Nigel
/
mindspore
forked from mindspore-Ecosystem/mindspore
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

!18782 Add Textcnn and Googlenet CPU Scripts 

Merge pull request !18782 from wuxuejian/textcnn
This commit is contained in:
i-robot 2021-06-25 08:25:49 +00:00 committed by Gitee
parent 246f96f21b 4b90d90e1f
commit d3a53ff1f6
10 changed files with 272 additions and 10 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

26
model_zoo/official/cv/googlenet/README.md
View File

@ -71,8 +71,8 @@ For FP16 operators, if the input data type is FP32, the backend of MindSpore wil
# [Environment Requirements](#contents)
- Hardware（Ascend/GPU）
- Prepare hardware environment with Ascend or GPU processor.
- Hardware（Ascend/GPU/CPU）
- Prepare hardware environment with Ascend/GPU/CPU processor.
- Framework
- [MindSpore](https://www.mindspore.cn/install/en)
- For more information, please check the resources below：
@ -125,6 +125,16 @@ After installing MindSpore via the official website, you can start training and
sh run_eval_gpu.sh [CHECKPOINT_PATH]
```
- running on CPU
```python
# run training example
nohup python train.py --config_path=cifar10_config_cpu.yaml --dataset_name=cifar10 > train.log 2>&1 &
# run evaluation example
nohup python eval.py --checkpoint_path=[CHECKPOINT_PATH] > eval.log 2>&1 &
```
We use CIFAR-10 dataset by default. Your can also pass `$dataset_type` to the scripts so that select different datasets. For more details, please refer the specify script.
- ModelArts (If you want to run in modelarts, please check the official documentation of [modelarts](https://support.huaweicloud.com/modelarts/), and you can start training as follows)
@ -232,9 +242,11 @@ We use CIFAR-10 dataset by default. Your can also pass `$dataset_type` to the sc
├── scripts
│ ├──run_train.sh // shell script for distributed on Ascend
│ ├──run_train_gpu.sh // shell script for distributed on GPU
│ ├──run_train_cpu.sh // shell script for training on CPU
│ ├──run_eval.sh // shell script for evaluation on Ascend
│ ├──run_infer_310.sh // shell script for 310 inference
│ ├──run_eval_gpu.sh // shell script for evaluation on GPU
│ ├──run_eval_cpu.sh // shell script for evaluation on CPU
├── src
│ ├──dataset.py // creating dataset
│ ├──googlenet.py // googlenet architecture
@ -338,6 +350,16 @@ For more configuration details, please refer the script `config.py`.
After training, you'll get some checkpoint files under the folder `./ckpt_0/` by default.
- running on CPU
```python
nohup python train.py --config_path=cifar10_config_cpu.yaml --dataset_name=cifar10 > train.log 2>&1 &
```
The python command above will run in the background, you can view the results through the file `train.log`.
After training, you'll get some checkpoint files under the folder defined in config.yaml.
### Distributed Training
- running on Ascend

26
model_zoo/official/cv/googlenet/README_CN.md
View File

@ -78,8 +78,8 @@ GoogleNet由多个inception模块串联起来，可以更加深入。 降维的
# 环境要求
- 硬件（Ascend/GPU）
- 使用Ascend或GPU处理器来搭建硬件环境。
- 硬件（Ascend/GPU/CPU）
- 使用Ascend/GPU/CPU处理器来搭建硬件环境。
- 框架
- [MindSpore](https://www.mindspore.cn/install/en)
- 如需查看详情，请参见如下资源：
@ -132,6 +132,16 @@ GoogleNet由多个inception模块串联起来，可以更加深入。 降维的
sh run_eval_gpu.sh [CHECKPOINT_PATH]
```
- CPU处理器环境运行
```python
# 运行训练示例
nohup python train.py --config_path=cifar10_config_cpu.yaml --dataset_name=cifar10 > train.log 2>&1 &
# 运行评估示例
nohup python eval.py --checkpoint_path=[CHECKPOINT_PATH] > eval.log 2>&1 &
```
默认使用CIFAR-10数据集。您也可以将`$dataset_type`传入脚本，以便选择其他数据集。如需查看更多详情，请参考指定脚本。
- 在 ModelArts 进行训练 (如果你想在modelarts上运行，可以参考以下文档 [modelarts](https://support.huaweicloud.com/modelarts/))
@ -239,9 +249,11 @@ GoogleNet由多个inception模块串联起来，可以更加深入。 降维的
├── scripts
│ ├──run_train.sh // 分布式到Ascend的shell脚本
│ ├──run_train_gpu.sh // 分布式到GPU处理器的shell脚本
│ ├──run_train_cpu.sh // CPU处理器训练的shell脚本
│ ├──run_eval.sh // Ascend评估的shell脚本
│ ├──run_infer_310.sh // Ascend推理shell脚本
│ ├──run_eval_gpu.sh // GPU处理器评估的shell脚本
│ ├──run_eval_cpu.sh // CPU处理器评估的shell脚本
├── src
│ ├──dataset.py // 创建数据集
│ ├──googlenet.py // googlenet架构
@ -313,6 +325,16 @@ GoogleNet由多个inception模块串联起来，可以更加深入。 降维的
训练结束后，您可在默认`./ckpt_0/`脚本文件夹下找到检查点文件。
- CPU处理器环境运行
```bash
nohup python train.py --config_path=cifar10_config_cpu.yaml --dataset_name=cifar10 > train.log 2>&1 &
```
上述python命令将在后台运行，您可以通过train.log文件查看结果。
训练结束后，您可在yaml文件中配置的文件夹下找到检查点文件。
### 分布式训练
- Ascend处理器环境运行

51
model_zoo/official/cv/googlenet/cifar10_config_cpu.yaml Normal file
View File

@ -0,0 +1,51 @@
# Builtin Configurations(DO NOT CHANGE THESE CONFIGURATIONS unless you know exactly what you are doing)
enable_modelarts: False
# Url for modelarts
data_url: ""
train_url: ""
checkpoint_url: ""
# Path for local
data_path: "/cache/data"
output_path: "/cache/train"
load_path: "/cache/checkpoint_path"
device_target: "CPU"
need_modelarts_dataset_unzip: True
modelarts_dataset_unzip_name: "cifar10"
# ==============================================================================
# options
dataset_name: "cifar10"
name: "cifar10"
pre_trained: False
num_classes: 10
lr_init: 0.1
batch_size: 128
epoch_size: 125
momentum: 0.9
weight_decay: 0.0005 #5e-4
image_height: 224
image_width: 224
train_data_path: "/cache/data/cifar10/"
val_data_path: "/cache/data/cifar10/"
keep_checkpoint_max: 10
checkpoint_path: "./train_googlenet_cifar10-125_390.ckpt"
onnx_filename: "googlenet"
air_filename: "googlenet"
ckpt_save_dir: "./ckpt/"
# export option
ckpt_file: ""
file_name: "googlenet"
file_format: "MINDIR"
#batch_size: 1
---
# Help description for each configuration
enable_modelarts: "Whether training on modelarts, default: False"
data_url: "Url for modelarts"
train_url: "Url for modelarts"
data_path: "The location of the input data."
output_path: "The location of the output file."
device_target: 'Target device type'
enable_profiling: 'Whether enable profiling while training, default: False'

38
model_zoo/official/cv/googlenet/scripts/run_eval_cpu.sh Normal file
View File

@ -0,0 +1,38 @@
#!/bin/bash
# Copyright 2021 Huawei Technologies Co., Ltd
#
# 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.
# ============================================================================
ulimit -u unlimited
# check checkpoint file
if [ ! -f $1 ]
then
echo "error: CHECKPOINT_PATH=$1 is not a file"
exit 1
fi
dataset_type='cifar10'
BASEPATH=$(cd "`dirname $0`" || exit; pwd)
export PYTHONPATH=${BASEPATH}:$PYTHONPATH
if [ -d "../eval" ];
then
rm -rf ../eval
fi
mkdir ../eval
cd ../eval || exit
config_path="${BASEPATH}/../${dataset_type}_config_cpu.yaml"
echo "config path is : ${config_path}"
nohup python ${BASEPATH}/../eval.py --config_path=$config_path --checkpoint_path=$1 --dataset_name=$dataset_type > ./eval.log 2>&1 &

30
model_zoo/official/cv/googlenet/scripts/run_train_cpu.sh Normal file
View File

@ -0,0 +1,30 @@
#!/bin/bash
# Copyright 2021 Huawei Technologies Co., Ltd
#
# 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.
# ============================================================================
BASEPATH=$(cd "`dirname $0`" || exit; pwd)
export PYTHONPATH=${BASEPATH}:$PYTHONPATH
if [ -d "./train" ];
then
rm -rf ./train
fi
mkdir ./train
cd ./train || exit
dataset_type='cifar10'
config_path="${BASEPATH}/../${dataset_type}_config_cpu.yaml"
echo "config path is : ${config_path}"
nohup python ${BASEPATH}/../train.py --config_path=$config_path --dataset_name=$dataset_type > train.log 2>&1 &

2
model_zoo/official/cv/googlenet/train.py
View File

@ -184,8 +184,6 @@ def run_train():
context.reset_auto_parallel_context()
context.set_auto_parallel_context(device_num=device_num, parallel_mode=ParallelMode.DATA_PARALLEL,
gradients_mean=True)
else:
raise ValueError("Unsupported platform.")
if cfg.dataset_name == "cifar10":
dataset = create_dataset_cifar10(cfg.train_data_path, 1, cifar_cfg=cfg)

11
model_zoo/official/nlp/textcnn/README.md
View File

@ -41,7 +41,7 @@ Dataset used: [Movie Review Data](<http://www.cs.cornell.edu/people/pabo/movie-r
# [Environment Requirements](#contents)
- Hardware（Ascend）
- Hardware（Ascend/CPU）
- Prepare hardware environment with Ascend processor.
- Framework
- [MindSpore](https://www.mindspore.cn/install/en)
@ -53,7 +53,7 @@ Dataset used: [Movie Review Data](<http://www.cs.cornell.edu/people/pabo/movie-r
After installing MindSpore via the official website, you can start training and evaluation as follows:
- running on Ascend
- running on Ascend/CPU
```python
# run training example
@ -110,6 +110,8 @@ If you want to run in modelarts, please check the official documentation of [mod
├──scripts
│ ├── run_train.sh // shell script for distributed on Ascend
│ ├── run_eval.sh // shell script for evaluation on Ascend
│ ├── run_train_cpu.sh // shell script for training on CPU
│ ├── run_eval_cpu.sh // shell script for evaluation on CPU
├── src
│ ├── dataset.py // Processing dataset
│ ├── textcnn.py // textcnn architecture
@ -119,6 +121,7 @@ If you want to run in modelarts, please check the official documentation of [mod
│ ├──moxing_adapter.py // moxing adapter
│ ├──config.py // parameter analysis
├── mr_config.yaml // parameter configuration
├── mr_config_cpu.yaml // parameter configuration
├── sst2_config.yaml // parameter configuration
├── subj_config.yaml // parameter configuration
├── train.py // training script
@ -152,7 +155,7 @@ For more configuration details, please refer the script `*.yaml`.
## [Training Process](#contents)
- running on Ascend
- running on Ascend/CPU
```python
# need set config_path in config.py file and set data_path in yaml file
@ -176,7 +179,7 @@ For more configuration details, please refer the script `*.yaml`.
## [Evaluation Process](#contents)
- evaluation on movie review dataset when running on Ascend
- evaluation on movie review dataset when running on Ascend/CPU
Before running the command below, please check the checkpoint path used for evaluation. Please set the checkpoint path to be the absolute full path, e.g., "username/textcnn/ckpt/train_textcnn.ckpt".

59
model_zoo/official/nlp/textcnn/mr_config_cpu.yaml Normal file
View File

@ -0,0 +1,59 @@
# Builtin Configurations(DO NOT CHANGE THESE CONFIGURATIONS unless you know exactly what you are doing)
enable_modelarts: False
# Url for modelarts
data_url: ""
train_url: ""
checkpoint_url: ""
# Path for local
data_path: "./data"
output_path: "./train"
load_path: "./checkpoint_path/"
device_target: 'CPU'
enable_profiling: False
# ==============================================================================
# Training options
dataset: 'MR'
pre_trained: False
num_classes: 2
batch_size: 64
epoch_size: 4
weight_decay: 3e-5
keep_checkpoint_max: 1
checkpoint_path: './checkpoint/'
checkpoint_file_path: './train/checkpoint/train_textcnn-4_149.ckpt'
word_len: 51
vec_length: 40
base_lr: 1e-3
label_dir: ""
result_dir: ""
result_path: './preprocess_Result/'
# Export options
device_id: 0
file_name: "textcnn"
file_format: "AIR"
---
# Help description for each configuration
enable_modelarts: 'Whether training on modelarts, default: False'
data_url: 'Dataset url for obs'
train_url: 'Training output url for obs'
checkpoint_url: 'The location of checkpoint for obs'
data_path: 'Dataset path for local'
output_path: 'Training output path for local'
load_path: 'The location of checkpoint for obs'
device_target: 'Target device type, available: [Ascend, GPU, CPU]'
enable_profiling: 'Whether enable profiling while training, default: False'
dataset: "Dataset to be trained and evaluated, choice: ['MR, SUBJ, SST2']"
train_epochs: "The number of epochs used to train."
pre_trained: 'If need load pre_trained checkpoint, default: False'
num_classes: 'Class for dataset'
batch_size: "Batch size for training and evaluation"
epoch_size: "Total training epochs."
weight_decay: "Weight decay."
keep_checkpoint_max: "keep the last keep_checkpoint_max checkpoint"
num_factors: "The Embedding size of MF model."
checkpoint_path: "The location of the checkpoint file."
eval_file_name: "Eval output file."
checkpoint_file_path: "The location of the checkpoint file."

19
model_zoo/official/nlp/textcnn/scripts/run_eval_cpu.sh Normal file
View File

@ -0,0 +1,19 @@
#!/bin/bash
# Copyright 2021 Huawei Technologies Co., Ltd
#
# 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.
# ============================================================================
BASE_PATH=$(dirname "$(dirname "$(readlink -f $0)")")
dataset_type='MR'
CONFIG_FILE="${BASE_PATH}/mr_config_cpu.yaml"
python eval.py --checkpoint_file_path="$1" --dataset=$dataset_type --config_path=$CONFIG_FILE > eval.log 2>&1 &

20
model_zoo/official/nlp/textcnn/scripts/run_train_cpu.sh Normal file
View File

@ -0,0 +1,20 @@
#!/bin/bash
# Copyright 2021 Huawei Technologies Co., Ltd
#
# 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.
# ============================================================================
BASE_PATH=$(dirname "$(dirname "$(readlink -f $0)")")
dataset_type='MR'
echo $BASE_PATH
CONFIG_FILE="${BASE_PATH}/mr_config_cpu.yaml"
python train.py --dataset=$dataset_type --config_path=$CONFIG_FILE > train.log 2>&1 &