public static async Task <HttpResponseMessage> CreateSingleStructurePridiction(this OsdrWebClient client, RunSingleStructurePrediction ssp)
{
var postParameters = JsonConvert.SerializeObject(ssp);
var response = await client.PostData("api/machinelearning/predictions/structure", postParameters);
return(response);
}
public static async Task <HttpResponseMessage> MachineLearningPredict(this OsdrWebClient client,
Guid parentId, Guid modelBlobId, Guid datasetBlobId, Guid userId, string datasetBucket, Guid modelBucket, string folderName)
{
var postData = $@"
{{
'TargetFolderId': '{parentId}',
'DatasetBlobId': '{datasetBlobId}',
'DatasetBucket': '{datasetBucket}',
'ModelBlobId': '{modelBlobId}',
'ModelBucket': '{modelBucket}',
'UserId': '{userId}'
}}" ;
return(await client.PostData("/api/machinelearning/predictions", postData));
}
public static async Task <HttpResponseMessage> MachineLearningTrain(this OsdrWebClient client,
Guid sourceBlobId, string sourceBucket, Guid userId, Guid parentId, bool optimize)
{
var folderId = Guid.NewGuid();
var createMLModel = new CreateMachineLearningModel
{
TargetFolderId = folderId,
SourceBlobId = sourceBlobId,
Scaler = "some sting named as Scaler",
SourceBucket = sourceBucket,
UserId = userId,
SourceFileName = "combined lysomotrophic.sdf",
Methods = new List <string> {
"NaiveBayes"
},
TrainingParameter = "Soluble",
SubSampleSize = 1,
TestDataSize = new decimal(.1),
KFold = 2,
ModelType = "Classification",
Fingerprints = new List <Fingerprint>
{
new Fingerprint {
Type = "ecfp",
Radius = 2,
Size = 512
}
},
Optimize = optimize
};
var postParameters = JsonConvert.SerializeObject(createMLModel);
var response = await client.PostData("/api/machinelearning/models", postParameters);
return(response);
}
public static async Task <HttpResponseMessage> CreateFolderEntity(this OsdrWebClient client, Guid parentNodeId, string name)
{
return(await client.PostData("api/entities/folders", $"{{'Name': '{name}', parentId: '{parentNodeId}'}}"));
}
public static async Task <HttpResponseMessage> MachineLearningCreate(this OsdrWebClient client, Guid parentNodeId, string name)
{
return(await client.PostData("/api/machinelearning/predictions", $"{{'Name': '{name}', parentId: '{parentNodeId}'}}"));
}