[train] KTransformers SFT as backend engine for LLaMA-Factory (#9400)

Co-authored-by: jimmy128 <jimmy128@noreply.gitcode.com>
Co-authored-by: Yaowei Zheng <hiyouga@buaa.edu.cn>

examples/train_lora/deepseek2_lora_sft_kt.yaml

@@ -0,0 +1,52 @@
+### model
+model_name_or_path: deepseek-ai/DeepSeek-V2-Lite
+trust_remote_code: true
+
+### method
+stage: sft
+do_train: true
+finetuning_type: lora
+lora_rank: 8
+lora_target: all
+
+### dataset
+dataset: identity
+template: deepseek
+cutoff_len: 2048
+max_samples: 100000
+overwrite_cache: true
+preprocessing_num_workers: 16
+dataloader_num_workers: 4
+
+### output
+output_dir: saves/Kllama_deepseekV2
+logging_steps: 10
+save_steps: 500
+plot_loss: true
+overwrite_output_dir: true
+save_only_model: false
+report_to: none  # choices: [none, wandb, tensorboard, swanlab, mlflow]
+
+### train
+per_device_train_batch_size: 1
+gradient_accumulation_steps: 8
+learning_rate: 1.0e-4
+num_train_epochs: 3.0
+lr_scheduler_type: cosine
+warmup_ratio: 0.1
+bf16: true
+ddp_timeout: 180000000
+resume_from_checkpoint: null
+
+### ktransformers
+use_kt: true # use KTransformers as LoRA sft backend
+kt_optimize_rule: examples/kt_optimize_rules/DeepSeek-V2-Lite-Chat-sft-amx.yaml
+cpu_infer: 32
+chunk_size: 8192
+
+### eval
+# eval_dataset: alpaca_en_demo
+# val_size: 0.1
+# per_device_eval_batch_size: 1
+# eval_strategy: steps
+# eval_steps: 500

examples/train_lora/deepseek3_lora_sft_kt.yaml

@@ -0,0 +1,52 @@
+### model
+model_name_or_path: opensourcerelease/DeepSeek-V3-bf16
+trust_remote_code: true
+
+### method
+stage: sft
+do_train: true
+finetuning_type: lora
+lora_rank: 8
+lora_target: all
+
+### dataset
+dataset: identity
+template: deepseek
+cutoff_len: 2048
+max_samples: 100000
+overwrite_cache: true
+preprocessing_num_workers: 16
+dataloader_num_workers: 4
+
+### output
+output_dir: saves/Kllama_deepseekV3
+logging_steps: 10
+save_steps: 500
+plot_loss: true
+overwrite_output_dir: true
+save_only_model: false
+report_to: none  # choices: [none, wandb, tensorboard, swanlab, mlflow]
+
+### train
+per_device_train_batch_size: 1
+gradient_accumulation_steps: 8
+learning_rate: 1.0e-4
+num_train_epochs: 3.0
+lr_scheduler_type: cosine
+warmup_ratio: 0.1
+bf16: true
+ddp_timeout: 180000000
+resume_from_checkpoint: null
+
+### ktransformers
+use_kt: true # use KTransformers as LoRA sft backend
+kt_optimize_rule: examples/kt_optimize_rules/DeepSeek-V3-Chat-sft-amx-multi-gpu.yaml
+cpu_infer: 32
+chunk_size: 8192
+
+### eval
+# eval_dataset: alpaca_en_demo
+# val_size: 0.1
+# per_device_eval_batch_size: 1
+# eval_strategy: steps
+# eval_steps: 500