Critical fix to models front

.gitignore

@@ -200,3 +200,5 @@ docs/node_modules/
 # Re-include docs content section named "build" (not a build output)
 !docs/docs/build/
 !docs/docs/build/**
+!docs/i18n/**/build/
+!docs/i18n/**/build/**

DashAI/back/dataloaders/classes/csv_dataloader.py

@@ -30,21 +30,6 @@ class CSVDataloaderSchema(BaseSchema):
     ``pandas.read_csv``.
     """
 
-    name: schema_field(
-        string_field(),
-        "",
-        description=MultilingualString(
-            en=(
-                "Custom name to register your dataset. If no name is specified, "
-                "the name of the uploaded file will be used."
-            ),
-            es=(
-                "Nombre personalizado para registrar su dataset. Si no se especifica "
-                "un nombre, se usará el nombre del archivo subido."
-            ),
-        ),
-        alias=MultilingualString(en="Name", es="Nombre"),
-    )  # type: ignore
     separator: schema_field(
         enum_field([",", ";", "blank space", "tab"]),
         ",",

DashAI/back/dataloaders/classes/dataloader.py

@@ -5,6 +5,7 @@
 from typing import TYPE_CHECKING, Any, Dict, Final
 
 from DashAI.back.config_object import ConfigObject
+from DashAI.back.core.utils import MultilingualString
 
 logger = logging.getLogger(__name__)
 
@@ -18,6 +19,16 @@ class BaseDataLoader(ConfigObject):
     """Abstract class with base methods for DashAI dataloaders."""
 
     TYPE: Final[str] = "DataLoader"
+    CATEGORY: Final = MultilingualString(
+        en="File Uploading",
+        es="Carga de Archivos",
+    )
+
+    @classmethod
+    def get_metadata(cls) -> Dict[str, Any]:
+        return {
+            "category": cls.CATEGORY if cls.CATEGORY else "File Uploading",
+        }
 
     @abstractmethod
     def load_data(

DashAI/back/dataloaders/classes/excel_dataloader.py

@@ -27,21 +27,6 @@ class ExcelDataloaderSchema(BaseSchema):
     leaving the sheet field empty selects the first sheet.
     """
 
-    name: schema_field(
-        string_field(),
-        "",
-        description=MultilingualString(
-            en=(
-                "Custom name to register your dataset. If no name is specified, "
-                "the name of the uploaded file will be used."
-            ),
-            es=(
-                "Nombre personalizado para registrar su dataset. Si no se especifica "
-                "un nombre, se usará el nombre del archivo subido."
-            ),
-        ),
-        alias=MultilingualString(en="Name", es="Nombre"),
-    )  # type: ignore
     sheet: schema_field(
         union_type(int_field(ge=0), string_field()),
         placeholder=0,

DashAI/back/dataloaders/classes/json_dataloader.py

@@ -19,21 +19,6 @@ class JSONDataloaderSchema(BaseSchema):
     ``None``, the entire JSON value is interpreted as the record list.
     """
 
-    name: schema_field(
-        string_field(),
-        "",
-        description=MultilingualString(
-            en=(
-                "Custom name to register your dataset. If no name is specified, "
-                "the name of the uploaded file will be used."
-            ),
-            es=(
-                "Nombre personalizado para registrar su dataset. Si no se especifica "
-                "un nombre, se usará el nombre del archivo subido."
-            ),
-        ),
-        alias=MultilingualString(en="Name", es="Nombre"),
-    )  # type: ignore
     data_key: schema_field(
         none_type(string_field()),
         placeholder="data",

DashAI/back/initial_components.py

@@ -123,21 +123,44 @@
 from DashAI.back.metrics.translation.ter import Ter
 
 # Models
+from DashAI.back.models.hugging_face.albert_transformer import AlbertTransformer
+from DashAI.back.models.hugging_face.bert_transformer import BertTransformer
+from DashAI.back.models.hugging_face.bertin_transformer import BertinTransformer
+from DashAI.back.models.hugging_face.beto_transformer import BetoTransformer
 from DashAI.back.models.hugging_face.deberta_v3_transformer import DebertaV3Transformer
 from DashAI.back.models.hugging_face.distilbert_transformer import DistilBertTransformer
+from DashAI.back.models.hugging_face.electra_transformer import ElectraTransformer
 from DashAI.back.models.hugging_face.llama_model import LlamaModel
+from DashAI.back.models.hugging_face.m2m100_transformer import M2M100Transformer
+from DashAI.back.models.hugging_face.minilm_transformer import MiniLMTransformer
 from DashAI.back.models.hugging_face.mistral_model import MistralModel
 from DashAI.back.models.hugging_face.mixtral_model import MixtralModel
 from DashAI.back.models.hugging_face.modernbert_transformer import ModernBertTransformer
+from DashAI.back.models.hugging_face.multilingual_bert_transformer import (
+    MultilingualBertTransformer,
+)
 from DashAI.back.models.hugging_face.nllb_transformer import NllbTransformer
+from DashAI.back.models.hugging_face.opus_mt_en_de_transformer import (
+    OpusMtEnDeTransformer,
+)
 from DashAI.back.models.hugging_face.opus_mt_en_es_transformer import (
     OpusMtEnESTransformer,
 )
+from DashAI.back.models.hugging_face.opus_mt_en_fr_transformer import (
+    OpusMtEnFrTransformer,
+)
+from DashAI.back.models.hugging_face.opus_mt_en_pt_transformer import (
+    OpusMtEnPtTransformer,
+)
 from DashAI.back.models.hugging_face.opus_mt_es_en_transformer import (
     OpusMtEsENTransformer,
 )
+from DashAI.back.models.hugging_face.opus_mt_fr_en_transformer import (
+    OpusMtFrEnTransformer,
+)
 from DashAI.back.models.hugging_face.pixart_sigma_model import PixArtSigmaModel
 from DashAI.back.models.hugging_face.qwen_model import QwenModel
+from DashAI.back.models.hugging_face.roberta_transformer import RobertaTransformer
 from DashAI.back.models.hugging_face.sd15_depth_controlnet_model import (
     SD15DepthControlNetModel,
 )
@@ -164,24 +187,52 @@
 from DashAI.back.models.hugging_face.stable_diffusion_xl_model import (
     StableDiffusionXLModel,
 )
+from DashAI.back.models.hugging_face.t5_small_transformer import T5SmallTransformer
 from DashAI.back.models.hugging_face.tongyi_z_image_model import TongyiZImageModel
+from DashAI.back.models.hugging_face.xlm_roberta_transformer import (
+    XlmRobertaTransformer,
+)
+from DashAI.back.models.hugging_face.xlnet_transformer import XlnetTransformer
+from DashAI.back.models.scikit_learn.adaboost_classifier import AdaBoostClassifier
+from DashAI.back.models.scikit_learn.adaboost_regression import AdaBoostRegression
+from DashAI.back.models.scikit_learn.bagging_classifier import BaggingClassifier
+from DashAI.back.models.scikit_learn.bayesian_ridge_regression import (
+    BayesianRidgeRegression,
+)
 from DashAI.back.models.scikit_learn.bow_text_classification_model import (
     BagOfWordsTextClassificationModel,
 )
 from DashAI.back.models.scikit_learn.decision_tree_classifier import (
     DecisionTreeClassifier,
 )
+from DashAI.back.models.scikit_learn.decision_tree_regression import (
+    DecisionTreeRegression,
+)
 from DashAI.back.models.scikit_learn.dummy_classifier import DummyClassifier
+from DashAI.back.models.scikit_learn.elastic_net_regression import ElasticNetRegression
+from DashAI.back.models.scikit_learn.extra_trees_classifier import ExtraTreesClassifier
+from DashAI.back.models.scikit_learn.extra_trees_regression import ExtraTreesRegression
+from DashAI.back.models.scikit_learn.gaussian_nb import GaussianNB
+from DashAI.back.models.scikit_learn.gradient_boosting_classifier import (
+    GradientBoostingClassifier,
+)
 from DashAI.back.models.scikit_learn.gradient_boosting_regression import (
     GradientBoostingR,
 )
 from DashAI.back.models.scikit_learn.hist_gradient_boosting_classifier import (
     HistGradientBoostingClassifier,
 )
+from DashAI.back.models.scikit_learn.hist_gradient_boosting_regression import (
+    HistGradientBoostingRegression,
+)
 from DashAI.back.models.scikit_learn.k_neighbors_classifier import KNeighborsClassifier
+from DashAI.back.models.scikit_learn.k_neighbors_regression import KNeighborsRegression
+from DashAI.back.models.scikit_learn.lasso_regression import LassoRegression
 from DashAI.back.models.scikit_learn.linear_regression import LinearRegression
+from DashAI.back.models.scikit_learn.linear_svc_classifier import LinearSVCClassifier
 from DashAI.back.models.scikit_learn.linearSVR import LinearSVR
 from DashAI.back.models.scikit_learn.logistic_regression import LogisticRegression
+from DashAI.back.models.scikit_learn.mlp_classifier import MLPClassifier
 from DashAI.back.models.scikit_learn.mlp_regression import MLPRegression
 from DashAI.back.models.scikit_learn.random_forest_classifier import (
     RandomForestClassifier,
@@ -190,7 +241,12 @@
     RandomForestRegression,
 )
 from DashAI.back.models.scikit_learn.ridge_regression import RidgeRegression
+from DashAI.back.models.scikit_learn.sgd_classifier import SGDClassifier
 from DashAI.back.models.scikit_learn.svc import SVC
+from DashAI.back.models.scikit_learn.svr import SVR
+from DashAI.back.models.scikit_learn.tfidf_logreg_text_classification_model import (
+    TfIdfLogRegTextClassificationModel,
+)
 
 # Optimizers
 from DashAI.back.optimizers.hyperopt_optimizer import HyperOptOptimizer
@@ -240,42 +296,76 @@ def get_initial_components():
         TextToTextGenerationTask,
         ControlNetTask,
         # Models
-        SVC,
+        AdaBoostClassifier,
+        AlbertTransformer,
+        AdaBoostRegression,
+        BaggingClassifier,
+        BagOfWordsTextClassificationModel,
+        BertTransformer,
+        BertinTransformer,
+        BetoTransformer,
+        BayesianRidgeRegression,
+        DebertaV3Transformer,
         DecisionTreeClassifier,
+        DecisionTreeRegression,
+        DistilBertTransformer,
         DummyClassifier,
+        ElasticNetRegression,
+        ElectraTransformer,
+        ExtraTreesClassifier,
+        ExtraTreesRegression,
+        GaussianNB,
+        GradientBoostingClassifier,
         GradientBoostingR,
         HistGradientBoostingClassifier,
+        HistGradientBoostingRegression,
         KNeighborsClassifier,
-        QwenModel,
+        KNeighborsRegression,
+        LassoRegression,
+        LinearRegression,
+        LinearSVCClassifier,
+        LinearSVR,
         LlamaModel,
+        LogisticRegression,
+        M2M100Transformer,
+        MiniLMTransformer,
         MistralModel,
         MixtralModel,
-        SmolLMModel,
-        StableDiffusionV2Model,
-        StableDiffusionV3Model,
-        StableDiffusionXLModel,
-        SDXLTurboModel,
-        PixArtSigmaModel,
-        TongyiZImageModel,
-        StableDiffusionXLV1ControlNet,
-        SD15DepthControlNetModel,
-        SD15OpenPoseControlNetModel,
-        SD15HEDControlNetModel,
-        SDXLCannyControlNetModel,
-        LogisticRegression,
+        MultilingualBertTransformer,
+        MLPClassifier,
         MLPRegression,
-        RandomForestClassifier,
-        RandomForestRegression,
-        DistilBertTransformer,
         ModernBertTransformer,
-        DebertaV3Transformer,
+        NllbTransformer,
+        OpusMtEnDeTransformer,
         OpusMtEnESTransformer,
+        OpusMtEnFrTransformer,
+        OpusMtEnPtTransformer,
         OpusMtEsENTransformer,
-        NllbTransformer,
-        BagOfWordsTextClassificationModel,
+        OpusMtFrEnTransformer,
+        PixArtSigmaModel,
+        QwenModel,
+        RandomForestClassifier,
+        RobertaTransformer,
+        RandomForestRegression,
         RidgeRegression,
-        LinearSVR,
-        LinearRegression,
+        SD15DepthControlNetModel,
+        SD15HEDControlNetModel,
+        SD15OpenPoseControlNetModel,
+        SDXLCannyControlNetModel,
+        SDXLTurboModel,
+        SGDClassifier,
+        SmolLMModel,
+        StableDiffusionV2Model,
+        StableDiffusionV3Model,
+        StableDiffusionXLModel,
+        StableDiffusionXLV1ControlNet,
+        SVC,
+        SVR,
+        T5SmallTransformer,
+        TfIdfLogRegTextClassificationModel,
+        TongyiZImageModel,
+        XlmRobertaTransformer,
+        XlnetTransformer,
         # Dataloaders
         CSVDataLoader,
         JSONDataLoader,

DashAI/back/metrics/classification/log_loss.py

@@ -86,4 +86,8 @@ def score(
         from sklearn.metrics import log_loss
 
         true_labels, _ = prepare_to_metric(true_labels, probs_pred_labels)
-        return log_loss(true_labels, probs_pred_labels)
+        # Pass all expected class indices so log_loss works even when a split
+        # happens to contain only one class (e.g. small validation sets).
+        n_classes = probs_pred_labels.shape[1]
+        labels = list(range(n_classes))
+        return log_loss(true_labels, probs_pred_labels, labels=labels)

DashAI/back/metrics/classification/roc_auc.py

@@ -80,14 +80,26 @@ def score(
         float
             RoC AUC score between true labels and predicted labels
         """
+        import numpy as np
+
         true_labels, _ = prepare_to_metric(true_labels, probs_pred_labels)
-        # Use the provided multiclass parameter or determine it using is_multiclass
-        if multiclass is None:
-            multiclass = ClassificationMetric.is_multiclass(true_labels)
+
+        # ROC AUC is undefined when only one class is present in y_true.
+        if len(np.unique(true_labels)) < 2:
+            return float("nan")
 
         from sklearn.metrics import roc_auc_score
 
+        n_classes = probs_pred_labels.shape[1]
+
+        if multiclass is None:
+            multiclass = n_classes > 2
         if multiclass:
-            return roc_auc_score(true_labels, probs_pred_labels, multi_class="ovr")
+            return roc_auc_score(
+                true_labels,
+                probs_pred_labels,
+                multi_class="ovr",
+                labels=list(range(n_classes)),
+            )
         else:
             return roc_auc_score(true_labels, probs_pred_labels[:, 1])

DashAI/back/models/hugging_face/albert_transformer.py

@@ -0,0 +1,45 @@
+"""DashAI implementation of ALBERT model for English text classification."""
+
+from DashAI.back.core.utils import MultilingualString
+from DashAI.back.models.hugging_face.base_text_classification_transformer import (
+    HuggingFaceTextClassificationTransformer,
+)
+from DashAI.back.models.hugging_face.distilbert_transformer import (
+    DistilBertTransformerSchema,
+)
+
+
+class AlbertTransformer(HuggingFaceTextClassificationTransformer):
+    """Pre-trained ALBERT model for efficient English text classification.
+
+    ALBERT (A Lite BERT) reduces BERT's parameters via cross-layer parameter
+    sharing and factorised embedding parametrisation, making it significantly
+    smaller and faster while retaining high accuracy. Requires the
+    ``sentencepiece`` package for its tokeniser.
+
+    References
+    ----------
+    - [1] Lan, Z. et al. (2020). "ALBERT: A Lite BERT for Self-supervised
+           Learning of Language Representations." ICLR 2020.
+    - [2] https://huggingface.co/albert-base-v2
+    """
+
+    DISPLAY_NAME: str = MultilingualString(
+        en="ALBERT Transformer",
+        es="Transformer ALBERT",
+    )
+    DESCRIPTION: str = MultilingualString(
+        en=(
+            "Parameter-efficient BERT variant for English text classification. "
+            "Downloads weights from Hugging Face on first use (internet required)."
+        ),
+        es=(
+            "Variante de BERT eficiente en parámetros para clasificación en inglés. "
+            "Descarga pesos de Hugging Face en el primer uso (requiere internet)."
+        ),
+    )
+    COLOR: str = "#00838F"
+    ICON: str = "Speed"
+    SCHEMA = DistilBertTransformerSchema
+    MODEL_NAME: str = "albert-base-v2"
+    TEMP_CHECKPOINT_DIR: str = "DashAI/back/user_models/temp_checkpoints_albert"