Skip to content

Instantly share code, notes, and snippets.

@ThomasDelteil
Last active May 4, 2018 20:50
Show Gist options
  • Save ThomasDelteil/88838cc0a8541109aac692c24a1eb382 to your computer and use it in GitHub Desktop.
Save ThomasDelteil/88838cc0a8541109aac692c24a1eb382 to your computer and use it in GitHub Desktop.
Display the source blob
Display the rendered blob
Raw
{
"cells": [
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# Using pre-trained models in MXNet"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"In this blog post I’ll show you how to use multiple pre-trained models with Apache MXNet. . In addition, prediction speed can vary, and that's an important factor for many applications. By trying a few pretrained models, you have an opportunity to find a model that can be a good fit for solving your business problem.\n",
"\n",
"First, let's download three image classification models from the Apache MXNet [model zoo](https://mxnet.incubator.apache.org/api/python/gluon/model_zoo.html).\n",
"* **VGG-16** ([research paper](https://arxiv.org/abs/1409.1556)), the 2014 classification winner at the [ImageNet Large Scale Visual Recognition Challenge](http://image-net.org/challenges/LSVRC).\n",
"* **MobileNet** ([research paper](https://arxiv.org/abs/1704.04861)), MobileNets are based on a streamlined architecture that uses depth-wise separable convolutions to build light weight deep neural networks, suitable for mobile applications.\n",
"* **ResNet-18** ([research paper](https://arxiv.org/abs/1512.03385v1)), the -152 version is the 2015 winner in multiple categories.\n",
"\n",
"Why would you want to try multiple models? Why not just pick the one with the best accuracy? As we will see later in the blog post, even though these models have been trained on the same data set and optimized for maximum accuracy, they do behave slightly differently on specific images. In addition, prediction speed can vary, and that’s an important factor for many applications. By trying a few pretrained models, you have an opportunity to find a model that can be a good fit for solving your business problem."
]
},
{
"cell_type": "code",
"execution_count": 4,
"metadata": {},
"outputs": [],
"source": [
"import mxnet as mx\n",
"from mxnet import gluon, nd\n",
"from mxnet.gluon.model_zoo import vision\n",
"import matplotlib.pyplot as plt\n",
"import numpy as np\n",
"import json\n",
"%matplotlib inline"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Loading the model\n",
"\n",
"The [Gluon Model Zoo](https://mxnet.incubator.apache.org/api/python/gluon/model_zoo.html) provides a collection of off-the-shelf models. You can get the ImageNet pre-trained model by using `pretrained=True`. \n",
"If you want to train on your own classification problem from scratch, you can get an untrained network with a specific number of classes using the `classes=10` for example"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"We can specify the *context* where we want to run the model: the default behavior is to use a CPU context. There are two reasons for this:\n",
"* First, this will allow you to test the notebook even if your machine is not equipped with a GPU :)\n",
"* Second, we're going to predict a single image and we don't have any specific performance requirements. For production applications where you'd want to predict large batches of images with the best possible throughput, a GPU could definitely be the way to go.\n",
"* If you want to use a GPU, make sure you have pip installed the right version of mxnet, or you will get an error when using the `mx.gpu()` context. Refer to the [install instructions](http://mxnet.incubator.apache.org/install/index.html)"
]
},
{
"cell_type": "code",
"execution_count": 121,
"metadata": {},
"outputs": [],
"source": [
"# We set the context to CPU, you can switch to GPU if you have one and installed a compatible version of MXNet \n",
"ctx = mx.cpu() "
]
},
{
"cell_type": "code",
"execution_count": 153,
"metadata": {},
"outputs": [],
"source": [
"# We can load three the three models\n",
"vgg16 = vision.vgg16(pretrained=True, ctx=ctx)\n",
"mobileNet = vision.mobilenet0_5(pretrained=True, ctx=ctx)\n",
"resnet18 = vision.resnet18_v1(pretrained=True, ctx=ctx)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"We can look at the description of the vgg16 network for example, which has a relatively simple architecture"
]
},
{
"cell_type": "code",
"execution_count": 114,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"VGG(\n",
" (features): HybridSequential(\n",
" (0): Conv2D(3 -> 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
" (1): Activation(relu)\n",
" (2): Conv2D(64 -> 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
" (3): Activation(relu)\n",
" (4): MaxPool2D(size=(2, 2), stride=(2, 2), padding=(0, 0), ceil_mode=False)\n",
" (5): Conv2D(64 -> 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
" (6): Activation(relu)\n",
" (7): Conv2D(128 -> 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
" (8): Activation(relu)\n",
" (9): MaxPool2D(size=(2, 2), stride=(2, 2), padding=(0, 0), ceil_mode=False)\n",
" (10): Conv2D(128 -> 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
" (11): Activation(relu)\n",
" (12): Conv2D(256 -> 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
" (13): Activation(relu)\n",
" (14): Conv2D(256 -> 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
" (15): Activation(relu)\n",
" (16): MaxPool2D(size=(2, 2), stride=(2, 2), padding=(0, 0), ceil_mode=False)\n",
" (17): Conv2D(256 -> 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
" (18): Activation(relu)\n",
" (19): Conv2D(512 -> 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
" (20): Activation(relu)\n",
" (21): Conv2D(512 -> 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
" (22): Activation(relu)\n",
" (23): MaxPool2D(size=(2, 2), stride=(2, 2), padding=(0, 0), ceil_mode=False)\n",
" (24): Conv2D(512 -> 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
" (25): Activation(relu)\n",
" (26): Conv2D(512 -> 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
" (27): Activation(relu)\n",
" (28): Conv2D(512 -> 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
" (29): Activation(relu)\n",
" (30): MaxPool2D(size=(2, 2), stride=(2, 2), padding=(0, 0), ceil_mode=False)\n",
" (31): Dense(25088 -> 4096, Activation(relu))\n",
" (32): Dropout(p = 0.5, axes=())\n",
" (33): Dense(4096 -> 4096, Activation(relu))\n",
" (34): Dropout(p = 0.5, axes=())\n",
" )\n",
" (output): Dense(4096 -> 1000, linear)\n",
")\n"
]
}
],
"source": [
"print(vgg16)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Count the `Conv2D` and `Dense` layers and you'll see that there are sixteen of them: thirteen convolutional layers and three fully connected layers. Now you know why this model is called VGG-16 :)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Let's have a closer look at the first convolution layer:"
]
},
{
"cell_type": "code",
"execution_count": 115,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"vgg3_conv0_ (\n",
" Parameter vgg3_conv0_weight (shape=(64, 3, 3, 3), dtype=<class 'numpy.float32'>)\n",
" Parameter vgg3_conv0_bias (shape=(64,), dtype=<class 'numpy.float32'>)\n",
")"
]
},
"execution_count": 115,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"print(vgg16.features[0].params)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"For each convolutional layer, there are two parameters: the weights of the convolutional masks and the biases. The first layer applies **`64`** different convolutional masks, of size **`InputChannels x 3 x 3`**. For the first convolution, there are **`3`** input channels, the `R`, `G`, `B` channels of the input image. That gives us the weight matrix of shape **`64 x 3 x 3 x 3`**, and bias vector of size **`64`**."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Let's have a look at the output layer now:"
]
},
{
"cell_type": "code",
"execution_count": 116,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"Dense(4096 -> 1000, linear)"
]
},
"execution_count": 116,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"print(vgg16.output)"
]
},
{
"cell_type": "code",
"execution_count": 142,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"vgg7_dense2_ (\n",
" Parameter vgg7_dense2_weight (shape=(1000, 4096), dtype=<class 'numpy.float32'>)\n",
" Parameter vgg7_dense2_bias (shape=(1000,), dtype=<class 'numpy.float32'>)\n",
")\n"
]
}
],
"source": [
"print(vgg16.output.params)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Did you notice the shape of the weight matrix? **1000 x 4096**. This layer contains 1,000 neurons: each of them will store the an activation representative of the probability of the image belonging to a specific category. Each neuron is also fully connected to all 4,096 neurons in the previous layer"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"OK, enough exploring! Now let's use these models to classify our own images."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Loading the data\n",
"All three models have been pre-trained on the ImageNet data set which includes over 1.2 million pictures of objects and animals sorted in 1,000 categories.\n",
"We get the imageNet list of labels. That way we have the mapping so when the model predicts for example category index `4`, we know it is predicting `hammerhead, hammerhead shark`"
]
},
{
"cell_type": "code",
"execution_count": 63,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"hammerhead, hammerhead shark\n"
]
}
],
"source": [
"mx.test_utils.download('https://raw.githubusercontent.com/dmlc/web-data/master/mxnet/doc/tutorials/onnx/image_net_labels.json')\n",
"categories = np.array(json.load(open('image_net_labels.json', 'r')))\n",
"print(categories[4])"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Get a test image"
]
},
{
"cell_type": "code",
"execution_count": 55,
"metadata": {},
"outputs": [],
"source": [
"mx.test_utils.download('https://github.com/dmlc/web-data/blob/master/mxnet/doc/tutorials/onnx/images/dog.jpg?raw=true', fname='dog.jpg')"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"If you want to use your own image for the test, upload it in the same folder and change the following line"
]
},
{
"cell_type": "code",
"execution_count": 138,
"metadata": {},
"outputs": [],
"source": [
"filename = 'dog.jpg'"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Load the image as a NDArray"
]
},
{
"cell_type": "code",
"execution_count": 139,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"<matplotlib.image.AxesImage at 0x7f9a2c211208>"
]
},
"execution_count": 139,
"metadata": {},
"output_type": "execute_result"
},
{
"data": {
"image/png": "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\n",
"text/plain": [
"<matplotlib.figure.Figure at 0x7f9a2c31d4a8>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"image = mx.image.imread(filename)\n",
"plt.imshow(image.asnumpy())"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Neural network expects input in a specific format. Usually images comes in the `Width x Height x Channels` format. Where channels are the RGB channels.\n",
"This network accepts images in the `BatchSize x 3 x 224 x 224`. `224 x 224` is the image resolution, that’s how the model was trained. `3` is the number of channels : Red, Green and Blue (in this order). In this case we use a `BatchSize` of `1` since we are predicting one image at a time.\n",
"Here are the transformation steps:\n",
"* Read the image: this will return a NDArray shaped as (image height, image width, 3), with the three channels in RGB order.\n",
"* Resize the shorter edge of the image 224.\n",
"* Crop, using a size of 224x224 from the center of the image.\n",
"* Shift the mean and standard deviation of our color channels to match the ones of the dataset the network has been trained on.\n",
"* Reshape the array from (Height, Width, 3) to (3, Height, Width).\n",
"* Add a fourth dimension, the batch dimension.\n"
]
},
{
"cell_type": "code",
"execution_count": 17,
"metadata": {},
"outputs": [],
"source": [
"def transform(image):\n",
" resized = mx.image.resize_short(image, 224) #minimum 224x224 images\n",
" cropped, crop_info = mx.image.center_crop(resized, (224, 224))\n",
" normalized = mx.image.color_normalize(cropped.astype(np.float32)/255,\n",
" mean=mx.nd.array([0.485, 0.456, 0.406]),\n",
" std=mx.nd.array([0.229, 0.224, 0.225])) \n",
" # the network expect batches of the form (N,3,224,224)\n",
" flipped_axis = normalized.transpose((2,0,1)) # Flipping from (224, 224, 3) to (3, 224, 224)\n",
" batchified = flipped_axis.expand_dims(axis=0) # change the shape from (3, 224, 224) to (1, 3, 224, 224)\n",
" return batchified"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Testing the different networks\n",
"We run the image through each pre-trained network. The models output a *NDArray* holding 1,000 activation values, which we convert to probabilities using the `softmax()` function, corresponding to the 1,000 categories it has been trained on. The *NDArray* has only one line since batch size is equal to 1"
]
},
{
"cell_type": "code",
"execution_count": 69,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"(1, 1000)\n"
]
}
],
"source": [
"predictions = resnet18(transform(image)).softmax()\n",
"print(predictions.shape)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"We then take the top `k` predictions for our image, here the top `3`."
]
},
{
"cell_type": "code",
"execution_count": 70,
"metadata": {},
"outputs": [],
"source": [
"top_pred = predictions.topk(k=3)[0].asnumpy()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"And we print the category predicted with the corresponding probability:"
]
},
{
"cell_type": "code",
"execution_count": 73,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"boxer: 92.89%\n",
"bull mastiff: 5.89%\n",
"Staffordshire bullterrier, Staffordshire bull terrier: 0.61%\n"
]
}
],
"source": [
"for index in top_pred:\n",
" probability = predictions[0][int(index)]\n",
" category = categories[int(index)]\n",
" print(\"{}: {:.2f}%\".format(category, probability.asscalar()*100))"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Let's turn this into a function. Our parameters are an image, a model, a list of categories and the number of top categories we'd like to print. "
]
},
{
"cell_type": "code",
"execution_count": 151,
"metadata": {},
"outputs": [],
"source": [
"def predict(model, image, categories, k):\n",
" predictions = model(transform(image)).softmax()\n",
" top_pred = predictions.topk(k=k)[0].asnumpy()\n",
" for index in top_pred:\n",
" probability = predictions[0][int(index)]\n",
" category = categories[int(index)]\n",
" print(\"{}: {:.2f}%\".format(category, probability.asscalar()*100))\n",
" print('')"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### VGG16"
]
},
{
"cell_type": "code",
"execution_count": 154,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"boxer: 97.13%\n",
"bull mastiff: 2.47%\n",
"Staffordshire bullterrier, Staffordshire bull terrier: 0.31%\n",
"\n",
"CPU times: user 916 ms, sys: 16 ms, total: 932 ms\n",
"Wall time: 587 ms\n"
]
}
],
"source": [
"%%time\n",
"predict(vgg16, image, categories, 3)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### MobileNet"
]
},
{
"cell_type": "code",
"execution_count": 135,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"boxer: 84.02%\n",
"bull mastiff: 13.63%\n",
"Rhodesian ridgeback: 0.66%\n",
"\n",
"CPU times: user 68 ms, sys: 0 ns, total: 68 ms\n",
"Wall time: 28.6 ms\n"
]
}
],
"source": [
"%%time\n",
"predict(mobileNet, image, categories, 3)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Resnet-18"
]
},
{
"cell_type": "code",
"execution_count": 136,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"boxer: 92.89%\n",
"bull mastiff: 5.89%\n",
"Staffordshire bullterrier, Staffordshire bull terrier: 0.61%\n",
"\n",
"CPU times: user 148 ms, sys: 0 ns, total: 148 ms\n",
"Wall time: 70.2 ms\n"
]
}
],
"source": [
"%%time\n",
"predict(resnet18, image, categories, 3)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"As you can see, pre-trained networks produce slightly different predictions, and have different run-time. In this case, MobileNet is almost **20 times faster** than VGG16!"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Fine-tuning using pre-trained models"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"You can modify the output layer of your pre-trained model to fit the right number of classes for your own image classification task like this, for example for 10 classes:"
]
},
{
"cell_type": "code",
"execution_count": 146,
"metadata": {},
"outputs": [],
"source": [
"NUM_CLASSES = 10\n",
"with resnet18.name_scope():\n",
" resnet18.output = gluon.nn.Dense(NUM_CLASSES)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Now you can train your model on your new data using the pre-trained weights on ImageNet as initialization. This is called transfer learning and it has shown to be very useful especially in the cases where you only have access to a small dataset. Your network will have already learned how to perform general pattern detection and feature extraction on the larger dataset.\n",
"You can learn more about transfer learning and fine-tuning with MXNet in these tutorials:\n",
"- [Transferring knowledge through fine-tuning](http://gluon.mxnet.io/chapter08_computer-vision/fine-tuning.html)\n",
"- [Fine Tuning an ONNX Model](https://mxnet.incubator.apache.org/tutorials/onnx/fine_tuning_gluon.html)\n"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"That's it! Explore the model zoo, have fun with pre-trained models!"
]
}
],
"metadata": {
"kernelspec": {
"display_name": "Environment (conda_mxnet_p36)",
"language": "python",
"name": "conda_mxnet_p36"
},
"language_info": {
"codemirror_mode": {
"name": "ipython",
"version": 3
},
"file_extension": ".py",
"mimetype": "text/x-python",
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
"version": "3.6.4"
}
},
"nbformat": 4,
"nbformat_minor": 2
}
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment