> ## Documentation Index
> Fetch the complete documentation index at: https://docs.monostate.ai/llms.txt
> Use this file to discover all available pages before exploring further.
# Configurações YAML
> Use arquivos de configuração para treinamento
# Arquivos de Configuração YAML
Arquivos de configuração tornam configurações de treinamento complexas reproduzíveis e compartilháveis.
## Uso Básico
```bash theme={null}
aitraining --config training.yaml
```
## Estrutura do Arquivo de Configuração
A configuração YAML tem uma estrutura aninhada específica:
```yaml theme={null}
# training.yaml
task: llm-sft # Task with trainer suffix
backend: local # Required: local, spaces-*, etc.
# Model settings
base_model: google/gemma-3-270m # Note: "base_model", not "model"
project_name: my-gemma-model
# Data settings (nested under "data:")
data:
path: ./data/conversations.jsonl # Note: nested under data
train_split: train
valid_split: null
chat_template: tokenizer # For LLM: tokenizer, chatml, zephyr, none
column_mapping: # Column names
text_column: text
# Logging
log: wandb
# Hub settings (optional)
hub:
username: ${HF_USERNAME}
token: ${HF_TOKEN}
push_to_hub: false
# All other training parameters go under "params:"
params:
epochs: 3
batch_size: 4
lr: 3e-5
mixed_precision: bf16
peft: true
lora_r: 16
lora_alpha: 32
lora_dropout: 0.05
```
**Notas Importantes sobre a Estrutura:**
* Use `base_model`, não `model`
* O caminho dos dados é `data.path`, não `data_path`
* Mapeamentos de colunas ficam sob `data.column_mapping`
* Parâmetros de treinamento ficam sob `params:`
* O campo `backend` é **obrigatório**
## Tipos de Tarefas
O campo task inclui o tipo de trainer:
```yaml theme={null}
# LLM tasks (trainer in task name)
task: llm-sft # SFT training
task: llm-dpo # DPO training
task: llm-orpo # ORPO training
task: llm-reward # Reward model training
task: llm-generic # Default/pretraining
# Other tasks
task: text-classification # Text classification
task: image-classification # Image classification
task: token-classification # NER
task: seq2seq # Sequence to sequence
task: tabular # Tabular data
task: vlm:vqa # Vision-language (VQA)
task: vlm:captioning # Vision-language (captioning)
task: sentence-transformers:pair_score # Sentence transformers
```
## Configurações de Treinamento LLM
### Treinamento SFT
```yaml theme={null}
task: llm-sft
backend: local
base_model: meta-llama/Llama-3.2-1B
project_name: llama-sft
data:
path: ./conversations.jsonl
train_split: train
valid_split: null
chat_template: tokenizer
column_mapping:
text_column: text
log: wandb
hub:
push_to_hub: false
params:
epochs: 3
batch_size: 2
lr: 3e-5
peft: true
lora_r: 16
lora_alpha: 32
```
### Treinamento DPO
```yaml theme={null}
task: llm-dpo
backend: local
base_model: meta-llama/Llama-3.2-1B
project_name: llama-dpo
data:
path: ./preferences.jsonl
train_split: train
valid_split: null
chat_template: tokenizer
column_mapping:
prompt_text_column: prompt
text_column: chosen
rejected_text_column: rejected
log: wandb
params:
dpo_beta: 0.1
max_prompt_length: 128
max_completion_length: null
epochs: 1
batch_size: 2
peft: true
lora_r: 16
```
### Distilação de Conhecimento
```yaml theme={null}
task: llm-sft
backend: local
base_model: google/gemma-3-270m
project_name: distilled-model
data:
path: ./prompts.jsonl
train_split: train
valid_split: null
chat_template: tokenizer
column_mapping:
text_column: text
log: wandb
params:
use_distillation: true
teacher_model: google/gemma-2-2b
distill_temperature: 3.0
distill_alpha: 0.7
epochs: 3
```
## Configuração de Classificação de Texto
```yaml theme={null}
task: text-classification
backend: local
base_model: bert-base-uncased
project_name: sentiment-classifier
data:
path: ./reviews.csv
train_split: train
valid_split: null
column_mapping:
text_column: text
target_column: target
log: wandb
params:
epochs: 5
batch_size: 16
lr: 5e-5
```
## Variáveis de Ambiente em Configurações
Referencie variáveis de ambiente com `${VAR_NAME}`:
```yaml theme={null}
hub:
token: ${HF_TOKEN}
username: ${HF_USERNAME}
```
Configure-as antes de executar:
```bash theme={null}
export HF_TOKEN="hf_..."
export HF_USERNAME="my-username"
aitraining --config training.yaml
```
## Exemplo de Configuração Completa
```yaml theme={null}
task: llm-sft
backend: local
base_model: meta-llama/Llama-3.2-1B
project_name: production-model
data:
path: ./conversations.jsonl
train_split: train
valid_split: validation
chat_template: tokenizer
column_mapping:
text_column: text
log: wandb
hub:
push_to_hub: true
username: ${HF_USERNAME}
token: ${HF_TOKEN}
params:
# Training
epochs: 3
batch_size: 4
gradient_accumulation: 4
lr: 3e-5
warmup_ratio: 0.1
mixed_precision: bf16
# LoRA
peft: true
lora_r: 32
lora_alpha: 64
lora_dropout: 0.05
target_modules: all-linear
# Distribution (for multi-GPU)
distributed_backend: null # null for auto (DDP), or "deepspeed"
# Optimization
use_flash_attention_2: true
packing: true
auto_find_batch_size: true
# Checkpointing
logging_steps: 10
save_strategy: steps
save_steps: 100
save_total_limit: 1
```
## Configuração Mínima
Os campos mínimos obrigatórios:
```yaml theme={null}
task: llm-sft
backend: local
base_model: google/gemma-3-270m
project_name: my-model
data:
path: ./data.jsonl
train_split: train
valid_split: null
chat_template: tokenizer
column_mapping:
text_column: text
log: wandb
```
## Próximos Passos
Referência de parâmetros LLM
Templates prontos para uso