Browse Source

Merge branch 'V20221228' into zpz-issue-1228

pull/3437/head
zhoupzh 2 years ago
parent
commit
31d1ae4736
22 changed files with 730 additions and 211 deletions
  1. +18
    -9
      models/cloudbrain.go
  2. +6
    -1
      models/user.go
  3. +1
    -0
      modules/cloudbrain/cloudbrain.go
  4. +7
    -5
      modules/convert/cloudbrain.go
  5. +1
    -83
      modules/modelarts/modelarts.go
  6. +1
    -0
      modules/modelarts_cd/modelarts.go
  7. +198
    -0
      modules/notebook/contentManager.go
  8. +28
    -27
      modules/structs/cloudbrain.go
  9. +2
    -1
      options/locale/locale_en-US.ini
  10. +3
    -2
      options/locale/locale_zh-CN.ini
  11. +5
    -1
      routers/api/v1/api.go
  12. +3
    -0
      routers/api/v1/repo/cloudbrain.go
  13. +32
    -0
      routers/api/v1/repo/modelmanage.go
  14. +20
    -0
      routers/api/v1/user/repo.go
  15. +20
    -13
      routers/repo/ai_model_manage.go
  16. +20
    -2
      routers/repo/cloudbrain.go
  17. +46
    -4
      routers/repo/modelarts.go
  18. +5
    -1
      routers/routes/routes.go
  19. +240
    -28
      services/cloudbrain/cloudbrainTask/notebook.go
  20. +3
    -3
      templates/repo/modelarts/trainjob/show.tmpl
  21. +11
    -2
      web_src/vuepages/apis/modules/notobook.js
  22. +60
    -29
      web_src/vuepages/pages/notebook/debug/index.vue

+ 18
- 9
models/cloudbrain.go View File

@@ -205,7 +205,7 @@ type Cloudbrain struct {
BenchmarkTypeRankLink string `xorm:"-"`
StartTime timeutil.TimeStamp
EndTime timeutil.TimeStamp
Cleared bool `xorm:"DEFAULT false"`
Cleared bool `xorm:"DEFAULT false"`
Spec *Specification `xorm:"-"`
}

@@ -1823,7 +1823,7 @@ func QueryModelTrainJobVersionList(jobId string) ([]*Cloudbrain, int, error) {
return cloudbrains, int(len(cloudbrains)), nil
}

func QueryModelTrainJobList(repoId int64) ([]*CloudbrainInfo, int, error) {
func QueryModelTrainJobList(repoId int64) ([]*Cloudbrain, int, error) {
sess := x.NewSession()
defer sess.Close()
var cond = builder.NewCond()
@@ -1840,14 +1840,14 @@ func QueryModelTrainJobList(repoId int64) ([]*CloudbrainInfo, int, error) {
// builder.In("type", 0, 1),
// )

cloudbrains := make([]*CloudbrainInfo, 0)
cloudbrains := make([]*Cloudbrain, 0)
if err := sess.Select("job_id,display_job_name").Table(&Cloudbrain{}).Where(cond).OrderBy("created_unix DESC").
Find(&cloudbrains); err != nil {
return nil, 0, fmt.Errorf("Find: %v", err)
}

keys := make(map[string]string)
uniqueElements := make([]*CloudbrainInfo, 0)
uniqueElements := make([]*Cloudbrain, 0)
for _, entry := range cloudbrains {
if _, value := keys[entry.JobID]; !value {
keys[entry.JobID] = entry.DisplayJobName
@@ -1988,7 +1988,7 @@ func GetCloudbrainByID(id string) (*Cloudbrain, error) {
return getRepoCloudBrain(cb)
}

func IsCloudbrainExistByJobName(jobName string)(bool,error){
func IsCloudbrainExistByJobName(jobName string) (bool, error) {
return x.Unscoped().Exist(&Cloudbrain{
JobName: jobName,
})
@@ -2126,6 +2126,15 @@ func GetCloudbrainByName(jobName string) (*Cloudbrain, error) {
cb := &Cloudbrain{JobName: jobName}
return getRepoCloudBrain(cb)
}
func GetWaitOrRunFileNotebookByRepo(repoId int64, cloudbrainType int) (*Cloudbrain, error) {
cloudBrain := new(Cloudbrain)
has, err := x.In("status", JobWaiting, JobRunning, ModelArtsCreateQueue, ModelArtsCreating, ModelArtsStarting,
ModelArtsReadyToStart, ModelArtsResizing, ModelArtsStartQueuing, ModelArtsRunning, ModelArtsDeleting, ModelArtsRestarting).Where("repo_id=? and type=? and boot_file!=''", repoId, cloudbrainType).Get(cloudBrain)
if has {
return cloudBrain, err
}
return nil, err
}

func CanDelJob(isSigned bool, user *User, job *CloudbrainInfo) bool {
if !isSigned || (job.Status != string(JobStopped) && job.Status != string(JobFailed) && job.Status != string(ModelArtsStartFailed) && job.Status != string(ModelArtsCreateFailed)) {
@@ -2172,25 +2181,25 @@ func GetCloudBrainOneStoppedNotDebugJobDaysAgo(days int, limit int) ([]*Cloudbra
Limit(limit).
Find(&cloudbrains)
}

/**
本方法考虑了再次调试的情况,多次调试取最后一次的任务的结束时间
*/
*/
func GetCloudBrainOneStoppedDebugJobDaysAgo(days int, limit int) ([]*Cloudbrain, error) {
cloudbrains := make([]*Cloudbrain, 0, 10)
endTimeBefore := time.Now().Unix() - int64(days)*24*3600
missEndTimeBefore := endTimeBefore - 24*3600
sql:=`SELECT id,job_name,job_id from (SELECT DISTINCT ON (job_name)
sql := `SELECT id,job_name,job_id from (SELECT DISTINCT ON (job_name)
id, job_name, job_id,status,end_time,updated_unix,cleared
FROM cloudbrain
where type=0 and job_type='DEBUG'
ORDER BY job_name, updated_unix DESC) a
where status in ('STOPPED','SUCCEEDED','FAILED') and (((end_time is null or end_time=0) and updated_unix<? and updated_unix != 0 ) or (end_time<? and end_time != 0)) and cleared=false`

return cloudbrains, x.Unscoped().SQL(sql,missEndTimeBefore, endTimeBefore).Limit(limit).Find(&cloudbrains)
return cloudbrains, x.Unscoped().SQL(sql, missEndTimeBefore, endTimeBefore).Limit(limit).Find(&cloudbrains)

}


func UpdateCloudBrainRecordsCleared(ids []int64) error {
pageSize := 150
n := len(ids) / pageSize


+ 6
- 1
models/user.go View File

@@ -16,6 +16,7 @@ import (
"fmt"
_ "image/jpeg" // Needed for jpeg support
"image/png"
"math/rand"
"os"
"path/filepath"
"regexp"
@@ -495,7 +496,11 @@ func (u *User) RealSizedAvatarLink(size int) string {
// may either be a sub-URL to this site, or a full URL to an external avatar
// service.
func (u *User) RelAvatarLink() string {
return u.SizedRelAvatarLink(base.DefaultAvatarSize)
append := ""
if u.UseCustomAvatar {
append = "?" + fmt.Sprint(rand.Intn(100))
}
return u.SizedRelAvatarLink(base.DefaultAvatarSize) + append
}

// AvatarLink returns user avatar absolute link.


+ 1
- 0
modules/cloudbrain/cloudbrain.go View File

@@ -709,6 +709,7 @@ type GenerateModelArtsNotebookReq struct {

ImageId string
AutoStopDurationMs int64
BranchName string

Spec *models.Specification
ModelName string


+ 7
- 5
modules/convert/cloudbrain.go View File

@@ -28,14 +28,13 @@ func ToCloudBrain(task *models.Cloudbrain) *api.Cloudbrain {
BootFile: task.BootFile,
Description: task.Description,
ModelName: task.ModelName,
ModelVersion: task.ModelVersion,
CkptName: task.CkptName,
VersionName: task.VersionName,
ModelVersion: task.ModelVersion,
CkptName: task.CkptName,

StartTime: int64(task.StartTime),
EndTime: int64(task.EndTime),

Spec: ToSpecification(task.Spec),
Spec: ToSpecification(task.Spec),
}
}
func ToAttachment(attachment *models.Attachment) *api.AttachmentShow {
@@ -89,6 +88,9 @@ func ToDataset(dataset *models.Dataset) *api.Dataset {
}

func ToSpecification(s *models.Specification) *api.SpecificationShow {
if s == nil {
return nil
}
return &api.SpecificationShow{
ID: s.ID,
AccCardsNum: s.AccCardsNum,


+ 1
- 83
modules/modelarts/modelarts.go View File

@@ -1,13 +1,9 @@
package modelarts

import (
"encoding/base64"
"encoding/json"
"errors"
"fmt"
"io/ioutil"
"net/http"
"path"
"strconv"
"strings"

@@ -239,6 +235,7 @@ func GenerateNotebook2(ctx *context.Context, req cloudbrain.GenerateModelArtsNot
ComputeResource: models.NPUResource,
Image: imageName,
BootFile: req.BootFile,
BranchName: req.BranchName,
Description: req.Description,
CreatedUnix: createTime,
UpdatedUnix: createTime,
@@ -830,10 +827,6 @@ func HandleNotebookInfo(task *models.Cloudbrain) error {
task.FlavorCode = result.Flavor
}

if oldStatus != task.Status && task.Status == string(models.ModelArtsRunning) && task.BootFile != "" {
uploadNoteBookFile(task, result)

}
err = models.UpdateJob(task)
if err != nil {
log.Error("UpdateJob(%s) failed:%v", task.DisplayJobName, err)
@@ -844,81 +837,6 @@ func HandleNotebookInfo(task *models.Cloudbrain) error {
return nil
}

func uploadNoteBookFile(task *models.Cloudbrain, result *models.GetNotebook2Result) {
jupyterUrl := result.Url + "?token=" + result.Token

cookies, xsrf := getCookiesAndCsrf(jupyterUrl)
if xsrf == "" {
log.Error("browser jupyterUrl failed:%v", task.DisplayJobName)
} else {

codePath := setting.JobPath + task.JobName + cloudbrain.CodeMountPath
fileContents, err := ioutil.ReadFile(codePath + "/" + task.BootFile)
if err != nil {
log.Error("read jupyter file failed:%v", task.DisplayJobName, err)
}

base64Content := base64.StdEncoding.EncodeToString(fileContents)
client := getRestyClient()
uploadUrl := getJupyterBaseUrl(result.Url) + "api/contents/" + path.Base(task.BootFile)
res, err := client.R().
SetCookies(cookies).
SetHeader("X-XSRFToken", xsrf).
SetBody(map[string]interface{}{
"type": "file",
"format": "base64",
"name": path.Base(task.BootFile),
"path": path.Base(task.BootFile),
"content": base64Content}).
Put(uploadUrl)
if err != nil {
log.Error("upload jupyter file failed:%v", task.DisplayJobName, err)
} else if res.StatusCode() != http.StatusCreated {
log.Error("upload jupyter file failed:%v", task.DisplayJobName, err)
}

}

}

func getJupyterBaseUrl(url string) string {
jupyterUrlLength := len(url)
baseUrl := url[0 : jupyterUrlLength-len(path.Base(url))]
return baseUrl
}

func getCookiesAndCsrf(jupyterUrl string) ([]*http.Cookie, string) {
log.Info("jupyter url:" + jupyterUrl)
var cookies []*http.Cookie
const retryTimes = 10
for i := 0; i < retryTimes; i++ {
res, err := http.Get(jupyterUrl)
if err != nil {
log.Error("browser jupyterUrl failed.", err)
if i == retryTimes-1 {
return cookies, ""
}

} else {
cookies = res.Cookies()
xsrf := ""
for _, cookie := range cookies {
if cookie.Name == "_xsrf" {
xsrf = cookie.Value
break
}

}
if xsrf != "" {
return cookies, xsrf
}

}
}
return cookies, ""

}

func SyncTempStatusJob() {
jobs, err := models.GetCloudBrainTempJobs()
if err != nil {


+ 1
- 0
modules/modelarts_cd/modelarts.go View File

@@ -148,6 +148,7 @@ func GenerateNotebook(ctx *context.Context, req cloudbrain.GenerateModelArtsNote
UpdatedUnix: createTime,
Spec: req.Spec,
BootFile: req.BootFile,
BranchName: req.BranchName,
ModelName: req.ModelName,
ModelVersion: req.ModelVersion,
LabelName: req.LabelName,


+ 198
- 0
modules/notebook/contentManager.go View File

@@ -0,0 +1,198 @@
package notebook

import (
"crypto/tls"
"encoding/base64"
"fmt"
"io/ioutil"
"net/http"
"path"
"strings"

"github.com/go-resty/resty/v2"

"code.gitea.io/gitea/models"
"code.gitea.io/gitea/modules/cloudbrain"
"code.gitea.io/gitea/modules/setting"

"code.gitea.io/gitea/modules/log"
)

var restyClient *resty.Client

type NotebookApiResponse struct {
Name string `json:"name"`
Path string `json:"path"`
}

type NotebookContent struct {
Url string
Path string
Cookies []*http.Cookie
Xsrf string
PathType string //file directory
Token string
}

func (c *NotebookContent) IsNotebookFileCanBrowser() bool {
if c.Xsrf == "" {
c.SetCookiesAndCsrf()
}
if c.Xsrf == "" {
log.Warn("xsrf is empty, can not broswer url:" + c.Url)
return false
}
return c.IsNoteBookContentsExist()

}

func (c *NotebookContent) SetCookiesAndCsrf() {
log.Info("jupyter url:" + c.Url)
var cookies []*http.Cookie
const retryTimes = 10
url := c.Url
if c.Token != "" {
url = c.Url + "?token=" + c.Token
}
for i := 0; i < retryTimes; i++ {
res, err := http.Get(url)
if err != nil {
log.Error("browser jupyterUrl failed.", err)
if i == retryTimes-1 {
c.Cookies = cookies
}

} else {
cookies = res.Cookies()
xsrf := ""
for _, cookie := range cookies {
if cookie.Name == "_xsrf" {
xsrf = cookie.Value
if len(cookies) > 1 {
break
}

}

}
if xsrf != "" {
c.Cookies = cookies
c.Xsrf = xsrf
}

}
}
c.Cookies = cookies

}

func (c *NotebookContent) IsNoteBookContentsExist() bool {
client := getRestyClient()
uploadUrl := getJupyterBaseUrl(c.Url) + "api/contents/" + c.Path + "?type=" + c.PathType
res, err := client.R().
SetCookies(c.Cookies).
SetHeader("X-XSRFToken", c.Xsrf).
Get(uploadUrl)
if err != nil {
log.Warn("browser url error:"+uploadUrl, err)
return false
}
return res.StatusCode() == http.StatusOK
}

func (c *NotebookContent) UploadNoteBookFile(task *models.Cloudbrain) error {

err := c.MakeNoteBookDir()
if err != nil {
return err
}

codePath := setting.JobPath + task.JobName + cloudbrain.CodeMountPath
fileContents, err := ioutil.ReadFile(codePath + "/" + c.Path)
if err != nil {
log.Error("read jupyter file failed:%v", task.DisplayJobName, err)
}

base64Content := base64.StdEncoding.EncodeToString(fileContents)
client := getRestyClient()
uploadUrl := getJupyterBaseUrl(c.Url) + "api/contents/" + c.Path
res, err := client.R().
SetCookies(c.Cookies).
SetHeader("X-XSRFToken", c.Xsrf).
SetBody(map[string]interface{}{
"type": "file",
"format": "base64",
"name": path.Base(c.Path),
"path": c.Path,
"content": base64Content}).
Put(uploadUrl)
if err != nil {
log.Error("upload jupyter file failed:%v", task.DisplayJobName, err)
return err
} else if res.StatusCode() != http.StatusCreated {
log.Error("upload jupyter file failed:%v, status is %s", task.DisplayJobName, res.Status(), err)
return fmt.Errorf("status:", res.StatusCode())
}
return nil
}

/**
if c.Path is a/b/c.txt
makedir a/b
if c.Path is a/b/c
makedir a/b
*/
func (c *NotebookContent) MakeNoteBookDir() error {
filePaths := strings.Split(c.Path, "/")

for i := 0; i < len(filePaths)-1; i++ {
cTemp := &NotebookContent{
Url: c.Url,
Cookies: c.Cookies,
Path: path.Join(filePaths[0 : i+1]...),
PathType: "directory",
Xsrf: c.Xsrf,
}
if !cTemp.IsNoteBookContentsExist() {

createTempDirUrl := getJupyterBaseUrl(cTemp.Url) + "api/contents/" + cTemp.Path
client := getRestyClient()
var jobResult NotebookApiResponse
res, err := client.R().
SetCookies(c.Cookies).
SetHeader("X-XSRFToken", c.Xsrf).
SetBody(map[string]interface{}{
"type": cTemp.PathType,
"path": cTemp.Path,
}).SetResult(&jobResult).
Put(createTempDirUrl)
if err != nil {
return err
}
if res.StatusCode() != http.StatusCreated {
return fmt.Errorf("status code:" + res.Status())
}

}

}
return nil
}

func getJupyterBaseUrl(url string) string {
jupyterUrlLength := len(url)
baseUrl := url
if strings.HasSuffix(url, "lab") {
baseUrl = url[0 : jupyterUrlLength-len(path.Base(url))]
}

return baseUrl
}

func getRestyClient() *resty.Client {
if restyClient == nil {
restyClient = resty.New()
restyClient.SetTLSClientConfig(&tls.Config{InsecureSkipVerify: true})
}
return restyClient
}

+ 28
- 27
modules/structs/cloudbrain.go View File

@@ -47,36 +47,37 @@ type CreateFileNotebookJobOption struct {
BranchName string `json:"branch_name" binding:"Required"`
OwnerName string `json:"owner_name" binding:"Required"`
ProjectName string `json:"project_name" binding:"Required"`
JobId string `json:"job_id"`
}

type Cloudbrain struct {
ID int64 `json:"id"`
JobID string `json:"job_id"`
JobType string `json:"job_type"`
Type int `json:"type"`
DisplayJobName string `json:"display_job_name"`
Status string `json:"status"`
CreatedUnix int64 `json:"created_unix"`
RepoID int64 `json:"repo_id"`
Duration int64 `json:"duration"` //运行时长 单位秒
TrainJobDuration string `json:"train_job_duration"`
ImageID string `json:"image_id"` //grampus image_id
Image string `json:"image"`
Uuid string `json:"uuid"` //数据集id
DatasetName string `json:"dataset_name"`
ComputeResource string `json:"compute_resource"` //计算资源,例如npu
AiCenter string `json:"ai_center"` //grampus ai center: center_id+center_name
BranchName string `json:"branch_name"` //分支名称
Parameters string `json:"parameters"` //传给modelarts的param参数
BootFile string `json:"boot_file"` //启动文件
Description string `json:"description"` //描述
ModelName string `json:"model_name"` //模型名称
ModelVersion string `json:"model_version"` //模型版本
CkptName string `json:"ckpt_name"` //权重文件名称
StartTime int64 `json:"start_time"`
EndTime int64 `json:"end_time"`
Spec *SpecificationShow `json:"spec"`
ID int64 `json:"id"`
JobID string `json:"job_id"`
JobType string `json:"job_type"`
Type int `json:"type"`
DisplayJobName string `json:"display_job_name"`
Status string `json:"status"`
CreatedUnix int64 `json:"created_unix"`
RepoID int64 `json:"repo_id"`
Duration int64 `json:"duration"` //运行时长 单位秒
TrainJobDuration string `json:"train_job_duration"`
ImageID string `json:"image_id"` //grampus image_id
Image string `json:"image"`
Uuid string `json:"uuid"` //数据集id
DatasetName string `json:"dataset_name"`
ComputeResource string `json:"compute_resource"` //计算资源,例如npu
AiCenter string `json:"ai_center"` //grampus ai center: center_id+center_name
BranchName string `json:"branch_name"` //分支名称
Parameters string `json:"parameters"` //传给modelarts的param参数
BootFile string `json:"boot_file"` //启动文件
Description string `json:"description"` //描述
ModelName string `json:"model_name"` //模型名称
ModelVersion string `json:"model_version"` //模型版本
CkptName string `json:"ckpt_name"` //权重文件名称
StartTime int64 `json:"start_time"`
EndTime int64 `json:"end_time"`
VersionName string `json:"version_name"`
Spec *SpecificationShow `json:"spec"`
}

type SpecificationShow struct {


+ 2
- 1
options/locale/locale_en-US.ini View File

@@ -577,6 +577,7 @@ static.CloudBrainTaskNum=CloudBrain Task Count
static.CloudBrainRunTime=CloudBrain Run Time
static.CommitDatasetNum=Commit Dataset Count
static.CommitModelCount=Commit Model Count
static.ModelConvertCount=Model Convert Count
static.UserIndex=Normalized user index
static.UserIndexPrimitive=User Index
static.countdate=Count Date
@@ -1061,7 +1062,7 @@ model_rename=Duplicate model name, please modify model name.
notebook_file_not_exist=Notebook file does not exist.
notebook_select_wrong=Please select a Notebook(.ipynb) file first.
notebook_file_no_right=You have no right to access the Notebook(.ipynb) file.
debug_again_fail=Fail to restart debug task, please try again later.
notebook_repo_conflict=The files in different branches of the same repository can not run together.

date=Date
repo_add=Project Increment


+ 3
- 2
options/locale/locale_zh-CN.ini View File

@@ -581,6 +581,7 @@ static.CloudBrainTaskNum=云脑任务数
static.CloudBrainRunTime=云脑运行时间(小时)
static.CommitDatasetNum=上传(提交)数据集文件数
static.CommitModelCount=提交模型数
static.ModelConvertCount=模型转换数
static.UserIndex=归一化用户指数
static.UserIndexPrimitive=用户指数
static.countdate=系统统计时间
@@ -1060,7 +1061,7 @@ model_rename=模型名称重复,请修改模型名称
notebook_file_not_exist=Notebook文件不存在。
notebook_select_wrong=请先选择Notebook(.ipynb)文件。
notebook_file_no_right=您没有这个Notebook文件的读权限。
debug_again_fail=再次调试失败,请稍后再试
notebook_repo_conflict=同一个仓库的不同分支文件不能同时运行

date=日期
repo_add=新增项目
@@ -1672,7 +1673,7 @@ issues.action_assignee_no_select=未指派
issues.opened_by=由 <a href="%[2]s">%[3]s</a> 于 %[1]s创建
pulls.merged_by=由 <a href="%[2]s">%[3]s</a> 于 %[1]s 合并
pulls.merged_by_fake=由 %[2]s 于 %[1]s 合并
issues.closed_by=按 <a href="%[2]s">%[3]s</a> 关闭%[1]s
issues.closed_by=由 <a href="%[2]s">%[3]s</a> 创建,被关闭于 %[1]s
issues.opened_by_fake=由 %[2]s 于 %[1]s创建
issues.closed_by_fake=通过 %[2]s 关闭 %[1]s
issues.previous=上一页


+ 5
- 1
routers/api/v1/api.go View File

@@ -549,6 +549,8 @@ func RegisterRoutes(m *macaron.Macaron) {

}, reqToken())

m.Get("/compute-nodes", reqToken(), user.GetComputeNodes)

// Notifications
m.Group("/notifications", func() {
m.Combo("").
@@ -745,7 +747,7 @@ func RegisterRoutes(m *macaron.Macaron) {
m.Group("/file_notebook", func() {
m.Get("", repo.GetFileNoteBookInfo)
m.Post("/create", reqToken(), reqWeChat(), bind(api.CreateFileNotebookJobOption{}), repo.CreateFileNoteBook)
m.Post("/status", reqToken(), bind(api.CreateFileNotebookJobOption{}), repo.FileNoteBookStatus)
})

m.Group("/repos", func() {
@@ -1024,6 +1026,8 @@ func RegisterRoutes(m *macaron.Macaron) {
m.Get("/query_model_byName", repo.QueryModelByName)
m.Get("/query_model_for_predict", repo.QueryModelListForPredict)
m.Get("/query_modelfile_for_predict", repo.QueryModelFileForPredict)
m.Get("/query_train_job", repo.QueryTrainJobList)
m.Get("/query_train_job_version", repo.QueryTrainJobVersionList)
m.Get("/query_train_model", repo.QueryTrainModelList)
m.Post("/create_model_convert", repo.CreateModelConvert)
m.Post("/convert_stop", repo.StopModelConvert)


+ 3
- 0
routers/api/v1/repo/cloudbrain.go View File

@@ -110,6 +110,9 @@ func GeneralCloudBrainJobStop(ctx *context.APIContext) {
func CreateFileNoteBook(ctx *context.APIContext, option api.CreateFileNotebookJobOption) {
cloudbrainTask.FileNotebookCreate(ctx.Context, option)
}
func FileNoteBookStatus(ctx *context.APIContext, option api.CreateFileNotebookJobOption) {
cloudbrainTask.FileNotebookStatus(ctx.Context, option)
}

func GetFileNoteBookInfo(ctx *context.APIContext) {
//image description spec description waiting count


+ 32
- 0
routers/api/v1/repo/modelmanage.go View File

@@ -4,8 +4,10 @@ import (
"net/http"

"code.gitea.io/gitea/modules/context"
"code.gitea.io/gitea/modules/convert"
"code.gitea.io/gitea/modules/log"
"code.gitea.io/gitea/modules/storage"
api "code.gitea.io/gitea/modules/structs"
routerRepo "code.gitea.io/gitea/routers/repo"
)

@@ -54,6 +56,21 @@ func QueryModelListForPredict(ctx *context.APIContext) {
routerRepo.QueryModelListForPredict(ctx.Context)
}

func QueryTrainJobList(ctx *context.APIContext) {
result, err := routerRepo.QueryTrainJobListApi(ctx.Context)
if err != nil {
log.Info("query error." + err.Error())
ctx.JSON(http.StatusOK, nil)
} else {
re := make([]*api.Cloudbrain, 0)
for _, task := range result {
conRe := convert.ToCloudBrain(task)
re = append(re, conRe)
}
ctx.JSON(http.StatusOK, re)
}
}

func QueryTrainModelList(ctx *context.APIContext) {
result, err := routerRepo.QueryTrainModelFileById(ctx.Context)
if err != nil {
@@ -63,6 +80,21 @@ func QueryTrainModelList(ctx *context.APIContext) {
ctx.JSON(http.StatusOK, re)
}

func QueryTrainJobVersionList(ctx *context.APIContext) {
result, err := routerRepo.QueryTrainJobVersionListApi(ctx.Context)
if err != nil {
log.Info("query error." + err.Error())
ctx.JSON(http.StatusOK, nil)
} else {
re := make([]*api.Cloudbrain, 0)
for _, task := range result {
conRe := convert.ToCloudBrain(task)
re = append(re, conRe)
}
ctx.JSON(http.StatusOK, re)
}
}

func convertFileFormat(result []storage.FileInfo) []FileInfo {
re := make([]FileInfo, 0)
if result != nil {


+ 20
- 0
routers/api/v1/user/repo.go View File

@@ -5,6 +5,7 @@
package user

import (
"code.gitea.io/gitea/modules/modelarts"
"net/http"

"code.gitea.io/gitea/models"
@@ -146,3 +147,22 @@ func ListOrgRepos(ctx *context.APIContext) {

listUserRepos(ctx, ctx.Org.Organization, ctx.IsSigned)
}

func GetComputeNodes(ctx *context.APIContext) {
taskeType := ctx.QueryInt("type")
if taskeType == 2 {
ctx.JSON(http.StatusOK, []int{1})
} else {
modelarts.InitMultiNode()
if modelarts.MultiNodeConfig != nil {
for _, info := range modelarts.MultiNodeConfig.Info {
if isInOrg, _ := models.IsOrganizationMemberByOrgName(info.Org, ctx.User.ID); isInOrg {
ctx.JSON(http.StatusOK, info.Node)
return
}
}
}
ctx.JSON(http.StatusOK, []int{1})
}

}

+ 20
- 13
routers/repo/ai_model_manage.go View File

@@ -2,7 +2,6 @@ package repo

import (
"archive/zip"
"code.gitea.io/gitea/services/repository"
"encoding/json"
"errors"
"fmt"
@@ -12,6 +11,8 @@ import (
"regexp"
"strings"

"code.gitea.io/gitea/services/repository"

"code.gitea.io/gitea/models"
"code.gitea.io/gitea/modules/context"
"code.gitea.io/gitea/modules/log"
@@ -710,36 +711,42 @@ func downloadFromCloudBrainTwo(path string, task *models.AiModelManage, ctx *con
}

func QueryTrainJobVersionList(ctx *context.Context) {

VersionListTasks, err := QueryTrainJobVersionListApi(ctx)
if err != nil {
ctx.JSON(200, nil)
} else {
ctx.JSON(200, VersionListTasks)
}
}

func QueryTrainJobVersionListApi(ctx *context.Context) ([]*models.Cloudbrain, error) {
log.Info("query train job version list. start.")
JobID := ctx.Query("jobId")
if JobID == "" {
JobID = ctx.Query("JobId")
}

VersionListTasks, count, err := models.QueryModelTrainJobVersionList(JobID)

log.Info("query return count=" + fmt.Sprint(count))

return VersionListTasks, err
}

func QueryTrainJobList(ctx *context.Context) {
VersionListTasks, err := QueryTrainJobListApi(ctx)
if err != nil {
ctx.ServerError("QueryTrainJobList:", err)
ctx.JSON(200, nil)
} else {
ctx.JSON(200, VersionListTasks)
}
}

func QueryTrainJobList(ctx *context.Context) {
log.Info("query train job list. start.")
func QueryTrainJobListApi(ctx *context.Context) ([]*models.Cloudbrain, error) {
repoId := ctx.QueryInt64("repoId")

VersionListTasks, count, err := models.QueryModelTrainJobList(repoId)
log.Info("query return count=" + fmt.Sprint(count))

if err != nil {
ctx.ServerError("QueryTrainJobList:", err)
} else {
ctx.JSON(200, VersionListTasks)
}

return VersionListTasks, err
}

func QueryTrainModelFileById(ctx *context.Context) ([]storage.FileInfo, error) {


+ 20
- 2
routers/repo/cloudbrain.go View File

@@ -911,10 +911,13 @@ func cloudBrainShow(ctx *context.Context, tpName base.TplName, jobType models.Jo
func CloudBrainDebug(ctx *context.Context) {
task := ctx.Cloudbrain
debugUrl := setting.DebugServerHost + "jpylab_" + task.JobID + "_" + task.SubTaskName
if task.BootFile != "" {
ctx.Redirect(getFileUrl(debugUrl, task.BootFile))

if ctx.QueryTrim("file") != "" {
ctx.Redirect(getFileUrl(debugUrl, ctx.QueryTrim("file")))
} else {
if task.BootFile != "" {
go cloudbrainTask.UploadNotebookFiles(task)
}
ctx.Redirect(debugUrl)
}

@@ -1638,6 +1641,21 @@ func CloudBrainDownloadModel(ctx *context.Context) {
ctx.Resp.Header().Set("Cache-Control", "max-age=0")
http.Redirect(ctx.Resp, ctx.Req.Request, url, http.StatusMovedPermanently)
}

func CloudBrainDownloadMultiModel(ctx *context.Context) {
parentDir := ctx.Query("parentDir")
jobName := ctx.Query("jobName")
filePath := "jobs/" + jobName + "/model/" + parentDir
allFile, err := storage.GetAllObjectByBucketAndPrefixMinio(setting.Attachment.Minio.Bucket, filePath)
if err == nil {
returnFileName := jobName + ".zip"
MinioDownloadManyFile(filePath, ctx, returnFileName, allFile)
} else {
log.Info("error,msg=" + err.Error())
ctx.ServerError("no file to download.", err)
}
}

func CloudBrainDownloadInferenceResult(ctx *context.Context) {
parentDir := ctx.Query("parentDir")
fileName := ctx.Query("fileName")


+ 46
- 4
routers/repo/modelarts.go View File

@@ -439,9 +439,13 @@ func NotebookDebug2(ctx *context.Context) {
ctx.RenderWithErr(err.Error(), tplModelArtsNotebookIndex, nil)
return
}
if task.BootFile != "" {
ctx.Redirect(getFileUrl(result.Url, task.BootFile) + "?token=" + result.Token)

if ctx.QueryTrim("file") != "" {
ctx.Redirect(getFileUrl(result.Url, ctx.QueryTrim("file")) + "?token=" + result.Token)
} else {
if task.BootFile != "" {
go cloudbrainTask.UploadNotebookFiles(task)
}
ctx.Redirect(result.Url + "?token=" + result.Token)
}

@@ -463,7 +467,7 @@ func getFileUrl(url string, filename string) string {
}
}

return url + middle + path.Base(filename)
return url + middle + filename
}

func NotebookRestart(ctx *context.Context) {
@@ -630,7 +634,7 @@ func NotebookStop(ctx *context.Context) {
if task.Status != string(models.ModelArtsRunning) {
log.Error("the job(%s) is not running", task.JobName, ctx.Data["MsgID"])
resultCode = "-1"
errorMsg = "the job is not running"
errorMsg = ctx.Tr("cloudbrain.Already_stopped")
break
}

@@ -2654,6 +2658,44 @@ func InferenceJobShow(ctx *context.Context) {
ctx.HTML(http.StatusOK, tplModelArtsInferenceJobShow)
}

func MultiModelDownload(ctx *context.Context) {
var (
err error
)
jobID := ctx.Params(":jobid")
versionName := ctx.Query("version_name")
parentDir := ctx.Query("parent_dir")

task, err := models.GetCloudbrainByJobIDAndVersionName(jobID, versionName)
if err != nil {
log.Error("GetCloudbrainByJobIDAndVersionName(%s) failed:%v", task.JobName, err.Error())
return
}

if task.ComputeResource == models.NPUResource {
path := strings.TrimPrefix(path.Join(setting.TrainJobModelPath, task.JobName, setting.OutPutPath, versionName, parentDir), "/")
allFile, err := storage.GetAllObjectByBucketAndPrefix(setting.Bucket, path)
if err == nil {
returnFileName := task.DisplayJobName + ".zip"
ObsDownloadManyFile(path, ctx, returnFileName, allFile)
} else {
log.Info("error,msg=" + err.Error())
ctx.ServerError("no file to download.", err)
}
} else if task.ComputeResource == models.GPUResource {
filePath := setting.CBCodePathPrefix + task.JobName + cloudbrain.ModelMountPath + "/" + parentDir
allFile, err := storage.GetAllObjectByBucketAndPrefixMinio(setting.Attachment.Minio.Bucket, filePath)
if err == nil {
returnFileName := task.DisplayJobName + ".zip"
MinioDownloadManyFile(filePath, ctx, returnFileName, allFile)
} else {
log.Info("error,msg=" + err.Error())
ctx.ServerError("no file to download.", err)
}
}

}

func ModelDownload(ctx *context.Context) {
var (
err error


+ 5
- 1
routers/routes/routes.go View File

@@ -1186,6 +1186,7 @@ func RegisterRoutes(m *macaron.Macaron) {
m.Get("/rate", reqRepoCloudBrainReader, repo.GetRate)
m.Get("/models", reqRepoCloudBrainReader, repo.CloudBrainShowModels)
m.Get("/download_model", cloudbrain.AdminOrJobCreaterRight, repo.CloudBrainDownloadModel)
m.Get("/download_multi_model", cloudbrain.AdminOrJobCreaterRight, repo.CloudBrainDownloadMultiModel)
})
m.Get("/create", reqWechatBind, reqRepoCloudBrainWriter, context.PointAccount(), repo.CloudBrainNew)
m.Post("/create", reqWechatBind, reqRepoCloudBrainWriter, bindIgnErr(auth.CreateCloudBrainForm{}), repo.CloudBrainCreate)
@@ -1209,6 +1210,7 @@ func RegisterRoutes(m *macaron.Macaron) {
m.Post("/del", cloudbrain.AdminOrOwnerOrJobCreaterRightForTrain, repo.CloudBrainTrainJobDel)
//m.Get("/models", reqRepoCloudBrainReader, repo.CloudBrainShowModels)
m.Get("/download_model", cloudbrain.AdminOrOwnerOrJobCreaterRightForTrain, repo.CloudBrainDownloadModel)
m.Get("/download_multi_model", cloudbrain.AdminOrJobCreaterRight, repo.CloudBrainDownloadMultiModel)
//m.Get("/get_log", cloudbrain.AdminOrJobCreaterRightForTrain, repo.GetLogFromModelDir)
//m.Post("/create_version", reqWechatBind, cloudbrain.AdminOrJobCreaterRightForTrain, bindIgnErr(auth.CreateModelArtsTrainJobForm{}), repo.TrainJobCreateVersion)
m.Get("/create_version", reqWechatBind, cloudbrain.AdminOrJobCreaterRightForTrain, repo.CloudBrainTrainJobVersionNew)
@@ -1221,7 +1223,7 @@ func RegisterRoutes(m *macaron.Macaron) {
m.Group("/:jobid", func() {
m.Get("", reqRepoCloudBrainReader, repo.InferenceCloudBrainJobShow)
m.Get("/result_download", cloudbrain.AdminOrJobCreaterRightForTrain, repo.CloudBrainDownloadInferenceResult)
m.Get("/download_multi_model", cloudbrain.AdminOrJobCreaterRight, repo.CloudBrainDownloadMultiModel)
m.Get("/downloadall", repo.DownloadInferenceResultFile)
})
m.Get("/create", reqWechatBind, reqRepoCloudBrainWriter, context.PointAccount(), repo.InferenceCloudBrainJobNew)
@@ -1248,6 +1250,7 @@ func RegisterRoutes(m *macaron.Macaron) {
m.Post("/stop", cloudbrain.AdminOrOwnerOrJobCreaterRightForTrain, repo.GrampusStopJob)
m.Post("/del", cloudbrain.AdminOrOwnerOrJobCreaterRightForTrain, repo.GrampusTrainJobDel)
m.Get("/model_download", cloudbrain.AdminOrJobCreaterRightForTrain, repo.ModelDownload)
m.Get("/download_multi_model", cloudbrain.AdminOrJobCreaterRightForTrain, repo.MultiModelDownload)
m.Get("/create_version", reqWechatBind, cloudbrain.AdminOrJobCreaterRightForTrain, repo.GrampusTrainJobVersionNew)
m.Post("/create_version", reqWechatBind, cloudbrain.AdminOrJobCreaterRightForTrain, bindIgnErr(auth.CreateGrampusTrainJobForm{}), repo.GrampusTrainJobVersionCreate)
})
@@ -1333,6 +1336,7 @@ func RegisterRoutes(m *macaron.Macaron) {
m.Post("/stop", cloudbrain.AdminOrOwnerOrJobCreaterRightForTrain, repo.TrainJobStop)
m.Post("/del", cloudbrain.AdminOrOwnerOrJobCreaterRightForTrain, repo.TrainJobDel)
m.Get("/model_download", cloudbrain.AdminOrJobCreaterRightForTrain, repo.ModelDownload)
m.Get("/download_multi_model", cloudbrain.AdminOrJobCreaterRightForTrain, repo.MultiModelDownload)
m.Get("/download_log_file", cloudbrain.AdminOrJobCreaterRightForTrain, repo.TrainJobDownloadLogFile)
m.Get("/create_version", reqWechatBind, cloudbrain.AdminOrJobCreaterRightForTrain, context.PointAccount(), repo.TrainJobNewVersion)
m.Post("/create_version", reqWechatBind, cloudbrain.AdminOrJobCreaterRightForTrain, bindIgnErr(auth.CreateModelArtsTrainJobForm{}), repo.TrainJobCreateVersion)


+ 240
- 28
services/cloudbrain/cloudbrainTask/notebook.go View File

@@ -4,6 +4,9 @@ import (
"fmt"
"net/http"
"path"
"strings"

"code.gitea.io/gitea/modules/notebook"

"code.gitea.io/gitea/modules/modelarts"
"code.gitea.io/gitea/modules/modelarts_cd"
@@ -29,6 +32,9 @@ import (
)

const NoteBookExtension = ".ipynb"
const CPUType = 0
const GPUType = 1
const NPUType = 2

func FileNotebookCreate(ctx *context.Context, option api.CreateFileNotebookJobOption) {

@@ -66,7 +72,7 @@ func FileNotebookCreate(ctx *context.Context, option api.CreateFileNotebookJobOp
}

//create repo if not exist
repo, err := models.GetRepositoryByName(ctx.User.ID, setting.FileNoteBook.ProjectName)
repo, _ := models.GetRepositoryByName(ctx.User.ID, setting.FileNoteBook.ProjectName)
if repo == nil {
repo, err = repo_service.CreateRepository(ctx.User, ctx.User, models.CreateRepoOptions{
Name: setting.FileNoteBook.ProjectName,
@@ -80,19 +86,222 @@ func FileNotebookCreate(ctx *context.Context, option api.CreateFileNotebookJobOp
AutoInit: true,
DefaultBranch: "master",
})
if err != nil {
ctx.JSON(http.StatusOK, models.BaseErrorMessageApi(ctx.Tr("repo.failed_to_create_notebook_repo", setting.FileNoteBook.ProjectName)))
return
}
} else {

noteBook, _ := models.GetWaitOrRunFileNotebookByRepo(repo.ID, getCloudbrainType(option.Type))
if noteBook != nil {

if isRepoConfilcts(option, noteBook) {
ctx.JSON(http.StatusOK, models.BaseErrorMessageApi(ctx.Tr("repo.notebook_repo_conflict")))
return
}

if isNotebookSpecMath(option, noteBook) {
if !isRepoMatch(option, noteBook) {
err = downloadCode(sourceRepo, getCodePath(noteBook.JobName, sourceRepo), option.BranchName)
if err != nil {
log.Error("download code failed", err)
ctx.JSON(http.StatusOK, models.BaseErrorMessageApi(ctx.Tr("cloudbrain.load_code_failed")))
return
}
}
if !isRepoFileMatch(option, noteBook) {
noteBook.BootFile += ";" + getBootFile(option.File, option.OwnerName, option.ProjectName)
noteBook.BranchName += ";" + option.BranchName
noteBook.Description += ";" + getDescription(option)

err := models.UpdateJob(noteBook)
if err != nil {
log.Error("GenerateNotebook2 failed, %v", err, ctx.Data["MsgID"])
ctx.JSON(http.StatusOK, models.BaseErrorMessageApi(err.Error()))
return
}
}

ctx.JSON(http.StatusOK, models.BaseMessageApi{
Code: 0,
Message: noteBook.JobID,
})
return
}

}
}

if option.Type <= GPUType {
cloudBrainFileNoteBookCreate(ctx, option, repo, sourceRepo)
} else {
modelartsFileNoteBookCreate(ctx, option, repo, sourceRepo)
}

}
func FileNotebookStatus(ctx *context.Context, option api.CreateFileNotebookJobOption) {
if ctx.Written() {
return
}

if path.Ext(option.File) != NoteBookExtension {
ctx.JSON(http.StatusOK, models.BaseErrorMessageApi(ctx.Tr("repo.notebook_select_wrong")))
return
}

isNotebookFileExist, _ := isNoteBookFileExist(ctx, option)
if !isNotebookFileExist {
ctx.JSON(http.StatusOK, models.BaseErrorMessageApi(ctx.Tr("repo.notebook_file_not_exist")))
return
}

task, err := models.GetCloudbrainByJobID(option.JobId)
if err != nil {
ctx.JSON(http.StatusOK, models.BaseErrorMessageApi(ctx.Tr("repo.failed_to_create_notebook_repo", setting.FileNoteBook.ProjectName)))
log.Error("job not found:"+option.JobId, err)
ctx.JSON(http.StatusOK, models.BaseErrorMessageApi("Job id may not be right. can not find job."))
return
}
if option.Type <= 1 {
cloudBrainFileNoteBookCreate(ctx, option, repo, sourceRepo)
if task.BootFile == "" || task.Status != string(models.ModelArtsRunning) {
log.Warn("Boot file is empty or status is running. ")
ctx.JSON(http.StatusOK, models.BaseErrorMessageApi("Boot file is empty or status is running."))
return
}
if !isRepoFileMatch(option, task) {
ctx.JSON(http.StatusOK, models.BaseErrorMessageApi("can not math repo file."))
return
}
debugBaseUrl, token, err := getBaseUrlAndToken(task)
if err != nil {
ctx.JSON(http.StatusOK, models.BaseErrorMessageApi(err.Error()))
return
}

if uploadNotebookFileIfCannotBroswer(debugBaseUrl, getBootFile(option.File, option.OwnerName, option.ProjectName), task, token) {
ctx.JSON(http.StatusOK, models.BaseOKMessageApi)
} else {
modelartsFileNoteBookCreate(ctx, option, repo, sourceRepo)
ctx.JSON(http.StatusOK, models.BaseErrorMessageApi("upload failed."))

}

}

func getBaseUrlAndToken(task *models.Cloudbrain) (string, string, error) {
var debugBaseUrl string
var token string
if task.Type == models.TypeCloudBrainOne {
debugBaseUrl = setting.DebugServerHost + "jpylab_" + task.JobID + "_" + task.SubTaskName + "/lab"

} else {
var result *models.GetNotebook2Result
var err error
if task.Type == models.TypeCloudBrainTwo {
result, err = modelarts.GetNotebook2(task.JobID)
} else if task.Type == models.TypeCDCenter {
result, err = modelarts_cd.GetNotebook(task.JobID)
}
if err != nil || result == nil || result.Status != string(models.ModelArtsRunning) || result.Url == "" {
log.Error("notebook job not found:"+task.JobID, err)
return "", "", fmt.Errorf("can not get job or job is invalid.")
}

debugBaseUrl = result.Url
token = result.Token

}
return debugBaseUrl, token, nil
}

func uploadNotebookFileIfCannotBroswer(debugBaseUrl string, bootFile string, task *models.Cloudbrain, token string) bool {
c := &notebook.NotebookContent{
Url: debugBaseUrl,
Path: bootFile,
PathType: "file",
Token: token,
}
if c.IsNotebookFileCanBrowser() {
return true
} else {
c.SetCookiesAndCsrf()
c.UploadNoteBookFile(task)
return c.IsNotebookFileCanBrowser()
}

}

func isNotebookSpecMath(option api.CreateFileNotebookJobOption, book *models.Cloudbrain) bool {
if option.Type == NPUType || option.Type == CPUType {
return true
}
spec, err := models.GetCloudbrainSpecByID(book.ID)
if err != nil {
log.Warn("can not get spec ", err)
return false
}
return spec.AccCardsNum > 0
}

func isRepoConfilcts(option api.CreateFileNotebookJobOption, book *models.Cloudbrain) bool {
bootFiles := strings.Split(book.BootFile, ";")
branches := strings.Split(book.BranchName, ";")

for i, bootFile := range bootFiles {
splits := strings.Split(bootFile, "/")
if len(splits) >= 3 {
if splits[0] == option.OwnerName && splits[1] == option.ProjectName && branches[i] != option.BranchName {
return true
}
}
}

return false

}

func isRepoMatch(option api.CreateFileNotebookJobOption, book *models.Cloudbrain) bool {
bootFiles := strings.Split(book.BootFile, ";")

for _, bootFile := range bootFiles {
splits := strings.Split(bootFile, "/")
if len(splits) >= 3 {
if splits[0] == option.OwnerName && splits[1] == option.ProjectName {
return true
}
}
}
return false

}

func isRepoFileMatch(option api.CreateFileNotebookJobOption, book *models.Cloudbrain) bool {
bootFiles := strings.Split(book.BootFile, ";")
branches := strings.Split(book.BranchName, ";")

for i, bootFile := range bootFiles {
if branches[i] == option.BranchName && getBootFile(option.File, option.OwnerName, option.ProjectName) == bootFile {
return true
}
}

return false

}
func UploadNotebookFiles(task *models.Cloudbrain) {
if task.Status == string(models.JobRunning) && task.BootFile != "" {

debugBaseUrl, token, err := getBaseUrlAndToken(task)
if err != nil {
log.Error("can not get base url:", err)
return
}
bootFiles := strings.Split(task.BootFile, ";")

for _, bootFile := range bootFiles {
uploadNotebookFileIfCannotBroswer(debugBaseUrl, bootFile, task, token)
}

}
}

func cloudBrainFileNoteBookCreate(ctx *context.Context, option api.CreateFileNotebookJobOption, repo *models.Repository, sourceRepo *models.Repository) {

displayJobName := cloudbrainService.GetDisplayJobName(ctx.User.Name)
@@ -131,17 +340,18 @@ func cloudBrainFileNoteBookCreate(ctx *context.Context, option api.CreateFileNot
} else {
if count >= 1 {
log.Error("the user already has running or waiting task", ctx.Data["MsgID"])
ctx.JSON(http.StatusOK,models.BaseMessageApi{
Code: 2,
ctx.JSON(http.StatusOK, models.BaseMessageApi{
Code: 2,
Message: ctx.Tr("repo.cloudbrain.morethanonejob"),
})
return
}
}

errStr := uploadCodeFile(sourceRepo, getCodePath(jobName), option.BranchName, option.File, jobName)
if errStr != "" {
ctx.JSON(http.StatusOK, models.BaseErrorMessageApi(ctx.Tr("repo.notebook_file_not_exist")))
err = downloadCode(sourceRepo, getCodePath(jobName, sourceRepo), option.BranchName)
if err != nil {
log.Error("download code failed", err)
ctx.JSON(http.StatusOK, models.BaseErrorMessageApi(ctx.Tr("cloudbrain.load_code_failed")))
return
}
command := cloudbrain.GetCloudbrainDebugCommand()
@@ -185,7 +395,7 @@ func cloudBrainFileNoteBookCreate(ctx *context.Context, option api.CreateFileNot
JobType: jobType,
Description: getDescription(option),
BranchName: option.BranchName,
BootFile: option.File,
BootFile: getBootFile(option.File, option.OwnerName, option.ProjectName),
Params: "{\"parameter\":[]}",
CommitID: "",
BenchmarkTypeID: 0,
@@ -206,8 +416,18 @@ func cloudBrainFileNoteBookCreate(ctx *context.Context, option api.CreateFileNot

}

func getCodePath(jobName string) string {
return setting.JobPath + jobName + cloudbrain.CodeMountPath
func getCloudbrainType(optionType int) int {
if optionType < 1 {
return models.TypeCloudBrainOne
}
if setting.ModelartsCD.Enabled {
return models.TypeCDCenter
}
return models.TypeCloudBrainTwo
}

func getCodePath(jobName string, repo *models.Repository) string {
return setting.JobPath + jobName + cloudbrain.CodeMountPath + "/" + repo.OwnerName + "/" + repo.Name
}

func getDescription(option api.CreateFileNotebookJobOption) string {
@@ -237,8 +457,8 @@ func modelartsFileNoteBookCreate(ctx *context.Context, option api.CreateFileNote
} else {
if count >= 1 {
log.Error("the user already has running or waiting task", ctx.Data["MsgID"])
ctx.JSON(http.StatusOK,models.BaseMessageApi{
Code: 2,
ctx.JSON(http.StatusOK, models.BaseMessageApi{
Code: 2,
Message: ctx.Tr("repo.cloudbrain.morethanonejob"),
})
return
@@ -260,7 +480,7 @@ func modelartsFileNoteBookCreate(ctx *context.Context, option api.CreateFileNote
}
}

err = downloadCode(sourceRepo, getCodePath(jobName), option.BranchName)
err = downloadCode(sourceRepo, getCodePath(jobName, sourceRepo), option.BranchName)
if err != nil {
ctx.JSON(http.StatusOK, models.BaseErrorMessageApi(ctx.Tr("cloudbrain.load_code_failed")))
return
@@ -297,8 +517,9 @@ func modelartsFileNoteBookCreate(ctx *context.Context, option api.CreateFileNote
Description: getDescription(option),
ImageId: setting.FileNoteBook.ImageIdNPU,
Spec: spec,
BootFile: "",
BootFile: getBootFile(option.File, option.OwnerName, option.ProjectName),
AutoStopDurationMs: modelarts.AutoStopDurationMs / 4,
BranchName: option.BranchName,
}

if setting.ModelartsCD.Enabled {
@@ -347,17 +568,8 @@ func isNoteBookFileExist(ctx *context.Context, option api.CreateFileNotebookJobO
return true, nil
}

func uploadCodeFile(repo *models.Repository, codePath string, branchName string, filePath string, jobName string) string {
err := downloadCode(repo, codePath, branchName)
if err != nil {
return "cloudbrain.load_code_failed"
}

err = uploadOneFileToMinio(codePath, filePath, jobName, cloudbrain.CodeMountPath+"/")
if err != nil {
return "cloudbrain.load_code_failed"
}
return ""
func getBootFile(filePath string, ownerName string, projectName string) string {
return ownerName + "/" + projectName + "/" + filePath
}

func fileExists(gitRepo *git.Repository, path string, branch string) (bool, error) {


+ 3
- 3
templates/repo/modelarts/trainjob/show.tmpl View File

@@ -423,9 +423,9 @@
<input type="hidden" name="model{{.VersionName}}" value="-1">
<input type="hidden" name="modelback{{.VersionName}}" value="-1">
<div class='ui breadcrumb model_file_bread' id='file_breadcrumb{{.VersionName}}'>
<div class="active section">{{.VersionName}}</div>
<div class="divider"> / </div>
<div class="active section">{{.VersionName}}</div>
<div class="divider"> / </div>
</div>
<div id="dir_list{{.VersionName}}" style="max-height: 500px;overflow:auto;">



+ 11
- 2
web_src/vuepages/apis/modules/notobook.js View File

@@ -29,7 +29,7 @@ export const getCb1Notebook = (path,jobid) => {
});
};

// Notebook获取云脑I调试任务状态
// Notebook获取云脑II调试任务状态
export const getCb2Notebook = (path,jobid) => {
return service({
url: `/api/v1/${path}/modelarts/notebook/${jobid}`,
@@ -37,7 +37,16 @@ export const getCb2Notebook = (path,jobid) => {
params: {},
});
};

// Notebook查询文件在环境中是否已准备好
// type, file, branch_name, owner_name, project_name,job_id
export const getFileInfoNotebook = (data) => {
return service({
url: "/api/v1/file_notebook/status",
method: "post",
data,
params: {},
});
};
export const stopNotebook = (url) => {
return service({
url: url,


+ 60
- 29
web_src/vuepages/pages/notebook/debug/index.vue View File

@@ -178,9 +178,9 @@
</div>
</template>
<script>
import { getFileNotebook,createNotebook,getCb1Notebook,getCb2Notebook,stopNotebook } from "~/apis/modules/notobook";
import { getFileNotebook,createNotebook,getCb1Notebook,getCb2Notebook,getFileInfoNotebook,stopNotebook } from "~/apis/modules/notobook";
import { Message } from "element-ui";
let timerCb1,timerCb2
let timerCb1,timerCb2,timerCb3
let {AppSubUrl} = window.config
const finalState = [
"STOPPED",
@@ -253,27 +253,28 @@ export default {
if(this.activeLoadFirst){
this.loading = true
}
getFileNotebook().then((res)=>{
if(res.data.code==0){
this.notebookInfo = res.data
}else{
Message.error(res.data.message)
}
this.loading = false
this.activeLoadFirst = false
}).catch((err)=>{
Message.error(err)
this.loading = false
this.activeLoadFirst = false
})
getFileNotebook().then((res)=>{
if(res.data.code==0){
this.notebookInfo = res.data
}else{
Message.error(res.data.message)
}
this.loading = false
this.activeLoadFirst = false
}).catch((err)=>{
Message.error(err)
this.loading = false
this.activeLoadFirst = false
})
},
getCb1NotebookInfo(path,id,index){
getCb1NotebookInfo(path,id,index,data){
getCb1Notebook(path,id).then((res)=>{
if(res.status===200){
if(res.data.JobStatus==="RUNNING"){
this.btnStatus[index]=2
this.deubgUrlGpu = `${AppSubUrl}/${this.fileInfo.sign_name}/${this.notebookInfo.projectName}/cloudbrain/${id}/debug`
this.deubgUrlGpuStop = `${AppSubUrl}/${this.fileInfo.sign_name}/${this.notebookInfo.projectName}/cloudbrain/${id}/stop`
let fileData = {job_id:id,...data}
timerCb3 = setInterval(() => {
setTimeout(this.getFileInfoReadyNotebook(fileData,index), 0)
}, 5000)
clearInterval(timerCb1)
}
if(finalState.includes(res.data.JobStatus)){
@@ -281,23 +282,53 @@ export default {
clearInterval(timerCb1)
}
}
})
}).catch((err)=>{
this.btnStatus[index]=0
clearInterval(timerCb1)
Message.error(err)
})
},
getCb2NotebookInfo(path,id){
getCb2NotebookInfo(path,id,data){
getCb2Notebook(path,id).then((res)=>{
if(res.status===200){
if(res.data.JobStatus==="RUNNING"){
this.btnStatus[2]=2
this.deubgUrlNpu = `${AppSubUrl}/${this.fileInfo.sign_name}/${this.notebookInfo.projectName}/modelarts/notebook/${id}/debug`
this.deubgUrlNpuStop = `${AppSubUrl}/${this.fileInfo.sign_name}/${this.notebookInfo.projectName}/modelarts/notebook/${id}/stop`
clearInterval(timerCb2)
if(res.data.JobStatus==="RUNNING"){
let fileData = {job_id:id,...data}
timerCb3 = setInterval(() => {
setTimeout(this.getFileInfoReadyNotebook(fileData,2), 0)
}, 5000)
clearInterval(timerCb2)
}
if(finalState.includes(res.data.JobStatus)){
this.btnStatus[2] = 0
clearInterval(timerCb2)
}
}
})
}).catch((err)=>{
this.btnStatus[index]=0
clearInterval(timerCb2)
Message.error(err)
})
},
getFileInfoReadyNotebook(data,index){
getFileInfoNotebook(data).then((res)=>{
console.log(res)
if(res.data.code===0){
if(index===2){
this.btnStatus[2]=2
this.deubgUrlNpu = `${AppSubUrl}/${this.fileInfo.sign_name}/${this.notebookInfo.projectName}/modelarts/notebook/${data.job_id}/debug?file=${this.fileInfo.owner_name}/${this.fileInfo.project_name}/${this.fileInfo.file}`
this.deubgUrlNpuStop = `${AppSubUrl}/${this.fileInfo.sign_name}/${this.notebookInfo.projectName}/modelarts/notebook/${data.job_id}/stop`
}else{
this.btnStatus[index]=2
this.deubgUrlGpu = `${AppSubUrl}/${this.fileInfo.sign_name}/${this.notebookInfo.projectName}/cloudbrain/${data.job_id}/debug?file=${this.fileInfo.owner_name}/${this.fileInfo.project_name}/${this.fileInfo.file}`
this.deubgUrlGpuStop = `${AppSubUrl}/${this.fileInfo.sign_name}/${this.notebookInfo.projectName}/cloudbrain/${data.job_id}/stop`
}
clearInterval(timerCb3)
}
}).catch((err)=>{
this.btnStatus[index]=0
clearInterval(timerCb3)
Message.error(err)
})
},
stopDebug(index){
this.btnStatus[index]=3
@@ -323,11 +354,11 @@ export default {
if(res.data.code===0 && res.status===200){
if(index===2){
timerCb2 = setInterval(() => {
setTimeout(this.getCb2NotebookInfo(repoPath,res.data.message), 0)
setTimeout(this.getCb2NotebookInfo(repoPath,res.data.message,data), 0)
}, 10000)
}else{
timerCb1 = setInterval(() => {
setTimeout(this.getCb1NotebookInfo(repoPath,res.data.message,index), 0)
setTimeout(this.getCb1NotebookInfo(repoPath,res.data.message,index,data), 0)
}, 10000)
}
this.alertCb = false


Loading…
Cancel
Save