disable valset by default (#6690)

Former-commit-id: a1a94f364e33d1d73852f74eda4fa581e6b16533

examples/train_lora/llama3_lora_dpo.yaml

@@ -6,6 +6,7 @@ trust_remote_code: true
 stage: dpo
 do_train: true
 finetuning_type: lora
+lora_rank: 8
 lora_target: all
 pref_beta: 0.1
 pref_loss: sigmoid  # choices: [sigmoid (dpo), orpo, simpo]
@@ -36,7 +37,7 @@ bf16: true
 ddp_timeout: 180000000
 
 ### eval
-val_size: 0.1
-per_device_eval_batch_size: 1
-eval_strategy: steps
-eval_steps: 500
+# val_size: 0.1
+# per_device_eval_batch_size: 1
+# eval_strategy: steps
+# eval_steps: 500

examples/train_lora/llama3_lora_kto.yaml

@@ -6,6 +6,7 @@ trust_remote_code: true
 stage: kto
 do_train: true
 finetuning_type: lora
+lora_rank: 8
 lora_target: all
 pref_beta: 0.1
 
@@ -35,7 +36,7 @@ bf16: true
 ddp_timeout: 180000000
 
 ### eval
-val_size: 0.1
-per_device_eval_batch_size: 1
-eval_strategy: steps
-eval_steps: 500
+# val_size: 0.1
+# per_device_eval_batch_size: 1
+# eval_strategy: steps
+# eval_steps: 500

examples/train_lora/llama3_lora_ppo.yaml

@@ -7,6 +7,7 @@ trust_remote_code: true
 stage: ppo
 do_train: true
 finetuning_type: lora
+lora_rank: 8
 lora_target: all
 
 ### dataset

examples/train_lora/llama3_lora_pretrain.yaml

@@ -6,6 +6,7 @@ trust_remote_code: true
 stage: pt
 do_train: true
 finetuning_type: lora
+lora_rank: 8
 lora_target: all
 
 ### dataset
@@ -33,7 +34,7 @@ bf16: true
 ddp_timeout: 180000000
 
 ### eval
-val_size: 0.1
-per_device_eval_batch_size: 1
-eval_strategy: steps
-eval_steps: 500
+# val_size: 0.1
+# per_device_eval_batch_size: 1
+# eval_strategy: steps
+# eval_steps: 500

examples/train_lora/llama3_lora_reward.yaml

@@ -6,6 +6,7 @@ trust_remote_code: true
 stage: rm
 do_train: true
 finetuning_type: lora
+lora_rank: 8
 lora_target: all
 
 ### dataset
@@ -34,7 +35,7 @@ bf16: true
 ddp_timeout: 180000000
 
 ### eval
-val_size: 0.1
-per_device_eval_batch_size: 1
-eval_strategy: steps
-eval_steps: 500
+# val_size: 0.1
+# per_device_eval_batch_size: 1
+# eval_strategy: steps
+# eval_steps: 500

examples/train_lora/llama3_lora_sft.yaml

@@ -6,6 +6,7 @@ trust_remote_code: true
 stage: sft
 do_train: true
 finetuning_type: lora
+lora_rank: 8
 lora_target: all
 
 ### dataset
@@ -34,7 +35,7 @@ bf16: true
 ddp_timeout: 180000000
 
 ### eval
-val_size: 0.1
-per_device_eval_batch_size: 1
-eval_strategy: steps
-eval_steps: 500
+# val_size: 0.1
+# per_device_eval_batch_size: 1
+# eval_strategy: steps
+# eval_steps: 500

examples/train_lora/llama3_lora_sft_ds3.yaml

@@ -6,6 +6,7 @@ trust_remote_code: true
 stage: sft
 do_train: true
 finetuning_type: lora
+lora_rank: 8
 lora_target: all
 deepspeed: examples/deepspeed/ds_z3_config.json  # choices: [ds_z0_config.json, ds_z2_config.json, ds_z3_config.json]
 
@@ -35,7 +36,7 @@ bf16: true
 ddp_timeout: 180000000
 
 ### eval
-val_size: 0.1
-per_device_eval_batch_size: 1
-eval_strategy: steps
-eval_steps: 500
+# val_size: 0.1
+# per_device_eval_batch_size: 1
+# eval_strategy: steps
+# eval_steps: 500

examples/train_lora/llama3_lora_sft_ray.yaml

@@ -6,6 +6,7 @@ trust_remote_code: true
 stage: sft
 do_train: true
 finetuning_type: lora
+lora_rank: 8
 lora_target: all
 
 ### dataset
@@ -24,6 +25,13 @@ save_steps: 500
 plot_loss: true
 overwrite_output_dir: true
 
+### ray
+ray_run_name: llama3_8b_sft_lora
+ray_num_workers: 4  # number of GPUs to use
+resources_per_worker:
+  GPU: 1
+placement_strategy: PACK
+
 ### train
 per_device_train_batch_size: 1
 gradient_accumulation_steps: 8
@@ -35,14 +43,7 @@ bf16: true
 ddp_timeout: 180000000
 
 ### eval
-val_size: 0.1
-per_device_eval_batch_size: 1
-eval_strategy: steps
-eval_steps: 500
-
-### ray
-ray_run_name: llama3_8b_sft_lora
-ray_num_workers: 4  # number of GPUs to use
-resources_per_worker:
-  GPU: 1
-placement_strategy: PACK
+# val_size: 0.1
+# per_device_eval_batch_size: 1
+# eval_strategy: steps
+# eval_steps: 500

examples/train_lora/llama3_preprocess.yaml

@@ -6,6 +6,7 @@ trust_remote_code: true
 stage: sft
 do_train: true
 finetuning_type: lora
+lora_rank: 8
 lora_target: all
 
 ### dataset

examples/train_lora/llava1_5_lora_sft.yaml

@@ -1,11 +1,14 @@
 ### model
 model_name_or_path: llava-hf/llava-1.5-7b-hf
+image_resolution: 262144
+video_resolution: 16384
 trust_remote_code: true
 
 ### method
 stage: sft
 do_train: true
 finetuning_type: lora
+lora_rank: 8
 lora_target: all
 
 ### dataset
@@ -34,7 +37,7 @@ bf16: true
 ddp_timeout: 180000000
 
 ### eval
-val_size: 0.1
-per_device_eval_batch_size: 1
-eval_strategy: steps
-eval_steps: 500
+# val_size: 0.1
+# per_device_eval_batch_size: 1
+# eval_strategy: steps
+# eval_steps: 500

examples/train_lora/qwen2vl_lora_dpo.yaml

@@ -1,11 +1,14 @@
 ### model
 model_name_or_path: Qwen/Qwen2-VL-7B-Instruct
+image_resolution: 262144
+video_resolution: 16384
 trust_remote_code: true
 
 ### method
 stage: dpo
 do_train: true
 finetuning_type: lora
+lora_rank: 8
 lora_target: all
 pref_beta: 0.1
 pref_loss: sigmoid  # choices: [sigmoid (dpo), orpo, simpo]
@@ -36,7 +39,7 @@ bf16: true
 ddp_timeout: 180000000
 
 ### eval
-val_size: 0.1
-per_device_eval_batch_size: 1
-eval_strategy: steps
-eval_steps: 500
+# val_size: 0.1
+# per_device_eval_batch_size: 1
+# eval_strategy: steps
+# eval_steps: 500

examples/train_lora/qwen2vl_lora_sft.yaml

@@ -1,11 +1,14 @@
 ### model
 model_name_or_path: Qwen/Qwen2-VL-7B-Instruct
+image_resolution: 262144
+video_resolution: 16384
 trust_remote_code: true
 
 ### method
 stage: sft
 do_train: true
 finetuning_type: lora
+lora_rank: 8
 lora_target: all
 
 ### dataset
@@ -34,7 +37,7 @@ bf16: true
 ddp_timeout: 180000000
 
 ### eval
-val_size: 0.1
-per_device_eval_batch_size: 1
-eval_strategy: steps
-eval_steps: 500
+# val_size: 0.1
+# per_device_eval_batch_size: 1
+# eval_strategy: steps
+# eval_steps: 500