Merge remote-tracking branch 'origin/0.0.1' into 0.0.1

.gitattributes

@@ -0,0 +1,35 @@
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ckpt filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.mlmodel filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.npy filter=lfs diff=lfs merge=lfs -text
+*.npz filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pickle filter=lfs diff=lfs merge=lfs -text
+*.pkl filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+*.safetensors filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tar filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.xz filter=lfs diff=lfs merge=lfs -text
+*.zip filter=lfs diff=lfs merge=lfs -text
+*.zst filter=lfs diff=lfs merge=lfs -text
+*tfevents* filter=lfs diff=lfs merge=lfs -text

20240219162750_指纹_ULcpof.fingerprint

@@ -0,0 +1 @@
+0fac1eeefb2ef8a3c7f136f0a6a570874709ff332a17af4b812270e4fc29148f

README.md

@@ -0,0 +1,3 @@
+# restNet50-sss20240819140609
+
+dd

gitea.go

@@ -0,0 +1,424 @@
+package gitea
+
+import (
+	"crypto/tls"
+	"encoding/base64"
+	"encoding/json"
+	"errors"
+	"fmt"
+	"io"
+	"model-repository/domain"
+	"net/http"
+	"regexp"
+	"strconv"
+	"time"
+
+	"github.com/go-resty/resty/v2"
+	"gopkg.in/yaml.v2"
+)
+
+const (
+	DEFALUT_REVISION  = "main"
+	TIME_FORMAT       = "2006-01-02T15:04:05.000Z"
+	REF_VERSION_LIMIT = 20
+)
+
+var (
+	TagList = []string{
+		"audio-classification",
+		"automatic-speech-recognition",
+		"conversational",
+		"depth-estimation",
+		"document-question-answering",
+		"feature-extraction",
+		"fill-mask",
+		"image-classification",
+		"image-feature-extraction",
+		"image-segmentation",
+		"image-to-image",
+		"image-to-text",
+		"mask-generation",
+		"ner",
+		"object-detection",
+		"question-answering",
+		"sentiment-analysis",
+		"summarization",
+		"table-question-answering",
+		"text-classification",
+		"text-generation",
+		"text-to-audio",
+		"text-to-speech",
+		"text2text-generation",
+		"token-classification",
+		"translation",
+		"video-classification",
+		"visual-question-answering",
+		"vqa",
+		"zero-shot-audio-classification",
+		"zero-shot-classification",
+		"zero-shot-image-classification",
+		"zero-shot-object-detection",
+		"translation_XX_to_YY",
+	}
+)
+
+type GiteaStorage struct {
+	client *resty.Client
+}
+
+func NewGiteaStorage(baseUrl string) *GiteaStorage {
+	client := resty.New().SetTLSClientConfig(&tls.Config{InsecureSkipVerify: true}).SetBaseURL(baseUrl).SetTimeout(5 * time.Minute)
+	return &GiteaStorage{
+		client: client,
+	}
+}
+
+func (gs *GiteaStorage) GetPipelineTag(hfModel *domain.HuggingFaceModelRequest) (*Metadata, error) {
+	url := fmt.Sprintf("/api/v1/repos/%s/%s/raw/README.md", hfModel.Project, hfModel.Name)
+	resp, err := gs.client.R().
+		SetHeader("Accept", "application/json").
+		SetHeader("access_token", hfModel.AccessToken).
+		SetQueryParams(map[string]string{
+			"ref": safeSubstring(hfModel.Version, REF_VERSION_LIMIT),
+		}).
+		Get(url)
+	if err != nil {
+		return nil, err
+	}
+	var metadata Metadata
+	err = yaml.Unmarshal(resp.Body(), &metadata)
+	if err != nil {
+		fmt.Println("Error parsing YAML:", err)
+		return nil, err
+	}
+	return &metadata, nil
+}
+
+func (gs *GiteaStorage) GetModelInfo(hfModel *domain.HuggingFaceModelRequest) (*domain.HuggingFaceModelInfo, error) {
+	url := ""
+	mir := &ModelInfoResp{}
+	hfmi := &domain.HuggingFaceModelInfo{}
+
+	if hfModel.Version == "" {
+		url = fmt.Sprintf("/api/v1/repos/%s/%s/git/trees/%s", hfModel.Project, hfModel.Name, DEFALUT_REVISION)
+	} else {
+		url = fmt.Sprintf("/api/v1/repos/%s/%s/git/trees/%s", hfModel.Project, hfModel.Name, hfModel.Version)
+	}
+	resp, err := gs.client.R().
+		SetHeader("Accept", "application/json").
+		SetHeader("access_token", hfModel.AccessToken).
+		SetQueryParams(map[string]string{
+			"recursive": "true",
+		}).
+		SetResult(mir).
+		Get(url)
+	defer resp.RawBody().Close()
+
+	if err != nil {
+		return nil, err
+	}
+	if resp.StatusCode() == 404 {
+		return nil, fmt.Errorf(domain.HUGGINGFACE_HEADER_X_Error_Code_REPO_NotFound)
+	}
+	for _, file := range mir.Tree {
+		hfmi.Siblings = append(hfmi.Siblings, domain.Sibling{Rfilename: file.Path})
+	}
+	hfmi.LastModified = time.Now().Format(TIME_FORMAT)
+	hfmi.CreatedAt = time.Now().Format(TIME_FORMAT)
+	hfmi.ModelInfoID = hfModel.Project + "/" + hfModel.Name
+	hfmi.ModelID = hfModel.Project + "/" + hfModel.Name
+	hfmi.SHA = mir.SHA
+	hfmi.LibraryName = "transformers"
+	readmeMeta, err := gs.GetPipelineTag(hfModel)
+	if err != nil {
+		return nil, err
+	}
+	hfmi.Tags = readmeMeta.Tags
+	if readmeMeta.PipLinetag == "" {
+		for _, rtag := range readmeMeta.Tags {
+			for _, tag := range TagList {
+				if rtag == tag {
+					hfmi.PipelineTag = tag
+					break
+				}
+			}
+		}
+	} else {
+		hfmi.PipelineTag = readmeMeta.PipLinetag
+	}
+	return hfmi, nil
+}
+
+func (gs *GiteaStorage) GetModelFileTree(hfModel *domain.HuggingFaceModelRequest) ([]domain.HuggingFaceModelFileInfo, error) {
+	url := ""
+	mir := &ModelInfoResp{}
+	var tree []domain.HuggingFaceModelFileInfo
+
+	if hfModel.Version == "" {
+		url = fmt.Sprintf("/api/v1/repos/%s/%s/git/trees/%s", hfModel.Project, hfModel.Name, DEFALUT_REVISION)
+	} else {
+		url = fmt.Sprintf("/api/v1/repos/%s/%s/git/trees/%s", hfModel.Project, hfModel.Name, hfModel.Version)
+	}
+	resp, err := gs.client.R().
+		SetHeader("Accept", "application/json").
+		SetHeader("access_token", hfModel.AccessToken).
+		SetQueryParams(map[string]string{
+			"recursive": "true",
+		}).
+		SetResult(mir).
+		Get(url)
+	if err != nil {
+		return nil, err
+	}
+	defer resp.RawBody().Close()
+	for _, file := range mir.Tree {
+		if domain.IsLfsFile(file.Path) {
+			content, err := gs.GetFileContent(&domain.HuggingFaceFileRequest{
+				Project:     hfModel.Project,
+				Name:        hfModel.Name,
+				File:        file.Path,
+				Version:     hfModel.Version,
+				AccessToken: hfModel.AccessToken,
+			})
+			if err != nil {
+				return nil, err
+			}
+			size, hash := gs.ExtractLfsMetadata(content)
+			tree = append(tree, domain.HuggingFaceModelFileInfo{
+				Type: string(file.Type),
+				Oid:  file.SHA,
+				Size: size,
+				Path: file.Path,
+				Lfs: &domain.HuggingFaceModelLfsFileInfo{
+					Oid:         hash,
+					Size:        size,
+					PointerSize: file.Size,
+				},
+			})
+		} else {
+			tree = append(tree, domain.HuggingFaceModelFileInfo{
+				Type: string(file.Type),
+				Oid:  file.SHA,
+				Size: file.Size,
+				Path: file.Path,
+			})
+		}
+	}
+	return tree, nil
+}
+
+func (gs *GiteaStorage) GetFileContent(hfi *domain.HuggingFaceFileRequest) (string, error) {
+	url := fmt.Sprintf("/api/v1/repos/%s/%s/raw/%s", hfi.Project, hfi.Name, hfi.File)
+	resp, err := gs.client.R().
+		SetHeader("access_token", hfi.AccessToken).
+		SetQueryParams(map[string]string{
+			"ref": safeSubstring(hfi.Version, REF_VERSION_LIMIT),
+		}).
+		Get(url)
+	if err != nil {
+		return "", err
+	}
+	return string(resp.Body()), nil
+}
+
+func (gs *GiteaStorage) GetFileMetadata(hfFile *domain.HuggingFaceFileRequest) (*domain.GitFileMetadata, error) {
+	// TODO Implement me
+	gfd := &domain.GitFileMetadata{}
+	metaresp := &FileMetaResp{}
+	url := fmt.Sprintf("/api/v1/repos/%s/%s/contents/%s", hfFile.Project, hfFile.Name, hfFile.File)
+	resp, err := gs.client.R().
+		SetHeader("Accept", "application/json").
+		SetHeader("access_token", hfFile.AccessToken).
+		SetQueryParams(map[string]string{
+			// 支持commitId(前10位)，branch，tag
+			"ref": safeSubstring(hfFile.Version, REF_VERSION_LIMIT),
+		}).
+		SetResult(metaresp).
+		Get(url)
+	if err != nil {
+		return nil, err
+	}
+	defer resp.RawBody().Close()
+
+	if domain.IsLfsFile(hfFile.File) {
+		decodeContent, err := base64.StdEncoding.DecodeString(metaresp.Content)
+		if err != nil {
+			return nil, err
+		}
+		match := Regexp(`size (\d+)`, string(decodeContent))
+		if len(match) < 2 {
+			return nil, errors.New("size not found")
+		}
+		size, err := strconv.Atoi(match[1])
+		if err != nil {
+			return nil, err
+		}
+		sha256 := Regexp("sha256:([a-f0-9]+)", string(decodeContent))
+		gfd.Size = int64(size)
+		gfd.SHA256 = sha256[1]
+
+	} else {
+		gfd.Size = int64(metaresp.Size)
+	}
+	gfd.LastCommitID = metaresp.LastCommitSHA
+	gfd.BlobID = metaresp.SHA
+	gfd.FileName = metaresp.Name
+	return gfd, nil
+
+}
+
+func safeSubstring(s string, n int) string {
+	runes := []rune(s)
+	if len(runes) > n {
+		return string(runes[:n])
+	}
+	return s
+}
+
+func (gs *GiteaStorage) ExtractLfsMetadata(content string) (int64, string) {
+	match := Regexp(`size (\d+)`, content)
+	if len(match) < 2 {
+		return 0, ""
+	}
+	size, err := strconv.Atoi(match[1])
+	if err != nil {
+		return 0, ""
+	}
+	sha256 := Regexp("sha256:([a-f0-9]+)", content)
+	return int64(size), sha256[1]
+}
+
+func (gs *GiteaStorage) GetLfsRawFile(hfFile *domain.HuggingFaceFileRequest, writer io.Writer) error {
+	url := fmt.Sprintf("/api/v1/repos/%s/%s/media/%s", hfFile.Project, hfFile.Name, hfFile.File)
+	resp, _ := gs.client.R().SetDoNotParseResponse(true).
+		SetHeader("Accept", "application/json").
+		SetHeader("access_token", hfFile.AccessToken).
+		SetQueryParams(map[string]string{
+			"ref": safeSubstring(hfFile.Version, REF_VERSION_LIMIT),
+		}).
+		Get(url)
+	defer resp.RawBody().Close() // 确保关闭流
+
+	// 使用 buffer 进行流式读写，提高性能
+	buffer := make([]byte, 4096*1024) // 4M buffer
+	_, err := io.CopyBuffer(writer, resp.RawBody(), buffer)
+	return err
+}
+
+func (gs *GiteaStorage) CreateRepo(repo *domain.CreateRepoInput, token string) (*domain.CreateRepoOutput, error) {
+	url := fmt.Sprintf("/api/v1/user/repos")
+
+	// 构建请求体
+	requestBody := map[string]interface{}{
+		"auto_init":      false,
+		"default_branch": DEFALUT_REVISION,
+		"name":           repo.Name,
+		"private":        true,
+		"template":       false,
+		"trust_model":    "default",
+	}
+	requestBodyBytes, _ := json.Marshal(requestBody)
+
+	resp, err := gs.client.R().
+		SetHeader("Authorization", fmt.Sprintf("token %s", token)).
+		SetHeader("Content-Type", "application/json").
+		SetBody(requestBodyBytes).
+		Post(url)
+
+	if err != nil {
+		return nil, err
+	}
+	defer resp.RawBody().Close() // 确保关闭流
+
+	if resp.StatusCode() == http.StatusConflict {
+		output := &domain.CreateRepoOutput{
+			URL: domain.HUGGINGFACE_ENDPOINT + "/" + repo.Organization + "/" + repo.Name,
+		}
+		return output, fmt.Errorf(domain.HUGGINGFACE_HEADER_X_Error_Code_REPO_EXIST)
+	}
+
+	if resp.StatusCode() != http.StatusCreated {
+		return nil, fmt.Errorf("failed to create repository, status code: %d", resp.StatusCode())
+	}
+	// 解析响应体
+	var responseMap map[string]interface{}
+	if err := json.Unmarshal(resp.Body(), &responseMap); err != nil {
+		return nil, fmt.Errorf("failed to parse response body: %w", err)
+	}
+
+	// 提取 id 字段
+	id, ok := responseMap["id"].(float64)
+	if !ok {
+		return nil, fmt.Errorf("id field not found or not a float64 in response")
+	}
+	output := &domain.CreateRepoOutput{
+		URL:  domain.HUGGINGFACE_ENDPOINT + "/" + repo.Organization + "/" + repo.Name,
+		Name: repo.Organization + "/" + repo.Name,
+		//从 resp id中获取
+		ID: fmt.Sprintf("%f", id),
+	}
+	return output, nil
+}
+
+func Regexp(reg string, input string) []string {
+	re := regexp.MustCompile(reg)
+	return re.FindStringSubmatch(input)
+}
+
+type FileMetaResp struct {
+	Name            string `json:"name"`
+	Path            string `json:"path"`
+	SHA             string `json:"sha"`
+	LastCommitSHA   string `json:"last_commit_sha"`
+	Type            string `json:"type"`
+	Size            int64  `json:"size"`
+	Encoding        string `json:"encoding"`
+	Content         string `json:"content"`
+	Target          string `json:"target"`
+	URL             string `json:"url"`
+	HTMLURL         string `json:"html_url"`
+	GitURL          string `json:"git_url"`
+	DownloadURL     string `json:"download_url"`
+	SubmoduleGitURL string `json:"submodule_git_url"`
+	Links           Links  `json:"_links"`
+}
+
+type Links struct {
+	Self string `json:"self"`
+	Git  string `json:"git"`
+	HTML string `json:"html"`
+}
+
+type ModelInfoResp struct {
+	SHA        string `json:"sha"`
+	URL        string `json:"url"`
+	Tree       []Tree `json:"tree"`
+	Truncated  bool   `json:"truncated"`
+	Page       int64  `json:"page"`
+	TotalCount int64  `json:"total_count"`
+}
+
+type Tree struct {
+	Path string `json:"path"`
+	Mode string `json:"mode"`
+	Type Type   `json:"type"`
+	Size int64  `json:"size"`
+	SHA  string `json:"sha"`
+	URL  string `json:"url"`
+}
+
+type Type string
+
+const (
+	Blob Type = "blob"
+)
+
+type Metadata struct {
+	License     string   `yaml:"license"`
+	LicenseName string   `yaml:"license_name"`
+	LicenseLink string   `yaml:"license_link"`
+	Tags        []string `yaml:"tags"`
+	Datasets    []string `yaml:"datasets"`
+	PipLinetag  string   `yaml:"pipeline_tag"`
+	Language    []string `yaml:"language"`
+}

preprocessor_config.json

@@ -0,0 +1,18 @@
+{
+  "crop_pct": 0.875,
+  "do_normalize": true,
+  "do_resize": true,
+  "feature_extractor_type": "ConvNextFeatureExtractor",
+  "image_mean": [
+    0.485,
+    0.456,
+    0.406
+  ],
+  "image_std": [
+    0.229,
+    0.224,
+    0.225
+  ],
+  "resample": 3,
+  "size": 224
+}

新建文本文档 (3).h5

@@ -0,0 +1,23 @@
+使用LFS进行大文件上传
+lfs下载地址：https://git-lfs.com/
+git lfs install
+初始化本地仓库
+gii init
+添加远程仓库
+git remote add origin http://10.0.101.71:3000/wsg123/Test-wsg-001.git
+切换到 main 分支（如果还没有 main 分支，先创建它）
+git checkout -b main
+跟踪大文件
+git lfs track ".zip" 文件后缀
+添加并提交本地文件
+git add .
+git commit -m "你的提交信息"
+从远程仓库获取最新的更改
+git fetch origin
+强制合并不相关的历史记录
+git merge origin/main --allow-unrelated-histories
+配置Git用户名和邮箱
+git config --global user.name "Your Name"
+git config --global user.email you@example.com
+将本地 main 分支推送到远程仓库
+git push -u origin main