config.yaml

# Configuración del Sistema de IA Multimodal

# Configuración de modelos
models:
  text_embedder: "sentence-transformers/all-MiniLM-L6-v2"
  sentiment_analyzer: "cardiffnlp/twitter-roberta-base-sentiment-latest"
  image_captioner: "Salesforce/blip-image-captioning-base"
  image_classifier: "google/vit-base-patch16-224"
  
# Configuración de procesamiento
processing:
  max_text_length: 512
  image_size: [224, 224]
  audio_sample_rate: 16000
  batch_size: 8
  max_concurrent_requests: 10

# Configuración de reducción dimensional
dimensionality_reduction:
  umap:
    n_components: 2
    n_neighbors: 15
    min_dist: 0.1
    random_state: 42
  tsne:
    n_components: 2
    perplexity: 30
    random_state: 42
  dbscan:
    eps: 0.5
    min_samples: 5

# Configuración de la interfaz
interface:
  title: "Sistema de IA Multimodal Avanzado"
  theme: "soft"
  max_file_size: "10MB"
  allowed_extensions: [".jpg", ".png", ".wav", ".mp3", ".txt"]

# Configuración de logging
logging:
  level: "INFO"
  format: "%(asctime)s - %(name)s - %(levelname)s - %(message)s"
  file: "multimodal_system.log"

# Configuración de recursos
resources:
  gpu_memory_fraction: 0.8
  cpu_threads: 4
  cache_size: "1GB"