From dd62d6fa4a85f18a29de009a972f5599b19ec946 Mon Sep 17 00:00:00 2001
From: natanielruiz <nataniel777@hotmail.com>
Date: 星期四, 14 九月 2017 00:51:53 +0800
Subject: [PATCH] Fixing hopenet
---
code/train.py | 2
practice/load_AFLW-KEPLER_split.ipynb | 190 ++++++++++++++
practice/.ipynb_checkpoints/load_AFLW-Copy1-checkpoint.ipynb | 30 +-
practice/remove_KEPLER_test_split.ipynb | 128 +++++++++
practice/create_filtered_datasets.ipynb | 12
code/batch_testing.py | 6
practice/.ipynb_checkpoints/load_AFLW-KEPLER_split-checkpoint.ipynb | 225 +++++++++++++++++
code/test.py | 7
practice/load_AFLW-Copy1.ipynb | 4
practice/.ipynb_checkpoints/remove_KEPLER_test_split-checkpoint.ipynb | 128 +++++++++
10 files changed, 704 insertions(+), 28 deletions(-)
diff --git a/code/batch_testing.py b/code/batch_testing.py
index 237ea54..db688a9 100644
--- a/code/batch_testing.py
+++ b/code/batch_testing.py
@@ -123,9 +123,9 @@
label_roll = labels[:,2].float()
pre_yaw, pre_pitch, pre_roll, angles = model(images)
- yaw = angles[args.iter_ref-1][:,0].cpu().data
- pitch = angles[args.iter_ref-1][:,1].cpu().data
- roll = angles[args.iter_ref-1][:,2].cpu().data
+ yaw = angles[args.iter_ref][:,0].cpu().data
+ pitch = angles[args.iter_ref][:,1].cpu().data
+ roll = angles[args.iter_ref][:,2].cpu().data
# Mean absolute error
yaw_error += torch.sum(torch.abs(yaw - label_yaw) * 3)
diff --git a/code/test.py b/code/test.py
index 41db842..7f76714 100644
--- a/code/test.py
+++ b/code/test.py
@@ -110,11 +110,12 @@
label_roll = labels[:,2].float()
pre_yaw, pre_pitch, pre_roll, angles = model(images)
- yaw = angles[args.iter_ref-1][:,0].cpu().data
- pitch = angles[args.iter_ref-1][:,1].cpu().data
- roll = angles[args.iter_ref-1][:,2].cpu().data
+ yaw = angles[args.iter_ref][:,0].cpu().data
+ pitch = angles[args.iter_ref][:,1].cpu().data
+ roll = angles[args.iter_ref][:,2].cpu().data
# Mean absolute error
+ print yaw.numpy(), label_yaw.numpy()
yaw_error += torch.sum(torch.abs(yaw - label_yaw) * 3)
pitch_error += torch.sum(torch.abs(pitch - label_pitch) * 3)
roll_error += torch.sum(torch.abs(roll - label_roll) * 3)
diff --git a/code/train.py b/code/train.py
index 50eeb82..6e1ae5b 100644
--- a/code/train.py
+++ b/code/train.py
@@ -241,7 +241,7 @@
# Finetuning loss
loss_seq = [loss_yaw, loss_pitch, loss_roll]
- for idx in xrange(args.iter_ref):
+ for idx in xrange(args.iter_ref+1):
loss_angles = reg_criterion(angles[idx], label_angles.float())
loss_seq.append(loss_angles)
diff --git a/practice/.ipynb_checkpoints/load_AFLW-Copy1-checkpoint.ipynb b/practice/.ipynb_checkpoints/load_AFLW-Copy1-checkpoint.ipynb
index 5c0a9b6..856fbb4 100644
--- a/practice/.ipynb_checkpoints/load_AFLW-Copy1-checkpoint.ipynb
+++ b/practice/.ipynb_checkpoints/load_AFLW-Copy1-checkpoint.ipynb
@@ -2,7 +2,7 @@
"cells": [
{
"cell_type": "code",
- "execution_count": 6,
+ "execution_count": 1,
"metadata": {
"collapsed": true
},
@@ -23,7 +23,7 @@
},
{
"cell_type": "code",
- "execution_count": 7,
+ "execution_count": 2,
"metadata": {
"collapsed": true
},
@@ -36,20 +36,16 @@
},
{
"cell_type": "code",
- "execution_count": 8,
+ "execution_count": 4,
"metadata": {
"collapsed": false
},
"outputs": [
{
- "ename": "KeyboardInterrupt",
- "evalue": "",
- "output_type": "error",
- "traceback": [
- "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
- "\u001b[0;31mKeyboardInterrupt\u001b[0m Traceback (most recent call last)",
- "\u001b[0;32m<ipython-input-8-1f2606c2a679>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[1;32m 34\u001b[0m \u001b[0;32mif\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mos\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mpath\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0misfile\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0minput_path\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m==\u001b[0m \u001b[0mTrue\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 35\u001b[0m \u001b[0;31m#image = cv2.imread(input_path, 0) #load in grayscale\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 36\u001b[0;31m \u001b[0mimage\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mcv2\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mimread\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0minput_path\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 37\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 38\u001b[0m \u001b[0;31m#Image dimensions\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
- "\u001b[0;31mKeyboardInterrupt\u001b[0m: "
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Done\n"
]
}
],
@@ -148,11 +144,19 @@
},
{
"cell_type": "code",
- "execution_count": null,
+ "execution_count": 5,
"metadata": {
"collapsed": false
},
- "outputs": [],
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "test\n"
+ ]
+ }
+ ],
"source": [
"print 'test'"
]
diff --git a/practice/.ipynb_checkpoints/load_AFLW-KEPLER_split-checkpoint.ipynb b/practice/.ipynb_checkpoints/load_AFLW-KEPLER_split-checkpoint.ipynb
new file mode 100644
index 0000000..e882f99
--- /dev/null
+++ b/practice/.ipynb_checkpoints/load_AFLW-KEPLER_split-checkpoint.ipynb
@@ -0,0 +1,225 @@
+{
+ "cells": [
+ {
+ "cell_type": "code",
+ "execution_count": 2,
+ "metadata": {
+ "collapsed": true
+ },
+ "outputs": [],
+ "source": [
+ "%matplotlib inline\n",
+ "import numpy as np\n",
+ "import torch\n",
+ "from torch.utils.serialization import load_lua\n",
+ "import os\n",
+ "import scipy.io as sio\n",
+ "import cv2\n",
+ "import math\n",
+ "from matplotlib import pyplot as plt\n",
+ "\n",
+ "import sqlite3"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 3,
+ "metadata": {
+ "collapsed": true
+ },
+ "outputs": [],
+ "source": [
+ "#Change this paths according to your directories\n",
+ "images_path = \"/Data/nruiz9/data/facial_landmarks/AFLW/aflw/data/flickr/\"\n",
+ "storing_path = \"/Data/nruiz9/data/facial_landmarks/AFLW/aflw_cropped_loose/\""
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 20,
+ "metadata": {
+ "collapsed": false
+ },
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "(573, 1)\n",
+ "(232, 1)\n",
+ "(165, 1)\n",
+ "3\n"
+ ]
+ }
+ ],
+ "source": [
+ "#Load KEPLER split file\n",
+ "test_set = sio.loadmat('/Data/nruiz9/data/facial_landmarks/AFLW/testset.mat')\n",
+ "print test_set['test'][0][0][0].shape\n",
+ "print test_set['test'][0][0][1].shape\n",
+ "print test_set['test'][0][0][2].shape\n",
+ "print len(test_set['test'][0][0])\n",
+ "\n",
+ "# I will just use every single one for now. If results are not good I'll sample equally."
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 3,
+ "metadata": {
+ "collapsed": false
+ },
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Done\n"
+ ]
+ }
+ ],
+ "source": [
+ "#Image counter\n",
+ "counter = 1\n",
+ "\n",
+ "#Open the sqlite database\n",
+ "conn = sqlite3.connect('/Data/nruiz9/data/facial_landmarks/AFLW/aflw/data/aflw.sqlite')\n",
+ "c = conn.cursor()\n",
+ "\n",
+ "#Creating the query string for retriving: roll, pitch, yaw and faces position\n",
+ "#Change it according to what you want to retrieve\n",
+ "select_string = \"faceimages.filepath, faces.face_id, facepose.roll, facepose.pitch, facepose.yaw, facerect.x, facerect.y, facerect.w, facerect.h\"\n",
+ "from_string = \"faceimages, faces, facepose, facerect\"\n",
+ "where_string = \"faces.face_id = facepose.face_id and faces.file_id = faceimages.file_id and faces.face_id = facerect.face_id\"\n",
+ "query_string = \"SELECT \" + select_string + \" FROM \" + from_string + \" WHERE \" + where_string\n",
+ "\n",
+ "#It iterates through the rows returned from the query\n",
+ "for row in c.execute(query_string):\n",
+ "\n",
+ " #Using our specific query_string, the \"row\" variable will contain:\n",
+ " # row[0] = image path\n",
+ " # row[1] = face id\n",
+ " # row[2] = rollgma\n",
+ " # row[3] = pitch\n",
+ " # row[4] = yaw\n",
+ " # row[5] = face coord x\n",
+ " # row[6] = face coord y\n",
+ " # row[7] = face width\n",
+ " # row[8] = face heigh\n",
+ "\n",
+ " #Creating the full path names for input and output\n",
+ " input_path = images_path + str(row[0])\n",
+ " output_path = storing_path + str(row[0])\n",
+ "\n",
+ " #If the file exist then open it \n",
+ " if(os.path.isfile(input_path) == True):\n",
+ " #image = cv2.imread(input_path, 0) #load in grayscale\n",
+ " image = cv2.imread(input_path)\n",
+ "\n",
+ " #Image dimensions\n",
+ " image_h, image_w, _ = image.shape\n",
+ " #Roll, pitch and yaw\n",
+ " roll = row[2]\n",
+ " pitch = row[3]\n",
+ " yaw = row[4]\n",
+ " #Face rectangle coords\n",
+ " face_x = row[5]\n",
+ " face_y = row[6]\n",
+ " face_w = row[7]\n",
+ " face_h = row[8]\n",
+ "\n",
+ " #Error correction\n",
+ " k = 0.35\n",
+ " x_min = face_x - face_w * k * 0.6\n",
+ " x_max = face_x + face_w + face_w * k * 0.6\n",
+ " y_min = face_y - face_h * k * 2\n",
+ " y_max = face_y + face_h + face_h * k * 0.6\n",
+ " \n",
+ " x_min = int(max(0, x_min))\n",
+ " x_max = int(min(image_w, x_max))\n",
+ " y_min = int(max(0, y_min))\n",
+ " y_max = int(min(image_h, y_max))\n",
+ " \n",
+ " #Crop the face from the image\n",
+ " image_cropped = np.copy(image[y_min:y_max, x_min:x_max])\n",
+ " #Uncomment the lines below if you want to rescale the image to a particular size\n",
+ " to_size = 240\n",
+ " image_cropped = cv2.resize(image_cropped, (to_size,to_size), interpolation = cv2.INTER_AREA)\n",
+ " #Uncomment the line below if you want to use adaptive histogram normalisation\n",
+ " #clahe = cv2.createCLAHE(clipLimit=2.0, tileGridSize=(5,5))\n",
+ " #image_normalised = clahe.apply(image_rescaled)\n",
+ " #Save the image\n",
+ " #change \"image_cropped\" with the last uncommented variable name above\n",
+ " cv2.imwrite(output_path, image_cropped)\n",
+ " \n",
+ " txt_path = os.path.splitext(output_path)[0]+'.txt'\n",
+ " directory = os.path.dirname(output_path)\n",
+ " if not os.path.exists(directory):\n",
+ " os.makedirs(directory)\n",
+ " fid = open(txt_path, 'wb')\n",
+ " fid.write(input_path + ' ' + str(yaw) + ' ' + str(pitch) + ' ' + str(roll) + ' ' + str(face_x) + ' ' + str(face_y) + ' ' + str(face_w) + ' ' + str(face_h) + '\\n')\n",
+ " fid.close()\n",
+ " #Increasing the counter\n",
+ " counter = counter + 1 \n",
+ "\n",
+ " #if the file does not exits it return an exception\n",
+ " else:\n",
+ " raise ValueError('Error: I cannot find the file specified: ' + str(input_path))\n",
+ "\n",
+ "#Once finished the iteration it closes the database\n",
+ "c.close()\n",
+ "print 'Done'"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 4,
+ "metadata": {
+ "collapsed": false
+ },
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "test\n"
+ ]
+ }
+ ],
+ "source": [
+ "print 'test'"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {
+ "collapsed": true
+ },
+ "outputs": [],
+ "source": []
+ }
+ ],
+ "metadata": {
+ "anaconda-cloud": {},
+ "kernelspec": {
+ "display_name": "Python [conda root]",
+ "language": "python",
+ "name": "conda-root-py"
+ },
+ "language_info": {
+ "codemirror_mode": {
+ "name": "ipython",
+ "version": 2
+ },
+ "file_extension": ".py",
+ "mimetype": "text/x-python",
+ "name": "python",
+ "nbconvert_exporter": "python",
+ "pygments_lexer": "ipython2",
+ "version": "2.7.12"
+ }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 1
+}
diff --git a/practice/.ipynb_checkpoints/remove_KEPLER_test_split-checkpoint.ipynb b/practice/.ipynb_checkpoints/remove_KEPLER_test_split-checkpoint.ipynb
new file mode 100644
index 0000000..0628c13
--- /dev/null
+++ b/practice/.ipynb_checkpoints/remove_KEPLER_test_split-checkpoint.ipynb
@@ -0,0 +1,128 @@
+{
+ "cells": [
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {
+ "collapsed": true
+ },
+ "outputs": [],
+ "source": []
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 7,
+ "metadata": {
+ "collapsed": true
+ },
+ "outputs": [],
+ "source": [
+ "AFLW2000 = '/Data/nruiz9/data/facial_landmarks/AFLW2000/'\n",
+ "AFLW = '/Data/nruiz9/data/facial_landmarks/AFLW/aflw_cropped/'"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 8,
+ "metadata": {
+ "collapsed": true
+ },
+ "outputs": [],
+ "source": [
+ "list_2000 = '/Data/nruiz9/data/facial_landmarks/AFLW2000/filename_list_filtered.txt'\n",
+ "list_normal = '/Data/nruiz9/data/facial_landmarks/AFLW/aflw_cropped/filename_list.txt'"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 9,
+ "metadata": {
+ "collapsed": false
+ },
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "1969\n"
+ ]
+ }
+ ],
+ "source": [
+ "fid = open(list_2000, 'r')\n",
+ "dict_2000 = dict()\n",
+ "for line in fid:\n",
+ " line = line.strip('\\n').split('/')\n",
+ " dict_2000[line[-1]] = 1\n",
+ "print len(dict_2000)\n",
+ "fid.close()"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 10,
+ "metadata": {
+ "collapsed": false
+ },
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "1966 19109\n"
+ ]
+ }
+ ],
+ "source": [
+ "fid = open(list_normal, 'r')\n",
+ "naked = list_normal.strip('.txt')\n",
+ "train = open(naked + '_train.txt', 'wb')\n",
+ "test = open(naked + '_test.txt', 'wb')\n",
+ "test_dict = dict()\n",
+ "train_dict = dict()\n",
+ "for line in fid:\n",
+ " line = line.strip('\\n')\n",
+ " name = line.split('/')[-1]\n",
+ " if name in dict_2000.keys():\n",
+ " test.write(line + '\\n')\n",
+ " test_dict[line] = 1\n",
+ " else:\n",
+ " train.write(line + '\\n')\n",
+ " train_dict[line] = 1\n",
+ "\n",
+ "print len(test_dict), len(train_dict)"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {
+ "collapsed": true
+ },
+ "outputs": [],
+ "source": []
+ }
+ ],
+ "metadata": {
+ "anaconda-cloud": {},
+ "kernelspec": {
+ "display_name": "Python [conda root]",
+ "language": "python",
+ "name": "conda-root-py"
+ },
+ "language_info": {
+ "codemirror_mode": {
+ "name": "ipython",
+ "version": 2
+ },
+ "file_extension": ".py",
+ "mimetype": "text/x-python",
+ "name": "python",
+ "nbconvert_exporter": "python",
+ "pygments_lexer": "ipython2",
+ "version": "2.7.12"
+ }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 1
+}
diff --git a/practice/create_filtered_datasets.ipynb b/practice/create_filtered_datasets.ipynb
index 7e1fe4d..e5bbf42 100644
--- a/practice/create_filtered_datasets.ipynb
+++ b/practice/create_filtered_datasets.ipynb
@@ -96,19 +96,19 @@
},
{
"cell_type": "code",
- "execution_count": 2,
+ "execution_count": 4,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
- "AFLW = '/Data/nruiz9/data/facial_landmarks/AFLW/aflw_cropped/'\n",
- "filenames = '/Data/nruiz9/data/facial_landmarks/AFLW/aflw_cropped/filename_list_train.txt'"
+ "AFLW = '/Data/nruiz9/data/facial_landmarks/AFLW/aflw_cropped_loose/'\n",
+ "filenames = '/Data/nruiz9/data/facial_landmarks/AFLW/aflw_cropped_loose/filename_list.txt'"
]
},
{
"cell_type": "code",
- "execution_count": 11,
+ "execution_count": 5,
"metadata": {
"collapsed": false
},
@@ -117,13 +117,13 @@
"name": "stdout",
"output_type": "stream",
"text": [
- "246\n"
+ "243\n"
]
}
],
"source": [
"fid = open(filenames, 'r')\n",
- "out = open(os.path.join(AFLW, 'filename_list_train_filtered.txt'), 'wb')\n",
+ "out = open(os.path.join(AFLW, 'filename_list_filtered.txt'), 'wb')\n",
"counter = 0\n",
"for line in fid:\n",
" original_line = line\n",
diff --git a/practice/load_AFLW-Copy1.ipynb b/practice/load_AFLW-Copy1.ipynb
index 2b6bf67..856fbb4 100644
--- a/practice/load_AFLW-Copy1.ipynb
+++ b/practice/load_AFLW-Copy1.ipynb
@@ -36,7 +36,7 @@
},
{
"cell_type": "code",
- "execution_count": 3,
+ "execution_count": 4,
"metadata": {
"collapsed": false
},
@@ -144,7 +144,7 @@
},
{
"cell_type": "code",
- "execution_count": 4,
+ "execution_count": 5,
"metadata": {
"collapsed": false
},
diff --git a/practice/load_AFLW-KEPLER_split.ipynb b/practice/load_AFLW-KEPLER_split.ipynb
new file mode 100644
index 0000000..fa8ab5f
--- /dev/null
+++ b/practice/load_AFLW-KEPLER_split.ipynb
@@ -0,0 +1,190 @@
+{
+ "cells": [
+ {
+ "cell_type": "code",
+ "execution_count": 30,
+ "metadata": {
+ "collapsed": true
+ },
+ "outputs": [],
+ "source": [
+ "%matplotlib inline\n",
+ "import numpy as np\n",
+ "import torch\n",
+ "from torch.utils.serialization import load_lua\n",
+ "import os\n",
+ "import scipy.io as sio\n",
+ "import cv2\n",
+ "import math\n",
+ "from matplotlib import pyplot as plt\n",
+ "\n",
+ "import sqlite3"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 31,
+ "metadata": {
+ "collapsed": true
+ },
+ "outputs": [],
+ "source": [
+ "#Change this paths according to your directories\n",
+ "images_path = \"/Data/nruiz9/data/facial_landmarks/AFLW/aflw/data/flickr/\"\n",
+ "storing_path = \"/Data/nruiz9/data/facial_landmarks/AFLW/aflw_cropped_loose_test/\""
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 32,
+ "metadata": {
+ "collapsed": false
+ },
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "(573, 1)\n",
+ "(232, 1)\n",
+ "(165, 1)\n",
+ "3\n",
+ "(970, 1)\n"
+ ]
+ }
+ ],
+ "source": [
+ "#Load KEPLER split file\n",
+ "test_set = sio.loadmat('/Data/nruiz9/data/facial_landmarks/AFLW/testset.mat')\n",
+ "print test_set['test'][0][0][0].shape\n",
+ "print test_set['test'][0][0][1].shape\n",
+ "print test_set['test'][0][0][2].shape\n",
+ "print len(test_set['test'][0][0])\n",
+ "\n",
+ "id_test_set = np.concatenate([test_set['test'][0][0][0], test_set['test'][0][0][1], test_set['test'][0][0][2]])\n",
+ "print id_test_set.shape\n",
+ "\n",
+ "# I will just use every single one for now. If results are not good I'll sample equally."
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 38,
+ "metadata": {
+ "collapsed": false
+ },
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "970\n",
+ "Done\n"
+ ]
+ }
+ ],
+ "source": [
+ "#Image counter\n",
+ "counter = 0\n",
+ "\n",
+ "#Open the sqlite database\n",
+ "conn = sqlite3.connect('/Data/nruiz9/data/facial_landmarks/AFLW/aflw/data/aflw.sqlite')\n",
+ "c = conn.cursor()\n",
+ "\n",
+ "#Creating the query string for retriving: roll, pitch, yaw and faces position\n",
+ "#Change it according to what you want to retrieve\n",
+ "select_string = \"faceimages.filepath, faces.face_id, facepose.roll, facepose.pitch, facepose.yaw, facerect.x, facerect.y, facerect.w, facerect.h\"\n",
+ "from_string = \"faceimages, faces, facepose, facerect\"\n",
+ "where_string = \"faces.face_id = facepose.face_id and faces.file_id = faceimages.file_id and faces.face_id = facerect.face_id\"\n",
+ "query_string = \"SELECT \" + select_string + \" FROM \" + from_string + \" WHERE \" + where_string\n",
+ "\n",
+ "test_file_txt = '/Data/nruiz9/data/facial_landmarks/AFLW/KEPLER_test_split.txt'\n",
+ "out_txt = open(test_file_txt, 'w')\n",
+ "#It iterates through the rows returned from the query\n",
+ "for row in c.execute(query_string):\n",
+ "\n",
+ " #Using our specific query_string, the \"row\" variable will contain:\n",
+ " # row[0] = image path\n",
+ " # row[1] = face id\n",
+ " # row[2] = rollgma\n",
+ " # row[3] = pitch\n",
+ " # row[4] = yaw\n",
+ " # row[5] = face coord x\n",
+ " # row[6] = face coord y\n",
+ " # row[7] = face width\n",
+ " # row[8] = face heigh\n",
+ " if row[1] in id_test_set:\n",
+ " #Creating the full path names for input and output\n",
+ " input_path = images_path + str(row[0])\n",
+ " output_path = storing_path + str(row[0])\n",
+ "\n",
+ " #If the file exist then open it \n",
+ " if(os.path.isfile(input_path) == True):\n",
+ "\n",
+ " out_txt.write(str(row[0]) + '.jpg\\n')\n",
+ " #Increasing the counter\n",
+ " counter = counter + 1 \n",
+ "\n",
+ " #if the file does not exits it return an exception\n",
+ " else:\n",
+ " raise ValueError('Error: I cannot find the file specified: ' + str(input_path))\n",
+ "\n",
+ "#Once finished the iteration it closes the database\n",
+ "print counter\n",
+ "out_txt.close()\n",
+ "c.close()\n",
+ "print 'Done'"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 4,
+ "metadata": {
+ "collapsed": false
+ },
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "test\n"
+ ]
+ }
+ ],
+ "source": [
+ "print 'test'"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {
+ "collapsed": true
+ },
+ "outputs": [],
+ "source": []
+ }
+ ],
+ "metadata": {
+ "anaconda-cloud": {},
+ "kernelspec": {
+ "display_name": "Python [conda root]",
+ "language": "python",
+ "name": "conda-root-py"
+ },
+ "language_info": {
+ "codemirror_mode": {
+ "name": "ipython",
+ "version": 2
+ },
+ "file_extension": ".py",
+ "mimetype": "text/x-python",
+ "name": "python",
+ "nbconvert_exporter": "python",
+ "pygments_lexer": "ipython2",
+ "version": "2.7.12"
+ }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 1
+}
diff --git a/practice/remove_KEPLER_test_split.ipynb b/practice/remove_KEPLER_test_split.ipynb
new file mode 100644
index 0000000..239e56a
--- /dev/null
+++ b/practice/remove_KEPLER_test_split.ipynb
@@ -0,0 +1,128 @@
+{
+ "cells": [
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {
+ "collapsed": true
+ },
+ "outputs": [],
+ "source": []
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 7,
+ "metadata": {
+ "collapsed": true
+ },
+ "outputs": [],
+ "source": [
+ "AFLW2000 = '/Data/nruiz9/data/facial_landmarks/AFLW2000/'\n",
+ "AFLW = '/Data/nruiz9/data/facial_landmarks/AFLW/aflw_cropped_loose/'"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 8,
+ "metadata": {
+ "collapsed": true
+ },
+ "outputs": [],
+ "source": [
+ "list_2000 = '/Data/nruiz9/data/facial_landmarks/AFLW/KEPLER_test_split.txt'\n",
+ "list_normal = '/Data/nruiz9/data/facial_landmarks/AFLW/aflw_cropped_loose/filename_list_filtered.txt'"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 10,
+ "metadata": {
+ "collapsed": false
+ },
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "961\n"
+ ]
+ }
+ ],
+ "source": [
+ "fid = open(list_2000, 'r')\n",
+ "dict_2000 = dict()\n",
+ "for line in fid:\n",
+ " line = line.strip('\\n').split('/')\n",
+ " dict_2000[line[-1]] = 1\n",
+ "print len(dict_2000)\n",
+ "fid.close()"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 12,
+ "metadata": {
+ "collapsed": false
+ },
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "954 19842\n"
+ ]
+ }
+ ],
+ "source": [
+ "fid = open(list_normal, 'r')\n",
+ "naked = list_normal.strip('.txt')\n",
+ "train = open(naked + '_train.txt', 'wb')\n",
+ "test = open(naked + '_test.txt', 'wb')\n",
+ "test_dict = dict()\n",
+ "train_dict = dict()\n",
+ "for line in fid:\n",
+ " line = line.strip('\\n')\n",
+ " name = line.split('/')[-1]\n",
+ " if name+'.jpg' in dict_2000.keys():\n",
+ " test.write(line + '\\n')\n",
+ " test_dict[line] = 1\n",
+ " else:\n",
+ " train.write(line + '\\n')\n",
+ " train_dict[line] = 1\n",
+ "\n",
+ "print len(test_dict), len(train_dict)"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {
+ "collapsed": true
+ },
+ "outputs": [],
+ "source": []
+ }
+ ],
+ "metadata": {
+ "anaconda-cloud": {},
+ "kernelspec": {
+ "display_name": "Python [conda root]",
+ "language": "python",
+ "name": "conda-root-py"
+ },
+ "language_info": {
+ "codemirror_mode": {
+ "name": "ipython",
+ "version": 2
+ },
+ "file_extension": ".py",
+ "mimetype": "text/x-python",
+ "name": "python",
+ "nbconvert_exporter": "python",
+ "pygments_lexer": "ipython2",
+ "version": "2.7.12"
+ }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 1
+}
--
Gitblit v1.8.0