Multilabel-Classification

Repository containing Keras code for the blog post titled "How to Perform Multi Label Classification using Deep Learning". You can checkout the blog post here.

Using Keras

This section lists out the steps involved in training a Keras model (with TensorFlow backend) for Multi Label Classification.

Method 1: Google Colab

• You can explore this notebook on Colab to directly experiment with training the models.

Method 2: Local Setup

Follow these steps to train and use a model for Multilabel Classification. You can also directly use a sample trained model (mobilenet.h5) without training, which can be downloaded from here (skip to Step 4 in that case).

Step 1: Clone the Repo

git clone https://github.com/thatbrguy/Multilabel-Classification.git
cd Multilabel-Classification

Step 2: Download the Dataset

• Download data.tar.gz and place it in the current directory.
• Extract the dataset using tar -xzvf data.tar.gz
• Move the contents of ./data/keras/ to the current directory by using mv ./data/keras/* ./

Step 3: Train the Model

• Run train.py --model ResNet50 to train the model.
• The --model argument can take one among ResNet50, MobileNet, DenseNet121 or Xception.

Step 4: Inference

• Run predict.py --image PATH_TO_FILE --saved_model PATH_TO_h5 to obtain a prediction once the model is trained.
• PATH_TO_FILE refers to the path of the image.
• PATH_TO_h5 refers to the path of the h5 file.

Using Nanonets

This section lists out the steps involved in training a Nanonets model for Multi Label Classification.

Step 1: Clone the Repo

git clone https://github.com/thatbrguy/Multilabel-Classification.git
cd Multilabel-Classification/nanonets

Step 2: Get your free API Key

Get your free API Key from http://app.nanonets.com/user/api_key

Step 3: Set the API key as an Environment Variable

export NANONETS_API_KEY=YOUR_API_KEY_GOES_HERE

Step 4: Create a New Model

python ./code/create_model.py

_Note: This generates a MODEL_ID that you need for the next step

Step 5: Add Model Id as Environment Variable

export NANONETS_MODEL_ID=YOUR_MODEL_ID

_Note: you will get YOUR_MODEL_ID from the previous step

Step 6: Upload the Training Data

• Download data.tar.gz and place it in the current directory.
• Extract the dataset using tar -xzvf data.tar.gz
• Move the contents of ./data/nanonets/ to the current directory by using mv ./data/nanonets/* ./
• Run python ./code/upload_training.py to upload the data.

Step 7: Train Model

Once the Images have been uploaded, begin training the Model

python ./code/train_model.py

Step 8: Get Model State

The model takes ~2 hours to train. You will get an email once the model is trained. In the meanwhile you check the state of the model

python ./code/model_state.py

Step 9: Make Prediction

Once the model is trained. You can make predictions using the model

python ./code/prediction.py PATH_TO_YOUR_IMAGE.jpg

Sample Usage:

python ./code/prediction.py ./multilabel_data/ImageSets/2_my_caesar_salad_hostedLargeUrl.jpg

References

1. Recipes5k
2. Nanonets