training_config.json

{
  "model_name": "deepseek-ai/DeepSeek-OCR",
  "output_dir": "/app/trained_model",
  "batch_size": 6,
  "num_epochs": 3,
  "gradient_accumulation_steps": 2,
  "max_length": 512,
  "mixed_precision": "bf16",
  "optimizer": {
    "learning_rate": 2e-05,
    "weight_decay": 0.0,
    "warmup_steps": 0,
    "warmup_ratio": null,
    "scheduler_type": "cosine_with_warmup"
  },
  "log_interval": 10,
  "save_interval": 500,
  "push_to_hub": true,
  "hub_model_id": "baconnier/deepsynth-ocr-finetuned",
  "hub_private": false,
  "evaluation_split": "validation",
  "save_checkpoints_to_hub": true,
  "resume_from_checkpoint": null,
  "metrics_output_path": null,
  "save_metrics_to_hub": true,
  "max_train_samples": null,
  "max_eval_samples": null,
  "expert_dropout_rate": 0.0,
  "expert_dropout_min_keep": 1,
  "bidrop_passes": 1,
  "gate_dropout_rate": 0.0,
  "gate_dropout_keywords": [
    "gate",
    "router"
  ],
  "target_resolution": "base",
  "use_augmentation": true,
  "random_resize_min": null,
  "random_resize_max": null,
  "rotation_degrees": 3.0,
  "perspective_distortion": 0.1,
  "perspective_prob": 0.3,
  "color_jitter_brightness": 0.1,
  "color_jitter_contrast": 0.1,
  "horizontal_flip_prob": 0.3,
  "use_lora": true,
  "lora_rank": 64,
  "lora_alpha": 128,
  "lora_dropout": 0.05,
  "lora_target_modules": null,
  "lora_bias": "none",
  "use_qlora": false,
  "qlora_bits": 4,
  "qlora_type": "nf4",
  "qlora_double_quant": true,
  "lora_modules_to_save": null,
  "use_text_encoder": false,
  "text_encoder_type": null,
  "text_encoder_model": null,
  "text_encoder_trainable": true,
  "instruction_prompt": "Summarize this text:",
  "use_text_projection": false
}