initial complete for recognize_digits

recognize_digits/data/get_mnist_data.sh

+#!/usr/bin/env sh
+# Copyright (c) 2016 PaddlePaddle Authors. All Rights Reserved
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+# This scripts downloads the mnist data and unzips it.
+set -e
+DIR="$( cd "$(dirname "$0")" ; pwd -P )"
+rm -rf "$DIR/raw_data"
+mkdir "$DIR/raw_data"
+cd "$DIR/raw_data"
+
+echo "Downloading..."
+
+for fname in train-images-idx3-ubyte train-labels-idx1-ubyte t10k-images-idx3-ubyte t10k-labels-idx1-ubyte
+do
+    if [ ! -e $fname ]; then
+        wget --no-check-certificate http://yann.lecun.com/exdb/mnist/${fname}.gz
+        gunzip ${fname}.gz
+    fi
+done
+
+cd $DIR
+rm -f *.list
+echo "./data/raw_data/train\n" > "$DIR/train.list"
+echo "./data/raw_data/t10k\n" > "$DIR/test.list"

recognize_digits/src/cnn_mnist.py

+# Copyright (c) 2016 PaddlePaddle Authors. All Rights Reserved
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from paddle.trainer_config_helpers import *
+
+is_predict = get_config_arg("is_predict", bool, False)
+
+####################Data Configuration ##################
+
+if not is_predict:
+    data_dir = './data/'
+    define_py_data_sources2(
+        train_list=data_dir + 'train.list',
+        test_list=data_dir + 'test.list',
+        module='mnist_provider',
+        obj='process')
+
+######################Algorithm Configuration #############
+settings(
+    batch_size=128,
+    learning_rate=0.1 / 128.0,
+    learning_method=MomentumOptimizer(0.9),
+    regularization=L2Regularization(0.0005 * 128))
+
+#######################Network Configuration #############
+
+data_size = 1 * 28 * 28
+label_size = 10
+img = data_layer(name='pixel', size=data_size)
+
+# first conv layer
+conv_pool_1 = simple_img_conv_pool(
+    input=img,
+    filter_size=5,
+    num_filters=20,
+    num_channel=1,
+    pool_size=2,
+    pool_stride=2,
+    act=TanhActivation())
+# second conv layer
+conv_pool_2 = simple_img_conv_pool(
+    input=conv_pool_1,
+    filter_size=5,
+    num_filters=50,
+    num_channel=20,
+    pool_size=2,
+    pool_stride=2,
+    act=TanhActivation())
+# The first fully-connected layer
+fc1 = fc_layer(input=conv_pool_2, size=128, act=TanhActivation())
+# The softmax layer, note that the hidden size should be 10,
+# which is the number of unique digits
+predict = fc_layer(input=fc1, size=10, act=SoftmaxActivation())
+
+if not is_predict:
+    lbl = data_layer(name="label", size=label_size)
+    inputs(img, lbl)
+    outputs(classification_cost(input=predict, label=lbl))
+else:
+    outputs(predict)

recognize_digits/src/evaluate.py

+#!/usr/bin/python
+# Copyright (c) 2016 PaddlePaddle Authors. All Rights Reserved
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+import sys
+import re
+import math
+
+
+def get_best_pass(filename):
+    with open(filename, 'r') as f:
+        text = f.read()
+        pattern = re.compile(
+            'Test.*? cost=([0-9]+\.[0-9]+).*?classification_error_evaluator=([0-9]+\.[0-9]+).*?pass-([0-9]+)',
+            re.S)
+        results = re.findall(pattern, text)
+        sorted_results = sorted(results, key=lambda result: float(result[0]))
+        return sorted_results[0]
+
+
+filename = sys.argv[1]
+log = get_best_pass(filename)
+predict_error = math.sqrt(float(log[0])) / 2
+classification_accuracy = (1 - float(log[1])) * 100
+print 'Best pass is %s, error is %s, which means predict get error as %f' % (
+    log[2], log[0], predict_error)
+print 'The classification accuracy is %.2f%%' % classification_accuracy

recognize_digits/src/load_data.py

+# Copyright (c) 2016 PaddlePaddle Authors. All Rights Reserved
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+import numpy as np
+import matplotlib.pyplot as plt
+import random
+
+
+def read_data(path, filename):
+    imgf = path + filename + "-images-idx3-ubyte"
+    labelf = path + filename + "-labels-idx1-ubyte"
+    f = open(imgf, "rb")
+    l = open(labelf, "rb")
+
+    f.read(16)
+    l.read(8)
+
+    # Define number of samples for train/test
+    n = 60000 if "train" in filename else 10000
+
+    rows = 28
+    cols = 28
+
+    images = np.fromfile(
+        f, 'ubyte',
+        count=n * rows * cols).reshape(n, rows, cols).astype('float32')
+    labels = np.fromfile(l, 'ubyte', count=n).astype("int")
+
+    return images, labels
+
+
+if __name__ == "__main__":
+    train_images, train_labels = read_data("./raw_data/", "train")
+    test_images, test_labels = read_data("./raw_data/", "t10k")
+    label_list = []
+    for i in range(10):
+        index = random.randint(0, train_images.shape[0] - 1)
+        label_list.append(train_labels[index])
+        plt.subplot(1, 10, i + 1)
+        plt.imshow(train_images[index], cmap="Greys_r")
+        plt.axis('off')
+    print('label: %s' % (label_list, ))
+    plt.show()

recognize_digits/src/mlp_mnist.py

+# Copyright (c) 2016 PaddlePaddle Authors. All Rights Reserved
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from paddle.trainer_config_helpers import *
+
+is_predict = get_config_arg("is_predict", bool, False)
+
+####################Data Configuration ##################
+
+if not is_predict:
+    data_dir = './data/'
+    define_py_data_sources2(
+        train_list=data_dir + 'train.list',
+        test_list=data_dir + 'test.list',
+        module='mnist_provider',
+        obj='process')
+
+######################Algorithm Configuration #############
+settings(
+    batch_size=128,
+    learning_rate=0.1 / 128.0,
+    learning_method=MomentumOptimizer(0.9),
+    regularization=L2Regularization(0.0005 * 128))
+
+#######################Network Configuration #############
+
+data_size = 1 * 28 * 28
+label_size = 10
+img = data_layer(name='pixel', size=data_size)
+
+# The first fully-connected layer
+hidden1 = fc_layer(input=img, size=128, act=ReluActivation())
+# The second fully-connected layer and the according activation function
+hidden2 = fc_layer(input=hidden1, size=64, act=ReluActivation())
+# The thrid fully-connected layer, note that the hidden size should be 10,
+# which is the number of unique digits
+predict = fc_layer(input=hidden2, size=10, act=SoftmaxActivation())
+
+if not is_predict:
+    lbl = data_layer(name="label", size=label_size)
+    inputs(img, lbl)
+    outputs(classification_cost(input=predict, label=lbl))
+else:
+    outputs(predict)

recognize_digits/src/mnist_provider.py

+# Copyright (c) 2016 PaddlePaddle Authors. All Rights Reserved
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from paddle.trainer.PyDataProvider2 import *
+import numpy
+
+
+# Define a py data provider
+@provider(
+    input_types={'pixel': dense_vector(28 * 28),
+                 'label': integer_value(10)},
+    cache=CacheType.CACHE_PASS_IN_MEM)
+def process(settings, filename):  # settings is not used currently.
+    imgf = filename + "-images-idx3-ubyte"
+    labelf = filename + "-labels-idx1-ubyte"
+    f = open(imgf, "rb")
+    l = open(labelf, "rb")
+
+    f.read(16)
+    l.read(8)
+
+    # Define number of samples for train/test
+    if "train" in filename:
+        n = 60000
+    else:
+        n = 10000
+
+    images = numpy.fromfile(
+        f, 'ubyte', count=n * 28 * 28).reshape((n, 28 * 28)).astype('float32')
+    images = images / 255.0 * 2.0 - 1.0
+    labels = numpy.fromfile(l, 'ubyte', count=n).astype("int")
+
+    for i in xrange(n):
+        yield {"pixel": images[i, :], 'label': labels[i]}
+
+    f.close()
+    l.close()

recognize_digits/src/plot_error.py

+# Copyright (c) 2016 PaddlePaddle Authors. All Rights Reserved
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+import matplotlib.pyplot as plt
+import re
+import sys
+
+
+def plot_log(filename):
+    with open(filename, 'r') as f:
+        text = f.read()
+        pattern = re.compile('Test.*? cost=([0-9]+\.[0-9]+).*?pass-([0-9]+)',
+                             re.S)
+        results = re.findall(pattern, text)
+        cost, pass_ = zip(*results)
+        cost_float = map(float, cost)
+        pass_int = map(int, pass_)
+        plt.plot(pass_int, cost_float, 'bo', pass_, cost_float, 'k')
+        plt.ylabel('AvgCost')
+        plt.xlabel('epoch')
+        plt.show()
+
+
+if __name__ == '__main__':
+    plot_log(sys.argv[1])

recognize_digits/src/predict.py

+# Copyright (c) 2016 PaddlePaddle Authors. All Rights Reserved
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+"""Usage: predict.py -c CONF -d DATA -m MODEL
+
+
+Arguments:
+    CONF        train conf
+    DATA        MNIST Data
+    MODEL       Model
+
+Options:
+    -h      --help
+    -c      conf
+    -d      data
+    -m      model
+"""
+
+import os
+import sys
+from docopt import docopt
+import matplotlib.pyplot as plt
+import numpy as np
+
+from py_paddle import swig_paddle, DataProviderConverter
+from paddle.trainer.PyDataProvider2 import dense_vector
+from paddle.trainer.config_parser import parse_config
+
+from load_data import read_data
+
+
+class Prediction():
+    def __init__(self, train_conf, data_dir, model_dir):
+
+        conf = parse_config(train_conf, 'is_predict=1')
+        self.network = swig_paddle.GradientMachine.createFromConfigProto(
+            conf.model_config)
+        self.network.loadParameters(model_dir)
+
+        self.images, self.labels = read_data(data_dir, "t10k")
+
+        slots = [dense_vector(28 * 28)]
+        self.converter = DataProviderConverter(slots)
+
+    def predict(self, index):
+        input = self.converter([[self.images[index].flatten().tolist()]])
+        output = self.network.forwardTest(input)
+        prob = output[0]["value"]
+        predict = np.argsort(-prob)
+        print prob
+        print predict[0][0], self.labels[index]
+
+
+def main():
+    arguments = docopt(__doc__)
+    train_conf = arguments['CONF']
+    data_dir = arguments['DATA']
+    model_dir = arguments['MODEL']
+    swig_paddle.initPaddle("--use_gpu=0")
+    predictor = Prediction(train_conf, data_dir, model_dir)
+    while True:
+        index = int(raw_input("Input image_id [0~9999]: "))
+        predictor.predict(index)
+
+
+if __name__ == '__main__':
+    main()

recognize_digits/src/softmax_mnist.py

+# Copyright (c) 2016 PaddlePaddle Authors. All Rights Reserved
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from paddle.trainer_config_helpers import *
+
+is_predict = get_config_arg("is_predict", bool, False)
+
+####################Data Configuration ##################
+
+if not is_predict:
+    data_dir = './data/'
+    define_py_data_sources2(
+        train_list=data_dir + 'train.list',
+        test_list=data_dir + 'test.list',
+        module='mnist_provider',
+        obj='process')
+
+######################Algorithm Configuration #############
+settings(
+    batch_size=128,
+    learning_rate=0.1 / 128.0,
+    learning_method=MomentumOptimizer(0.9),
+    regularization=L2Regularization(0.0005 * 128))
+
+#######################Network Configuration #############
+
+data_size = 1 * 28 * 28
+label_size = 10
+img = data_layer(name='pixel', size=data_size)
+
+# mlp is used
+predict = fc_layer(input=img, size=10, act=SoftmaxActivation())
+
+if not is_predict:
+    lbl = data_layer(name="label", size=label_size)
+    inputs(img, lbl)
+    outputs(classification_cost(input=predict, label=lbl))
+else:
+    outputs(predict)

recognize_digits/src/train.sh

+#!/bin/bash
+# Copyright (c) 2016 PaddlePaddle Authors. All Rights Reserved
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+set -e
+
+config=cnn_mnist.py
+output=./cnn_mnist_model
+log=cnn_train.log
+
+
+
+paddle train \
+--config=$config \
+--dot_period=10 \
+--log_period=100 \
+--test_all_data_in_one_period=1 \
+--use_gpu=0 \
+--trainer_count=1 \
+--num_passes=100 \
+--save_dir=$output \
+2>&1 | tee $log
+
+python -m paddle.utils.plotcurve -i $log > plot.png