├── .gitignore ├── 01_install_packages.R ├── 02_download_datasets.R ├── 03_declare_mlp_model.R ├── 04_prepare_data_iterators.R ├── 05_fit_mlp_model.R ├── 06_restart_mlp_model.R ├── 07_predict_mlp_model.R ├── LICENSE └── README.md /.gitignore: -------------------------------------------------------------------------------- 1 | .RData 2 | .Rhistory 3 | -------------------------------------------------------------------------------- /01_install_packages.R: -------------------------------------------------------------------------------- 1 | .libPaths("libs") 2 | 3 | install.packages("checkpoint") 4 | repo <- checkpoint:::getSnapshotUrl(snapshotDate = "2016-09-20") 5 | 6 | install.packages("devtools", repos = repo) 7 | install.packages("drat", repos = repo) 8 | 9 | drat:::addRepo("dmlc") 10 | install.packages("mxnet") 11 | -------------------------------------------------------------------------------- /02_download_datasets.R: -------------------------------------------------------------------------------- 1 | data_dir <- "mnist" 2 | 3 | dir.create(data_dir, showWarnings = FALSE) 4 | 5 | if ((!file.exists(file.path(data_dir, 'train-images-idx3-ubyte'))) || 6 | (!file.exists(file.path(data_dir, 'train-labels-idx1-ubyte'))) || 7 | (!file.exists(file.path(data_dir, 't10k-images-idx3-ubyte'))) || 8 | (!file.exists(file.path(data_dir, 't10k-labels-idx1-ubyte')))) { 9 | download.file(url = 'http://data.dmlc.ml/mxnet/data/mnist.zip', 10 | destfile = 'mnist.zip', 11 | method = 'internal') 12 | unzip("mnist.zip", exdir = data_dir) 13 | file.remove("mnist.zip") 14 | } 15 | -------------------------------------------------------------------------------- /03_declare_mlp_model.R: -------------------------------------------------------------------------------- 1 | library(mxnet) 2 | 3 | #Ten digits 4 | num_classes <- 10 5 | 6 | # Data layer 7 | data <- mx.symbol.Variable('data') 8 | 9 | # First layer 10 | fc1 <- mx.symbol.FullyConnected(data = data, name = 'fc1', num_hidden = 128) 11 | # relu - activation 12 | act1 <- mx.symbol.Activation(data = fc1, name = 'relu1', act_type = "relu") 13 | 14 | # Second layer 15 | fc2 <- mx.symbol.FullyConnected(data = act1, name = 'fc2', num_hidden = 64) 16 | # relu - activation 17 | act2 <- mx.symbol.Activation(data = fc2, name = 'relu2', act_type = "relu") 18 | 19 | # Third layer 20 | fc3 <- mx.symbol.FullyConnected(data = act2, name = 'fc3', num_hidden = num_classes) 21 | 22 | # Softmax output layer 23 | mlp <- mx.symbol.SoftmaxOutput(data = fc3, name = 'softmax') 24 | 25 | -------------------------------------------------------------------------------- /04_prepare_data_iterators.R: -------------------------------------------------------------------------------- 1 | library(mxnet) 2 | 3 | data_dir <- "/home/ubuntu/R/mnist" 4 | 5 | #Train data iterator 6 | 7 | data_shape <- 28*28 #784 8 | batch_size <- 128 9 | flat <- TRUE 10 | 11 | mnist_train <- mx.io.MNISTIter( 12 | image = file.path(data_dir, "train-images-idx3-ubyte"), 13 | label = file.path(data_dir, "train-labels-idx1-ubyte"), 14 | input_shape = data_shape, 15 | batch_size = batch_size, 16 | shuffle = TRUE, 17 | flat = flat) 18 | 19 | mnist_validate <- mx.io.MNISTIter( 20 | image = file.path(data_dir, "t10k-images-idx3-ubyte"), 21 | label = file.path(data_dir, "t10k-labels-idx1-ubyte"), 22 | input_shape = data_shape, 23 | batch_size = batch_size, 24 | flat = flat) 25 | 26 | 27 | -------------------------------------------------------------------------------- /05_fit_mlp_model.R: -------------------------------------------------------------------------------- 1 | library(mxnet) 2 | 3 | #Devices 4 | #Using CPU 5 | #devs <- mx.cpu() 6 | 7 | #For graphic devices 8 | #use nvidia-smi to check status 9 | gpus_cnt <- 0 10 | devs <- lapply(X = 0:gpus_cnt, FUN = function(i) mx.gpu(i)) 11 | 12 | #How many learning rounds? 13 | num_round <- 50 14 | 15 | #How fast follow the gradient? 16 | learning_rate <- 0.1 17 | 18 | #How to deal with distributed learning? 19 | kv_store <- "local" 20 | 21 | #Checkpoint model 22 | ?mx.callback.save.checkpoint 23 | ?mx.model.FeedForward.create 24 | 25 | 26 | 27 | checkpoint_mlp <- function(iteration, nbatch, env, verbose=TRUE) { 28 | print(paste("Iteration=",iteration,format(Sys.time(), "%H:%M:%S"))) 29 | if (iteration %% 5 == 0) { 30 | mx.model.save(env$model, "mlp", iteration) 31 | cat(sprintf("Model checkpoint saved to %s-%04d.params\n", "mlp", iteration)) 32 | } 33 | return(TRUE) 34 | } 35 | 36 | 37 | 38 | 39 | model <- mx.model.FeedForward.create( 40 | X = mnist_train, 41 | eval.data = mnist_validate, 42 | ctx = devs, 43 | symbol = mlp, 44 | eval.metric = mx.metric.accuracy, 45 | num.round = num_round, 46 | learning.rate = learning_rate, 47 | momentum = 0.9, 48 | wd = 0.00001, 49 | kvstore = kv_store, 50 | array.batch.size = batch_size, 51 | epoch.end.callback = checkpoint_mlp, 52 | batch.end.callback = mx.callback.log.train.metric(150)) 53 | 54 | 55 | 56 | -------------------------------------------------------------------------------- /06_restart_mlp_model.R: -------------------------------------------------------------------------------- 1 | library(mxnet) 2 | 3 | model <- mx.model.load("mlp", 35) 4 | 5 | #Devices 6 | #Using CPU 7 | #devs <- mx.cpu() 8 | 9 | #For graphic devices 10 | gpus_cnt <- 0 11 | devs <- lapply(X = 0:gpus_cnt, FUN = function(i) mx.gpu(i)) 12 | 13 | #How many learning rounds? 14 | num_round <- 100 15 | 16 | #How fast follow the gradient? 17 | learning_rate <- 0.1 18 | 19 | #How to deal with distributed learning? 20 | kv_store <- "local" 21 | 22 | #Checkpoint model 23 | 24 | checkpoint_mlp <- mx.callback.save.checkpoint(prefix = "mlp", 25 | period = 5) 26 | 27 | 28 | model2 <- mx.model.FeedForward.create( 29 | X = mnist_train, 30 | eval.data = mnist_validate, 31 | ctx = devs, 32 | arg.params = model$arg.params, 33 | aux.params = model$aux.params, 34 | symbol = model$symbol, 35 | eval.metric = mx.metric.accuracy, 36 | num.round = num_round, 37 | learning.rate = 0.001, 38 | momentum = 0.9, 39 | wd = 0.00001, 40 | kvstore = kv_store, 41 | array.batch.size = batch_size, 42 | epoch.end.callback = checkpoint_mlp, 43 | batch.end.callback = mx.callback.log.train.metric(50)) 44 | -------------------------------------------------------------------------------- /07_predict_mlp_model.R: -------------------------------------------------------------------------------- 1 | library(mxnet) 2 | 3 | #model <- mx.model.load("mlp", 35) 4 | 5 | 6 | mnist_validate$reset() 7 | mnist_validate$iter.next() 8 | 9 | val <- mnist_validate$value() 10 | val$data <- as.array(val$data) 11 | val$label <- as.array(val$label) 12 | 13 | #preds <- predict(model, 14 | # X = val$data, 15 | # ctx = mx.cpu()) 16 | #ypred = max.col(t(as.array(preds))) 17 | #head(ypred,20) 18 | 19 | preds <- predict(model, 20 | X = mnist_validate, 21 | ctx = mx.cpu()) 22 | ypred = max.col(t(as.array(preds)))-1 23 | head(ypred,20) 24 | 25 | head(val$label,20) 26 | 27 | 28 | t=1 29 | image(t(apply(matrix(val$data[, t], nrow = 28, byrow = TRUE),2,rev)), col = grey(seq(0,1,length.out = 256))) 30 | title(main =sprintf("Record number=[%s]. Label = %s. Pred = %s", t, val$label[t],ypred[t])) 31 | 32 | 33 | for (t in 1:25) { 34 | image(t(apply(matrix(val$data[, t], nrow = 28, byrow = TRUE),2,rev)), col = grey(seq(0,1,length.out = 256))) 35 | title(main =sprintf("Record number=[%s]. Label = %s. Pred = %s", t, val$label[t],ypred[t])) 36 | } 37 | 38 | -------------------------------------------------------------------------------- /LICENSE: -------------------------------------------------------------------------------- 1 | Apache License 2 | Version 2.0, January 2004 3 | http://www.apache.org/licenses/ 4 | 5 | TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION 6 | 7 | 1. Definitions. 8 | 9 | "License" shall mean the terms and conditions for use, reproduction, 10 | and distribution as defined by Sections 1 through 9 of this document. 11 | 12 | "Licensor" shall mean the copyright owner or entity authorized by 13 | the copyright owner that is granting the License. 14 | 15 | "Legal Entity" shall mean the union of the acting entity and all 16 | other entities that control, are controlled by, or are under common 17 | control with that entity. For the purposes of this definition, 18 | "control" means (i) the power, direct or indirect, to cause the 19 | direction or management of such entity, whether by contract or 20 | otherwise, or (ii) ownership of fifty percent (50%) or more of the 21 | outstanding shares, or (iii) beneficial ownership of such entity. 22 | 23 | "You" (or "Your") shall mean an individual or Legal Entity 24 | exercising permissions granted by this License. 25 | 26 | "Source" form shall mean the preferred form for making modifications, 27 | including but not limited to software source code, documentation 28 | source, and configuration files. 29 | 30 | "Object" form shall mean any form resulting from mechanical 31 | transformation or translation of a Source form, including but 32 | not limited to compiled object code, generated documentation, 33 | and conversions to other media types. 34 | 35 | "Work" shall mean the work of authorship, whether in Source or 36 | Object form, made available under the License, as indicated by a 37 | copyright notice that is included in or attached to the work 38 | (an example is provided in the Appendix below). 39 | 40 | "Derivative Works" shall mean any work, whether in Source or Object 41 | form, that is based on (or derived from) the Work and for which the 42 | editorial revisions, annotations, elaborations, or other modifications 43 | represent, as a whole, an original work of authorship. For the purposes 44 | of this License, Derivative Works shall not include works that remain 45 | separable from, or merely link (or bind by name) to the interfaces of, 46 | the Work and Derivative Works thereof. 47 | 48 | "Contribution" shall mean any work of authorship, including 49 | the original version of the Work and any modifications or additions 50 | to that Work or Derivative Works thereof, that is intentionally 51 | submitted to Licensor for inclusion in the Work by the copyright owner 52 | or by an individual or Legal Entity authorized to submit on behalf of 53 | the copyright owner. For the purposes of this definition, "submitted" 54 | means any form of electronic, verbal, or written communication sent 55 | to the Licensor or its representatives, including but not limited to 56 | communication on electronic mailing lists, source code control systems, 57 | and issue tracking systems that are managed by, or on behalf of, the 58 | Licensor for the purpose of discussing and improving the Work, but 59 | excluding communication that is conspicuously marked or otherwise 60 | designated in writing by the copyright owner as "Not a Contribution." 61 | 62 | "Contributor" shall mean Licensor and any individual or Legal Entity 63 | on behalf of whom a Contribution has been received by Licensor and 64 | subsequently incorporated within the Work. 65 | 66 | 2. Grant of Copyright License. Subject to the terms and conditions of 67 | this License, each Contributor hereby grants to You a perpetual, 68 | worldwide, non-exclusive, no-charge, royalty-free, irrevocable 69 | copyright license to reproduce, prepare Derivative Works of, 70 | publicly display, publicly perform, sublicense, and distribute the 71 | Work and such Derivative Works in Source or Object form. 72 | 73 | 3. Grant of Patent License. Subject to the terms and conditions of 74 | this License, each Contributor hereby grants to You a perpetual, 75 | worldwide, non-exclusive, no-charge, royalty-free, irrevocable 76 | (except as stated in this section) patent license to make, have made, 77 | use, offer to sell, sell, import, and otherwise transfer the Work, 78 | where such license applies only to those patent claims licensable 79 | by such Contributor that are necessarily infringed by their 80 | Contribution(s) alone or by combination of their Contribution(s) 81 | with the Work to which such Contribution(s) was submitted. If You 82 | institute patent litigation against any entity (including a 83 | cross-claim or counterclaim in a lawsuit) alleging that the Work 84 | or a Contribution incorporated within the Work constitutes direct 85 | or contributory patent infringement, then any patent licenses 86 | granted to You under this License for that Work shall terminate 87 | as of the date such litigation is filed. 88 | 89 | 4. Redistribution. You may reproduce and distribute copies of the 90 | Work or Derivative Works thereof in any medium, with or without 91 | modifications, and in Source or Object form, provided that You 92 | meet the following conditions: 93 | 94 | (a) You must give any other recipients of the Work or 95 | Derivative Works a copy of this License; and 96 | 97 | (b) You must cause any modified files to carry prominent notices 98 | stating that You changed the files; and 99 | 100 | (c) You must retain, in the Source form of any Derivative Works 101 | that You distribute, all copyright, patent, trademark, and 102 | attribution notices from the Source form of the Work, 103 | excluding those notices that do not pertain to any part of 104 | the Derivative Works; and 105 | 106 | (d) If the Work includes a "NOTICE" text file as part of its 107 | distribution, then any Derivative Works that You distribute must 108 | include a readable copy of the attribution notices contained 109 | within such NOTICE file, excluding those notices that do not 110 | pertain to any part of the Derivative Works, in at least one 111 | of the following places: within a NOTICE text file distributed 112 | as part of the Derivative Works; within the Source form or 113 | documentation, if provided along with the Derivative Works; or, 114 | within a display generated by the Derivative Works, if and 115 | wherever such third-party notices normally appear. The contents 116 | of the NOTICE file are for informational purposes only and 117 | do not modify the License. You may add Your own attribution 118 | notices within Derivative Works that You distribute, alongside 119 | or as an addendum to the NOTICE text from the Work, provided 120 | that such additional attribution notices cannot be construed 121 | as modifying the License. 122 | 123 | You may add Your own copyright statement to Your modifications and 124 | may provide additional or different license terms and conditions 125 | for use, reproduction, or distribution of Your modifications, or 126 | for any such Derivative Works as a whole, provided Your use, 127 | reproduction, and distribution of the Work otherwise complies with 128 | the conditions stated in this License. 129 | 130 | 5. Submission of Contributions. Unless You explicitly state otherwise, 131 | any Contribution intentionally submitted for inclusion in the Work 132 | by You to the Licensor shall be under the terms and conditions of 133 | this License, without any additional terms or conditions. 134 | Notwithstanding the above, nothing herein shall supersede or modify 135 | the terms of any separate license agreement you may have executed 136 | with Licensor regarding such Contributions. 137 | 138 | 6. Trademarks. This License does not grant permission to use the trade 139 | names, trademarks, service marks, or product names of the Licensor, 140 | except as required for reasonable and customary use in describing the 141 | origin of the Work and reproducing the content of the NOTICE file. 142 | 143 | 7. Disclaimer of Warranty. Unless required by applicable law or 144 | agreed to in writing, Licensor provides the Work (and each 145 | Contributor provides its Contributions) on an "AS IS" BASIS, 146 | WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or 147 | implied, including, without limitation, any warranties or conditions 148 | of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A 149 | PARTICULAR PURPOSE. You are solely responsible for determining the 150 | appropriateness of using or redistributing the Work and assume any 151 | risks associated with Your exercise of permissions under this License. 152 | 153 | 8. Limitation of Liability. In no event and under no legal theory, 154 | whether in tort (including negligence), contract, or otherwise, 155 | unless required by applicable law (such as deliberate and grossly 156 | negligent acts) or agreed to in writing, shall any Contributor be 157 | liable to You for damages, including any direct, indirect, special, 158 | incidental, or consequential damages of any character arising as a 159 | result of this License or out of the use or inability to use the 160 | Work (including but not limited to damages for loss of goodwill, 161 | work stoppage, computer failure or malfunction, or any and all 162 | other commercial damages or losses), even if such Contributor 163 | has been advised of the possibility of such damages. 164 | 165 | 9. Accepting Warranty or Additional Liability. While redistributing 166 | the Work or Derivative Works thereof, You may choose to offer, 167 | and charge a fee for, acceptance of support, warranty, indemnity, 168 | or other liability obligations and/or rights consistent with this 169 | License. However, in accepting such obligations, You may act only 170 | on Your own behalf and on Your sole responsibility, not on behalf 171 | of any other Contributor, and only if You agree to indemnify, 172 | defend, and hold each Contributor harmless for any liability 173 | incurred by, or claims asserted against, such Contributor by reason 174 | of your accepting any such warranty or additional liability. 175 | 176 | END OF TERMS AND CONDITIONS 177 | 178 | APPENDIX: How to apply the Apache License to your work. 179 | 180 | To apply the Apache License to your work, attach the following 181 | boilerplate notice, with the fields enclosed by brackets "{}" 182 | replaced with your own identifying information. (Don't include 183 | the brackets!) The text should be enclosed in the appropriate 184 | comment syntax for the file format. We also recommend that a 185 | file or class name and description of purpose be included on the 186 | same "printed page" as the copyright notice for easier 187 | identification within third-party archives. 188 | 189 | Copyright {yyyy} {name of copyright owner} 190 | 191 | Licensed under the Apache License, Version 2.0 (the "License"); 192 | you may not use this file except in compliance with the License. 193 | You may obtain a copy of the License at 194 | 195 | http://www.apache.org/licenses/LICENSE-2.0 196 | 197 | Unless required by applicable law or agreed to in writing, software 198 | distributed under the License is distributed on an "AS IS" BASIS, 199 | WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. 200 | See the License for the specific language governing permissions and 201 | limitations under the License. 202 | -------------------------------------------------------------------------------- /README.md: -------------------------------------------------------------------------------- 1 | # Showcase: Deep learning in the cloud. Building a handwritten character recognition model with [GNU R](https://www.r-project.org/) and [MXNet](http://mxnet.io/) with [Amazon Web Services](https://aws.amazon.com/) on [GPU instances](http://docs.aws.amazon.com/AWSEC2/latest/UserGuide/using_cluster_computing.html). 2 | 3 | ## Business story behind the showcase 4 | 5 | Optical character recognition (OCR) for handwritten text is applicable in areas where several documents are being used in business for processing large amounts of paper documents. One example could be a traditional post operator that wants to automatically process information on the envelopes. The envelopes are usually not neatly adressed and standard OCR solutions fail in this area. A custom-made character recognition engine designed for this task might receive much better quality level. 6 | 7 | ## Approach 8 | The showcase presents two main things: 9 | 10 | * Construction of deep laerning models with MXNet library 11 | * How to run GPU computing in the cloud with Amazon Web Services 12 | 13 | 14 | 15 | ## Prerequisites 16 | 17 | We use bitfusion.io [Scientific Computing AMI](https://aws.amazon.com/marketplace/seller-profile?id=3b372560-86bf-4e3d-9ec0-016892a64bed) 18 | 19 | The AMI contains Ubuntu 14 along with a R installation along with CUDA drivers. 20 | Additionally we have installed MXNet running the following commands: 21 | 22 | * `sudo apt-get update` 23 | * `sudo apt-get install -y build-essential git libblas-dev libopencv-dev` 24 | * `git clone --recursive https://github.com/dmlc/mxnet` 25 | 26 | Next modify config.mk by setting the following keys: 27 | 28 | USE_CUDA = 1 29 | USE_CUDA_PATH = /usr/local/cuda 30 | USE_BLAS = atlas 31 | 32 | Finally, compile mxnet with the command `make –j4` 33 | 34 | ## Usage instruction 35 | 36 | 1. Install R packages `01_install_packages.R` 37 | 2. Prepare the dataset `02_download_datasets.R` 38 | 3. Declare layers for the deep neural network `03_declare_mlp_model.R` 39 | 4. Create data iterators for seuential data reading `04_prepare_data_iterators.R` 40 | 5. Fit the model to the data `05_fit_mlp_model.R` 41 | 6. If the fitting process is interrupted a script for resuming computation state can be used: `06_restart_mlp_model.R` 42 | 7. Perform predictions and observe the results `07_predict_mlp_model.R` 43 | 44 | ## What next 45 | 46 | This example presents one possible usage of deep learning models for classification of images. 47 | One important problem is selection of an optimal structure for a deep neural network. 48 | This requires execution of several experiments for measuring predictive capabilities for various network topologies. 49 | Amazon Web Services comes forward to this need and offers very large GPU instances. The flagship offering is a p2.16xlarge offering 16 x GPU Nvidia TESLA K80 with a total of 80'000 GPU cores. This machine availabe from around $2.10 on AWS spot market would make it to the list of Top Supercomputers just 10 years ago. 50 | 51 | 52 | 53 | 54 | 55 | --------------------------------------------------------------------------------