Renaming Structured Data Classification task to Tabular Classification

api_inference_community/validation.py

@@ -133,7 +133,7 @@ def all_rows_must_have_same_length(cls, table: Dict[str, List[str]]):
         raise ValueError("All rows in the table must be the same length")
 
 
-class StructuredDataClassificationInputsCheck(BaseModel):
+class TabularClassificationInputsCheck(BaseModel):
     data: Dict[str, List[str]]
 
     @validator("data")
@@ -177,7 +177,7 @@ class StringInput(BaseModel):
     "feature-extraction": StringOrStringBatchInputCheck,
     "sentence-similarity": SentenceSimilarityInputsCheck,
     "table-question-answering": TableQuestionAnsweringInputsCheck,
-    "structured-data-classification": StructuredDataClassificationInputsCheck,
+    "tabular-classification": TabularClassificationInputsCheck,
     "fill-mask": StringInput,
     "summarization": StringInput,
     "text2text-generation": StringInput,
@@ -228,7 +228,7 @@ def check_inputs(inputs, tag):
     "sentence-similarity",
     "fill-mask",
     "table-question-answering",
-    "structured-data-classification",
+    "tabular-classification",
     "summarization",
     "text2text-generation",
     "text-classification",

docker_images/adapter_transformers/app/pipelines/__init__.py

@@ -9,9 +9,7 @@
 from app.pipelines.question_answering import QuestionAnsweringPipeline
 from app.pipelines.sentence_similarity import SentenceSimilarityPipeline
 from app.pipelines.speech_segmentation import SpeechSegmentationPipeline
-from app.pipelines.structured_data_classification import (
-    StructuredDataClassificationPipeline,
-)
+from app.pipelines.tabular_classification import TabularClassificationPipeline
 from app.pipelines.text_classification import TextClassificationPipeline
 from app.pipelines.text_to_speech import TextToSpeechPipeline
 from app.pipelines.token_classification import TokenClassificationPipeline

docker_images/adapter_transformers/app/pipelines/structured_data_classification.py → docker_images/adapter_transformers/app/pipelines/tabular_classification.py

@@ -3,14 +3,14 @@
 from app.pipelines import Pipeline
 
 
-class StructuredDataClassificationPipeline(Pipeline):
+class TabularClassificationPipeline(Pipeline):
     def __init__(self, model_id: str):
         # IMPLEMENT_THIS
         # Preload all the elements you are going to need at inference.
         # For instance your model, processors, tokenizer that might be needed.
         # This function is only called once, so do all the heavy processing I/O here
         raise NotImplementedError(
-            "Please implement StructuredDataClassificationPipeline __init__ function"
+            "Please implement TabularClassificationPipeline __init__ function"
         )
 
     def __call__(
@@ -26,5 +26,5 @@ def __call__(
         """
         # IMPLEMENT_THIS
         raise NotImplementedError(
-            "Please implement StructuredDataClassificationPipeline __init__ function"
+            "Please implement TabularClassificationPipeline __init__ function"
         )

docker_images/stanza/tests/test_api_structured_data_classification.py → docker_images/adapter_transformers/tests/test_api_tabular_classification.py

@@ -8,16 +8,16 @@
 
 
 @skipIf(
-    "structured-data-classification" not in ALLOWED_TASKS,
-    "structured-data-classification not implemented",
+    "tabular-classification" not in ALLOWED_TASKS,
+    "tabular-classification not implemented",
 )
-class StructuredDataClassificationTestCase(TestCase):
+class TabularClassificationTestCase(TestCase):
     def setUp(self):
-        model_id = TESTABLE_MODELS["structured-data-classification"]
+        model_id = TESTABLE_MODELS["tabular-classification"]
         self.old_model_id = os.getenv("MODEL_ID")
         self.old_task = os.getenv("TASK")
         os.environ["MODEL_ID"] = model_id
-        os.environ["TASK"] = "structured-data-classification"
+        os.environ["TASK"] = "tabular-classification"
 
         from app.main import app
 

docker_images/common/app/pipelines/__init__.py

@@ -10,8 +10,6 @@
 from app.pipelines.question_answering import QuestionAnsweringPipeline
 from app.pipelines.sentence_similarity import SentenceSimilarityPipeline
 from app.pipelines.speech_segmentation import SpeechSegmentationPipeline
-from app.pipelines.structured_data_classification import (
-    StructuredDataClassificationPipeline,
-)
+from app.pipelines.tabular_classification import TabularClassificationPipeline
 from app.pipelines.text_to_speech import TextToSpeechPipeline
 from app.pipelines.token_classification import TokenClassificationPipeline

docker_images/pyannote/app/pipelines/structured_data_classification.py → docker_images/common/app/pipelines/tabular_classification.py

@@ -3,14 +3,14 @@
 from app.pipelines import Pipeline
 
 
-class StructuredDataClassificationPipeline(Pipeline):
+class TabularClassificationPipeline(Pipeline):
     def __init__(self, model_id: str):
         # IMPLEMENT_THIS
         # Preload all the elements you are going to need at inference.
         # For instance your model, processors, tokenizer that might be needed.
         # This function is only called once, so do all the heavy processing I/O here
         raise NotImplementedError(
-            "Please implement StructuredDataClassificationPipeline __init__ function"
+            "Please implement TabularClassificationPipeline __init__ function"
         )
 
     def __call__(
@@ -26,5 +26,5 @@ def __call__(
         """
         # IMPLEMENT_THIS
         raise NotImplementedError(
-            "Please implement StructuredDataClassificationPipeline __init__ function"
+            "Please implement TabularClassificationPipeline __init__ function"
         )

docker_images/common/tests/test_api.py

@@ -24,7 +24,7 @@
     "question-answering",
     "sentence-similarity",
     "speech-segmentation",
-    "structured-data-classification",
+    "tabular-classification",
     "text-classification",
     "text-to-image",
     "text-to-speech",
@@ -35,7 +35,7 @@
     "sentence-similarity",
     "fill-mask",
     "table-question-answering",
-    "structured-data-classification",
+    "tabular-classification",
     "summarization",
     "text2text-generation",
     "text-classification",

docker_images/adapter_transformers/tests/test_api_structured_data_classification.py → docker_images/common/tests/test_api_tabular_classification.py

@@ -8,16 +8,16 @@
 
 
 @skipIf(
-    "structured-data-classification" not in ALLOWED_TASKS,
-    "structured-data-classification not implemented",
+    "tabular-classification" not in ALLOWED_TASKS,
+    "tabular-classification not implemented",
 )
-class StructuredDataClassificationTestCase(TestCase):
+class TabularClassificationTestCase(TestCase):
     def setUp(self):
-        model_id = TESTABLE_MODELS["structured-data-classification"]
+        model_id = TESTABLE_MODELS["tabular-classification"]
         self.old_model_id = os.getenv("MODEL_ID")
         self.old_task = os.getenv("TASK")
         os.environ["MODEL_ID"] = model_id
-        os.environ["TASK"] = "structured-data-classification"
+        os.environ["TASK"] = "tabular-classification"
 
         from app.main import app
 

docker_images/generic/tests/test_api.py

@@ -10,7 +10,7 @@
     # This is very slow the first time as fasttext model is large.
     "feature-extraction": ["osanseviero/fasttext_english"],
     "image-classification": ["osanseviero/fastai_cat_vs_dog"],
-    "structured-data-classification": ["osanseviero/wine-quality"],
+    "tabular-classification": ["osanseviero/wine-quality"],
     "text-classification": ["osanseviero/fasttext_nearest"],
     "text-to-image": ["osanseviero/BigGAN-deep-128"],
     "token-classification": ["osanseviero/en_core_web_sm"],

docker_images/generic/tests/test_api_structured_data_classification.py → docker_images/generic/tests/test_api_tabular_classification.py

@@ -8,21 +8,18 @@
 
 
 @skipIf(
-    "structured-data-classification" not in TESTABLE_MODELS,
-    "structured-data-classification not implemented",
+    "tabular-classification" not in TESTABLE_MODELS,
+    "tabular-classification not implemented",
 )
 @parameterized_class(
-    [
-        {"model_id": model_id}
-        for model_id in TESTABLE_MODELS["structured-data-classification"]
-    ]
+    [{"model_id": model_id} for model_id in TESTABLE_MODELS["tabular-classification"]]
 )
-class StructuredDataClassificationTestCase(TestCase):
+class TabularClassificationTestCase(TestCase):
     def setUp(self):
         self.old_model_id = os.getenv("MODEL_ID")
         self.old_task = os.getenv("TASK")
         os.environ["MODEL_ID"] = self.model_id
-        os.environ["TASK"] = "structured-data-classification"
+        os.environ["TASK"] = "tabular-classification"
 
         from app.main import app, get_pipeline
 

docker_images/nemo/tests/test_api.py

@@ -23,7 +23,7 @@
     "question-answering",
     "sentence-similarity",
     "speech-segmentation",
-    "structured-data-classification",
+    "tabular-classification",
     "text-classification",
     "text-to-image",
     "text-to-speech",
@@ -34,7 +34,7 @@
     "sentence-similarity",
     "fill-mask",
     "table-question-answering",
-    "structured-data-classification",
+    "tabular-classification",
     "summarization",
     "text2text-generation",
     "text-classification",

docker_images/pyannote/app/pipelines/__init__.py

@@ -10,9 +10,7 @@
 from app.pipelines.question_answering import QuestionAnsweringPipeline
 from app.pipelines.sentence_similarity import SentenceSimilarityPipeline
 from app.pipelines.speech_segmentation import SpeechSegmentationPipeline
-from app.pipelines.structured_data_classification import (
-    StructuredDataClassificationPipeline,
-)
+from app.pipelines.tabular_classification import TabularClassificationPipeline
 from app.pipelines.text_classification import TextClassificationPipeline
 from app.pipelines.text_to_speech import TextToSpeechPipeline
 from app.pipelines.token_classification import TokenClassificationPipeline

docker_images/common/app/pipelines/structured_data_classification.py → docker_images/pyannote/app/pipelines/tabular_classification.py

@@ -3,14 +3,14 @@
 from app.pipelines import Pipeline
 
 
-class StructuredDataClassificationPipeline(Pipeline):
+class TabularClassificationPipeline(Pipeline):
     def __init__(self, model_id: str):
         # IMPLEMENT_THIS
         # Preload all the elements you are going to need at inference.
         # For instance your model, processors, tokenizer that might be needed.
         # This function is only called once, so do all the heavy processing I/O here
         raise NotImplementedError(
-            "Please implement StructuredDataClassificationPipeline __init__ function"
+            "Please implement TabularClassificationPipeline __init__ function"
         )
 
     def __call__(
@@ -26,5 +26,5 @@ def __call__(
         """
         # IMPLEMENT_THIS
         raise NotImplementedError(
-            "Please implement StructuredDataClassificationPipeline __init__ function"
+            "Please implement TabularClassificationPipeline __init__ function"
         )

docker_images/pyannote/tests/test_api.py

@@ -22,7 +22,7 @@
     "question-answering",
     "sentence-similarity",
     "speech-segmentation",
-    "structured-data-classification",
+    "tabular-classification",
     "text-classification",
     "text-to-image",
     "text-to-speech",
@@ -33,7 +33,7 @@
     "sentence-similarity",
     "fill-mask",
     "table-question-answering",
-    "structured-data-classification",
+    "tabular-classification",
     "summarization",
     "text2text-generation",
     "text-classification",

docker_images/pyannote/tests/test_api_structured_data_classification.py → docker_images/pyannote/tests/test_api_tabular_classification.py

@@ -8,16 +8,16 @@
 
 
 @skipIf(
-    "structured-data-classification" not in ALLOWED_TASKS,
-    "structured-data-classification not implemented",
+    "tabular-classification" not in ALLOWED_TASKS,
+    "tabular-classification not implemented",
 )
-class StructuredDataClassificationTestCase(TestCase):
+class TabularClassificationTestCase(TestCase):
     def setUp(self):
-        model_id = TESTABLE_MODELS["structured-data-classification"]
+        model_id = TESTABLE_MODELS["tabular-classification"]
         self.old_model_id = os.getenv("MODEL_ID")
         self.old_task = os.getenv("TASK")
         os.environ["MODEL_ID"] = model_id
-        os.environ["TASK"] = "structured-data-classification"
+        os.environ["TASK"] = "tabular-classification"
 
         from app.main import app
 

docker_images/sklearn/app/main.py

@@ -4,7 +4,7 @@
 from typing import Dict, Type
 
 from api_inference_community.routes import pipeline_route, status_ok
-from app.pipelines import Pipeline, StructuredDataClassificationPipeline
+from app.pipelines import Pipeline, TabularClassificationPipeline
 from starlette.applications import Starlette
 from starlette.middleware import Middleware
 from starlette.middleware.gzip import GZipMiddleware
@@ -34,7 +34,7 @@
 # directories. Implement directly within the directories.
 ALLOWED_TASKS: Dict[str, Type[Pipeline]] = {
     # IMPLEMENT_THIS: Add your implemented tasks here!
-    "structured-data-classification": StructuredDataClassificationPipeline
+    "tabular-classification": TabularClassificationPipeline
 }
 
 

docker_images/sklearn/app/pipelines/__init__.py

@@ -8,8 +8,6 @@
 from app.pipelines.image_classification import ImageClassificationPipeline
 from app.pipelines.question_answering import QuestionAnsweringPipeline
 from app.pipelines.sentence_similarity import SentenceSimilarityPipeline
-from app.pipelines.structured_data_classification import (
-    StructuredDataClassificationPipeline,
-)
+from app.pipelines.tabular_classification import TabularClassificationPipeline
 from app.pipelines.text_to_speech import TextToSpeechPipeline
 from app.pipelines.token_classification import TokenClassificationPipeline

docker_images/sklearn/app/pipelines/structured_data_classification.py → docker_images/sklearn/app/pipelines/tabular_classification.py

@@ -8,7 +8,7 @@
 DEFAULT_FILENAME = "sklearn_model.joblib"
 
 
-class StructuredDataClassificationPipeline(Pipeline):
+class TabularClassificationPipeline(Pipeline):
     def __init__(self, model_id: str):
         self.model = joblib.load(
             open(cached_download(hf_hub_url(model_id, DEFAULT_FILENAME)), "rb")

docker_images/sklearn/tests/test_api.py

@@ -8,9 +8,7 @@
 # Must contain at least one example of each implemented pipeline
 # Tests do not check the actual values of the model output, so small dummy
 # models are recommended for faster tests.
-TESTABLE_MODELS: Dict[str, str] = {
-    "structured-data-classification": "julien-c/wine-quality"
-}
+TESTABLE_MODELS: Dict[str, str] = {"tabular-classification": "julien-c/wine-quality"}
 
 
 ALL_TASKS = {
@@ -20,7 +18,7 @@
     "image-classification",
     "question-answering",
     "sentence-similarity",
-    "structured-data-classification",
+    "tabular-classification",
     "text-generation",
     "text-to-speech",
     "token-classification",

docker_images/sklearn/tests/test_api_structured_data_classification.py → docker_images/sklearn/tests/test_api_tabular_classification.py

@@ -8,16 +8,16 @@
 
 
 @skipIf(
-    "structured-data-classification" not in ALLOWED_TASKS,
-    "structured-data-classification not implemented",
+    "tabular-classification" not in ALLOWED_TASKS,
+    "tabular-classification not implemented",
 )
-class StructuredDataClassificationTestCase(TestCase):
+class TabularClassificationTestCase(TestCase):
     def setUp(self):
-        model_id = TESTABLE_MODELS["structured-data-classification"]
+        model_id = TESTABLE_MODELS["tabular-classification"]
         self.old_model_id = os.getenv("MODEL_ID")
         self.old_task = os.getenv("TASK")
         os.environ["MODEL_ID"] = model_id
-        os.environ["TASK"] = "structured-data-classification"
+        os.environ["TASK"] = "tabular-classification"
 
         from app.main import app
 

docker_images/stanza/app/pipelines/__init__.py

@@ -10,8 +10,6 @@
 from app.pipelines.question_answering import QuestionAnsweringPipeline
 from app.pipelines.sentence_similarity import SentenceSimilarityPipeline
 from app.pipelines.speech_segmentation import SpeechSegmentationPipeline
-from app.pipelines.structured_data_classification import (
-    StructuredDataClassificationPipeline,
-)
+from app.pipelines.tabular_classification import TabularClassificationPipeline
 from app.pipelines.text_to_speech import TextToSpeechPipeline
 from app.pipelines.token_classification import TokenClassificationPipeline

docker_images/stanza/app/pipelines/structured_data_classification.py → docker_images/stanza/app/pipelines/tabular_classification.py

@@ -3,14 +3,14 @@
 from app.pipelines import Pipeline
 
 
-class StructuredDataClassificationPipeline(Pipeline):
+class TabularClassificationPipeline(Pipeline):
     def __init__(self, model_id: str):
         # IMPLEMENT_THIS
         # Preload all the elements you are going to need at inference.
         # For instance your model, processors, tokenizer that might be needed.
         # This function is only called once, so do all the heavy processing I/O here
         raise NotImplementedError(
-            "Please implement StructuredDataClassificationPipeline __init__ function"
+            "Please implement TabularClassificationPipeline __init__ function"
         )
 
     def __call__(
@@ -26,5 +26,5 @@ def __call__(
         """
         # IMPLEMENT_THIS
         raise NotImplementedError(
-            "Please implement StructuredDataClassificationPipeline __init__ function"
+            "Please implement TabularaClassificationPipeline __init__ function"
         )