
|
||||
shape: [4, 4]
|
||||
},
|
||||
expected: {
|
||||
data: [0.562988, 0.168418, -0.14658, -0.369311, 0.562988, 0.168418, -0.14658, -0.369311, 0.562988, 0.168418, -0.14658, -0.369311, 0.653777, 0.806859, -0.922124, 0.830445, 0.653777, 0.806859, -0.922124, 0.830445, 0.653777, 0.806859, -0.922124, 0.830445, -0.878989, -0.638546, -0.855401, -0.082476, -0.878989, -0.638546, -0.855401, -0.082476, -0.878989, -0.638546, -0.855401, -0.082476, 0.416718, -0.03353, -0.949107, -0.866195, 0.416718, -0.03353, -0.949107, -0.866195, 0.416718, -0.03353, -0.949107, -0.866195],
|
||||
shape: [12, 4]
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
window.TEST_DATA = Object.assign({}, window.TEST_DATA, DATA)
|
||||
})()
|
||||
@@ -0,0 +1,50 @@
|
||||
// TEST DATA
|
||||
// Keyed by mocha test ID
|
||||
// Python code for generating test data can be found in the matching jupyter notebook in folder `notebooks/`.
|
||||
|
||||
(function () {
|
||||
var DATA = {
|
||||
'convolutional.UpSampling2D.0': {
|
||||
input: {
|
||||
data: [-0.570441, -0.454673, -0.285321, 0.237249, 0.282682, 0.428035, 0.160547, -0.332203, 0.546391, 0.272735, 0.010827, -0.763164, -0.442696, 0.381948, -0.676994, 0.753553, -0.031788, 0.915329, -0.738844, 0.269075, 0.434091, 0.991585, -0.944288, 0.258834, 0.162138, 0.565201, -0.492094],
|
||||
shape: [3, 3, 3]
|
||||
},
|
||||
expected: {
|
||||
data: [-0.570441, -0.454673, -0.285321, -0.570441, -0.454673, -0.285321, 0.237249, 0.282682, 0.428035, 0.237249, 0.282682, 0.428035, 0.160547, -0.332203, 0.546391, 0.160547, -0.332203, 0.546391, -0.570441, -0.454673, -0.285321, -0.570441, -0.454673, -0.285321, 0.237249, 0.282682, 0.428035, 0.237249, 0.282682, 0.428035, 0.160547, -0.332203, 0.546391, 0.160547, -0.332203, 0.546391, 0.272735, 0.010827, -0.763164, 0.272735, 0.010827, -0.763164, -0.442696, 0.381948, -0.676994, -0.442696, 0.381948, -0.676994, 0.753553, -0.031788, 0.915329, 0.753553, -0.031788, 0.915329, 0.272735, 0.010827, -0.763164, 0.272735, 0.010827, -0.763164, -0.442696, 0.381948, -0.676994, -0.442696, 0.381948, -0.676994, 0.753553, -0.031788, 0.915329, 0.753553, -0.031788, 0.915329, -0.738844, 0.269075, 0.434091, -0.738844, 0.269075, 0.434091, 0.991585, -0.944288, 0.258834, 0.991585, -0.944288, 0.258834, 0.162138, 0.565201, -0.492094, 0.162138, 0.565201, -0.492094, -0.738844, 0.269075, 0.434091, -0.738844, 0.269075, 0.434091, 0.991585, -0.944288, 0.258834, 0.991585, -0.944288, 0.258834, 0.162138, 0.565201, -0.492094, 0.162138, 0.565201, -0.492094],
|
||||
shape: [6, 6, 3]
|
||||
}
|
||||
},
|
||||
'convolutional.UpSampling2D.1': {
|
||||
input: {
|
||||
data: [-0.570441, -0.454673, -0.285321, 0.237249, 0.282682, 0.428035, 0.160547, -0.332203, 0.546391, 0.272735, 0.010827, -0.763164, -0.442696, 0.381948, -0.676994, 0.753553, -0.031788, 0.915329, -0.738844, 0.269075, 0.434091, 0.991585, -0.944288, 0.258834, 0.162138, 0.565201, -0.492094],
|
||||
shape: [3, 3, 3]
|
||||
},
|
||||
expected: {
|
||||
data: [-0.570441, -0.570441, -0.454673, -0.454673, -0.285321, -0.285321, -0.570441, -0.570441, -0.454673, -0.454673, -0.285321, -0.285321, 0.237249, 0.237249, 0.282682, 0.282682, 0.428035, 0.428035, 0.237249, 0.237249, 0.282682, 0.282682, 0.428035, 0.428035, 0.160547, 0.160547, -0.332203, -0.332203, 0.546391, 0.546391, 0.160547, 0.160547, -0.332203, -0.332203, 0.546391, 0.546391, 0.272735, 0.272735, 0.010827, 0.010827, -0.763164, -0.763164, 0.272735, 0.272735, 0.010827, 0.010827, -0.763164, -0.763164, -0.442696, -0.442696, 0.381948, 0.381948, -0.676994, -0.676994, -0.442696, -0.442696, 0.381948, 0.381948, -0.676994, -0.676994, 0.753553, 0.753553, -0.031788, -0.031788, 0.915329, 0.915329, 0.753553, 0.753553, -0.031788, -0.031788, 0.915329, 0.915329, -0.738844, -0.738844, 0.269075, 0.269075, 0.434091, 0.434091, -0.738844, -0.738844, 0.269075, 0.269075, 0.434091, 0.434091, 0.991585, 0.991585, -0.944288, -0.944288, 0.258834, 0.258834, 0.991585, 0.991585, -0.944288, -0.944288, 0.258834, 0.258834, 0.162138, 0.162138, 0.565201, 0.565201, -0.492094, -0.492094, 0.162138, 0.162138, 0.565201, 0.565201, -0.492094, -0.492094],
|
||||
shape: [3, 6, 6]
|
||||
}
|
||||
},
|
||||
'convolutional.UpSampling2D.2': {
|
||||
input: {
|
||||
data: [0.275222, -0.793967, -0.468107, -0.841484, -0.295362, 0.78175, 0.068787, -0.261747, -0.625733, -0.042907, 0.861141, 0.85267, 0.956439, 0.717838, -0.99869, -0.963008],
|
||||
shape: [4, 2, 2]
|
||||
},
|
||||
expected: {
|
||||
data: [0.275222, -0.793967, 0.275222, -0.793967, -0.468107, -0.841484, -0.468107, -0.841484, 0.275222, -0.793967, 0.275222, -0.793967, -0.468107, -0.841484, -0.468107, -0.841484, 0.275222, -0.793967, 0.275222, -0.793967, -0.468107, -0.841484, -0.468107, -0.841484, -0.295362, 0.78175, -0.295362, 0.78175, 0.068787, -0.261747, 0.068787, -0.261747, -0.295362, 0.78175, -0.295362, 0.78175, 0.068787, -0.261747, 0.068787, -0.261747, -0.295362, 0.78175, -0.295362, 0.78175, 0.068787, -0.261747, 0.068787, -0.261747, -0.625733, -0.042907, -0.625733, -0.042907, 0.861141, 0.85267, 0.861141, 0.85267, -0.625733, -0.042907, -0.625733, -0.042907, 0.861141, 0.85267, 0.861141, 0.85267, -0.625733, -0.042907, -0.625733, -0.042907, 0.861141, 0.85267, 0.861141, 0.85267, 0.956439, 0.717838, 0.956439, 0.717838, -0.99869, -0.963008, -0.99869, -0.963008, 0.956439, 0.717838, 0.956439, 0.717838, -0.99869, -0.963008, -0.99869, -0.963008, 0.956439, 0.717838, 0.956439, 0.717838, -0.99869, -0.963008, -0.99869, -0.963008],
|
||||
shape: [12, 4, 2]
|
||||
}
|
||||
},
|
||||
'convolutional.UpSampling2D.3': {
|
||||
input: {
|
||||
data: [-0.989173, -0.133618, -0.505338, 0.023259, 0.503982, -0.303769, -0.436321, 0.793911, 0.416102, 0.806405, -0.098342, -0.738022, -0.982676, 0.805073, 0.741244, -0.941634, -0.253526, -0.136544, -0.295772, 0.207565, -0.517246, -0.686963, -0.176235, -0.354111],
|
||||
shape: [4, 3, 2]
|
||||
},
|
||||
expected: {
|
||||
data: [-0.989173, -0.989173, -0.989173, -0.133618, -0.133618, -0.133618, -0.505338, -0.505338, -0.505338, 0.023259, 0.023259, 0.023259, 0.503982, 0.503982, 0.503982, -0.303769, -0.303769, -0.303769, -0.436321, -0.436321, -0.436321, 0.793911, 0.793911, 0.793911, 0.416102, 0.416102, 0.416102, 0.806405, 0.806405, 0.806405, -0.098342, -0.098342, -0.098342, -0.738022, -0.738022, -0.738022, -0.982676, -0.982676, -0.982676, 0.805073, 0.805073, 0.805073, 0.741244, 0.741244, 0.741244, -0.941634, -0.941634, -0.941634, -0.253526, -0.253526, -0.253526, -0.136544, -0.136544, -0.136544, -0.295772, -0.295772, -0.295772, 0.207565, 0.207565, 0.207565, -0.517246, -0.517246, -0.517246, -0.686963, -0.686963, -0.686963, -0.176235, -0.176235, -0.176235, -0.354111, -0.354111, -0.354111],
|
||||
shape: [4, 3, 6]
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
window.TEST_DATA = Object.assign({}, window.TEST_DATA, DATA)
|
||||
})()
|
||||
@@ -0,0 +1,50 @@
|
||||
// TEST DATA
|
||||
// Keyed by mocha test ID
|
||||
// Python code for generating test data can be found in the matching jupyter notebook in folder `notebooks/`.
|
||||
|
||||
(function () {
|
||||
var DATA = {
|
||||
'convolutional.UpSampling3D.0': {
|
||||
input: {
|
||||
data: [-0.806777, -0.564841, -0.481331, 0.559626, 0.274958, -0.659222, -0.178541, 0.689453, -0.028873, 0.053859, -0.446394, -0.53406, 0.776897, -0.700858, -0.802179, -0.616515, 0.718677, 0.303042, -0.080606, -0.850593, -0.795971, 0.860487, -0.90685, 0.89858],
|
||||
shape: [2, 2, 2, 3]
|
||||
},
|
||||
expected: {
|
||||
data: [-0.806777, -0.564841, -0.481331, -0.806777, -0.564841, -0.481331, 0.559626, 0.274958, -0.659222, 0.559626, 0.274958, -0.659222, -0.806777, -0.564841, -0.481331, -0.806777, -0.564841, -0.481331, 0.559626, 0.274958, -0.659222, 0.559626, 0.274958, -0.659222, -0.178541, 0.689453, -0.028873, -0.178541, 0.689453, -0.028873, 0.053859, -0.446394, -0.53406, 0.053859, -0.446394, -0.53406, -0.178541, 0.689453, -0.028873, -0.178541, 0.689453, -0.028873, 0.053859, -0.446394, -0.53406, 0.053859, -0.446394, -0.53406, -0.806777, -0.564841, -0.481331, -0.806777, -0.564841, -0.481331, 0.559626, 0.274958, -0.659222, 0.559626, 0.274958, -0.659222, -0.806777, -0.564841, -0.481331, -0.806777, -0.564841, -0.481331, 0.559626, 0.274958, -0.659222, 0.559626, 0.274958, -0.659222, -0.178541, 0.689453, -0.028873, -0.178541, 0.689453, -0.028873, 0.053859, -0.446394, -0.53406, 0.053859, -0.446394, -0.53406, -0.178541, 0.689453, -0.028873, -0.178541, 0.689453, -0.028873, 0.053859, -0.446394, -0.53406, 0.053859, -0.446394, -0.53406, 0.776897, -0.700858, -0.802179, 0.776897, -0.700858, -0.802179, -0.616515, 0.718677, 0.303042, -0.616515, 0.718677, 0.303042, 0.776897, -0.700858, -0.802179, 0.776897, -0.700858, -0.802179, -0.616515, 0.718677, 0.303042, -0.616515, 0.718677, 0.303042, -0.080606, -0.850593, -0.795971, -0.080606, -0.850593, -0.795971, 0.860487, -0.90685, 0.89858, 0.860487, -0.90685, 0.89858, -0.080606, -0.850593, -0.795971, -0.080606, -0.850593, -0.795971, 0.860487, -0.90685, 0.89858, 0.860487, -0.90685, 0.89858, 0.776897, -0.700858, -0.802179, 0.776897, -0.700858, -0.802179, -0.616515, 0.718677, 0.303042, -0.616515, 0.718677, 0.303042, 0.776897, -0.700858, -0.802179, 0.776897, -0.700858, -0.802179, -0.616515, 0.718677, 0.303042, -0.616515, 0.718677, 0.303042, -0.080606, -0.850593, -0.795971, -0.080606, -0.850593, -0.795971, 0.860487, -0.90685, 0.89858, 0.860487, -0.90685, 0.89858, -0.080606, -0.850593, -0.795971, -0.080606, -0.850593, -0.795971, 0.860487, -0.90685, 0.89858, 0.860487, -0.90685, 0.89858],
|
||||
shape: [4, 4, 4, 3]
|
||||
}
|
||||
},
|
||||
'convolutional.UpSampling3D.1': {
|
||||
input: {
|
||||
data: [-0.806777, -0.564841, -0.481331, 0.559626, 0.274958, -0.659222, -0.178541, 0.689453, -0.028873, 0.053859, -0.446394, -0.53406, 0.776897, -0.700858, -0.802179, -0.616515, 0.718677, 0.303042, -0.080606, -0.850593, -0.795971, 0.860487, -0.90685, 0.89858],
|
||||
shape: [2, 2, 2, 3]
|
||||
},
|
||||
expected: {
|
||||
data: [-0.806777, -0.806777, -0.564841, -0.564841, -0.481331, -0.481331, -0.806777, -0.806777, -0.564841, -0.564841, -0.481331, -0.481331, 0.559626, 0.559626, 0.274958, 0.274958, -0.659222, -0.659222, 0.559626, 0.559626, 0.274958, 0.274958, -0.659222, -0.659222, -0.806777, -0.806777, -0.564841, -0.564841, -0.481331, -0.481331, -0.806777, -0.806777, -0.564841, -0.564841, -0.481331, -0.481331, 0.559626, 0.559626, 0.274958, 0.274958, -0.659222, -0.659222, 0.559626, 0.559626, 0.274958, 0.274958, -0.659222, -0.659222, -0.178541, -0.178541, 0.689453, 0.689453, -0.028873, -0.028873, -0.178541, -0.178541, 0.689453, 0.689453, -0.028873, -0.028873, 0.053859, 0.053859, -0.446394, -0.446394, -0.53406, -0.53406, 0.053859, 0.053859, -0.446394, -0.446394, -0.53406, -0.53406, -0.178541, -0.178541, 0.689453, 0.689453, -0.028873, -0.028873, -0.178541, -0.178541, 0.689453, 0.689453, -0.028873, -0.028873, 0.053859, 0.053859, -0.446394, -0.446394, -0.53406, -0.53406, 0.053859, 0.053859, -0.446394, -0.446394, -0.53406, -0.53406, 0.776897, 0.776897, -0.700858, -0.700858, -0.802179, -0.802179, 0.776897, 0.776897, -0.700858, -0.700858, -0.802179, -0.802179, -0.616515, -0.616515, 0.718677, 0.718677, 0.303042, 0.303042, -0.616515, -0.616515, 0.718677, 0.718677, 0.303042, 0.303042, 0.776897, 0.776897, -0.700858, -0.700858, -0.802179, -0.802179, 0.776897, 0.776897, -0.700858, -0.700858, -0.802179, -0.802179, -0.616515, -0.616515, 0.718677, 0.718677, 0.303042, 0.303042, -0.616515, -0.616515, 0.718677, 0.718677, 0.303042, 0.303042, -0.080606, -0.080606, -0.850593, -0.850593, -0.795971, -0.795971, -0.080606, -0.080606, -0.850593, -0.850593, -0.795971, -0.795971, 0.860487, 0.860487, -0.90685, -0.90685, 0.89858, 0.89858, 0.860487, 0.860487, -0.90685, -0.90685, 0.89858, 0.89858, -0.080606, -0.080606, -0.850593, -0.850593, -0.795971, -0.795971, -0.080606, -0.080606, -0.850593, -0.850593, -0.795971, -0.795971, 0.860487, 0.860487, -0.90685, -0.90685, 0.89858, 0.89858, 0.860487, 0.860487, -0.90685, -0.90685, 0.89858, 0.89858],
|
||||
shape: [2, 4, 4, 6]
|
||||
}
|
||||
},
|
||||
'convolutional.UpSampling3D.2': {
|
||||
input: {
|
||||
data: [-0.989173, -0.133618, -0.505338, 0.023259, 0.503982, -0.303769, -0.436321, 0.793911, 0.416102, 0.806405, -0.098342, -0.738022],
|
||||
shape: [2, 1, 3, 2]
|
||||
},
|
||||
expected: {
|
||||
data: [-0.989173, -0.133618, -0.989173, -0.133618, -0.505338, 0.023259, -0.505338, 0.023259, 0.503982, -0.303769, 0.503982, -0.303769, -0.989173, -0.133618, -0.989173, -0.133618, -0.505338, 0.023259, -0.505338, 0.023259, 0.503982, -0.303769, 0.503982, -0.303769, -0.989173, -0.133618, -0.989173, -0.133618, -0.505338, 0.023259, -0.505338, 0.023259, 0.503982, -0.303769, 0.503982, -0.303769, -0.436321, 0.793911, -0.436321, 0.793911, 0.416102, 0.806405, 0.416102, 0.806405, -0.098342, -0.738022, -0.098342, -0.738022, -0.436321, 0.793911, -0.436321, 0.793911, 0.416102, 0.806405, 0.416102, 0.806405, -0.098342, -0.738022, -0.098342, -0.738022, -0.436321, 0.793911, -0.436321, 0.793911, 0.416102, 0.806405, 0.416102, 0.806405, -0.098342, -0.738022, -0.098342, -0.738022],
|
||||
shape: [2, 3, 6, 2]
|
||||
}
|
||||
},
|
||||
'convolutional.UpSampling3D.3': {
|
||||
input: {
|
||||
data: [-0.47588, 0.366985, 0.040173, 0.015578, -0.906159, 0.241982, -0.771299, -0.443554, -0.56404, -0.17751, 0.541277, -0.233327, 0.024369, 0.858275, 0.496191, 0.980574, -0.59522, 0.480899],
|
||||
shape: [2, 1, 3, 3]
|
||||
},
|
||||
expected: {
|
||||
data: [-0.47588, -0.47588, 0.366985, 0.366985, 0.040173, 0.040173, 0.015578, 0.015578, -0.906159, -0.906159, 0.241982, 0.241982, -0.771299, -0.771299, -0.443554, -0.443554, -0.56404, -0.56404, -0.47588, -0.47588, 0.366985, 0.366985, 0.040173, 0.040173, 0.015578, 0.015578, -0.906159, -0.906159, 0.241982, 0.241982, -0.771299, -0.771299, -0.443554, -0.443554, -0.56404, -0.56404, -0.17751, -0.17751, 0.541277, 0.541277, -0.233327, -0.233327, 0.024369, 0.024369, 0.858275, 0.858275, 0.496191, 0.496191, 0.980574, 0.980574, -0.59522, -0.59522, 0.480899, 0.480899, -0.17751, -0.17751, 0.541277, 0.541277, -0.233327, -0.233327, 0.024369, 0.024369, 0.858275, 0.858275, 0.496191, 0.496191, 0.980574, 0.980574, -0.59522, -0.59522, 0.480899, 0.480899],
|
||||
shape: [2, 2, 3, 6]
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
window.TEST_DATA = Object.assign({}, window.TEST_DATA, DATA)
|
||||
})()
|
||||
Reference in New Issue
Block a user