From f4685972ee4c675edab068970d343c6d8181ee30 Mon Sep 17 00:00:00 2001 From: zhanghuiyao <1814619459@qq.com> Date: Fri, 4 Jun 2021 10:06:36 +0800 Subject: [PATCH] modify bgcf network for clould --- model_zoo/official/cv/vgg16/eval.py | 3 +- model_zoo/official/cv/vgg16/train.py | 7 +- model_zoo/official/gnn/bgcf/README.md | 87 ++++++++++-- model_zoo/official/gnn/bgcf/README_CN.md | 83 +++++++++-- .../official/gnn/bgcf/default_config.yaml | 61 +++++++++ model_zoo/official/gnn/bgcf/eval.py | 129 +++++++++++++----- .../official/gnn/bgcf/model_utils/__init__.py | 0 .../official/gnn/bgcf/model_utils/config.py | 126 +++++++++++++++++ .../gnn/bgcf/model_utils/device_adapter.py | 27 ++++ .../gnn/bgcf/model_utils/local_adapter.py | 36 +++++ .../gnn/bgcf/model_utils/moxing_adapter.py | 116 ++++++++++++++++ .../gnn/bgcf/scripts/run_eval_ascend.sh | 11 +- .../official/gnn/bgcf/scripts/run_eval_gpu.sh | 2 + .../gnn/bgcf/scripts/run_train_ascend.sh | 11 +- .../gnn/bgcf/scripts/run_train_gpu.sh | 2 + model_zoo/official/gnn/bgcf/train.py | 116 ++++++++++++---- .../bgcf/test_BGCF_amazon_beauty.py | 8 +- 17 files changed, 729 insertions(+), 96 deletions(-) create mode 100644 model_zoo/official/gnn/bgcf/default_config.yaml create mode 100644 model_zoo/official/gnn/bgcf/model_utils/__init__.py create mode 100644 model_zoo/official/gnn/bgcf/model_utils/config.py create mode 100644 model_zoo/official/gnn/bgcf/model_utils/device_adapter.py create mode 100644 model_zoo/official/gnn/bgcf/model_utils/local_adapter.py create mode 100644 model_zoo/official/gnn/bgcf/model_utils/moxing_adapter.py diff --git a/model_zoo/official/cv/vgg16/eval.py b/model_zoo/official/cv/vgg16/eval.py index 4ee219e2b08..6b618ef4cf4 100644 --- a/model_zoo/official/cv/vgg16/eval.py +++ b/model_zoo/official/cv/vgg16/eval.py @@ -21,7 +21,7 @@ import numpy as np import mindspore.nn as nn from mindspore import Tensor, context -from mindspore.communication.management import get_rank, get_group_size +from mindspore.communication.management import init, get_rank, get_group_size from mindspore.nn.optim.momentum import Momentum from mindspore.train.model import Model from mindspore.train.serialization import load_checkpoint, load_param_into_net @@ -95,6 +95,7 @@ def modelarts_pre_process(): # Each server contains 8 devices as most. if config.device_target == "GPU": + init() device_id = get_rank() device_num = get_group_size() elif config.device_target == "Ascend": diff --git a/model_zoo/official/cv/vgg16/train.py b/model_zoo/official/cv/vgg16/train.py index aed70d4af9e..91930685c9f 100644 --- a/model_zoo/official/cv/vgg16/train.py +++ b/model_zoo/official/cv/vgg16/train.py @@ -82,6 +82,7 @@ def modelarts_pre_process(): # Each server contains 8 devices as most. if config.device_target == "GPU": + init() device_id = get_rank() device_num = get_group_size() elif config.device_target == "Ascend": @@ -129,7 +130,11 @@ def run_train(): init() context.set_context(device_id=config.device_id) elif config.device_target == "GPU": - init() + if not config.enable_modelarts: + init() + else: + if not config.need_modelarts_dataset_unzip: + init() device_num = config.group_size context.reset_auto_parallel_context() diff --git a/model_zoo/official/gnn/bgcf/README.md b/model_zoo/official/gnn/bgcf/README.md index d0bc2c5d3c2..f2d693dc58c 100644 --- a/model_zoo/official/gnn/bgcf/README.md +++ b/model_zoo/official/gnn/bgcf/README.md @@ -90,17 +90,17 @@ To ultilize the strong computation power of Ascend chip, and accelerate the trai After installing MindSpore via the official website and Dataset is correctly generated, you can start training and evaluation as follows. -- running on Ascend +- Running on Ascend ```python # run training example with Amazon-Beauty dataset - sh run_train_ascend.sh + sh run_train_ascend.sh dataset_path # run evaluation example with Amazon-Beauty dataset - sh run_eval_ascend.sh + sh run_eval_ascend.sh dataset_path ``` -- running on GPU +- Running on GPU ```python # run training example with Amazon-Beauty dataset @@ -110,6 +110,62 @@ After installing MindSpore via the official website and Dataset is correctly gen sh run_eval_gpu.sh 0 dataset_path ``` +- Running on 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) + + - Train 1p on ModelArts Ascend/GPU + + ```python + # (1) Perform a or b. + # a. Set "enable_modelarts=True" on default_config.yaml file. + # Set "datapath='/cache/data/amazon_beauty/data_mr'" on default_config.yaml file. + # Set "ckptpath='./ckpts'" on default_config.yaml file. + # (options)Set "device_target='GPU'" on default_config.yaml file if run on GPU. + # (options)Set "num_epoch=680" on default_config.yaml file if run on GPU. + # (options)Set "dist_reg=0" on default_config.yaml file if run on GPU. + # Set other parameters on default_config.yaml file you need. + # b. Add "enable_modelarts=True" on the website UI interface. + # Add "datapath=/cache/data/amazon_beauty/data_mr" on the website UI interface. + # Add "ckptpath=./ckpts" on the website UI interface. + # (options)Add "device_target=GPU" on the website UI interface if run on GPU. + # (options)Add "num_epoch=680" on the website UI interface if run on GPU. + # (options)Add "dist_reg=0" on the website UI interface if run on GPU. + # Add other parameters on the website UI interface. + # (2) Prepare the converted dataset and zip it to one file like "amazon_beauty.zip" locally. (The conversion process can refer to the above data set processing code.) + # (3) Upload a zip dataset to S3 bucket. (you could also upload the origin dataset, but it can be so slow.) + # (4) Set the code directory to "/path/bgcf" on the website UI interface. + # (5) Set the startup file to "train.py" on the website UI interface. + # (6) Set the "Dataset path" and "Output file path" and "Job log path" to your path on the website UI interface. + # (7) Create your job. + ``` + + - Eval 1p on ModelArts Ascend/GPU + + ```python + # (1) Perform a or b. + # a. Set "enable_modelarts=True" on default_config.yaml file. + # Set "datapath='/cache/data/amazon_beauty/data_mr'" on default_config.yaml file. + # Set "ckptpath='/cache/checkpoint_path'" on default_config.yaml file. + # Set "checkpoint_url='s3://dir_to_your_trained_ckpt/'" on default_config.yaml file. + # (options)Set "device_target='GPU'" on default_config.yaml file if run on GPU. + # (options)Set "num_epoch=680" on default_config.yaml file if run on GPU. + # (options)Set "dist_reg=0" on default_config.yaml file if run on GPU. + # Set other parameters on default_config.yaml file you need. + # b. Add "enable_modelarts=True" on the website UI interface. + # Add "datapath=/cache/data/amazon_beauty/data_mr" on the website UI interface. + # Add "ckptpath='/cache/checkpoint_path'" on the website UI interface. + # Add "checkpoint_url='s3://dir_to_your_trained_ckpt/'" on the website UI interface. + # (options)Add "device_target=GPU" on the website UI interface if run on GPU. + # (options)Add "num_epoch=680" on the website UI interface if run on GPU. + # (options)Add "dist_reg=0" on the website UI interface if run on GPU. + # Add other parameters on the website UI interface. + # (2) Prepare the converted dataset and zip it to one file like "amazon_beauty.zip" locally. (The conversion process can refer to the above data set processing code.) + # (3) Upload a zip dataset to S3 bucket. (you could also upload the origin dataset, but it can be so slow.) + # (4) Set the code directory to "/path/bgcf" on the website UI interface. + # (5) Set the startup file to "eval.py" on the website UI interface. + # (6) Set the "Dataset path" and "Output file path" and "Job log path" to your path on the website UI interface. + # (7) Create your job. + ``` + ## [Script Description](#contents) ### [Script and Sample Code](#contents) @@ -118,28 +174,35 @@ After installing MindSpore via the official website and Dataset is correctly gen . └─bgcf ├─README.md + ├─README_CN.md + ├─model_utils + | ├─__init__.py # Module init file + | ├─config.py # Parse arguments + | ├─device_adapter.py # Device adapter for ModelArts + | ├─local_adapter.py # Local adapter + | └─moxing_adapter.py # Moxing adapter for ModelArts ├─scripts | ├─run_eval_ascend.sh # Launch evaluation in ascend | ├─run_eval_gpu.sh # Launch evaluation in gpu | ├─run_process_data_ascend.sh # Generate dataset in mindrecord format | └─run_train_ascend.sh # Launch training in ascend | └─run_train_gpu.sh # Launch training in gpu - | ├─src | ├─bgcf.py # BGCF model | ├─callback.py # Callback function - | ├─config.py # Training configurations | ├─dataset.py # Data preprocessing | ├─metrics.py # Recommendation metrics | └─utils.py # Utils for training bgcf - | + ├─default_config.yaml # Configurations file + ├─mindspore_hub_conf.py # Mindspore hub file + ├─export.py # Export net ├─eval.py # Evaluation net └─train.py # Train net ``` ### [Script Parameters](#contents) -Parameters for both training and evaluation can be set in config.py. +Parameters for both training and evaluation can be set in default_config.yaml. - config for BGCF dataset @@ -154,7 +217,7 @@ Parameters for both training and evaluation can be set in config.py. "neighbor_dropout": [0.0, 0.2, 0.3] # Dropout ratio for different aggregation layer ``` - config.py for more configuration. + default_config.yaml for more configuration. ### [Training Process](#contents) @@ -163,7 +226,7 @@ Parameters for both training and evaluation can be set in config.py. - running on Ascend ```python - sh run_train_ascend.sh + sh run_train_ascend.sh dataset_path ``` Training result will be stored in the scripts path, whose folder name begins with "train". You can find the result like the @@ -206,7 +269,7 @@ Parameters for both training and evaluation can be set in config.py. - Evaluation on Ascend ```python - sh run_eval_ascend.sh + sh run_eval_ascend.sh dataset_path ``` Evaluation result will be stored in the scripts path, whose folder name begins with "eval". You can find the result like the @@ -279,7 +342,7 @@ Parameters for both training and evaluation can be set in config.py. ## [Description of random situation](#contents) -BGCF model contains lots of dropout operations, if you want to disable dropout, set the neighbor_dropout to [0.0, 0.0, 0.0] in src/config.py. +BGCF model contains lots of dropout operations, if you want to disable dropout, set the neighbor_dropout to [0.0, 0.0, 0.0] in default_config.yaml. ## [ModelZoo Homepage](#contents) diff --git a/model_zoo/official/gnn/bgcf/README_CN.md b/model_zoo/official/gnn/bgcf/README_CN.md index f9040c9f89b..64e2c8ac1b7 100644 --- a/model_zoo/official/gnn/bgcf/README_CN.md +++ b/model_zoo/official/gnn/bgcf/README_CN.md @@ -104,10 +104,10 @@ BGCF包含两个主要模块。首先是抽样，它生成基于节点复制的 ```text # 使用Amazon-Beauty数据集运行训练示例 - sh run_train_ascend.sh + sh run_train_ascend.sh dataset_path # 使用Amazon-Beauty数据集运行评估示例 - sh run_eval_ascend.sh + sh run_eval_ascend.sh dataset_path ``` @@ -123,37 +123,96 @@ BGCF包含两个主要模块。首先是抽样，它生成基于节点复制的 ``` +- 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/)) + + - 在 ModelArts 上使用单卡训练（GPU or Ascend） + + ```python + # (1) 执行a或者b + # a. 在 default_config.yaml 文件中设置 "enable_modelarts=True" + # 在 default_config.yaml 文件中设置 "datapath='/cache/data/amazon_beauty/data_mr'" + # 在 default_config.yaml 文件中设置 "ckptpath='./ckpts'" + # (可选)如果选择GPU运行，在 default_config.yaml 文件中设置 "device_target='GPU'" + # (可选)如果选择GPU运行，在 default_config.yaml 文件中设置 "num_epoch=680" + # (可选)如果选择GPU运行，在 default_config.yaml 文件中设置 "dist_reg=0" + # 在 default_config.yaml 文件中设置 其他参数 + # b. 在网页上设置 "enable_modelarts=True" + # 在网页上设置 "datapath=/cache/data/amazon_beauty/data_mr" + # 在网页上设置 "ckptpath=./ckpts" + # (可选)如果选择GPU运行，在网页上设置 "device_target=GPU" + # (可选)如果选择GPU运行，在网页上设置 "num_epoch=680" + # (可选)如果选择GPU运行，在网页上设置 "dist_reg=0" + # 在网页上设置 其他参数 + # (2) 在本地准备转换好的数据集并将其压缩为一个文件，如："amazon_beauty.zip" (数据集转换代码可以参考上面的Dataset章节) + # (3) 上传你的压缩数据集到 S3 桶上 (你也可以上传原始的数据集，但那可能会很慢。) + # (4) 在网页上设置你的代码路径为 "/path/googlenet" + # (5) 在网页上设置启动文件为 "train.py" + # (6) 在网页上设置"训练数据集"、"训练输出文件路径"、"作业日志路径"等 + # (7) 创建训练作业 + ``` + + - 在 ModelArts 上使用单卡验证（GPU or Ascend） + + ```python + # (1) 执行a或者b + # a. 在 default_config.yaml 文件中设置 "enable_modelarts=True" + # 在 default_config.yaml 文件中设置 "datapath='/cache/data/amazon_beauty/data_mr'" + # 在 default_config.yaml 文件中设置 "ckptpath='./ckpts'" + # (可选)如果选择GPU运行，在 default_config.yaml 文件中设置 "device_target='GPU'" + # (可选)如果选择GPU运行，在 default_config.yaml 文件中设置 "num_epoch=680" + # (可选)如果选择GPU运行，在 default_config.yaml 文件中设置 "dist_reg=0" + # 在 default_config.yaml 文件中设置 其他参数 + # b. 在网页上设置 "enable_modelarts=True" + # 在网页上设置 "datapath=/cache/data/amazon_beauty/data_mr" + # 在网页上设置 "ckptpath=./ckpts" + # (可选)如果选择GPU运行，在网页上设置 "device_target=GPU" + # (可选)如果选择GPU运行，在网页上设置 "num_epoch=680" + # (可选)如果选择GPU运行，在网页上设置 "dist_reg=0" + # 在网页上设置 其他参数 + # (2) 在本地准备转换好的数据集并将其压缩为一个文件，如："amazon_beauty.zip" (数据集转换代码可以参考上面的Dataset章节) + # (3) 上传你的压缩数据集到 S3 桶上 (你也可以上传原始的数据集，但那可能会很慢。) + # (4) 在网页上设置你的代码路径为 "/path/googlenet" + # (5) 在网页上设置启动文件为 "eval.py" + # (6) 在网页上设置"训练数据集"、"训练输出文件路径"、"作业日志路径"等 + # (7) 创建训练作业 + ``` + ## 脚本说明 ### 脚本及样例代码 ```shell - └─bgcf ├─README.md + ├─README_CN.md + ├─model_utils + | ├─__init__.py # 初始化文件 + | ├─config.py # 参数获取文件 + | ├─device_adapter.py # modelarts 设备适配文件 + | ├─local_adapter.py # 本地适配文件 + | └─moxing_adapter.py # modelarts 模型适配文件 ├─scripts | ├─run_eval_ascend.sh # Ascend启动评估 | ├─run_eval_gpu.sh # GPU启动评估 | ├─run_process_data_ascend.sh # 生成MindRecord格式的数据集 | └─run_train_ascend.sh # Ascend启动训练 | └─run_train_gpu.sh # GPU启动训练 - | ├─src | ├─bgcf.py # BGCF模型 | ├─callback.py # 回调函数 - | ├─config.py # 训练配置 | ├─dataset.py # 数据预处理 | ├─metrics.py # 推荐指标 | └─utils.py # 训练BGCF的工具 - | + ├─default_config.yaml # 参数配置文件 + ├─mindspore_hub_conf.py # Mindspore hub文件 + ├─export.py # 导出网络 ├─eval.py # 评估网络 └─train.py # 训练网络 - ``` ### 脚本参数 -在config.py中可以同时配置训练参数和评估参数。 +在 default_config.yaml 中可以同时配置训练参数和评估参数。 - BGCF数据集配置 @@ -171,7 +230,7 @@ BGCF包含两个主要模块。首先是抽样，它生成基于节点复制的 ``` - 在config.py中以获取更多配置。 + 在 default_config.yaml 中以获取更多配置。 ### 训练过程 @@ -181,7 +240,7 @@ BGCF包含两个主要模块。首先是抽样，它生成基于节点复制的 ```python - sh run_train_ascend.sh + sh run_train_ascend.sh dataset_path ``` @@ -229,7 +288,7 @@ BGCF包含两个主要模块。首先是抽样，它生成基于节点复制的 ```python - sh run_eval_ascend.sh + sh run_eval_ascend.sh dataset_path ``` @@ -305,7 +364,7 @@ BGCF包含两个主要模块。首先是抽样，它生成基于节点复制的 ## 随机情况说明 -BGCF模型中有很多的dropout操作，如果想关闭dropout，可以在src/config.py中将neighbor_dropout设置为[0.0, 0.0, 0.0] 。 +BGCF模型中有很多的dropout操作，如果想关闭dropout，可以在 ./default_config.yaml 中将neighbor_dropout设置为[0.0, 0.0, 0.0] 。 ## ModelZoo主页 diff --git a/model_zoo/official/gnn/bgcf/default_config.yaml b/model_zoo/official/gnn/bgcf/default_config.yaml new file mode 100644 index 00000000000..afd8bfaf914 --- /dev/null +++ b/model_zoo/official/gnn/bgcf/default_config.yaml @@ -0,0 +1,61 @@ +# 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: "Ascend" +need_modelarts_dataset_unzip: True +modelarts_dataset_unzip_name: "amazon_beauty" + +# ============================================================================== +# options +dataset: "Beauty" +datapath: "./scripts/data_mr" +Ks: [5, 10, 20, 100] +workers: 8 +ckptpath: "../ckpts" +epsilon: 0.00000001 # 1e-8 +learning_rate: 0.001 # 1e-3 +l2: 0.03 +activation: "tanh" +neighbor_dropout: [0.0, 0.2, 0.3] +log_name: "test" +num_epoch: 600 +input_dim: 64 +batch_pairs: 5000 +eval_interval: 20 +num_neg: 10 +raw_neighs: 40 +gnew_neighs: 20 +embedded_dimension: 64 +dist_reg: 0.003 + +--- + +# Help description for each configuration +dataset: "choose which dataset" +datapath: "minddata path" +Ks: "top K" +workers: "number of process to generate data" +ckptpath: "checkpoint path" +epsilon: "optimizer parameter" +learning_rate: "learning rate" +l2: "l2 coefficient" +activation: "activation function, choices in ['relu', 'tanh']." +neighbor_dropout: "dropout ratio for different aggregation layer" +log_name: "log name" +num_epoch: "epoch sizes for training" +input_dim: "user and item embedding dimension, choices in [64, 128]" +batch_pairs: "batch size" +eval_interval: "evaluation interval" +num_neg: "negative sampling rate " +raw_neighs: "num of sampling neighbors in raw graph" +gnew_neighs: "num of sampling neighbors in sample graph" +embedded_dimension: "output embedding dim" +dist_reg: "distance loss coefficient" +device_target: "device target, choices in ['Ascend', GPU]" \ No newline at end of file diff --git a/model_zoo/official/gnn/bgcf/eval.py b/model_zoo/official/gnn/bgcf/eval.py index 5aaca4aa2d6..4b84f3335e3 100644 --- a/model_zoo/official/gnn/bgcf/eval.py +++ b/model_zoo/official/gnn/bgcf/eval.py @@ -15,6 +15,8 @@ """ BGCF evaluation script. """ +import os +import time import datetime import mindspore.context as context @@ -23,38 +25,117 @@ from mindspore.common import set_seed from src.bgcf import BGCF from src.utils import BGCFLogger -from src.config import parser_args from src.metrics import BGCFEvaluate from src.callback import ForwardBGCF, TestBGCF from src.dataset import TestGraphDataset, load_graph +from model_utils.config import config +from model_utils.moxing_adapter import moxing_wrapper +from model_utils.device_adapter import get_device_id, get_device_num + set_seed(1) +def modelarts_pre_process(): + '''modelarts pre process function.''' + def unzip(zip_file, save_dir): + import zipfile + s_time = time.time() + if not os.path.exists(os.path.join(save_dir, config.modelarts_dataset_unzip_name)): + zip_isexist = zipfile.is_zipfile(zip_file) + if zip_isexist: + fz = zipfile.ZipFile(zip_file, 'r') + data_num = len(fz.namelist()) + print("Extract Start...") + print("unzip file num: {}".format(data_num)) + data_print = int(data_num / 100) if data_num > 100 else 1 + i = 0 + for file in fz.namelist(): + if i % data_print == 0: + print("unzip percent: {}%".format(int(i * 100 / data_num)), flush=True) + i += 1 + fz.extract(file, save_dir) + print("cost time: {}min:{}s.".format(int((time.time() - s_time) / 60), + int(int(time.time() - s_time) % 60))) + print("Extract Done.") + else: + print("This is not zip.") + else: + print("Zip has been extracted.") + + if config.need_modelarts_dataset_unzip: + zip_file_1 = os.path.join(config.data_path, config.modelarts_dataset_unzip_name + ".zip") + save_dir_1 = os.path.join(config.data_path) + + sync_lock = "/tmp/unzip_sync.lock" + + # Each server contains 8 devices as most. + if get_device_id() % min(get_device_num(), 8) == 0 and not os.path.exists(sync_lock): + print("Zip file path: ", zip_file_1) + print("Unzip file save dir: ", save_dir_1) + unzip(zip_file_1, save_dir_1) + print("===Finish extract data synchronization===") + try: + os.mknod(sync_lock) + except IOError: + pass + + while True: + if os.path.exists(sync_lock): + break + time.sleep(1) + + print("Device: {}, Finish sync unzip data from {} to {}.".format(get_device_id(), zip_file_1, save_dir_1)) + + +@moxing_wrapper(pre_process=modelarts_pre_process) def evaluation(): """evaluation""" + context.set_context(mode=context.GRAPH_MODE, + device_target=config.device_target, + save_graphs=False) + if config.device_target == "Ascend": + context.set_context(device_id=get_device_id()) + + train_graph, test_graph, sampled_graph_list = load_graph(config.datapath) + test_graph_dataset = TestGraphDataset(train_graph, sampled_graph_list, num_samples=config.raw_neighs, + num_bgcn_neigh=config.gnew_neighs, + num_neg=config.num_neg) + + if config.log_name: + now = datetime.datetime.now().strftime("%b_%d_%H_%M_%S") + name = "bgcf" + '-' + config.log_name + '-' + config.dataset + log_save_path = './log-files/' + name + '/' + now + log = BGCFLogger(logname=name, now=now, foldername='log-files', copy=False) + log.open(log_save_path + '/log.train.txt', mode='a') + for arg in vars(config): + log.write(arg + '=' + str(getattr(config, arg)) + '\n') + else: + for arg in vars(config): + print(arg + '=' + str(getattr(config, arg))) + num_user = train_graph.graph_info()["node_num"][0] num_item = train_graph.graph_info()["node_num"][1] - eval_class = BGCFEvaluate(parser, train_graph, test_graph, parser.Ks) - for _epoch in range(parser.eval_interval, parser.num_epoch+1, parser.eval_interval) \ - if parser.device_target == "Ascend" else range(parser.num_epoch, parser.num_epoch+1): - bgcfnet_test = BGCF([parser.input_dim, num_user, num_item], - parser.embedded_dimension, - parser.activation, + eval_class = BGCFEvaluate(config, train_graph, test_graph, config.Ks) + for _epoch in range(config.eval_interval, config.num_epoch+1, config.eval_interval) \ + if config.device_target == "Ascend" else range(config.num_epoch, config.num_epoch+1): + bgcfnet_test = BGCF([config.input_dim, num_user, num_item], + config.embedded_dimension, + config.activation, [0.0, 0.0, 0.0], num_user, num_item, - parser.input_dim) + config.input_dim) - load_checkpoint(parser.ckptpath + "/bgcf_epoch{}.ckpt".format(_epoch), net=bgcfnet_test) + load_checkpoint(config.ckptpath + "/bgcf_epoch{}.ckpt".format(_epoch), net=bgcfnet_test) forward_net = ForwardBGCF(bgcfnet_test) - user_reps, item_reps = TestBGCF(forward_net, num_user, num_item, parser.input_dim, test_graph_dataset) + user_reps, item_reps = TestBGCF(forward_net, num_user, num_item, config.input_dim, test_graph_dataset) test_recall_bgcf, test_ndcg_bgcf, \ - test_sedp, test_nov = eval_class.eval_with_rep(user_reps, item_reps, parser) + test_sedp, test_nov = eval_class.eval_with_rep(user_reps, item_reps, config) - if parser.log_name: + if config.log_name: log.write( 'epoch:%03d, recall_@10:%.5f, recall_@20:%.5f, ndcg_@10:%.5f, ndcg_@20:%.5f, ' 'sedp_@10:%.5f, sedp_@20:%.5f, nov_@10:%.5f, nov_@20:%.5f\n' % (_epoch, @@ -80,28 +161,4 @@ def evaluation(): if __name__ == "__main__": - parser = parser_args() - context.set_context(mode=context.GRAPH_MODE, - device_target=parser.device_target, - save_graphs=False) - if parser.device_target == "Ascend": - context.set_context(device_id=int(parser.device)) - - train_graph, test_graph, sampled_graph_list = load_graph(parser.datapath) - test_graph_dataset = TestGraphDataset(train_graph, sampled_graph_list, num_samples=parser.raw_neighs, - num_bgcn_neigh=parser.gnew_neighs, - num_neg=parser.num_neg) - - if parser.log_name: - now = datetime.datetime.now().strftime("%b_%d_%H_%M_%S") - name = "bgcf" + '-' + parser.log_name + '-' + parser.dataset - log_save_path = './log-files/' + name + '/' + now - log = BGCFLogger(logname=name, now=now, foldername='log-files', copy=False) - log.open(log_save_path + '/log.train.txt', mode='a') - for arg in vars(parser): - log.write(arg + '=' + str(getattr(parser, arg)) + '\n') - else: - for arg in vars(parser): - print(arg + '=' + str(getattr(parser, arg))) - evaluation() diff --git a/model_zoo/official/gnn/bgcf/model_utils/__init__.py b/model_zoo/official/gnn/bgcf/model_utils/__init__.py new file mode 100644 index 00000000000..e69de29bb2d diff --git a/model_zoo/official/gnn/bgcf/model_utils/config.py b/model_zoo/official/gnn/bgcf/model_utils/config.py new file mode 100644 index 00000000000..ad0d7497a8e --- /dev/null +++ b/model_zoo/official/gnn/bgcf/model_utils/config.py @@ -0,0 +1,126 @@ +# 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. +# ============================================================================ + +"""Parse arguments""" + +import os +import ast +import argparse +from pprint import pformat +import yaml + +class Config: + """ + Configuration namespace. Convert dictionary to members. + """ + def __init__(self, cfg_dict): + for k, v in cfg_dict.items(): + if isinstance(v, (list, tuple)): + setattr(self, k, [Config(x) if isinstance(x, dict) else x for x in v]) + else: + setattr(self, k, Config(v) if isinstance(v, dict) else v) + + def __str__(self): + return pformat(self.__dict__) + + def __repr__(self): + return self.__str__() + + +def parse_cli_to_yaml(parser, cfg, helper=None, choices=None, cfg_path="default_config.yaml"): + """ + Parse command line arguments to the configuration according to the default yaml. + + Args: + parser: Parent parser. + cfg: Base configuration. + helper: Helper description. + cfg_path: Path to the default yaml config. + """ + parser = argparse.ArgumentParser(description="[REPLACE THIS at config.py]", + parents=[parser]) + helper = {} if helper is None else helper + choices = {} if choices is None else choices + for item in cfg: + if not isinstance(cfg[item], list) and not isinstance(cfg[item], dict): + help_description = helper[item] if item in helper else "Please reference to {}".format(cfg_path) + choice = choices[item] if item in choices else None + if isinstance(cfg[item], bool): + parser.add_argument("--" + item, type=ast.literal_eval, default=cfg[item], choices=choice, + help=help_description) + else: + parser.add_argument("--" + item, type=type(cfg[item]), default=cfg[item], choices=choice, + help=help_description) + args = parser.parse_args() + return args + + +def parse_yaml(yaml_path): + """ + Parse the yaml config file. + + Args: + yaml_path: Path to the yaml config. + """ + with open(yaml_path, 'r') as fin: + try: + cfgs = yaml.load_all(fin.read(), Loader=yaml.FullLoader) + cfgs = [x for x in cfgs] + if len(cfgs) == 1: + cfg_helper = {} + cfg = cfgs[0] + cfg_choices = {} + elif len(cfgs) == 2: + cfg, cfg_helper = cfgs + cfg_choices = {} + elif len(cfgs) == 3: + cfg, cfg_helper, cfg_choices = cfgs + else: + raise ValueError("At most 3 docs (config, description for help, choices) are supported in config yaml") + print(cfg_helper) + except: + raise ValueError("Failed to parse yaml") + return cfg, cfg_helper, cfg_choices + + +def merge(args, cfg): + """ + Merge the base config from yaml file and command line arguments. + + Args: + args: Command line arguments. + cfg: Base configuration. + """ + args_var = vars(args) + for item in args_var: + cfg[item] = args_var[item] + return cfg + + +def get_config(): + """ + Get Config according to the yaml file and cli arguments. + """ + parser = argparse.ArgumentParser(description="default name", add_help=False) + current_dir = os.path.dirname(os.path.abspath(__file__)) + parser.add_argument("--config_path", type=str, default=os.path.join(current_dir, "../default_config.yaml"), + help="Config file path") + path_args, _ = parser.parse_known_args() + default, helper, choices = parse_yaml(path_args.config_path) + args = parse_cli_to_yaml(parser=parser, cfg=default, helper=helper, choices=choices, cfg_path=path_args.config_path) + final_config = merge(args, default) + return Config(final_config) + +config = get_config() diff --git a/model_zoo/official/gnn/bgcf/model_utils/device_adapter.py b/model_zoo/official/gnn/bgcf/model_utils/device_adapter.py new file mode 100644 index 00000000000..7c5d7f837dd --- /dev/null +++ b/model_zoo/official/gnn/bgcf/model_utils/device_adapter.py @@ -0,0 +1,27 @@ +# 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. +# ============================================================================ + +"""Device adapter for ModelArts""" + +from .config import config + +if config.enable_modelarts: + from .moxing_adapter import get_device_id, get_device_num, get_rank_id, get_job_id +else: + from .local_adapter import get_device_id, get_device_num, get_rank_id, get_job_id + +__all__ = [ + "get_device_id", "get_device_num", "get_rank_id", "get_job_id" +] diff --git a/model_zoo/official/gnn/bgcf/model_utils/local_adapter.py b/model_zoo/official/gnn/bgcf/model_utils/local_adapter.py new file mode 100644 index 00000000000..769fa6dc78e --- /dev/null +++ b/model_zoo/official/gnn/bgcf/model_utils/local_adapter.py @@ -0,0 +1,36 @@ +# 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. +# ============================================================================ + +"""Local adapter""" + +import os + +def get_device_id(): + device_id = os.getenv('DEVICE_ID', '0') + return int(device_id) + + +def get_device_num(): + device_num = os.getenv('RANK_SIZE', '1') + return int(device_num) + + +def get_rank_id(): + global_rank_id = os.getenv('RANK_ID', '0') + return int(global_rank_id) + + +def get_job_id(): + return "Local Job" diff --git a/model_zoo/official/gnn/bgcf/model_utils/moxing_adapter.py b/model_zoo/official/gnn/bgcf/model_utils/moxing_adapter.py new file mode 100644 index 00000000000..25838a7da99 --- /dev/null +++ b/model_zoo/official/gnn/bgcf/model_utils/moxing_adapter.py @@ -0,0 +1,116 @@ +# 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. +# ============================================================================ + +"""Moxing adapter for ModelArts""" + +import os +import functools +from mindspore import context +from .config import config + +_global_sync_count = 0 + +def get_device_id(): + device_id = os.getenv('DEVICE_ID', '0') + return int(device_id) + + +def get_device_num(): + device_num = os.getenv('RANK_SIZE', '1') + return int(device_num) + + +def get_rank_id(): + global_rank_id = os.getenv('RANK_ID', '0') + return int(global_rank_id) + + +def get_job_id(): + job_id = os.getenv('JOB_ID') + job_id = job_id if job_id != "" else "default" + return job_id + +def sync_data(from_path, to_path): + """ + Download data from remote obs to local directory if the first url is remote url and the second one is local path + Upload data from local directory to remote obs in contrast. + """ + import moxing as mox + import time + global _global_sync_count + sync_lock = "/tmp/copy_sync.lock" + str(_global_sync_count) + _global_sync_count += 1 + + # Each server contains 8 devices as most. + if get_device_id() % min(get_device_num(), 8) == 0 and not os.path.exists(sync_lock): + print("from path: ", from_path) + print("to path: ", to_path) + mox.file.copy_parallel(from_path, to_path) + print("===finish data synchronization===") + try: + os.mknod(sync_lock) + except IOError: + pass + print("===save flag===") + + while True: + if os.path.exists(sync_lock): + break + time.sleep(1) + + print("Finish sync data from {} to {}.".format(from_path, to_path)) + + +def moxing_wrapper(pre_process=None, post_process=None): + """ + Moxing wrapper to download dataset and upload outputs. + """ + def wrapper(run_func): + @functools.wraps(run_func) + def wrapped_func(*args, **kwargs): + # Download data from data_url + if config.enable_modelarts: + if config.data_url: + sync_data(config.data_url, config.data_path) + print("Dataset downloaded: ", os.listdir(config.data_path)) + if config.checkpoint_url: + sync_data(config.checkpoint_url, config.load_path) + print("Preload downloaded: ", os.listdir(config.load_path)) + if config.train_url: + sync_data(config.train_url, config.output_path) + print("Workspace downloaded: ", os.listdir(config.output_path)) + + context.set_context(save_graphs_path=os.path.join(config.output_path, str(get_rank_id()))) + config.device_num = get_device_num() + config.device_id = get_device_id() + if not os.path.exists(config.output_path): + os.makedirs(config.output_path) + + if pre_process: + pre_process() + + # Run the main function + run_func(*args, **kwargs) + + # Upload data to train_url + if config.enable_modelarts: + if post_process: + post_process() + + if config.train_url: + print("Start to copy output directory") + sync_data(config.output_path, config.train_url) + return wrapped_func + return wrapper diff --git a/model_zoo/official/gnn/bgcf/scripts/run_eval_ascend.sh b/model_zoo/official/gnn/bgcf/scripts/run_eval_ascend.sh index 18e28b7bdc7..335e2701c51 100644 --- a/model_zoo/official/gnn/bgcf/scripts/run_eval_ascend.sh +++ b/model_zoo/official/gnn/bgcf/scripts/run_eval_ascend.sh @@ -14,6 +14,13 @@ # limitations under the License. # ============================================================================ +if [ $# != 1 ] +then + echo "Usage: sh run_train_ascend.sh [DATASET_PATH]" + exit 1 +fi +DATASET_PATH=$1 + ulimit -u unlimited export DEVICE_NUM=1 export RANK_SIZE=$DEVICE_NUM @@ -26,12 +33,14 @@ fi mkdir ./eval cp ../*.py ./eval +cp ../*.yaml ./eval cp *.sh ./eval cp -r ../src ./eval +cp -r ../model_utils ./eval cd ./eval || exit env > env.log echo "start evaluation" -python eval.py --datapath=../data_mr --ckptpath=../ckpts &> log & +python eval.py --datapath=$DATASET_PATH --ckptpath=../ckpts &> log & cd .. diff --git a/model_zoo/official/gnn/bgcf/scripts/run_eval_gpu.sh b/model_zoo/official/gnn/bgcf/scripts/run_eval_gpu.sh index 5088286ab42..7c8de03f661 100644 --- a/model_zoo/official/gnn/bgcf/scripts/run_eval_gpu.sh +++ b/model_zoo/official/gnn/bgcf/scripts/run_eval_gpu.sh @@ -32,8 +32,10 @@ fi mkdir ./eval cp ../*.py ./eval +cp ../*.yaml ./eval cp *.sh ./eval cp -r ../src ./eval +cp -r ../model_utils ./eval cd ./eval || exit env > env.log echo "start evaluation" diff --git a/model_zoo/official/gnn/bgcf/scripts/run_train_ascend.sh b/model_zoo/official/gnn/bgcf/scripts/run_train_ascend.sh index eef920ba32c..c6dc9b6fea1 100644 --- a/model_zoo/official/gnn/bgcf/scripts/run_train_ascend.sh +++ b/model_zoo/official/gnn/bgcf/scripts/run_train_ascend.sh @@ -14,6 +14,13 @@ # limitations under the License. # ============================================================================ +if [ $# != 1 ] +then + echo "Usage: sh run_train_ascend.sh [DATASET_PATH]" + exit 1 +fi +DATASET_PATH=$1 + ulimit -u unlimited export DEVICE_NUM=1 export RANK_SIZE=$DEVICE_NUM @@ -32,12 +39,14 @@ fi mkdir ./ckpts cp ../*.py ./train +cp ../*.yaml ./train cp *.sh ./train cp -r ../src ./train +cp -r ../model_utils ./train cd ./train || exit env > env.log echo "start training" -python train.py --datapath=../data_mr --ckptpath=../ckpts &> log & +python train.py --datapath=$DATASET_PATH --ckptpath=../ckpts &> log & cd .. diff --git a/model_zoo/official/gnn/bgcf/scripts/run_train_gpu.sh b/model_zoo/official/gnn/bgcf/scripts/run_train_gpu.sh index 0a54e835b01..aa46ae3b7d9 100644 --- a/model_zoo/official/gnn/bgcf/scripts/run_train_gpu.sh +++ b/model_zoo/official/gnn/bgcf/scripts/run_train_gpu.sh @@ -36,8 +36,10 @@ fi mkdir ./ckpts cp ../*.py ./train +cp ../*.yaml ./train cp *.sh ./train cp -r ../src ./train +cp -r ../model_utils ./train cd ./train || exit env > env.log echo "start training" diff --git a/model_zoo/official/gnn/bgcf/train.py b/model_zoo/official/gnn/bgcf/train.py index 30ba81b326f..61bedf96f17 100644 --- a/model_zoo/official/gnn/bgcf/train.py +++ b/model_zoo/official/gnn/bgcf/train.py @@ -15,6 +15,7 @@ """ BGCF training script. """ +import os import time from mindspore import Tensor @@ -24,35 +25,107 @@ from mindspore.train.serialization import save_checkpoint from mindspore.common import set_seed from src.bgcf import BGCF -from src.config import parser_args from src.utils import convert_item_id from src.callback import TrainBGCF from src.dataset import load_graph, create_dataset +from model_utils.config import config +from model_utils.moxing_adapter import moxing_wrapper +from model_utils.device_adapter import get_device_id, get_device_num + set_seed(1) -def train(): + +def modelarts_pre_process(): + '''modelarts pre process function.''' + def unzip(zip_file, save_dir): + import zipfile + s_time = time.time() + if not os.path.exists(os.path.join(save_dir, config.modelarts_dataset_unzip_name)): + zip_isexist = zipfile.is_zipfile(zip_file) + if zip_isexist: + fz = zipfile.ZipFile(zip_file, 'r') + data_num = len(fz.namelist()) + print("Extract Start...") + print("unzip file num: {}".format(data_num)) + data_print = int(data_num / 100) if data_num > 100 else 1 + i = 0 + for file in fz.namelist(): + if i % data_print == 0: + print("unzip percent: {}%".format(int(i * 100 / data_num)), flush=True) + i += 1 + fz.extract(file, save_dir) + print("cost time: {}min:{}s.".format(int((time.time() - s_time) / 60), + int(int(time.time() - s_time) % 60))) + print("Extract Done.") + else: + print("This is not zip.") + else: + print("Zip has been extracted.") + + if config.need_modelarts_dataset_unzip: + zip_file_1 = os.path.join(config.data_path, config.modelarts_dataset_unzip_name + ".zip") + save_dir_1 = os.path.join(config.data_path) + + sync_lock = "/tmp/unzip_sync.lock" + + # Each server contains 8 devices as most. + if get_device_id() % min(get_device_num(), 8) == 0 and not os.path.exists(sync_lock): + print("Zip file path: ", zip_file_1) + print("Unzip file save dir: ", save_dir_1) + unzip(zip_file_1, save_dir_1) + print("===Finish extract data synchronization===") + try: + os.mknod(sync_lock) + except IOError: + pass + + while True: + if os.path.exists(sync_lock): + break + time.sleep(1) + + print("Device: {}, Finish sync unzip data from {} to {}.".format(get_device_id(), zip_file_1, save_dir_1)) + + config.ckptpath = os.path.join(config.output_path, config.ckptpath) + if not os.path.isdir(config.ckptpath): + os.makedirs(config.ckptpath) + + +@moxing_wrapper(pre_process=modelarts_pre_process) +def run_train(): """Train""" + context.set_context(mode=context.GRAPH_MODE, + device_target=config.device_target, + save_graphs=False) + + if config.device_target == "Ascend": + context.set_context(device_id=get_device_id()) + + train_graph, _, sampled_graph_list = load_graph(config.datapath) + train_ds = create_dataset(train_graph, sampled_graph_list, config.workers, batch_size=config.batch_pairs, + num_samples=config.raw_neighs, num_bgcn_neigh=config.gnew_neighs, num_neg=config.num_neg) + num_user = train_graph.graph_info()["node_num"][0] num_item = train_graph.graph_info()["node_num"][1] num_pairs = train_graph.graph_info()['edge_num'][0] - bgcfnet = BGCF([parser.input_dim, num_user, num_item], - parser.embedded_dimension, - parser.activation, - parser.neighbor_dropout, + bgcfnet = BGCF([config.input_dim, num_user, num_item], + config.embedded_dimension, + config.activation, + config.neighbor_dropout, num_user, num_item, - parser.input_dim) + config.input_dim) - train_net = TrainBGCF(bgcfnet, parser.num_neg, parser.l2, parser.learning_rate, - parser.epsilon, parser.dist_reg) + train_net = TrainBGCF(bgcfnet, config.num_neg, config.l2, config.learning_rate, + config.epsilon, config.dist_reg) train_net.set_train(True) - itr = train_ds.create_dict_iterator(parser.num_epoch, output_numpy=True) - num_iter = int(num_pairs / parser.batch_pairs) + itr = train_ds.create_dict_iterator(config.num_epoch, output_numpy=True) + num_iter = int(num_pairs / config.batch_pairs) - for _epoch in range(1, parser.num_epoch + 1): + for _epoch in range(1, config.num_epoch + 1): epoch_start = time.time() iter_num = 1 @@ -98,22 +171,9 @@ def train(): '{}, cost:{:.4f}'.format(train_loss, time.time() - epoch_start)) iter_num += 1 - if _epoch % parser.eval_interval == 0: - save_checkpoint(bgcfnet, parser.ckptpath + "/bgcf_epoch{}.ckpt".format(_epoch)) + if _epoch % config.eval_interval == 0: + save_checkpoint(bgcfnet, config.ckptpath + "/bgcf_epoch{}.ckpt".format(_epoch)) if __name__ == "__main__": - parser = parser_args() - - context.set_context(mode=context.GRAPH_MODE, - device_target=parser.device_target, - save_graphs=False) - - if parser.device_target == "Ascend": - context.set_context(device_id=int(parser.device)) - - train_graph, _, sampled_graph_list = load_graph(parser.datapath) - train_ds = create_dataset(train_graph, sampled_graph_list, parser.workers, batch_size=parser.batch_pairs, - num_samples=parser.raw_neighs, num_bgcn_neigh=parser.gnew_neighs, num_neg=parser.num_neg) - - train() + run_train() diff --git a/tests/st/model_zoo_tests/bgcf/test_BGCF_amazon_beauty.py b/tests/st/model_zoo_tests/bgcf/test_BGCF_amazon_beauty.py index 0b167db106b..39d9dd690a2 100644 --- a/tests/st/model_zoo_tests/bgcf/test_BGCF_amazon_beauty.py +++ b/tests/st/model_zoo_tests/bgcf/test_BGCF_amazon_beauty.py @@ -29,10 +29,10 @@ def test_BGCF_amazon_beauty(): utils.copy_files(model_path, cur_path, model_name) cur_model_path = os.path.join(cur_path, model_name) - old_list = ["default=600,"] - new_list = ["default=50,"] - utils.exec_sed_command(old_list, new_list, os.path.join(cur_model_path, "src/config.py")) - old_list = ["context.set_context(device_id=int(parser.device))", + old_list = ["num_epoch: 600"] + new_list = ["num_epoch: 50"] + utils.exec_sed_command(old_list, new_list, os.path.join(cur_model_path, "default_config.yaml")) + old_list = ["context.set_context(device_id=get_device_id())", "save_checkpoint("] new_list = ["context.set_context()", "pass \\# save_checkpoint("]