diff options
Diffstat (limited to 'backend')
24 files changed, 286 insertions, 142 deletions
diff --git a/backend/api/api/Controllers/ExperimentController.cs b/backend/api/api/Controllers/ExperimentController.cs index 3fa02943..eecbe756 100644 --- a/backend/api/api/Controllers/ExperimentController.cs +++ b/backend/api/api/Controllers/ExperimentController.cs @@ -88,5 +88,28 @@ namespace api.Controllers var experiments=_experimentService.GetMyExperiments(uploaderId); return Ok(experiments); } + + // PUT api/<ExperimentController>/{name} + [HttpPut("{id}")] + [Authorize(Roles = "User")] + public ActionResult Put(string id, [FromBody] Experiment experiment) + { + string uploaderId = getUserId(); + + if (uploaderId == null) + return BadRequest(); + + var existingDataset = _experimentService.GetOneExperiment(uploaderId, id); + + //ne mora da se proverava + if (existingDataset == null) + return NotFound($"Experiment with ID = {id} or user with ID = {uploaderId} not found"); + + experiment.lastUpdated = DateTime.UtcNow; + + _experimentService.Update(uploaderId, id, experiment); + + return Ok($"Experiment with ID = {id} updated"); + } } } diff --git a/backend/api/api/Controllers/FileController.cs b/backend/api/api/Controllers/FileController.cs index 9baf6294..99d98a78 100644 --- a/backend/api/api/Controllers/FileController.cs +++ b/backend/api/api/Controllers/FileController.cs @@ -94,9 +94,9 @@ namespace api.Controllers return Ok(fileModel._id); } - [HttpGet("csvread/{hasHeader}/{fileId}/{skipRows}/{takeRows}")] + [HttpGet("csvread/{fileId}/{skipRows}/{takeRows}")] [Authorize(Roles = "User,Guest")] - public ActionResult<string> CsvRead(bool hasHeader, string fileId, int skipRows = 0, int takeRows = 10) + public ActionResult<string> CsvRead(string fileId, int skipRows = 0, int takeRows = 10) { string uploaderId = getUserId(); @@ -109,10 +109,7 @@ namespace api.Controllers - if (hasHeader) - return String.Join("\n", System.IO.File.ReadLines(filePath).Skip(skipRows+1).Take(takeRows)); - else - return String.Join("\n", System.IO.File.ReadLines(filePath).Skip(skipRows).Take(takeRows)); + return String.Join("\n", System.IO.File.ReadLines(filePath).Skip(skipRows+1).Take(takeRows)); } diff --git a/backend/api/api/Controllers/ModelController.cs b/backend/api/api/Controllers/ModelController.cs index f279bf7a..1c1ea364 100644 --- a/backend/api/api/Controllers/ModelController.cs +++ b/backend/api/api/Controllers/ModelController.cs @@ -187,8 +187,11 @@ namespace api.Controllers /*if (_modelService.CheckHyperparameters(1, model.hiddenLayerNeurons, model.hiddenLayers, model.outputNeurons) == false) return BadRequest("Bad parameters!");*/ + model.uploaderId = getUserId(); + var existingModel = _modelService.GetOneModel(model.uploaderId, model.name); + if (existingModel != null && !overwrite) return NotFound($"Model with name = {model.name} exisits"); else diff --git a/backend/api/api/Services/IAuthService.cs b/backend/api/api/Interfaces/IAuthService.cs index 9a109208..9a109208 100644 --- a/backend/api/api/Services/IAuthService.cs +++ b/backend/api/api/Interfaces/IAuthService.cs diff --git a/backend/api/api/Services/IDatasetService.cs b/backend/api/api/Interfaces/IDatasetService.cs index bb06208d..f493a2ec 100644 --- a/backend/api/api/Services/IDatasetService.cs +++ b/backend/api/api/Interfaces/IDatasetService.cs @@ -5,7 +5,7 @@ namespace api.Services { public interface IDatasetService { - Dataset GetOneDataset(string userId, string id); + Dataset GetOneDataset(string userId, string name); Dataset GetOneDataset(string id); List<Dataset> SearchDatasets(string name); List<Dataset> GetMyDatasets(string userId); diff --git a/backend/api/api/Services/IExperimentService.cs b/backend/api/api/Interfaces/IExperimentService.cs index 47c86046..2a69cff9 100644 --- a/backend/api/api/Services/IExperimentService.cs +++ b/backend/api/api/Interfaces/IExperimentService.cs @@ -8,5 +8,8 @@ namespace api.Services public Experiment Get(string id); public List<Experiment> GetMyExperiments(string id); public Experiment Get(string uploaderId, string name); + Experiment GetOneExperiment(string userId, string name); + void Update(string userId, string id, Experiment experiment); + } }
\ No newline at end of file diff --git a/backend/api/api/Services/IFileService.cs b/backend/api/api/Interfaces/IFileService.cs index e061dfdb..e061dfdb 100644 --- a/backend/api/api/Services/IFileService.cs +++ b/backend/api/api/Interfaces/IFileService.cs diff --git a/backend/api/api/Models/IJwtToken.cs b/backend/api/api/Interfaces/IJwtToken.cs index 5c54e4e3..5c54e4e3 100644 --- a/backend/api/api/Models/IJwtToken.cs +++ b/backend/api/api/Interfaces/IJwtToken.cs diff --git a/backend/api/api/Services/IMLWebSocketService.cs b/backend/api/api/Interfaces/IMLWebSocketService.cs index 52efb7fc..52efb7fc 100644 --- a/backend/api/api/Services/IMLWebSocketService.cs +++ b/backend/api/api/Interfaces/IMLWebSocketService.cs diff --git a/backend/api/api/Services/IMlConnectionService.cs b/backend/api/api/Interfaces/IMlConnectionService.cs index d5dda9f2..d5dda9f2 100644 --- a/backend/api/api/Services/IMlConnectionService.cs +++ b/backend/api/api/Interfaces/IMlConnectionService.cs diff --git a/backend/api/api/Services/IModelService.cs b/backend/api/api/Interfaces/IModelService.cs index 00299979..00299979 100644 --- a/backend/api/api/Services/IModelService.cs +++ b/backend/api/api/Interfaces/IModelService.cs diff --git a/backend/api/api/Services/IPredictorService.cs b/backend/api/api/Interfaces/IPredictorService.cs index 16f0432a..16f0432a 100644 --- a/backend/api/api/Services/IPredictorService.cs +++ b/backend/api/api/Interfaces/IPredictorService.cs diff --git a/backend/api/api/Services/IUserService.cs b/backend/api/api/Interfaces/IUserService.cs index d34d410a..d34d410a 100644 --- a/backend/api/api/Services/IUserService.cs +++ b/backend/api/api/Interfaces/IUserService.cs diff --git a/backend/api/api/Models/ColumnInfo.cs b/backend/api/api/Models/ColumnInfo.cs index 04450fef..dcf5171c 100644 --- a/backend/api/api/Models/ColumnInfo.cs +++ b/backend/api/api/Models/ColumnInfo.cs @@ -4,9 +4,10 @@ { public ColumnInfo() { } - public ColumnInfo(string columnName, bool isNumber, int numNulls, float mean, float min, float max, float median, string[] uniqueValues) + public ColumnInfo(string columnName, string columnType, bool isNumber, int numNulls, float mean, float min, float max, float median, string[] uniqueValues, int[]uniqueValuesCount, float[] uniqueValuesPercent, float q1, float q3) { this.columnName = columnName; + this.columnType = columnType; this.isNumber = isNumber; this.numNulls = numNulls; this.mean = mean; @@ -14,9 +15,14 @@ this.max = max; this.median = median; this.uniqueValues = uniqueValues; + this.uniqueValuesPercent = uniqueValuesPercent; + this.uniqueValuesCount = uniqueValuesCount; + this.q1 = q1; + this.q3 = q3; } public string columnName { get; set; } + public string columnType { get; set; } public bool isNumber { get; set; } public int numNulls { get; set; } public float mean { get; set; } @@ -24,6 +30,12 @@ public float max { get; set; } public float median { get; set; } public string[] uniqueValues { get; set; } + public int[] uniqueValuesCount { get; set; } + public float[] uniqueValuesPercent { get; set; } + + + public float q1 { get; set; } + public float q3 { get; set; } } } diff --git a/backend/api/api/Models/Dataset.cs b/backend/api/api/Models/Dataset.cs index cc7185f0..7acd4382 100644 --- a/backend/api/api/Models/Dataset.cs +++ b/backend/api/api/Models/Dataset.cs @@ -12,9 +12,9 @@ namespace api.Models [BsonId] [BsonRepresentation(BsonType.ObjectId)]//mongo data type to .net public string _id { get; set; } - public string name { get; set; } + public string description { get; set; } - public string[] header { get; set; } + public string name { get; set; } public string fileId { get; set; } public string extension { get; set; } public bool isPublic { get; set; } @@ -22,8 +22,6 @@ namespace api.Models public DateTime dateCreated { get; set; } public DateTime lastUpdated { get; set; } public string delimiter { get; set; } - public bool hasHeader { get; set; } - public ColumnInfo[] columnInfo { get; set; } public int rowCount { get; set; } public int nullCols { get; set; } diff --git a/backend/api/api/Models/Experiment.cs b/backend/api/api/Models/Experiment.cs index 6f665c52..3af063be 100644 --- a/backend/api/api/Models/Experiment.cs +++ b/backend/api/api/Models/Experiment.cs @@ -10,15 +10,15 @@ namespace api.Models public string _id { get; set; } public string name { get; set; } public string description { get; set; } + public string type { get; set; } public List<string> ModelIds { get; set; } public string datasetId { get; set; } public string uploaderId { get; set; } public string[] inputColumns { get; set; } public string outputColumn { get; set; } - public bool randomOrder { get; set; } - public bool randomTestSet { get; set; } - public float randomTestSetDistribution { get; set; } public string nullValues { get; set; } + public DateTime dateCreated { get; set; } + public DateTime lastUpdated { get; set; } public NullValues[] nullValuesReplacers { get; set; } public ColumnEncoding[] encodings { get; set; } diff --git a/backend/api/api/Models/Model.cs b/backend/api/api/Models/Model.cs index a9dbfbdd..d8921713 100644 --- a/backend/api/api/Models/Model.cs +++ b/backend/api/api/Models/Model.cs @@ -26,17 +26,30 @@ namespace api.Models public string optimizer { get; set; } public string lossFunction { get; set; } //public int inputNeurons { get; set; } - public int hiddenLayerNeurons { get; set; } public int hiddenLayers { get; set; } public int batchSize { get; set; } // na izlazu je moguce da bude vise neurona (klasifikacioni problem sa vise od 2 klase) public int outputNeurons { get; set; } - public string[] hiddenLayerActivationFunctions { get; set; } + public Layer[] layers { get; set; } public string outputLayerActivationFunction { get; set; } public string[] metrics { get; set; } public int epochs { get; set; } //public bool isTrained { get; set; } //public NullValues[] nullValues { get; set; } + public bool randomOrder { get; set; } + public bool randomTestSet { get; set; } + public float randomTestSetDistribution { get; set; } + } + + public class Layer + { + public int layerNumber { get; set; } + public string activationFunction { get; set; } + public int neurons { get; set; } + public string regularisation { get; set; } + public float regularisationRate { get; set; } } } + + diff --git a/backend/api/api/Services/DatasetService.cs b/backend/api/api/Services/DatasetService.cs index fe177d77..f39cac29 100644 --- a/backend/api/api/Services/DatasetService.cs +++ b/backend/api/api/Services/DatasetService.cs @@ -62,9 +62,9 @@ namespace api.Services return _dataset.Find(dataset => dataset.isPublic == true && dataset.isPreProcess).ToList(); } - public Dataset GetOneDataset(string userId, string id) + public Dataset GetOneDataset(string userId, string name) { - return _dataset.Find(dataset => dataset.uploaderId == userId && dataset._id == id && dataset.isPreProcess).FirstOrDefault(); + return _dataset.Find(dataset => dataset.uploaderId == userId && dataset.name == name && dataset.isPreProcess).FirstOrDefault(); } //odraditi za pretragu getOne diff --git a/backend/api/api/Services/ExperimentService.cs b/backend/api/api/Services/ExperimentService.cs index 7bdf9a6e..539e4c08 100644 --- a/backend/api/api/Services/ExperimentService.cs +++ b/backend/api/api/Services/ExperimentService.cs @@ -36,5 +36,15 @@ namespace api.Services return _experiment.Find(e=>e.uploaderId==id).ToList(); } + + public Experiment GetOneExperiment(string userId, string name) + { + return _experiment.Find(experiment => experiment.uploaderId == userId && experiment.name == name).FirstOrDefault(); + } + + public void Update(string userId, string id, Experiment experiment) + { + _experiment.ReplaceOne(experiment => experiment.uploaderId == userId && experiment._id == id, experiment); + } } } diff --git a/backend/api/api/Services/FillAnEmptyDb.cs b/backend/api/api/Services/FillAnEmptyDb.cs index 216e1221..99bbb91f 100644 --- a/backend/api/api/Services/FillAnEmptyDb.cs +++ b/backend/api/api/Services/FillAnEmptyDb.cs @@ -55,7 +55,6 @@ namespace api.Services dataset.uploaderId = "000000000000000000000000"; dataset.name = "Titanik dataset"; dataset.description = "Titanik dataset"; - dataset.header = new string[] { "PassengerId", "Survived", "Pclass", "Name", "Sex", "Age", "SibSp", "Parch", "Ticket", "Fare", "Cabin", "Embarked" }; dataset.fileId = _fileService.GetFileId(fullPath); dataset.extension = ".csv"; dataset.isPublic = true; @@ -63,22 +62,22 @@ namespace api.Services dataset.dateCreated = DateTime.Now; dataset.lastUpdated = DateTime.Now; dataset.delimiter = ""; - dataset.hasHeader = true; dataset.columnInfo = new ColumnInfo[] { }; dataset.columnInfo = new[] { - new ColumnInfo( "PassengerId", true, 0, 446, 1, 891, 446, new string[]{ }), - new ColumnInfo( "Survived", true, 0, 0.38383838534355164f, 0, 1, 0, new string[]{ }), - new ColumnInfo( "Pclass", true, 0, 2.3086419105529785f, 1, 3, 3, new string[]{ }), - new ColumnInfo( "Name", false, 0, 0, 0, 0, 0, new string[]{"Braund, Mr. Owen Harris", "Boulos, Mr. Hanna", "Frolicher-Stehli, Mr. Maxmillian", "Gilinski, Mr. Eliezer", "Murdlin, Mr. Joseph", "Rintamaki, Mr. Matti", "Stephenson, Mrs. Walter Bertram (Martha Eustis)", "Elsbury, Mr. William James", "Bourke, Miss. Mary", "Chapman, Mr. John Henry"}), - new ColumnInfo( "Sex", false, 0, 0, 0, 0, 0, new string[]{ "male", "female" }), - new ColumnInfo( "Age", true, 177, 29.69911766052246f, 0.41999998688697815f, 80, 28, new string[]{ }), - new ColumnInfo( "SibSp", true, 0, 0.523007869720459f, 0, 8, 0, new string[]{ }), - new ColumnInfo( "Parch", true, 0, 0.3815937042236328f, 0, 6, 0, new string[]{ }), - new ColumnInfo( "Ticket", false, 0, 0, 0, 0, 0, new string[]{ "347082", "CA. 2343", "1601", "3101295", "CA 2144", "347088", "S.O.C. 14879", "382652", "LINE", "PC 17757" }), - new ColumnInfo( "Fare", true, 0, 32.20420837402344f, 0, 512.3292236328125f, 14.45419979095459f, new string[]{ }), - new ColumnInfo( "Cabin", false, 687, 0, 0, 0, 0, new string[]{ "B96 B98", "G6", "C23 C25 C27", "C22 C26", "F33", "F2", "E101", "D", "C78", "C93" }), - new ColumnInfo( "Embarked", false, 2, 0.3815937042236328f, 0, 6, 0, new string[]{ "S", "C", "Q" }), + + new ColumnInfo( "PassengerId", "columnType", true, 0, 446, 1, 891, 446, new string[]{ }, new int[] {}, new float[] {}, 0.01f,0.1f ), + new ColumnInfo( "Survived", "columnType", true, 0, 0.38383838534355164f, 0, 1, 0, new string[]{ }, new int[] {}, new float[] {}, 0.01f,0.1f ), + new ColumnInfo( "Pclass", "columnType", true, 0, 2.3086419105529785f, 1, 3, 3, new string[]{ }, new int[] {}, new float[] {}, 0.01f,0.1f ), + new ColumnInfo( "Name", "columnType", false, 0, 0, 0, 0, 0, new string[]{"Braund, Mr. Owen Harris", "Boulos, Mr. Hanna", "Frolicher-Stehli, Mr. Maxmillian", "Gilinski, Mr. Eliezer", "Murdlin, Mr. Joseph", "Rintamaki, Mr. Matti", "Stephenson, Mrs. Walter Bertram (Martha Eustis)", "Elsbury, Mr. William James", "Bourke, Miss. Mary", "Chapman, Mr. John Henry"}, new int[] {}, new float[] {}, 0.01f,0.1f ), + new ColumnInfo( "Sex", "columnType", false, 0, 0, 0, 0, 0, new string[]{ "male", "female" }, new int[] {}, new float[] {}, 0.01f,0.1f ), + new ColumnInfo( "Age", "columnType", true, 177, 29.69911766052246f, 0.41999998688697815f, 80, 28, new string[]{ }, new int[] {}, new float[] {}, 0.01f,0.1f ), + new ColumnInfo( "SibSp", "columnType", true, 0, 0.523007869720459f, 0, 8, 0, new string[]{ }, new int[] {}, new float[] {}, 0.01f,0.1f ), + new ColumnInfo( "Parch", "columnType", true, 0, 0.3815937042236328f, 0, 6, 0, new string[]{ }, new int[] {}, new float[] {}, 0.01f,0.1f ), + new ColumnInfo( "Ticket", "columnType", false, 0, 0, 0, 0, 0, new string[]{ "347082", "CA. 2343", "1601", "3101295", "CA 2144", "347088", "S.O.C. 14879", "382652", "LINE", "PC 17757" }, new int[] {}, new float[] {}, 0.01f,0.1f ), + new ColumnInfo( "Fare", "columnType", true, 0, 32.20420837402344f, 0, 512.3292236328125f, 14.45419979095459f, new string[]{ }, new int[] {}, new float[] {}, 0.01f,0.1f ), + new ColumnInfo( "Cabin", "columnType", false, 687, 0, 0, 0, 0, new string[]{ "B96 B98", "G6", "C23 C25 C27", "C22 C26", "F33", "F2", "E101", "D", "C78", "C93" }, new int[] {}, new float[] {}, 0.01f,0.1f ), + new ColumnInfo( "Embarked", "columnType", false, 2, 0.3815937042236328f, 0, 6, 0, new string[]{ "S", "C", "Q" }, new int[] {}, new float[] {}, 0.01f,0.1f ), }; dataset.rowCount = 891; dataset.nullCols = 3; @@ -99,11 +98,9 @@ namespace api.Services model.type = "binarni-klasifikacioni"; model.optimizer = "Adam"; model.lossFunction = "mean_squared_error"; - model.hiddenLayerNeurons = 3; model.hiddenLayers = 5; model.batchSize = 8; model.outputNeurons = 0; - model.hiddenLayerActivationFunctions = new string[] { "relu", "relu", "relu", "relu", "relu" }; model.outputLayerActivationFunction = "sigmoid"; model.metrics = new string[] { }; model.epochs = 5; @@ -121,9 +118,8 @@ namespace api.Services experiment.uploaderId = "000000000000000000000000"; experiment.inputColumns = new string[] { "Embarked" }; experiment.outputColumn = "Survived"; - experiment.randomOrder = true; - experiment.randomTestSet = true; - experiment.randomTestSetDistribution = 0.30000001192092896f; + experiment.dateCreated = DateTime.Now; + experiment.lastUpdated = DateTime.Now; experiment.nullValues = "delete_rows"; experiment.nullValuesReplacers = new NullValues[] { }; experiment.encodings = new[] @@ -179,20 +175,19 @@ namespace api.Services dataset.dateCreated = DateTime.Now; dataset.lastUpdated = DateTime.Now; dataset.delimiter = ""; - dataset.hasHeader = true; dataset.columnInfo = new[] { - new ColumnInfo( "Unnamed: 0", true, 0, 26969.5f, 0, 53939, 26969.5f, new string[]{ }), - new ColumnInfo( "carat", true, 0, 0.7979397773742676f, 0.20000000298023224f, 5.010000228881836f, 0.699999988079071f, new string[]{ }), - new ColumnInfo( "cut", false, 0, 0, 0, 0, 0, new string[]{ "Ideal", "Premium", "Very Good", "Good", "Fair" }), - new ColumnInfo( "color", false, 0, 0, 0, 0, 0, new string[]{"G", "E", "F", "H", "D", "I", "I", "J"}), - new ColumnInfo( "clarity", false, 0, 0, 0, 0, 0, new string[]{ "SI1", "VS2","SI2", "VS1", "VVS2", "VVS1", "IF", "I1" }), - new ColumnInfo( "depth", true, 0, 61.74940490722656f, 43, 79, 61.79999923706055f, new string[]{ }), - new ColumnInfo( "table", true, 0, 57.457183837890625f, 43, 95, 57, new string[]{ }), - new ColumnInfo( "price", true, 0, 3932.7998046875f, 326, 18823, 2401, new string[]{ }), - new ColumnInfo( "x", true, 0, 5.731157302856445f, 0, 10.739999771118164f, 5.699999809265137f, new string[]{ }), - new ColumnInfo( "y", true, 0, 5.73452615737915f, 0, 58.900001525878906f, 5.710000038146973f, new string[]{ }), - new ColumnInfo( "z", true, 0, 3.538733720779419f, 0, 31.799999237060547f, 3.5299999713897705f, new string[]{ }) + new ColumnInfo( "Unnamed: 0", "columnType", true, 0, 26969.5f, 0, 53939, 26969.5f, new string[]{ }, new int[] {}, new float[] {}, 0.01f,0.1f ), + new ColumnInfo( "carat", "columnType", true, 0, 0.7979397773742676f, 0.20000000298023224f, 5.010000228881836f, 0.699999988079071f, new string[]{ }, new int[] {}, new float[] {}, 0.01f,0.1f ), + new ColumnInfo( "cut", "columnType", false, 0, 0, 0, 0, 0, new string[]{ "Ideal", "Premium", "Very Good", "Good", "Fair" }, new int[] {}, new float[] {}, 0.01f,0.1f ), + new ColumnInfo( "color", "columnType", false, 0, 0, 0, 0, 0, new string[]{"G", "E", "F", "H", "D", "I", "I", "J"}, new int[] {}, new float[] {}, 0.01f,0.1f ), + new ColumnInfo( "clarity", "columnType", false, 0, 0, 0, 0, 0, new string[]{ "SI1", "VS2","SI2", "VS1", "VVS2", "VVS1", "IF", "I1" }, new int[] {}, new float[] {}, 0.01f,0.1f ), + new ColumnInfo( "depth", "columnType", true, 0, 61.74940490722656f, 43, 79, 61.79999923706055f, new string[]{ }, new int[] {}, new float[] {}, 0.01f,0.1f ), + new ColumnInfo( "table", "columnType", true, 0, 57.457183837890625f, 43, 95, 57, new string[]{ }, new int[] {}, new float[] {}, 0.01f,0.1f ), + new ColumnInfo( "price", "columnType", true, 0, 3932.7998046875f, 326, 18823, 2401, new string[]{ }, new int[] {}, new float[] {}, 0.01f,0.1f ), + new ColumnInfo( "x", "columnType", true, 0, 5.731157302856445f, 0, 10.739999771118164f, 5.699999809265137f, new string[]{ }, new int[] {}, new float[] {}, 0.01f,0.1f ), + new ColumnInfo( "y", "columnType", true, 0, 5.73452615737915f, 0, 58.900001525878906f, 5.710000038146973f, new string[]{ }, new int[] {}, new float[] {}, 0.01f,0.1f ), + new ColumnInfo( "z", "columnType", true, 0, 3.538733720779419f, 0, 31.799999237060547f, 3.5299999713897705f, new string[]{ }, new int[] {}, new float[] {}, 0.01f,0.1f ) }; dataset.rowCount = 53940; dataset.nullCols = 0; @@ -214,11 +209,9 @@ namespace api.Services model.type = "regresioni"; model.optimizer = "Adam"; model.lossFunction = "mean_absolute_error"; - model.hiddenLayerNeurons = 2; model.hiddenLayers = 4; model.batchSize = 5; model.outputNeurons = 0; - model.hiddenLayerActivationFunctions = new string[] { "relu", "relu", "relu", "relu" }; model.outputLayerActivationFunction = "relu"; model.metrics = new string[] { }; model.epochs = 5; @@ -236,9 +229,8 @@ namespace api.Services experiment.uploaderId = "000000000000000000000000"; experiment.inputColumns = new string[] { "Unnamed: 0", "carat", "cut", "color", "clarity", "depth", "table", "x", "y", "z" }; experiment.outputColumn = "price"; - experiment.randomOrder = true; - experiment.randomTestSet = true; - experiment.randomTestSetDistribution = 0.30000001192092896f; + experiment.dateCreated = DateTime.Now; + experiment.lastUpdated = DateTime.Now; experiment.nullValues = "delete_rows"; experiment.nullValuesReplacers = new NullValues[] { }; experiment.encodings = new[] @@ -297,14 +289,13 @@ namespace api.Services dataset.dateCreated = DateTime.Now; dataset.lastUpdated = DateTime.Now; dataset.delimiter = ""; - dataset.hasHeader = true; dataset.columnInfo = new[] { - new ColumnInfo( "sepal_length", true, 0, 5.8433332443237305f, 4.300000190734863f, 7.900000095367432f, 5.800000190734863f, new string[]{ }), - new ColumnInfo( "sepal_width", true, 0, 3.053999900817871f, 2, 4.400000095367432f, 3, new string[]{ }), - new ColumnInfo( "petal_length", true, 0, 3.758666753768921f, 1, 6.900000095367432f, 4.349999904632568f, new string[]{ }), - new ColumnInfo( "petal_width", true, 0, 1.1986666917800903f, 0.10000000149011612f, 2.5f, 1.2999999523162842f, new string[]{}), - new ColumnInfo( "class", false, 0, 0, 0, 0, 0, new string[]{ "Iris-setosa", "Iris-versicolor", "Iris-virginica" }), + new ColumnInfo( "sepal_length", "columnType", true, 0, 5.8433332443237305f, 4.300000190734863f, 7.900000095367432f, 5.800000190734863f, new string[]{ }, new int[] {}, new float[] {}, 0.01f, 0.1f ), + new ColumnInfo( "sepal_width", "columnType", true, 0, 3.053999900817871f, 2, 4.400000095367432f, 3, new string[]{ }, new int[] {}, new float[] {}, 0.01f,0.1f ), + new ColumnInfo( "petal_length", "columnType", true, 0, 3.758666753768921f, 1, 6.900000095367432f, 4.349999904632568f, new string[]{ }, new int[] {}, new float[] {}, 0.01f,0.1f ), + new ColumnInfo( "petal_width", "columnType", true, 0, 1.1986666917800903f, 0.10000000149011612f, 2.5f, 1.2999999523162842f, new string[]{}, new int[] {}, new float[] {}, 0.01f,0.1f ), + new ColumnInfo( "class", "columnType", false, 0, 0, 0, 0, 0, new string[]{ "Iris-setosa", "Iris-versicolor", "Iris-virginica" }, new int[] {}, new float[] {}, 0.01f,0.1f ), }; dataset.nullCols = 150; dataset.nullRows = 0; @@ -324,11 +315,9 @@ namespace api.Services model.type = "multi-klasifikacioni"; model.optimizer = "Adam"; model.lossFunction = "sparse_categorical_crossentropy"; - model.hiddenLayerNeurons = 3; model.hiddenLayers = 3; model.batchSize = 4; model.outputNeurons = 0; - model.hiddenLayerActivationFunctions = new string[] { "relu", "relu", "softmax" }; model.outputLayerActivationFunction = "softmax"; model.metrics = new string[] { }; model.epochs = 1; @@ -346,9 +335,8 @@ namespace api.Services experiment.uploaderId = "000000000000000000000000"; experiment.inputColumns = new string[] { "sepal_length", "sepal_width", "petal_length", "petal_width" }; experiment.outputColumn = "class"; - experiment.randomOrder = true; - experiment.randomTestSet = true; - experiment.randomTestSetDistribution = 0.20000000298023224f; + experiment.dateCreated = DateTime.Now; + experiment.lastUpdated = DateTime.Now; experiment.nullValues = "delete_rows"; experiment.nullValuesReplacers = new NullValues[] { }; experiment.encodings = new[] diff --git a/backend/api/api/Models/JwtToken.cs b/backend/api/api/Services/JwtToken.cs index 20b0bc73..20b0bc73 100644 --- a/backend/api/api/Models/JwtToken.cs +++ b/backend/api/api/Services/JwtToken.cs diff --git a/backend/api/api/Models/PasswordCrypt.cs b/backend/api/api/Services/PasswordCrypt.cs index 016fde51..016fde51 100644 --- a/backend/api/api/Models/PasswordCrypt.cs +++ b/backend/api/api/Services/PasswordCrypt.cs diff --git a/backend/microservice/api/controller.py b/backend/microservice/api/controller.py index 9b83b8e7..41035cc7 100644 --- a/backend/microservice/api/controller.py +++ b/backend/microservice/api/controller.py @@ -107,26 +107,32 @@ def predict(): @app.route('/preprocess',methods=['POST']) def returnColumnsInfo(): print("********************************PREPROCESS*******************************") + dataset = json.loads(request.form["dataset"]) file = request.files.get("file") data=pd.read_csv(file) - - #dataset={} + ''' #f = request.json['filepath'] #data=pd.read_csv(f) - + dataset={} + ''' preprocess = newmlservice.returnColumnsInfo(data) #samo 10 jedinstvenih posto ih ima previse, bilo bi dobro da promenimo ovo da to budu 10 najzastupljenijih vrednosti + for col in preprocess["columnInfo"]: - col["uniqueValues"] = col["uniqueValues"][0:10] - col["uniqueValuesCount"] = col["uniqueValuesCount"][0:10] + col["uniqueValues"] = col["uniqueValues"][0:6] + col["uniqueValuesCount"] = col["uniqueValuesCount"][0:6] + col['uniqueValuesPercent']=col['uniqueValuesPercent'][0:6] dataset["columnInfo"] = preprocess["columnInfo"] dataset["nullCols"] = preprocess["allNullColl"] dataset["nullRows"] = preprocess["allNullRows"] dataset["colCount"] = preprocess["colCount"] dataset["rowCount"] = preprocess["rowCount"] dataset["isPreProcess"] = True - print(dataset) + #print(dataset) + + + return jsonify(dataset) print("App loaded.") diff --git a/backend/microservice/api/newmlservice.py b/backend/microservice/api/newmlservice.py index 604e4d3c..9e26b03a 100644 --- a/backend/microservice/api/newmlservice.py +++ b/backend/microservice/api/newmlservice.py @@ -1,6 +1,7 @@ from enum import unique from itertools import count import os +from sys import breakpointhook import pandas as pd from sklearn import datasets, multiclass import tensorflow as tf @@ -38,27 +39,38 @@ def returnColumnsInfo(dataset): unique=datafront[kolona].value_counts() uniquevalues=[] uniquevaluescount=[] + uniquevaluespercent=[] for val, count in unique.iteritems(): - uniquevalues.append(val) - uniquevaluescount.append(count) + if(val): + uniquevalues.append(val) + uniquevaluescount.append(count) + percent=count/rowCount + uniquevaluespercent.append(percent) #print(uniquevalues) #print(uniquevaluescount) mean=0 median=0 minimum=0 maximum=0 + q1=0 + q3=0 nullCount=datafront[kolona].isnull().sum() if(nullCount>0): allNullCols=allNullCols+1 - frontreturn={'columnName':kolona, + frontreturn={ + 'columnName':kolona, 'isNumber':False, 'uniqueValues':uniquevalues, 'uniqueValuesCount':uniquevaluescount, - 'median':float(mean), - 'mean':float(median), + 'uniqueValuesPercent':uniquevaluespercent, + 'mean':float(mean), + 'median':float(median), 'numNulls':int(nullCount), 'min':float(minimum), 'max':float(maximum), + 'q1':float(q1), + 'q3':float(q3), + } dict.append(frontreturn) else: @@ -66,18 +78,39 @@ def returnColumnsInfo(dataset): maximum=max(datafront[kolona]) mean=datafront[kolona].mean() median=s.median(datafront[kolona].copy().dropna()) + q1= np.percentile(datafront[kolona].copy().dropna(), 25) + q3= np.percentile(datafront[kolona].copy().dropna(), 75) nullCount=datafront[kolona].isnull().sum() if(nullCount>0): allNullCols=allNullCols+1 - frontreturn={'columnName':kolona, + + #pretvaranje u kategorijsku + datafront = datafront.astype({kolona: str}) + print(datafront.dtypes) + unique=datafront[kolona].value_counts() + uniquevaluesn=[] + uniquevaluescountn=[] + uniquevaluespercentn=[] + for val, count in unique.iteritems(): + if(val): + uniquevaluesn.append(val) + uniquevaluescountn.append(count) + percent=count/rowCount + uniquevaluespercentn.append(percent) + frontreturn={ + 'columnName':kolona, 'isNumber':1, - 'uniqueValues':[], - 'uniqueValuesCount':[], + 'uniqueValues':uniquevaluesn, + 'uniqueValuesCount':uniquevaluescountn, + 'uniqueValuesPercent':uniquevaluespercentn, 'mean':float(mean), 'median':float(median), 'numNulls':int(nullCount), 'min':float(minimum), 'max':float(maximum), + 'q1':float(q1), + 'q3':float(q3), + } dict.append(frontreturn) NullRows = datafront[datafront.isnull().any(axis=1)] @@ -115,17 +148,28 @@ class TrainingResult: ''' def train(dataset, paramsModel,paramsExperiment,paramsDataset,callback): + ###UCITAVANJE SETA problem_type = paramsModel["type"] #print(problem_type) data = pd.DataFrame() #print(data) for col in paramsExperiment["inputColumns"]: #print(col) - data[col]=dataset[col] + if(col!=paramsExperiment["outputColumn"]): + data[col]=dataset[col] output_column = paramsExperiment["outputColumn"] data[output_column] = dataset[output_column] #print(data) + ###KATEGORIJSKE KOLONE + kategorijskekolone=[] + ###PRETVARANJE NUMERICKIH U KATREGORIJSKE AKO JE KORISNIK TAKO OZNACIO + columnInfo=paramsDataset['columnInfo'] + for col in columnInfo: + if(col['columnType']=="Kategorijski"): + data[col['columnName']]=data[col['columnName']].apply(str) + kategorijskekolone.append(col['coumnName']) + ###NULL null_value_options = paramsExperiment["nullValues"] null_values_replacers = paramsExperiment["nullValuesReplacers"] @@ -149,16 +193,18 @@ def train(dataset, paramsModel,paramsExperiment,paramsDataset,callback): # # Brisanje kolona koje ne uticu na rezultat # + ''' num_rows=data.shape[0] for col in data.columns: if((data[col].nunique()==(num_rows)) and (data[col].dtype==np.object_)): data.pop(col) # + ''' ### Enkodiranje encodings=paramsExperiment["encodings"] - datafront=dataset.copy() - svekolone=datafront.columns - kategorijskekolone=datafront.select_dtypes(include=['object']).columns + #datafront=dataset.copy() + #svekolone=datafront.columns + #kategorijskekolone=datafront.select_dtypes(include=['object']).columns for kolonaEncoding in encodings: kolona = kolonaEncoding["columnName"] @@ -235,75 +281,75 @@ def train(dataset, paramsModel,paramsExperiment,paramsDataset,callback): # # ###OPTIMIZATORI - """ - if(params['optimizer']=='adam'): - opt=tf.keras.optimizers.Adam(learning_rate=params['learningRate']) + + if(paramsModel['optimizer']=='Adam'): + opt=tf.keras.optimizers.Adam(learning_rate=float(paramsModel['learningRate'])) - elif(params['optimizer']=='adadelta'): - opt=tf.keras.optimizers.Adadelta(learning_rate=params['learningRate']) + elif(paramsModel['optimizer']=='Adadelta'): + opt=tf.keras.optimizers.Adadelta(learning_rate=float(paramsModel['learningRate'])) - elif(params['optimizer']=='adagrad'): - opt=tf.keras.optimizers.Adagrad(learning_rate=params['learningRate']) + elif(paramsModel['optimizer']=='Adagrad'): + opt=tf.keras.optimizers.Adagrad(learning_rate=float(paramsModel['learningRate'])) - elif(params['optimizer']=='adamax'): - opt=tf.keras.optimizers.Adamax(learning_rate=params['learningRate']) + elif(paramsModel['optimizer']=='Adamax'): + opt=tf.keras.optimizers.Adamax(learning_rate=float(paramsModel['learningRate'])) - elif(params['optimizer']=='nadam'): - opt=tf.keras.optimizers.Nadam(learning_rate=params['learningRate']) + elif(paramsModel['optimizer']=='Nadam'): + opt=tf.keras.optimizers.Nadam(learning_rate=float(paramsModel['learningRate'])) - elif(params['optimizer']=='sgd'): - opt=tf.keras.optimizers.SGD(learning_rate=params['learningRate']) + elif(paramsModel['optimizer']=='Sgd'): + opt=tf.keras.optimizers.SGD(learning_rate=float(paramsModel['learningRate'])) - elif(params['optimizer']=='ftrl'): - opt=tf.keras.optimizers.Ftrl(learning_rate=params['learningRate']) + elif(paramsModel['optimizer']=='Ftrl'): + opt=tf.keras.optimizers.Ftrl(learning_rate=float(paramsModel['learningRate'])) - elif(params['optimizer']=='rmsprop'): - opt=tf.keras.optimizers.RMSprop(learning_rate=params['learningRate']) + elif(paramsModel['optimizer']=='Rmsprop'): + opt=tf.keras.optimizers.RMSprop(learning_rate=float(paramsModel['learningRate'])) ###REGULARIZACIJA #regularisation={'kernelType':'l1 ili l2 ili l1_l2','kernelRate':default=0.01 ili jedna od vrednosti(0.0001,0.001,0.1,1,2,3) ili neka koju je korisnik zadao,'biasType':'','biasRate':'','activityType','activityRate'} - reg=params['regularisation'] - - ###Kernel - if(reg['kernelType']=='l1'): - kernelreg=tf.keras.regularizers.l1(reg['kernelRate']) - elif(reg['kernelType']=='l2'): - kernelreg=tf.keras.regularizers.l2(reg['kernelRate']) - elif(reg['kernelType']=='l1l2'): - kernelreg=tf.keras.regularizers.l1_l2(l1=reg['kernelRate'][0],l2=reg['kernelRate'][1]) - - ###Bias - if(reg['biasType']=='l1'): - biasreg=tf.keras.regularizers.l1(reg['biasRate']) - elif(reg['biasType']=='l2'): - biasreg=tf.keras.regularizers.l2(reg['biasRate']) - elif(reg['biasType']=='l1l2'): - biasreg=tf.keras.regularizers.l1_l2(l1=reg['biasRate'][0],l2=reg['biasRate'][1]) - - ###Activity - if(reg['kernelType']=='l1'): - activityreg=tf.keras.regularizers.l1(reg['activityRate']) - elif(reg['kernelType']=='l2'): - activityreg=tf.keras.regularizers.l2(reg['activityRate']) - elif(reg['kernelType']=='l1l2'): - activityreg=tf.keras.regularizers.l1_l2(l1=reg['activityRate'][0],l2=reg['activityRate'][1]) - """ + + filepath=os.path.join("temp/",paramsExperiment['_id']+"_"+paramsModel['_id']+".h5") if(problem_type=='multi-klasifikacioni'): #print('multi') + + reg=paramsModel['regularisation'][0] + regRate=float(paramsModel['regularisationRate'][0]) + if(reg=='l1'): + kernelreg=tf.keras.regularizers.l1(regRate) + biasreg=tf.keras.regularizers.l1(regRate) + activityreg=tf.keras.regularizers.l1(regRate) + elif(reg=='l2'): + kernelreg=tf.keras.regularizers.l2(regRate) + biasreg=tf.keras.regularizers.l2(regRate) + activityreg=tf.keras.regularizers.l2(regRate) + classifier=tf.keras.Sequential() - - classifier.add(tf.keras.layers.Dense(units=paramsModel['hiddenLayerNeurons'], activation=paramsModel['hiddenLayerActivationFunctions'][0],input_dim=x_train.shape[1]))#prvi skriveni + definisanje prethodnog-ulaznog + classifier.add(tf.keras.layers.Dense(units=paramsModel['hiddenLayerNeurons'], activation=paramsModel['hiddenLayerActivationFunctions'][0],input_dim=x_train.shape[1], kernel_regularizer=kernelreg, bias_regularizer=biasreg, activity_regularizer=activityreg))#prvi skriveni + definisanje prethodnog-ulaznog + for i in range(paramsModel['hiddenLayers']-1):#ako postoji vise od jednog skrivenog sloja - #print(i) - classifier.add(tf.keras.layers.Dense(units=paramsModel['hiddenLayerNeurons'], activation=paramsModel['hiddenLayerActivationFunctions'][i+1]))#i-ti skriveni sloj + ###Kernel + reg=paramsModel['regularisation'][i+1] + regRate=float(paramsModel['regularisationRate'][i+1]) + if(reg=='l1'): + kernelreg=tf.keras.regularizers.l1(regRate) + biasreg=tf.keras.regularizers.l1(regRate) + activityreg=tf.keras.regularizers.l1(regRate) + elif(reg=='l2'): + kernelreg=tf.keras.regularizers.l2(regRate) + biasreg=tf.keras.regularizers.l2(regRate) + activityreg=tf.keras.regularizers.l2(regRate) + + classifier.add(tf.keras.layers.Dense(units=paramsModel['hiddenLayerNeurons'], activation=paramsModel['hiddenLayerActivationFunctions'][i+1],kernel_regularizer=kernelreg, bias_regularizer=biasreg, activity_regularizer=activityreg))#i-ti skriveni sloj + classifier.add(tf.keras.layers.Dense(units=5, activation=paramsModel['outputLayerActivationFunction']))#izlazni sloj - classifier.compile(loss =paramsModel["lossFunction"] , optimizer = paramsModel['optimizer'] , metrics =paramsModel['metrics']) + classifier.compile(loss =paramsModel["lossFunction"] , optimizer = opt, metrics =paramsModel['metrics']) - history=classifier.fit(x_train, y_train, epochs = paramsModel['epochs'],batch_size=paramsModel['batchSize'],callbacks=callback(x_test, y_test,paramsModel['_id'])) + history=classifier.fit(x_train, y_train, epochs = paramsModel['epochs'],batch_size=float(paramsModel['batchSize']),callbacks=callback(x_test, y_test,paramsModel['_id'])) hist=history.history #plt.plot(hist['accuracy']) @@ -325,17 +371,39 @@ def train(dataset, paramsModel,paramsExperiment,paramsDataset,callback): elif(problem_type=='binarni-klasifikacioni'): #print('*************************************************************************binarni') + reg=paramsModel['regularisation'][0] + regRate=float(paramsModel['regularisationRate'][0]) + if(reg=='l1'): + kernelreg=tf.keras.regularizers.l1(regRate) + biasreg=tf.keras.regularizers.l1(regRate) + activityreg=tf.keras.regularizers.l1(regRate) + elif(reg=='l2'): + kernelreg=tf.keras.regularizers.l2(regRate) + biasreg=tf.keras.regularizers.l2(regRate) + activityreg=tf.keras.regularizers.l2(regRate) classifier=tf.keras.Sequential() - classifier.add(tf.keras.layers.Dense(units=paramsModel['hiddenLayerNeurons'], activation=paramsModel['hiddenLayerActivationFunctions'][0],input_dim=x_train.shape[1]))#prvi skriveni + definisanje prethodnog-ulaznog + classifier.add(tf.keras.layers.Dense(units=paramsModel['hiddenLayerNeurons'], activation=paramsModel['hiddenLayerActivationFunctions'][0],input_dim=x_train.shape[1],kernel_regularizer=kernelreg, bias_regularizer=biasreg, activity_regularizer=activityreg))#prvi skriveni + definisanje prethodnog-ulaznog + for i in range(paramsModel['hiddenLayers']-1):#ako postoji vise od jednog skrivenog sloja #print(i) - classifier.add(tf.keras.layers.Dense(units=paramsModel['hiddenLayerNeurons'], activation=paramsModel['hiddenLayerActivationFunctions'][i+1]))#i-ti skriveni sloj + reg=paramsModel['regularisation'][i+1] + regRate=float(paramsModel['regularisationRate'][i+1]) + if(reg=='l1'): + kernelreg=tf.keras.regularizers.l1(regRate) + biasreg=tf.keras.regularizers.l1(regRate) + activityreg=tf.keras.regularizers.l1(regRate) + elif(reg=='l2'): + kernelreg=tf.keras.regularizers.l2(regRate) + biasreg=tf.keras.regularizers.l2(regRate) + activityreg=tf.keras.regularizers.l2(regRate) + classifier.add(tf.keras.layers.Dense(units=paramsModel['hiddenLayerNeurons'], activation=paramsModel['hiddenLayerActivationFunctions'][i+1],kernel_regularizer=kernelreg, bias_regularizer=biasreg, activity_regularizer=activityreg))#i-ti skriveni sloj + classifier.add(tf.keras.layers.Dense(units=1, activation=paramsModel['outputLayerActivationFunction']))#izlazni sloj - classifier.compile(loss =paramsModel["lossFunction"] , optimizer = paramsModel['optimizer'] , metrics =paramsModel['metrics']) + classifier.compile(loss =paramsModel["lossFunction"] , optimizer = opt , metrics =paramsModel['metrics']) - history=classifier.fit(x_train, y_train, epochs = paramsModel['epochs'],batch_size=paramsModel['batchSize'],callbacks=callback(x_test, y_test,paramsModel['_id'])) + history=classifier.fit(x_train, y_train, epochs = paramsModel['epochs'],batch_size=float(paramsModel['batchSize']),callbacks=callback(x_test, y_test,paramsModel['_id'])) hist=history.history y_pred=classifier.predict(x_test) y_pred=(y_pred>=0.5).astype('int') @@ -351,17 +419,40 @@ def train(dataset, paramsModel,paramsExperiment,paramsDataset,callback): return filepath,hist elif(problem_type=='regresioni'): + reg=paramsModel['regularisation'][0] + regRate=float(paramsModel['regularisationRate'][0]) + if(reg=='l1'): + kernelreg=tf.keras.regularizers.l1(regRate) + biasreg=tf.keras.regularizers.l1(regRate) + activityreg=tf.keras.regularizers.l1(regRate) + elif(reg=='l2'): + kernelreg=tf.keras.regularizers.l2(regRate) + biasreg=tf.keras.regularizers.l2(regRate) + activityreg=tf.keras.regularizers.l2(regRate) classifier=tf.keras.Sequential() - classifier.add(tf.keras.layers.Dense(units=paramsModel['hiddenLayerNeurons'], activation=paramsModel['hiddenLayerActivationFunctions'][0],input_dim=x_train.shape[1]))#prvi skriveni + definisanje prethodnog-ulaznog + classifier.add(tf.keras.layers.Dense(units=paramsModel['hiddenLayerNeurons'], activation=paramsModel['hiddenLayerActivationFunctions'][0],input_dim=x_train.shape[1],kernel_regularizer=kernelreg, bias_regularizer=biasreg, activity_regularizer=activityreg))#prvi skriveni + definisanje prethodnog-ulaznog + for i in range(paramsModel['hiddenLayers']-1):#ako postoji vise od jednog skrivenog sloja #print(i) - classifier.add(tf.keras.layers.Dense(units=paramsModel['hiddenLayerNeurons'], activation=paramsModel['hiddenLayerActivationFunctions'][i+1]))#i-ti skriveni sloj - classifier.add(tf.keras.layers.Dense(units=1)) + reg=paramsModel['regularisation'][i+1] + regRate=float(paramsModel['regularisationRate'][i+1]) + if(reg=='l1'): + kernelreg=tf.keras.regularizers.l1(regRate) + biasreg=tf.keras.regularizers.l1(regRate) + activityreg=tf.keras.regularizers.l1(regRate) + elif(reg=='l2'): + kernelreg=tf.keras.regularizers.l2(regRate) + biasreg=tf.keras.regularizers.l2(regRate) + activityreg=tf.keras.regularizers.l2(regRate) + + classifier.add(tf.keras.layers.Dense(units=paramsModel['hiddenLayerNeurons'], activation=paramsModel['hiddenLayerActivationFunctions'][i+1],kernel_regularizer=kernelreg, bias_regularizer=biasreg, activity_regularizer=activityreg))#i-ti skriveni sloj + + classifier.add(tf.keras.layers.Dense(units=1),activation=paramsModel['outputLayerActivationFunction']) - classifier.compile(loss =paramsModel["lossFunction"] , optimizer = paramsModel['optimizer'] , metrics =paramsModel['metrics']) + classifier.compile(loss =paramsModel["lossFunction"] , optimizer = opt , metrics =paramsModel['metrics']) - history=classifier.fit(x_train, y_train, epochs = paramsModel['epochs'],batch_size=paramsModel['batchSize'],callbacks=callback(x_test, y_test,paramsModel['_id'])) + history=classifier.fit(x_train, y_train, epochs = paramsModel['epochs'],batch_size=float(paramsModel['batchSize']),callbacks=callback(x_test, y_test,paramsModel['_id'])) hist=history.history y_pred=classifier.predict(x_test) #print(classifier.evaluate(x_test, y_test)) |