Upload 8 files

app.py

@@ -0,0 +1,50 @@
+import os
+
+import gradio as gr
+from huggingface_hub import login, CommitScheduler
+
+from lib.result import Result
+from lib.data_entry import DataEntry
+from lib.telemetry import TelemetryManager
+from lib.model import Model
+from lib.lang import Language
+from schemas.request import Request
+
+login(os.environ["HF_TOKEN"])
+
+models: dict[Language, Model] = {
+    Language.ENGLISH: Model.get_english_model(),
+    Language.SPANISH: Model.get_spanish_model()
+}
+
+telemetry = TelemetryManager()
+
+async def app_func(text: str, language: str) -> int:
+    try:
+        request = Request(text=text, language=language)
+    except ValueError as e:
+        raise gr.Error(e)
+    
+    result = models[request.language].analyze(request.text)
+    telemetry.write_data(DataEntry(text, result))
+
+    percentage = round(result.percentage * 100)
+    percentage = max(percentage, 0)
+    percentage = min(percentage, 100)
+
+    return percentage
+
+
+demo = gr.Interface(
+    fn=app_func,
+    inputs=[gr.Text(label="Texto"), gr.Radio(label="Idioma", choices=[Language.ENGLISH.value, Language.SPANISH.value])],
+    outputs=gr.Label(num_top_classes=1, label="Probabilidad de phishing"),
+    title="ConfIA Model Demo",
+    description="Demo que te permite probar nuestros modelos de forma muy sencilla",
+    examples=[["You have just Woned a free iPhone 16!! FOR FREE!!!", Language.ENGLISH], ["When will you review that PR? It's kinda urgent", Language.ENGLISH]],
+    cache_examples=True
+)
+
+if __name__ == "__main__":
+    demo.queue(max_size=5)
+    demo.launch(share=True, debug=True)

lib/data_entry.py

@@ -0,0 +1,16 @@
+from dataclasses import dataclass
+
+from lib.result import Result
+
+
+@dataclass(frozen=True, slots=True)
+class DataEntry:
+    text: str
+    result: Result
+
+    def to_dict(self) -> dict[str, str | float | bool]:
+        return {
+            "text": self.text,
+            "is_phishing": self.result.is_phishing(),
+            "percentage": self.result.percentage
+        }

lib/lang.py

@@ -0,0 +1,6 @@
+from enum import Enum
+
+
+class Language(Enum):
+    ENGLISH = "Inglés"
+    SPANISH = "Español"

lib/model.py

@@ -0,0 +1,40 @@
+import os
+
+import tensorflow as tf
+import keras
+from huggingface_hub import login, hf_hub_download
+
+from lib.result import Result
+
+os.environ["KERAS_BACKEND"] = "jax"
+
+login(token=os.getenv("HF_TOKEN"))
+
+
+class Model:
+    
+    @staticmethod
+    def get_english_model() -> "Model":
+        return Model("elsamueldev/confia-97-english", "confia-97-english.keras")
+    
+    @staticmethod
+    def get_spanish_model() -> "Model":
+        return Model("elsamueldev/confia-97-spanish", "confia-97-spanish.keras")
+
+    def __init__(self, repo_id: str, filename: str) -> None:
+        path = hf_hub_download(
+            repo_id=repo_id,
+            filename=filename,
+            local_dir="./",
+            local_dir_use_symlinks=False
+        )
+        if path is None:
+            raise RuntimeError("Model could not be downloaded")
+        
+        self.__model = keras.saving.load_model(path)
+
+    def analyze(self, text: str) -> Result:
+        raw_result = self.__model.predict(tf.constant([text]))
+        result = round(float(raw_result[0][0]), 3)
+
+        return Result(percentage=result)

lib/result.py

@@ -0,0 +1,9 @@
+from dataclasses import dataclass
+
+
+@dataclass(frozen=True, slots=True)
+class Result:
+    percentage: float
+
+    def is_phishing(self) -> bool:
+        return self.percentage >= 0.5

lib/telemetry.py

@@ -0,0 +1,29 @@
+import os
+from uuid import uuid4
+import json
+
+from huggingface_hub import login, CommitScheduler
+
+from lib.data_entry import DataEntry
+
+login(os.environ["HF_TOKEN"])
+
+
+class TelemetryManager:
+    def __init__(self) -> None:
+        self.__file_path = f"data/{uuid4()}.jsonl"
+        self.__scheduler = CommitScheduler(
+            repo_id="elsamueldev/confia-demo-data",
+            repo_type="dataset",
+            private=True,
+            folder_path="data/",
+            path_in_repo="/",
+            every=5
+        )
+    
+    def write_data(self, data: DataEntry) -> None:
+        content = json.dumps(data.to_dict())
+        
+        with self.__scheduler.lock:
+            with open(self.__file_path, "a") as file:
+                file.write(f"{content}\n")

requirements.txt

@@ -0,0 +1,73 @@
+absl-py==2.1.0
+aiofiles==23.2.1
+annotated-types==0.7.0
+anyio==4.8.0
+astunparse==1.6.3
+certifi==2024.12.14
+charset-normalizer==3.4.1
+click==8.1.8
+fastapi==0.115.6
+ffmpy==0.5.0
+filelock==3.16.1
+flatbuffers==24.12.23
+fsspec==2024.12.0
+gast==0.6.0
+google-pasta==0.2.0
+gradio==5.12.0
+gradio_client==1.5.4
+grpcio==1.69.0
+h11==0.14.0
+h5py==3.12.1
+httpcore==1.0.7
+httpx==0.28.1
+huggingface-hub==0.27.1
+idna==3.10
+Jinja2==3.1.5
+keras==3.8.0
+libclang==18.1.1
+Markdown==3.7
+markdown-it-py==3.0.0
+MarkupSafe==2.1.5
+mdurl==0.1.2
+ml-dtypes==0.4.1
+namex==0.0.8
+numpy==2.0.2
+opt_einsum==3.4.0
+optree==0.14.0
+orjson==3.10.15
+packaging==24.2
+pandas==2.2.3
+pillow==11.1.0
+protobuf==5.29.3
+pydantic==2.10.5
+pydantic_core==2.27.2
+pydub==0.25.1
+Pygments==2.19.1
+python-dateutil==2.9.0.post0
+python-multipart==0.0.20
+pytz==2024.2
+PyYAML==6.0.2
+requests==2.32.3
+rich==13.9.4
+ruff==0.9.2
+safehttpx==0.1.6
+semantic-version==2.10.0
+shellingham==1.5.4
+six==1.17.0
+sniffio==1.3.1
+starlette==0.41.3
+tensorboard==2.18.0
+tensorboard-data-server==0.7.2
+tensorflow==2.18.0
+tensorflow-io-gcs-filesystem==0.37.1
+termcolor==2.5.0
+tomlkit==0.13.2
+tqdm==4.67.1
+typer==0.15.1
+typing_extensions==4.12.2
+tzdata==2024.2
+urllib3==2.3.0
+uvicorn==0.34.0
+websockets==14.2
+Werkzeug==3.1.3
+wrapt==1.17.2

schemas/request.py

@@ -0,0 +1,8 @@
+from pydantic import BaseModel, Field
+
+from lib.lang import Language
+
+
+class Request(BaseModel):
+    text: str = Field(min_length=3, max_length=500)
+    language: Language