chore: import upstream snapshot with attribution

examples/finetune/reranker/decoder_only/base.sh

@@ -0,0 +1,70 @@
+export WANDB_MODE=disabled
+
+train_data="\
+    ../example_data/prompt_based/examples.jsonl "
+
+# set large epochs and small batch size for testing
+num_train_epochs=1
+per_device_train_batch_size=2
+gradient_accumulation_steps=1
+train_group_size=8
+
+# set num_gpus to 2 for testing
+num_gpus=2
+
+if [ -z "$HF_HUB_CACHE" ]; then
+    export HF_HUB_CACHE="$HOME/.cache/huggingface/hub"
+fi
+
+model_args="\
+    --model_name_or_path BAAI/bge-reranker-v2-gemma \
+    --cache_dir $HF_HUB_CACHE \
+    --use_lora True \
+    --lora_rank 32 \
+    --lora_alpha 64 \
+    --use_flash_attn True \
+    --target_modules q_proj k_proj v_proj o_proj \
+    --save_merged_lora_model True \
+    --model_type decoder \
+"
+
+data_args="\
+    --train_data $train_data \
+    --cache_path ~/.cache \
+    --train_group_size $train_group_size \
+    --query_max_len 512 \
+    --passage_max_len 512 \
+    --pad_to_multiple_of 8 \
+    --knowledge_distillation True \
+    --query_instruction_for_rerank 'A: ' \
+    --query_instruction_format '{}{}' \
+    --passage_instruction_for_rerank 'B: ' \
+    --passage_instruction_format '{}{}' \
+"
+
+training_args="\
+    --output_dir ./test_decoder_only_base_bge-reranker-v2-gemma \
+    --overwrite_output_dir \
+    --learning_rate 2e-4 \
+    --bf16 \
+    --num_train_epochs $num_train_epochs \
+    --per_device_train_batch_size $per_device_train_batch_size \
+    --gradient_accumulation_steps $gradient_accumulation_steps \
+    --dataloader_drop_last True \
+    --warmup_ratio 0.1 \
+    --gradient_checkpointing \
+    --weight_decay 0.01 \
+    --deepspeed ../../ds_stage0.json \
+    --logging_steps 1 \
+    --save_steps 1000 \
+"
+
+cmd="torchrun --nproc_per_node $num_gpus \
+    -m FlagEmbedding.finetune.reranker.decoder_only.base \
+    $model_args \
+    $data_args \
+    $training_args \
+"
+
+echo $cmd
+eval $cmd

examples/finetune/reranker/decoder_only/layerwise.sh

@@ -0,0 +1,76 @@
+export WANDB_MODE=disabled
+
+train_data="\
+    ../example_data/prompt_based/examples.jsonl "
+
+# set large epochs and small batch size for testing
+num_train_epochs=1
+per_device_train_batch_size=2
+gradient_accumulation_steps=1
+train_group_size=8
+
+# set num_gpus to 2 for testing
+num_gpus=2
+
+if [ -z "$HF_HUB_CACHE" ]; then
+    export HF_HUB_CACHE="$HOME/.cache/huggingface/hub"
+fi
+
+model_args="\
+    --model_name_or_path BAAI/bge-reranker-v2-minicpm-layerwise \
+    --cache_dir $HF_HUB_CACHE \
+    --use_lora True \
+    --lora_rank 32 \
+    --lora_alpha 64 \
+    --use_flash_attn True \
+    --target_modules q_proj k_proj v_proj o_proj \
+    --save_merged_lora_model True \
+    --model_type decoder \
+    --model_type from_finetuned_model \
+    --start_layer 8 \
+    --head_multi True \
+    --head_type simple \
+    --trust_remote_code True \
+"
+
+data_args="\
+    --train_data $train_data \
+    --cache_path ~/.cache \
+    --train_group_size $train_group_size \
+    --query_max_len 512 \
+    --passage_max_len 512 \
+    --pad_to_multiple_of 8 \
+    --knowledge_distillation True \
+    --query_instruction_for_rerank 'A: ' \
+    --query_instruction_format '{}{}' \
+    --passage_instruction_for_rerank 'B: ' \
+    --passage_instruction_format '{}{}' \
+"
+
+training_args="\
+    --output_dir ./test_decoder_only_base_bge-reranker-v2-minicpm-layerwise \
+    --overwrite_output_dir \
+    --learning_rate 2e-4 \
+    --bf16 \
+    --num_train_epochs $num_train_epochs \
+    --per_device_train_batch_size $per_device_train_batch_size \
+    --gradient_accumulation_steps $gradient_accumulation_steps \
+    --dataloader_drop_last True \
+    --warmup_ratio 0.1 \
+    --gradient_checkpointing \
+    --weight_decay 0.01 \
+    --deepspeed ../../ds_stage0.json \
+    --logging_steps 1 \
+    --save_steps 1000 \
+"
+
+cmd="torchrun --nproc_per_node $num_gpus \
+    -m FlagEmbedding.finetune.reranker.decoder_only.layerwise \
+    $model_args \
+    $data_args \
+    $training_args \
+"
+
+echo $cmd
+eval $cmd
+