圖像分類數據庫

This article is a tutorial on how to use the Monk library to classify house room types like the living room, dining room, etc.

本文是有關如何使用Monk庫對房屋房間類型(例如客廳，飯廳等)進行分類的教程。

A detailed tutorial is available on GitHub.

GitHub上有詳細的教程。

關于數據集 (About the dataset)

• The dataset contains a total of 145436 images divided into seven classes namely, ‘Exterior’, ‘bedroom’, ‘kitchen’, ‘living_room’, ‘Interior’, ‘bathroom’, ‘dining_room’.
  數據集包含總共145436張圖像，分為7類，即“外部”，“臥室”，“廚房”，“客廳”，“內部”，“浴室”，“餐廳”。

• It can be downloaded from here.

  可以從這里下載。

關于僧侶圖書館和僧侶如何使圖像分類容易 (About Monk Library and How does Monk make image classification easy)

• Write less code and create an end to end applications.
  編寫更少的代碼并創建端到端應用程序。
• Learn only one syntax and create applications using any deep learning library — PyTorch, Mxnet, Keras, TensorFlow, etc.
  使用任何深度學習庫(PyTorch，Mxnet，Keras，TensorFlow等)僅學習一種語法并創建應用程序。
• Manage your entire project easily with multiple experiments.
  通過多個實驗輕松管理整個項目。

該庫是為誰而建的 (For whom this library is built)

• Students: Seamlessly learn computer vision using our comprehensive study road-maps.
  學生：使用我們全面的學習路線圖無縫學習計算機視覺。
• Researchers and Developers: Create and Manage multiple deep learning projects.
  研究人員和開發人員：創建和管理多個深度學習項目。
• Competition participants on Kaggle, Codalab, HackerEarth, AiCrowd, etc.
  Kaggle，Codalab，HackerEarth，AiCrowd等競賽參與者。

目錄 (Table of contents)

Install Monk

安裝和尚

2. Using the pre-trained model for part of house classification dataset

2.將預訓練模型用于房屋分類數據集的一部分

3. Training a classifier from scratch

3.從頭開始訓練分類器

• Train a part of house classifier using ResNet variants

  使用ResNet變體訓練房屋分類器的一部分

• Understand what all differences happen when switching between ResNets variants.

  了解在ResNets變體之間切換時發生的所有差異。

• Understand a bigger and deeper network not always means better results

  了解更大更深的網絡并不總是意味著更好的結果

• For this experiment, you will be using mxnet backend

  對于本實驗，您將使用mxnet后端

4. Conclusions.

4。結論。

1.安裝和尚 (1. Installing Monk)

#using cuda10.0 download monk using following commandpip install -U monk-gluon-cuda100

For more ways to install visit Monk Library.

有關更多安裝方式的信息，請訪問Monk Library 。

2.使用預訓練的模型進行演示 (2. Using a pre-trained model for the demo)

First, download the pre-trained models from google drive.

首先，從Google Drive下載預訓練的模型。

! wget --load-cookies /tmp/cookies.txt "https://docs.google.com/uc?export=download&confirm=$(wget --save-cookies /tmp/cookies.txt --keep-session-cookies --no-check-certificate 'https://docs.google.com/uc?export=download&id=10SrowcOJp8GWqEB21BfCIinqUCHS7PMv' -O- | sed -rn 's/.*confirm=([0-9A-Za-z_]+).*/\1\n/p')&id=10SrowcOJp8GWqEB21BfCIinqUCHS7PMv" -O cls_house_scene_trained.zip && rm -rf /tmp/cookies.txt

The above will download the zip file of pre-trained models and name them as cls_house_scene_trained.zip. Now we have to unzip the file.

上面的代碼將下載預訓練模型的zip文件，并將其命名為cls_house_scene_trained.zip。 現在我們必須解壓縮文件。

! unzip -qq cls_house_scene_trained.zip

Importing monk.gluon_prototype for working with the monk library.

導入monk.gluon_prototype 與和尚圖書館一起工作。

#Using mxnet-gluon backend# When installed using pipfrom monk.gluon_prototype import prototype

Let’s load the model in infer mode to classify the demo data.

讓我們以推斷模式加載模型以對演示數據進行分類。

# Load project in inference modegtf = prototype(verbose=1);
gtf.Prototype("Task", "gluon_resnet18_v1_train_all_layers", eval_infer=True);

推斷一些樣品 (Infer on some samples)

img_name = "workspace/test/2.jpg"
predictions = gtf.Infer(img_name=img_name);
from IPython.display import Image
Image(filename=img_name, height=300, width=300)img_name = "workspace/test/3.jpg"
predictions = gtf.Infer(img_name=img_name);
from IPython.display import Image
Image(filename=img_name, height=300, width=300)img_name = "workspace/test/6.jpg"
predictions = gtf.Infer(img_name=img_name);
from IPython.display import Image
Image(filename=img_name, height=300, width=300)

For more examples visit the notebook.

有關更多示例，請訪問筆記本 。

3. 從頭開始訓練自定義分類器 (3. Training custom classifier from scratch)

什么是ResNet？ (What is ResNet?)

Points from https://towardsdatascience.com/an-overview-of-resnet-and-its-variants-5281e2f56035

來自https://towardsdatascience.com/an-overview-of-resnet-and-its-variants-5281e2f56035的點

- The core idea of ResNet is introducing a so-called “identity shortcut connection” that skips one or more layers
- The deeper model should not produce a training error higher than its shallower counterparts.
- solves the problem of vanishing gradients as network depth increasedResNet18 ArchitectureResNet18體系結構

即將到來的內容 (Upcoming contents)

Load the Dataset

加載數據集

Train model using resnet18_v1(training only last layer) architecture for transfer learning.

使用resnet18_v1(僅訓練最后一層)體系結構訓練模型以進行遷移學習。

Train model using resnet18_v1(training all the layers) architecture for transfer learning.

使用resnet18_v1(訓練所有層)體系結構訓練模型以進行遷移學習。

Train model using resnet18_v2(training only last layer) architecture for transfer learning.

使用resnet18_v2(僅訓練最后一層)體系結構訓練模型以進行遷移學習。

Train model using resnet34_v1(training only last layer) architecture for transfer learning.

使用resnet34_v1(僅訓練最后一層)體系結構訓練模型以進行遷移學習。

Compare all the models.

比較所有模型。

1.加載數據集 (1. Loading the Dataset)

Dataset Credits: https://omidpoursaeed.github.io/publication/vision-based-real-estate-price-estimation/

數據集積分： https : //omidpoursaeed.github.io/publication/vision-based-real-estate-price-estimation/

Step-1: Downloading the dataset.

步驟1：下載數據集。

! wget --load-cookies /tmp/cookies.txt "https://docs.google.com/uc?export=download&confirm=$(wget --save-cookies /tmp/cookies.txt --keep-session-cookies --no-check-certificate 'https://docs.google.com/uc?export=download&id=0BxDIywue_VABY1dRcFVvZ3BodnM' -O- | sed -rn 's/.*confirm=([0-9A-Za-z_]+).*/\1\n/p')&id=0BxDIywue_VABY1dRcFVvZ3BodnM" -O dataset.zip && rm -rf /tmp/cookies.txt

Step-2: Unzipping the dataset.

步驟2：解壓縮數據集。

! unzip -qq dataset.zip

2.使用resnet18_v1(僅訓練最后一層)體系結構訓練模型以進行遷移學習。 (2. Train model using resnet18_v1(training only last layer) architecture for transfer learning.)

Step-1: Load experiment and insert data. For more details visit.

步驟1：加載實驗并插入數據。 有關更多詳細信息， 請訪問 。

# Load experimentmonk_gln = prototype(verbose=1);
monk_gln.Prototype("Task", "gluon_resnet18_v1");# Insert data and set params in default modemonk_gln.Default(dataset_path="Train",
model_name="resnet18_v1",
freeze_base_network=True,
num_epochs=10);

Step-2: Training the model.

步驟2：訓練模型。

#Start Trainingmonk_gln.Train();

3.使用resnet18_v1(訓練所有層)體系結構訓練模型以進行遷移學習。 (3. Train model using resnet18_v1(training all the layers) architecture for transfer learning.)

Step-1: Load experiment and insert data. For more details visit. Parameter freeze_base_network will be set to false to train all the layers.

步驟1：加載實驗并插入數據。 有關更多詳細信息， 請訪問 。 參數freeze_base_network 將設置為false以訓練所有圖層。

# Load experimentmonk_gln = prototype(verbose=1);
monk_gln.Prototype("Task", "gluon_resnet18_v1_train_all_layers");# Insert data and set params in default modemonk_gln.Default(dataset_path="Train",
model_name="resnet18_v1",
freeze_base_network=False,
num_epochs=10);

Step-2: Training the model.

步驟2：訓練模型。

#Start Trainingmonk_gln.Train();

4.使用resnet18_v2(僅訓練最后一層)體系結構訓練模型以進行遷移學習。 (4. Train model using resnet18_v2(training only last layer) architecture for transfer learning.)

Step-1: Load experiment and insert data. For more details visit.

步驟1：加載實驗并插入數據。 有關更多詳細信息， 請訪問 。

# Load experiment
monk_gln = prototype(verbose=1);
monk_gln.Prototype("Task", "gluon-resnet18_v2");# Insert data and set params in default modemonk_gln.Default(dataset_path="Train",
model_name="resnet18_v2",
freeze_base_network=True,
num_epochs=10);

Step-2: Training the model.

步驟2：訓練模型。

#Start Trainingmonk_gln.Train()

5.使用resnet34_v1(僅訓練最后一層) 體系結構來訓練模型以 進行遷移學習。 (5. Train model using resnet34_v1(training only last layer) architecture for transfer learning.)

Step-1: Load experiment and insert data. For more details visit.

步驟1：加載實驗并插入數據。 有關更多詳細信息， 請訪問 。

# Load experiment
monk_gln = prototype(verbose=1);
monk_gln.Prototype("Task", "gluon-resnet18_v2");# Insert data and set params in default modemonk_gln.Default(dataset_path="Train",
model_name="resnet34_v1",
freeze_base_network=True,
num_epochs=10);

Step-2: Training the model.

步驟2：訓練模型。

#Start Trainingmonk_gln.Train()

6.比較所有模型。 (6. Comparing all the models.)

Step-1: Using comparison class of Monk Library. For more details visit.

步驟1：使用僧侶庫的比較類。 有關更多詳細信息， 請訪問 。

# Invoke the comparison classfrom monk.compare_prototype import compare

Step-2: Creating and managing comparison experiments. Provide the project name.

步驟2：創建和管理比較實驗。 提供項目名稱。

gtf = compare(verbose=1);
gtf.Comparison("Comparison-1");

This creates files and directories as per the following structure

這將按照以下結構創建文件和目錄

workspace
|
|--------comparison
|
|
|-----Compare-1
|
|------stats_best_val_acc.png
|------stats_max_gpu_usage.png
|------stats_training_time.png
|------train_accuracy.png
|------train_loss.png
|------val_accuracy.png
|------val_loss.png
|-----comparison.csv (Contains necessary details of all experiments)

Step-3: Adding the experiments to the comparison object.

步驟3：將實驗添加到比較對象。

• The first argument - Project name.
  第一個參數-項目名稱。
• The second argument - Experiment name.
  第二個參數-實驗名稱。

#add the models trained above.
gtf.Add_Experiment("Task", "gluon_resnet18_v1");
gtf.Add_Experiment("Task", "gluon_resnet18_v1_train_all_layers");
gtf.Add_Experiment("Task", "gluon-resnet18_v2");
gtf.Add_Experiment("Task", "gluon-resnet34_v1");

Step-4: Run analysis. This step generates seven images.

步驟4：運行分析。 此步驟將生成七個圖像。

gtf.Generate_Statistics();

Step-5: Visualize the study comparison metrics.

步驟5：可視化研究比較指標。

• Training Accuracy curve
  訓練精度曲線

from IPython.display import Image
Image(filename="workspace/comparison/Comparison-1/train_accuracy.png")

• Training loss curve
  訓練損失曲線

Image(filename="workspace/comparison/Comparison-1/train_loss.png")

• Validation accuracy curve
  驗證精度曲線

Image(filename="workspace/comparison/Comparison-1/val_accuracy.png")

• Validation loss curve
  驗證損失曲線

Image(filename="workspace/comparison/Comparison-1/val_loss.png")

• Training time is taken
  花費訓練時間

Image(filename="workspace/comparison/Comparison-1/stats_training_time.png")

• Max GPU usage
  最大GPU使用率

Image(filename="workspace/comparison/Comparison-1/stats_max_gpu_usage.png")

• Best validation accuracy
  最佳驗證精度

Image(filename="workspace/comparison/Comparison-1/stats_best_val_acc.png")

結論 (Conclusions)

Monk library makes it very easy for students, researchers, and competitors to create deep learning models and try different hyper-parameter tuning to increase the accuracy of the model.

僧侶庫使學生，研究人員和競爭者很容易創建深度學習模型，并嘗試不同的超參數調整以提高模型的準確性。

The pre-trained model can be downloaded used directly without getting into the model creation part.

可以直接下載使用預訓練的模型，而無需進入模型創建部分。

From the above graphs, we can see that in this dataset model in which all the layers are trainable gives more accuracy than the rest of the models. But this is not always true so till advisable to try both.

從上面的圖表中，我們可以看到，在所有層都可訓練的數據集模型中，該模型比其他模型具有更高的準確性。 但這并非總是如此，因此建議您同時嘗試兩者。

Also, the time taken when all the layers are trainable is around 40% more when compared with the model in which only the last layer is trainable.

而且，與僅最后一層是可訓練的模型相比，所有層都可訓練的時間要多40％左右。

謝謝閱讀。 請提出您的建議。 (Thanks for reading. Please give your suggestions.)

Rohit is a BTech. final year student. He has knowledge of machine learning and deep learning. He is interested to work in the field of AI and ML. Currently, he is working as a computer vision intern at Tessellate Imaging. Connect with him on LinkedIn.

Rohit是一名BTech。 最后一年的學生。 他具有機器學習和深度學習的知識。 他對從事AI和ML領域感興趣。 目前，他在Tessellate Imaging擔任計算機視覺實習生。 在LinkedIn上與他聯系。

翻譯自: https://towardsdatascience.com/image-classifier-house-room-type-classification-using-monk-library-d633795a42ef

圖像分類數據庫

總結

以上是生活随笔為你收集整理的图像分类数据库_图像分类器-使用僧侣库对房屋房间类型进行分类的全部內容，希望文章能夠幫你解決所遇到的問題。

如果覺得生活随笔網站內容還不錯，歡迎將生活随笔推薦給好友。