forked from mindspore-Ecosystem/mindspore
!4937 vgg16: modify readme format and replace callback
Merge pull request !4937 from ms_yan/vgg_format
This commit is contained in:
Gitee
commit
38c366306c
|
|
@ -1,36 +1,196 @@
|
|||
# VGG16 Example
|
||||
# Contents
|
||||
|
||||
## Description
|
||||
- [VGG Description](#vgg-description)
|
||||
- [Model Architecture](#model-architecture)
|
||||
- [Dataset](#dataset)
|
||||
- [Features](#features)
|
||||
- [Mixed Precision](#mixed-precision)
|
||||
- [Environment Requirements](#environment-requirements)
|
||||
- [Quick Start](#quick-start)
|
||||
- [Script Description](#script-description)
|
||||
- [Script and Sample Code](#script-and-sample-code)
|
||||
- [Script Parameters](#script-parameters)
|
||||
- [Parameter configuration](#parameter-configuration)
|
||||
- [Training Process](#training-process)
|
||||
- [Training](#training)
|
||||
- [Evaluation Process](#evaluation-process)
|
||||
- [Evaluation](#evaluation)
|
||||
- [Model Description](#model-description)
|
||||
- [Performance](#performance)
|
||||
- [Training Performance](#training-performance)
|
||||
- [Evaluation Performance](#evaluation-performance)
|
||||
- [Description of Random Situation](#description-of-random-situation)
|
||||
- [ModelZoo Homepage](#modelzoo-homepage)
|
||||
|
||||
This example is for VGG16 model training and evaluation.
|
||||
|
||||
## Requirements
|
||||
# [VGG Description](#contents)
|
||||
|
||||
- Install [MindSpore](https://www.mindspore.cn/install/en).
|
||||
VGG, a very deep convolutional networks for large-scale image recognition, was proposed in 2014 and won the 1th place in object localization and 2th place in image classification task in ImageNet Large-Scale Visual Recognition Challenge 2014 (ILSVRC14).
|
||||
|
||||
- Download the dataset CIFAR-10 or ImageNet2012.
|
||||
[Paper](): Simonyan K, zisserman A. Very Deep Convolutional Networks for Large-Scale Image Recognition[J]. arXiv preprint arXiv:1409.1556, 2014.
|
||||
|
||||
CIFAR-10
|
||||
# [Model Architecture](#contents)
|
||||
VGG 16 network is mainly consisted by several basic modules (including convolution and pooling layer) and three continuous Dense layer.
|
||||
here basic modules mainly include basic operation like: **3×3 conv** and **2×2 max pooling**.
|
||||
|
||||
> Unzip the CIFAR-10 dataset to any path you want and the folder structure should be as follows:
|
||||
> ```
|
||||
> .
|
||||
> ├── cifar-10-batches-bin # train dataset
|
||||
> └── cifar-10-verify-bin # infer dataset
|
||||
> ```
|
||||
|
||||
ImageNet2012
|
||||
# [Dataset](#contents)
|
||||
|
||||
> Unzip the ImageNet2012 dataset to any path you want and the folder should include train and eval dataset as follows:
|
||||
>
|
||||
> ```
|
||||
> .
|
||||
> └─dataset
|
||||
> ├─ilsvrc # train dataset
|
||||
> └─validation_preprocess # evaluate dataset
|
||||
> ```
|
||||
#### Dataset used: [CIFAR-10](<http://www.cs.toronto.edu/~kriz/cifar.html>)
|
||||
|
||||
## Parameter configuration
|
||||
- CIFAR-10 Dataset size：175M，60,000 32*32 colorful images in 10 classes
|
||||
- Train：146M，50,000 images
|
||||
- Test：29.3M，10,000 images
|
||||
- Data format: binary files
|
||||
- Note: Data will be processed in src/dataset.py
|
||||
|
||||
#### Dataset used: [ImageNet2012](http://www.image-net.org/)
|
||||
- Dataset size: ~146G, 1.28 million colorful images in 1000 classes
|
||||
- Train: 140G, 1,281,167 images
|
||||
- Test: 6.4G, 50, 000 images
|
||||
- Data format: RGB images
|
||||
- Note: Data will be processed in src/dataset.py
|
||||
|
||||
#### Dataset organize way
|
||||
|
||||
CIFAR-10
|
||||
|
||||
> Unzip the CIFAR-10 dataset to any path you want and the folder structure should be as follows:
|
||||
> ```
|
||||
> .
|
||||
> ├── cifar-10-batches-bin # train dataset
|
||||
> └── cifar-10-verify-bin # infer dataset
|
||||
> ```
|
||||
|
||||
ImageNet2012
|
||||
|
||||
> Unzip the ImageNet2012 dataset to any path you want and the folder should include train and eval dataset as follows:
|
||||
>
|
||||
> ```
|
||||
> .
|
||||
> └─dataset
|
||||
> ├─ilsvrc # train dataset
|
||||
> └─validation_preprocess # evaluate dataset
|
||||
> ```
|
||||
|
||||
|
||||
# [Features](#contents)
|
||||
|
||||
## Mixed Precision
|
||||
|
||||
The [mixed precision](https://www.mindspore.cn/tutorial/zh-CN/master/advanced_use/mixed_precision.html) training method accelerates the deep learning neural network training process by using both the single-precision and half-precision data formats, and maintains the network precision achieved by the single-precision training at the same time. Mixed precision training can accelerate the computation process, reduce memory usage, and enable a larger model or batch size to be trained on specific hardware.
|
||||
For FP16 operators, if the input data type is FP32, the backend of MindSpore will automatically handle it with reduced precision. Users could check the reduced-precision operators by enabling INFO log and then searching ‘reduce precision’.
|
||||
|
||||
|
||||
# [Environment Requirements](#contents)
|
||||
|
||||
- Hardware（Ascend/GPU）
|
||||
- Prepare hardware environment with Ascend or GPU processor. If you want to try Ascend , please send the [application form](https://obs-9be7.obs.cn-east-2.myhuaweicloud.com/file/other/Ascend%20Model%20Zoo%E4%BD%93%E9%AA%8C%E8%B5%84%E6%BA%90%E7%94%B3%E8%AF%B7%E8%A1%A8.docx) to ascend@huawei.com. Once approved, you can get the resources.
|
||||
- Framework
|
||||
- [MindSpore](https://www.mindspore.cn/install/en)
|
||||
- For more information, please check the resources below：
|
||||
- [MindSpore tutorials](https://www.mindspore.cn/tutorial/zh-CN/master/index.html)
|
||||
- [MindSpore API](https://www.mindspore.cn/api/zh-CN/master/index.html)
|
||||
|
||||
|
||||
# [Quick Start](#contents)
|
||||
|
||||
After installing MindSpore via the official website, you can start training and evaluation as follows:
|
||||
|
||||
- Running on Ascend
|
||||
```python
|
||||
# run training example
|
||||
python train.py --data_path=[DATA_PATH] --device_id=[DEVICE_ID] > output.train.log 2>&1 &
|
||||
|
||||
# run distributed training example
|
||||
sh run_distribute_train.sh [RANL_TABLE_JSON] [DATA_PATH]
|
||||
|
||||
# run evaluation example
|
||||
python eval.py --data_path=[DATA_PATH] --pre_trained=[PRE_TRAINED] > output.eval.log 2>&1 &
|
||||
```
|
||||
For distributed training, a hccl configuration file with JSON format needs to be created in advance.
|
||||
Please follow the instructions in the link below:
|
||||
https://gitee.com/mindspore/mindspore/tree/master/model_zoo/utils/hccl_tools
|
||||
|
||||
- Running on GPU
|
||||
```
|
||||
# run training example
|
||||
python train.py --device_target="GPU" --device_id=[DEVICE_ID] --dataset=[DATASET_TYPE] --data_path=[DATA_PATH] > output.train.log 2>&1 &
|
||||
|
||||
# run distributed training example
|
||||
sh run_distribute_train_gpu.sh [DATA_PATH]
|
||||
|
||||
# run evaluation example
|
||||
python eval.py --device_target="GPU" --device_id=[DEVICE_ID] --dataset=[DATASET_TYPE] --data_path=[DATA_PATH] --pre_trained=[PRE_TRAINED] > output.eval.log 2>&1 &
|
||||
```
|
||||
|
||||
# [Script Description](#contents)
|
||||
|
||||
## [Script and Sample Code](#contents)
|
||||
|
||||
|
||||
```
|
||||
├── model_zoo
|
||||
├── README.md // descriptions about all the models
|
||||
├── vgg16
|
||||
├── README.md // descriptions about googlenet
|
||||
├── scripts
|
||||
│ ├── run_distribute_train.sh // shell script for distributed training on Ascend
|
||||
│ ├── run_distribute_train_gpu.sh // shell script for distributed training on GPU
|
||||
├── src
|
||||
│ ├── utils
|
||||
│ │ ├── logging.py // logging format setting
|
||||
│ │ ├── sampler.py // create sampler for dataset
|
||||
│ │ ├── util.py // util function
|
||||
│ │ ├── var_init.py // network parameter init method
|
||||
│ ├── config.py // parameter configuration
|
||||
│ ├── crossentropy.py // loss caculation
|
||||
│ ├── dataset.py // creating dataset
|
||||
│ ├── linear_warmup.py // linear leanring rate
|
||||
│ ├── warmup_cosine_annealing_lr.py // consine anealing learning rate
|
||||
│ ├── warmup_step_lr.py // step or multi step learning rate
|
||||
│ ├──vgg.py // vgg architecture
|
||||
├── train.py // training script
|
||||
├── eval.py // evaluation script
|
||||
```
|
||||
|
||||
## [Script Parameters](#contents)
|
||||
|
||||
### Training
|
||||
```
|
||||
usage: train.py [--device_target TARGET][--data_path DATA_PATH]
|
||||
[--dataset DATASET_TYPE][--is_distributed VALUE]
|
||||
[--device_id DEVICE_ID][--pre_trained PRE_TRAINED]
|
||||
[--ckpt_path CHECKPOINT_PATH][--ckpt_interval INTERVAL_STEP]
|
||||
|
||||
parameters/options:
|
||||
--device_target the training backend type, Ascend or GPU, default is Ascend.
|
||||
--dataset the dataset type, cifar10 or imagenet2012.
|
||||
--is_distributed the way of traing, whether do distribute traing, value can be 0 or 1.
|
||||
--data_path the storage path of dataset
|
||||
--device_id the device which used to train model.
|
||||
--pre_trained the pretrained checkpoint file path.
|
||||
--ckpt_path the path to save checkpoint.
|
||||
--ckpt_interval the epoch interval for saving checkpoint.
|
||||
|
||||
```
|
||||
|
||||
### Evaluation
|
||||
|
||||
```
|
||||
usage: eval.py [--device_target TARGET][--data_path DATA_PATH]
|
||||
[--dataset DATASET_TYPE][--pre_trained PRE_TRAINED]
|
||||
[--device_id DEVICE_ID]
|
||||
|
||||
parameters/options:
|
||||
--device_target the evaluation backend type, Ascend or GPU, default is Ascend.
|
||||
--dataset the dataset type, cifar10 or imagenet2012.
|
||||
--data_path the storage path of dataset.
|
||||
--device_id the device which used to evaluate model.
|
||||
--pre_trained the checkpoint file path used to evaluate model.
|
||||
```
|
||||
|
||||
## [Parameter configuration](#contents)
|
||||
|
||||
Parameters for both training and evaluation can be set in config.py.
|
||||
|
||||
|
|
@ -90,12 +250,13 @@ Parameters for both training and evaluation can be set in config.py.
|
|||
"has_dropout": True # wether using Dropout layer
|
||||
```
|
||||
|
||||
## Running the Example
|
||||
## [Training Process](#contents)
|
||||
|
||||
### Training
|
||||
**Run vgg16, using CIFAR-10 dataset**
|
||||
|
||||
- Training using single device(1p)
|
||||
#### Run vgg16 on Ascend
|
||||
|
||||
- Training using single device(1p), using CIFAR-10 dataset in default
|
||||
```
|
||||
python train.py --data_path=your_data_path --device_id=6 > out.train.log 2>&1 &
|
||||
```
|
||||
|
|
@ -105,13 +266,13 @@ After training, you'll get some checkpoint files in specified ckpt_path, default
|
|||
|
||||
You will get the loss value as following:
|
||||
```
|
||||
# grep "loss is " out.train.log
|
||||
# grep "loss is " output.train.log
|
||||
epoch: 1 step: 781, loss is 2.093086
|
||||
epcoh: 2 step: 781, loss is 1.827582
|
||||
...
|
||||
```
|
||||
|
||||
- Distribute Training
|
||||
- Distributed Training
|
||||
```
|
||||
sh run_distribute_train.sh rank_table.json your_data_path
|
||||
```
|
||||
|
|
@ -131,37 +292,35 @@ train_parallel1/log:epcoh: 2 step: 97, loss is 1.7133579
|
|||
> About rank_table.json, you can refer to the [distributed training tutorial](https://www.mindspore.cn/tutorial/en/master/advanced_use/distributed_training.html).
|
||||
|
||||
|
||||
**Run vgg16, using imagenet2012 dataset**
|
||||
#### Run vgg16 on GPU
|
||||
|
||||
- Training using single device(1p)
|
||||
```
|
||||
python train.py --device_target="GPU" --dataset="imagenet2012" --is_distributed=0 --data_path=$DATA_PATH > output.train.log 2>&1 &
|
||||
```
|
||||
|
||||
- Distribute Training
|
||||
- Distributed Training
|
||||
```
|
||||
# distributed training(8p)
|
||||
bash scripts/run_distribute_train_gpu.sh /path/ImageNet2012/train"
|
||||
```
|
||||
|
||||
## [Evaluation Process](#contents)
|
||||
|
||||
### Evaluation
|
||||
|
||||
- Do eval as follows, need to specify dataset type as "cifar10" or "imagenet2012"
|
||||
```
|
||||
# when using cifar10 dataset
|
||||
python eval.py --data_path=your_data_path --dataset="cifar10" --device_target="Ascend" --pre_trained=./*-70-781.ckpt > out.eval.log 2>&1 &
|
||||
python eval.py --data_path=your_data_path --dataset="cifar10" --device_target="Ascend" --pre_trained=./*-70-781.ckpt > output.eval.log 2>&1 &
|
||||
|
||||
# when using imagenet2012 dataset
|
||||
python eval.py --data_path=your_data_path --dataset="imagenet2012" --device_target="GPU" --pre_trained=./*-150-5004.ckpt > out.eval.log 2>&1 &
|
||||
python eval.py --data_path=your_data_path --dataset="imagenet2012" --device_target="GPU" --pre_trained=./*-150-5004.ckpt > output.eval.log 2>&1 &
|
||||
```
|
||||
- If the using dataset is
|
||||
The above python command will run in the background, you can view the results through the file `out.eval.log`.
|
||||
|
||||
You will get the accuracy as following:
|
||||
- The above python command will run in the background, you can view the results through the file `output.eval.log`. You will get the accuracy as following:
|
||||
```
|
||||
# when using cifar10 dataset
|
||||
# grep "result: " out.eval.log
|
||||
# grep "result: " output.eval.log
|
||||
result: {'acc': 0.92}
|
||||
|
||||
# when using the imagenet2012 dataset
|
||||
|
|
@ -169,57 +328,46 @@ after allreduce eval: top1_correct=36636, tot=50000, acc=73.27%
|
|||
after allreduce eval: top5_correct=45582, tot=50000, acc=91.16%
|
||||
```
|
||||
|
||||
## Usage:
|
||||
|
||||
### Training
|
||||
```
|
||||
usage: train.py [--device_target TARGET][--data_path DATA_PATH]
|
||||
[--dataset DATASET_TYPE][--is_distributed VALUE]
|
||||
[--device_id DEVICE_ID][--pre_trained PRE_TRAINED]
|
||||
[--ckpt_path CHECKPOINT_PATH][--ckpt_interval INTERVAL_STEP]
|
||||
# [Model Description](#contents)
|
||||
## [Performance](#contents)
|
||||
|
||||
parameters/options:
|
||||
--device_target the training backend type, Ascend or GPU, default is Ascend.
|
||||
--dataset the dataset type, cifar10 or imagenet2012.
|
||||
--is_distributed the way of traing, whether do distribute traing, value can be 0 or 1.
|
||||
--data_path the storage path of dataset
|
||||
--device_id the device which used to train model.
|
||||
--pre_trained the pretrained checkpoint file path.
|
||||
--ckpt_path the path to save checkpoint.
|
||||
--ckpt_interval the epoch interval for saving checkpoint.
|
||||
### Training Performance
|
||||
|
||||
```
|
||||
| Parameters | VGG16(Ascend) | VGG16(GPU) |
|
||||
| -------------------------- | ---------------------------------------------- |------------------------------------|
|
||||
| Model Version | VGG16 | VGG16 |
|
||||
| Resource | Ascend 910 ；CPU 2.60GHz，56cores；Memory，314G |NV SMX2 V100-32G |
|
||||
| uploaded Date | 08/20/2020 |08/20/2020 |
|
||||
| MindSpore Version | 0.5.0-alpha |0.5.0-alpha |
|
||||
| Dataset | CIFAR-10 |ImageNet2012 |
|
||||
| Training Parameters | epoch=70, steps=781, batch_size = 64, lr=0.1 |epoch=150, steps=40036, batch_size = 32, lr=0.1 |
|
||||
| Optimizer | Momentum |Momentum |
|
||||
| Loss Function | SoftmaxCrossEntropy |SoftmaxCrossEntropy |
|
||||
| outputs | probability |probability |
|
||||
| Loss | 0.01 |1.5~2.0 |
|
||||
| Speed | 1pc: 79 ms/step; 8pcs: 104 ms/step |1pc: 81 ms/step; 8pcs 94.4ms/step |
|
||||
| Total time | 1pc: 72 mins; 8pcs: 11.8 mins |8pcs: 19.7 hours |
|
||||
| Checkpoint for Fine tuning | 1.1G(.ckpt file) |1.1G(.ckpt file) |
|
||||
| Scripts |[vgg16](https://gitee.com/mindspore/mindspore/tree/master/model_zoo/official/cv/vgg16) | |
|
||||
|
||||
### Evaluation
|
||||
|
||||
```
|
||||
usage: eval.py [--device_target TARGET][--data_path DATA_PATH]
|
||||
[--dataset DATASET_TYPE][--pre_trained PRE_TRAINED]
|
||||
[--device_id DEVICE_ID]
|
||||
### Evaluation Performance
|
||||
|
||||
parameters/options:
|
||||
--device_target the evaluation backend type, Ascend or GPU, default is Ascend.
|
||||
--dataset the dataset type, cifar10 or imagenet2012.
|
||||
--data_path the storage path of dataset.
|
||||
--device_id the device which used to evaluate model.
|
||||
--pre_trained the checkpoint file path used to evaluate model.
|
||||
```
|
||||
| Parameters | VGG16(Ascend) | VGG16(GPU)
|
||||
| ------------------- | --------------------------- |---------------------
|
||||
| Model Version | VGG16 | VGG16 |
|
||||
| Resource | Ascend 910 | GPU |
|
||||
| Uploaded Date | 08/20/2020 | 08/20/2020 |
|
||||
| MindSpore Version | 0.5.0-alpha |0.5.0-alpha |
|
||||
| Dataset | CIFAR-10, 10,000 images |ImageNet2012, 5000 images |
|
||||
| batch_size | 64 | 32 |
|
||||
| outputs | probability | probability |
|
||||
| Accuracy | 1pc: 93.4% |1pc: 73.0%; |
|
||||
|
||||
### Distribute Training
|
||||
- Train on Ascend.
|
||||
# [Description of Random Situation](#contents)
|
||||
|
||||
```
|
||||
Usage: sh script/run_distribute_train.sh [RANK_TABLE_FILE] [DATA_PATH]
|
||||
In dataset.py, we set the seed inside “create_dataset" function. We also use random seed in train.py.
|
||||
|
||||
parameters/options:
|
||||
RANK_TABLE_FILE HCCL configuration file path.
|
||||
DATA_PATH the storage path of dataset.
|
||||
```
|
||||
|
||||
- Train on GPU.
|
||||
```
|
||||
Usage: bash run_distribute_train_gpu.sh [DATA_PATH]
|
||||
|
||||
parameters/options:
|
||||
DATA_PATH the storage path of dataset.
|
||||
```
|
||||
# [ModelZoo Homepage](#contents)
|
||||
Please check the official [homepage](https://gitee.com/mindspore/mindspore/tree/master/model_zoo).
|
||||
|
|
@ -15,7 +15,7 @@
|
|||
# ============================================================================
|
||||
|
||||
echo "=============================================================================================================="
|
||||
echo "Please run the scipt as: "
|
||||
echo "Please run the script as: "
|
||||
echo "bash run_distribute_train_gpu.sh DATA_PATH"
|
||||
echo "for example: bash run_distribute_train_gpu.sh /path/ImageNet2012/train"
|
||||
echo "=============================================================================================================="
|
||||
|
|
|
|||
|
|
@ -0,0 +1,32 @@
|
|||
#!/bin/bash
|
||||
# Copyright 2020 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.
|
||||
# ============================================================================
|
||||
|
||||
echo "=============================================================================================================="
|
||||
echo "Please run the script as: "
|
||||
echo "bash run_eval.sh DATA_PATH DATASET_TYPE DEVICE_TYPE CHECKPOINT_PATH"
|
||||
echo "for example: bash run_eval.sh /path/ImageNet2012/train cifar10 Ascend /path/a.ckpt "
|
||||
echo "=============================================================================================================="
|
||||
|
||||
DATA_PATH=&1
|
||||
DATASET_TYPE=$2
|
||||
DEVICE_TYPE=$3
|
||||
CHECKPOINT_PATH=$4
|
||||
|
||||
python eval.py \
|
||||
--data_path=$DATA_PATH \
|
||||
--dataset=$DATASET_TYPE \
|
||||
--device_target=$DEVICE_TYPE \
|
||||
--pre_trained=$CHECKPOINT_PATH > output.eval.log 2>&1 &
|
||||
|
|
@ -18,7 +18,6 @@ python train.py --data_path=$DATA_HOME --device_id=$DEVICE_ID
|
|||
"""
|
||||
import argparse
|
||||
import datetime
|
||||
import time
|
||||
import os
|
||||
import random
|
||||
|
||||
|
|
@ -29,7 +28,7 @@ from mindspore import Tensor
|
|||
from mindspore import context
|
||||
from mindspore.communication.management import init, get_rank, get_group_size
|
||||
from mindspore.nn.optim.momentum import Momentum
|
||||
from mindspore.train.callback import Callback, ModelCheckpoint, CheckpointConfig
|
||||
from mindspore.train.callback import ModelCheckpoint, CheckpointConfig, LossMonitor, TimeMonitor
|
||||
from mindspore.train.model import Model, ParallelMode
|
||||
from mindspore.train.serialization import load_param_into_net, load_checkpoint
|
||||
from mindspore.train.loss_scale_manager import FixedLossScaleManager
|
||||
|
|
@ -49,63 +48,6 @@ random.seed(1)
|
|||
np.random.seed(1)
|
||||
|
||||
|
||||
class ProgressMonitor(Callback):
|
||||
"""monitor loss and time"""
|
||||
def __init__(self, args_param):
|
||||
super(ProgressMonitor, self).__init__()
|
||||
self.me_epoch_start_time = 0
|
||||
self.me_epoch_start_step_num = 0
|
||||
self.args = args_param
|
||||
self.ckpt_history = []
|
||||
|
||||
def begin(self, run_context):
|
||||
self.args.logger.info('start network train...')
|
||||
|
||||
def epoch_begin(self, run_context):
|
||||
pass
|
||||
|
||||
def epoch_end(self, run_context):
|
||||
"""
|
||||
Called after each epoch finished.
|
||||
|
||||
Args:
|
||||
run_context (RunContext): Include some information of the model.
|
||||
"""
|
||||
cb_params = run_context.original_args()
|
||||
me_step = cb_params.cur_step_num - 1
|
||||
|
||||
real_epoch = me_step // self.args.steps_per_epoch
|
||||
time_used = time.time() - self.me_epoch_start_time
|
||||
fps_mean = self.args.per_batch_size * (me_step-self.me_epoch_start_step_num) * self.args.group_size / time_used
|
||||
self.args.logger.info('epoch[{}], iter[{}], loss:{}, mean_fps:{:.2f}'
|
||||
'imgs/sec'.format(real_epoch, me_step, cb_params.net_outputs, fps_mean))
|
||||
|
||||
if self.args.rank_save_ckpt_flag:
|
||||
import glob
|
||||
ckpts = glob.glob(os.path.join(self.args.outputs_dir, '*.ckpt'))
|
||||
for ckpt in ckpts:
|
||||
ckpt_fn = os.path.basename(ckpt)
|
||||
if not ckpt_fn.startswith('{}-'.format(self.args.rank)):
|
||||
continue
|
||||
if ckpt in self.ckpt_history:
|
||||
continue
|
||||
self.ckpt_history.append(ckpt)
|
||||
self.args.logger.info('epoch[{}], iter[{}], loss:{}, ckpt:{},'
|
||||
'ckpt_fn:{}'.format(real_epoch, me_step, cb_params.net_outputs, ckpt, ckpt_fn))
|
||||
|
||||
self.me_epoch_start_step_num = me_step
|
||||
self.me_epoch_start_time = time.time()
|
||||
|
||||
def step_begin(self, run_context):
|
||||
pass
|
||||
|
||||
def step_end(self, run_context, *me_args):
|
||||
pass
|
||||
|
||||
def end(self, run_context):
|
||||
self.args.logger.info('end network train...')
|
||||
|
||||
|
||||
def parse_args(cloud_args=None):
|
||||
"""parameters"""
|
||||
parser = argparse.ArgumentParser('mindspore classification training')
|
||||
|
|
@ -279,9 +221,10 @@ if __name__ == '__main__':
|
|||
loss_scale_manager = FixedLossScaleManager(args.loss_scale, drop_overflow_update=False)
|
||||
model = Model(network, loss_fn=loss, optimizer=opt, loss_scale_manager=loss_scale_manager, amp_level="O2")
|
||||
|
||||
# checkpoint save
|
||||
progress_cb = ProgressMonitor(args)
|
||||
callbacks = [progress_cb,]
|
||||
# define callbacks
|
||||
time_cb = TimeMonitor(data_size=batch_num)
|
||||
loss_cb = LossMonitor(per_print_times=batch_num)
|
||||
callbacks = [time_cb, loss_cb]
|
||||
if args.rank_save_ckpt_flag:
|
||||
ckpt_config = CheckpointConfig(save_checkpoint_steps=args.ckpt_interval * args.steps_per_epoch,
|
||||
keep_checkpoint_max=args.ckpt_save_max)
|
||||
|
|
|
|||
Loading…
Reference in New Issue