mindspore/model_zoo/official/nlp/lstm/README.md

• Contents
• LSTM Description
• Model Architecture
• Dataset
• Environment Requirements
• Quick Start
• Script Description
• • Script and Sample Code
  • Script Parameters
  • • Training Script Parameters
    • Running Options
    • Network Parameters
  • Dataset Preparation
  • Training Process
  • Evaluation Process
• Model Description
• • Performance
  • • Training Performance
    • Evaluation Performance
• Description of Random Situation
• ModelZoo Homepage

Contents

• LSTM Description
• Model Architecture
• Dataset
• Environment Requirements
• Quick Start
• Script Description
  • Script and Sample Code
  • Script Parameters
  • Dataset Preparation
  • Training Process
  • Evaluation Process
• Model Description
  • Performance
    • Training Performance
    • Evaluation Performance
• Description of Random Situation
• ModelZoo Homepage

LSTM Description

This example is for LSTM model training and evaluation.

Paper: Andrew L. Maas, Raymond E. Daly, Peter T. Pham, Dan Huang, Andrew Y. Ng, Christopher Potts. Learning Word Vectors for Sentiment Analysis. Proceedings of the 49th Annual Meeting of the Association for Computational Linguistics: Human Language Technologies. 2011

Model Architecture

LSTM contains embeding, encoder and decoder modules. Encoder module consists of LSTM layer. Decoder module consists of fully-connection layer.

Dataset

• aclImdb_v1 for training evaluation.Large Movie Review Dataset
• GloVe: Vector representations for words.GloVe: Global Vectors for Word Representation

Environment Requirements

• Hardware（GPU/CPU）
• Framework
  • MindSpore
• For more information, please check the resources below：
  • MindSpore tutorials
  • MindSpore API

Quick Start

• runing on GPU

  # run training example
  bash run_train_gpu.sh 0 ./aclimdb ./glove_dir
  
  # run evaluation example
  bash run_eval_gpu.sh 0 ./aclimdb ./glove_dir lstm-20_390.ckpt
  

• runing on CPU

  # run training example
  bash run_train_cpu.sh ./aclimdb ./glove_dir
  
  # run evaluation example
  bash run_eval_cpu.sh ./aclimdb ./glove_dir lstm-20_390.ckpt
  

Script Description

Script and Sample Code

.
├── lstm
    ├── README.md               # descriptions about LSTM
    ├── script
    │   ├── run_eval_gpu.sh     # shell script for evaluation on GPU
    │   ├── run_eval_cpu.sh     # shell script for evaluation on CPU
    │   ├── run_train_gpu.sh    # shell script for training on GPU
    │   └── run_train_cpu.sh    # shell script for training on CPU
    ├── src
    │   ├── config.py           # parameter configuration
    │   ├── dataset.py          # dataset preprocess
    │   ├── imdb.py             # imdb dataset read script
    │   └── lstm.py             # Sentiment model
    ├── eval.py                 # evaluation script on both GPU and CPU
    └── train.py                # training script on both GPU and CPU

Script Parameters

Training Script Parameters

usage: train.py  [-h] [--preprocess {true, false}] [--aclimdb_path ACLIMDB_PATH]
                 [--glove_path GLOVE_PATH] [--preprocess_path PREPROCESS_PATH]
                 [--ckpt_path CKPT_PATH] [--pre_trained PRE_TRAINING]
                 [--device_target {GPU, CPU}]

Mindspore LSTM Example

options:
  -h, --help                          # show this help message and exit
  --preprocess {true, false}          # whether to preprocess data.
  --aclimdb_path ACLIMDB_PATH         # path where the dataset is stored.
  --glove_path GLOVE_PATH             # path where the GloVe is stored.
  --preprocess_path PREPROCESS_PATH   # path where the pre-process data is stored.
  --ckpt_path CKPT_PATH               # the path to save the checkpoint file.
  --pre_trained                       # the pretrained checkpoint file path.
  --device_target                     # the target device to run, support "GPU", "CPU". Default: "GPU".

Running Options

config.py:
    num_classes                   # classes num
    learning_rate                 # value of learning rate
    momentum                      # value of momentum
    num_epochs                    # epoch size
    batch_size                    # batch size of input dataset
    embed_size                    # the size of each embedding vector
    num_hiddens                   # number of features of hidden layer
    num_layers                    # number of layers of stacked LSTM
    bidirectional                 # specifies whether it is a bidirectional LSTM
    save_checkpoint_steps         # steps for saving checkpoint files

Network Parameters

Dataset Preparation

• Download the dataset aclImdb_v1.

Unzip the aclImdb_v1 dataset to any path you want and the folder structure should be as follows:

.
├── train  # train dataset
└── test   # infer dataset

• Download the GloVe file.

Unzip the glove.6B.zip to any path you want and the folder structure should be as follows:

.
├── glove.6B.100d.txt
├── glove.6B.200d.txt
├── glove.6B.300d.txt    # we will use this one later.
└── glove.6B.50d.txt

Adding a new line at the beginning of the file which named glove.6B.300d.txt.
It means reading a total of 400,000 words, each represented by a 300-latitude word vector.

400000    300

Training Process

• Set options in config.py, including learning rate and network hyperparameters.

• runing on GPU

  Run sh run_train_gpu.sh for training.

  bash run_train_gpu.sh 0 ./aclimdb ./glove_dir
  

  The above shell script will run distribute training in the background. You will get the loss value as following:

  # grep "loss is " log.txt
  epoch: 1 step: 390, loss is 0.6003723
  epcoh: 2 step: 390, loss is 0.35312173
  ...
  

• runing on CPU

  Run sh run_train_cpu.sh for training.

  bash run_train_cpu.sh ./aclimdb ./glove_dir
  

  The above shell script will train in the background. You will get the loss value as following:

  # grep "loss is " log.txt
  epoch: 1 step: 390, loss is 0.6003723
  epcoh: 2 step: 390, loss is 0.35312173
  ...
  

Evaluation Process

• evaluation on GPU

  Run bash run_eval_gpu.sh for evaluation.

  bash run_eval_gpu.sh 0 ./aclimdb ./glove_dir lstm-20_390.ckpt
  

• evaluation on CPU

  Run bash run_eval_cpu.sh for evaluation.

  bash run_eval_cpu.sh ./aclimdb ./glove_dir lstm-20_390.ckpt
  

Model Description

Performance

Training Performance

Parameters	LSTM (GPU)	LSTM (CPU)
Resource	Tesla V100-SMX2-16GB	Ubuntu X86-i7-8565U-16GB
uploaded Date	08/06/2020 (month/day/year)	08/06/2020 (month/day/year)
MindSpore Version	0.6.0-beta	0.6.0-beta
Dataset	aclimdb_v1	aclimdb_v1
Training Parameters	epoch=20, batch_size=64	epoch=20, batch_size=64
Optimizer	Momentum	Momentum
Loss Function	Softmax Cross Entropy	Softmax Cross Entropy
Speed	1022 (1pcs)	20
Loss	0.12	0.12
Params (M)	6.45	6.45
Checkpoint for inference	292.9M (.ckpt file)	292.9M (.ckpt file)
Scripts	lstm script	lstm script

Evaluation Performance

Parameters	LSTM (GPU)	LSTM (CPU)
Resource	Tesla V100-SMX2-16GB	Ubuntu X86-i7-8565U-16GB
uploaded Date	08/06/2020 (month/day/year)	08/06/2020 (month/day/year)
MindSpore Version	0.6.0-beta	0.6.0-beta
Dataset	aclimdb_v1	aclimdb_v1
batch_size	64	64
Accuracy	84%	83%

Description of Random Situation

There are three random situations:

• Shuffle of the dataset.
• Initialization of some model weights.

ModelZoo Homepage

Please check the official homepage.