feat: Unified HPC Toolchain — merge Nexa_Vortex runtime + Nexa_Inference apps + CUTLASS kernels into PyC

apps/inference_api/models/schemas.py

@@ -0,0 +1,39 @@
+from pydantic import BaseModel, Field
+from typing import List, Optional
+
+class BiologyRequest(BaseModel):
+    sequence: str = Field(..., description="Protein sequence")
+    model_version: str = Field(default="2", pattern="^[12]$")
+    confidence_threshold: float = Field(default=0.8, ge=0.0, le=1.0)
+    explain: bool = Field(default=False, description="Enable XAI explanation")
+    mode: Optional[str] = Field(default=None, description="Inference mode, e.g., 'speculative'")
+
+class MaterialsRequest(BaseModel):
+    structure: str = Field(..., description="Material structure")
+    model_version: str = Field(default="2", pattern="^[12]$")
+    energy_threshold: float = Field(default=0.5, ge=0.0)
+    explain: bool = Field(default=False, description="Enable XAI explanation")
+    mode: Optional[str] = Field(default=None, description="Inference mode, e.g., 'speculative'")
+
+class BatchBiologyRequest(BaseModel):
+    requests: List[BiologyRequest]
+
+class BatchMaterialsRequest(BaseModel):
+    requests: List[MaterialsRequest]
+
+class PipelineRequest(BaseModel):
+    query: str = Field(..., description="The natural language query to initiate the pipeline.")
+
+class LLMRequest(BaseModel):
+    prompt: str = Field(..., description="Prompt for the LLM.")
+    max_tokens: int = Field(default=100, ge=10, le=512)
+
+class BenchmarkRequest(BaseModel):
+    model_type: str = Field(..., description="Model to benchmark", pattern="^(bio|materials|llm)$")
+    model_version: str = Field(default="1", description="Model version to benchmark")
+    test_duration_seconds: int = Field(default=10, ge=5, le=60, description="Duration of the benchmark test in seconds.")
+
+class DatasetRequest(BaseModel):
+    model_type: str = Field(..., pattern="^(bio|materials)$")
+    model_version: str = Field(default="2", pattern="^[12]$")
+    size: int = Field(default=100, ge=10, le=1000)

apps/inference_api/src/auth.py

@@ -0,0 +1,17 @@
+from fastapi import HTTPException, Security
+from fastapi.security.api_key import APIKeyHeader
+import os
+
+API_KEY_HEADER = APIKeyHeader(name="X-API-Key", auto_error=True)
+
+async def verify_api_key(api_key: str = Security(API_KEY_HEADER)):
+    """Verify API key from header"""
+    # For development, use a hardcoded key. In production, use environment variables
+    VALID_API_KEY = os.getenv("API_KEY", "development_key")
+
+    if api_key != VALID_API_KEY:
+        raise HTTPException(
+            status_code=401,
+            detail="Invalid API Key"
+        )
+    return api_key

apps/inference_api/src/config.py

@@ -0,0 +1,48 @@
+# src/Config.py
+import os
+import json
+
+class Config:
+    """
+    Configuration class to manage settings for the application.
+    Loads configuration from a JSON file and provides access to settings.
+    """
+
+    def __init__(self, config_file='config.json'):
+        self.config_file = config_file
+        self.settings = {}
+        self.load_config()
+
+    def load_config(self):
+        """Load configuration from a JSON file."""
+        if not os.path.exists(self.config_file):
+            raise FileNotFoundError(f"Configuration file not found: {self.config_file}")
+
+        with open(self.config_file, 'r') as f:
+            self.settings = json.load(f)
+
+    def get(self, key, default=None):
+        """Get a configuration setting by key."""
+        return self.settings.get(key, default)
+
+    def set(self, key, value):
+        """Set a configuration setting by key and save it."""
+        self.settings[key] = value
+        self.save_config()
+
+    def save_config(self):
+        """Save the current settings to the configuration file."""
+        with open(self.config_file, 'w') as f:
+            json.dump(self.settings, f, indent=4)
+
+    def __getitem__(self, key):
+        """Get a configuration setting using dictionary-like access."""
+        return self.get(key)
+
+    def __setitem__(self, key, value):
+        """Set a configuration setting using dictionary-like access."""
+        self.set(key, value)
+
+    def __repr__(self):
+        """String representation of the configuration settings."""
+        return f"Config(settings={self.settings})"