chore: import upstream snapshot with attribution

examples/train/seq_cls/multi_label/infer.py

@@ -0,0 +1,35 @@
+import os
+from typing import List
+
+from swift import BaseArguments, InferRequest, TransformersEngine, get_template
+
+os.environ['IMAGE_MAX_TOKEN_NUM'] = '1024'
+os.environ['VIDEO_MAX_TOKEN_NUM'] = '128'
+os.environ['FPS_MAX_FRAMES'] = '16'
+
+infer_request = InferRequest(
+    messages=[{
+        'role':
+        'user',
+        'content':
+        "多标签分类，类别包括：['aeroplane', 'bicycle', 'bird', 'boat', 'bottle', "
+        "'bus', 'car', 'cat', 'chair', 'cow', 'diningtable', 'dog', 'horse', "
+        "'motorbike', 'person', 'pottedplant', 'sheep', 'sofa', 'train', 'tvmonitor']"
+    }],
+    images=['xxx.jpg'])
+adapter_path = 'output/vx-xxx/checkpoint-xxx'
+args = BaseArguments.from_pretrained(adapter_path)
+
+engine = TransformersEngine(
+    args.model,
+    adapters=[adapter_path],
+    task_type='seq_cls',
+    num_labels=args.num_labels,
+    problem_type=args.problem_type)
+template = get_template(
+    engine.processor, args.system, template_type=args.template, use_chat_template=args.use_chat_template)
+engine.template = template
+
+resp_list = engine.infer([infer_request])
+response: List[int] = resp_list[0].choices[0].message.content
+print(f'response: {response}')

examples/train/seq_cls/multi_label/infer.sh

@@ -0,0 +1,7 @@
+CUDA_VISIBLE_DEVICES=0 \
+IMAGE_MAX_TOKEN_NUM=1024 \
+VIDEO_MAX_TOKEN_NUM=128 \
+FPS_MAX_FRAMES=16 \
+swift infer \
+    --adapters output/vx-xxx/checkpoint-xxx \
+    --load_data_args true

examples/train/seq_cls/multi_label/sft.sh

@@ -0,0 +1,30 @@
+# Custom dataset format reference: https://swift.readthedocs.io/en/latest/Customization/Custom-dataset.html
+CUDA_VISIBLE_DEVICES=0 \
+swift sft \
+    --model Qwen/Qwen2.5-0.5B \
+    --tuner_type lora \
+    --dataset '<your-dataset>' \
+    --load_from_cache_file true \
+    --split_dataset_ratio 0.01 \
+    --torch_dtype bfloat16 \
+    --num_train_epochs 1 \
+    --per_device_train_batch_size 16 \
+    --per_device_eval_batch_size 16 \
+    --learning_rate 1e-4 \
+    --lora_rank 8 \
+    --lora_alpha 32 \
+    --target_modules all-linear \
+    --gradient_accumulation_steps 1 \
+    --eval_steps 100 \
+    --save_steps 100 \
+    --save_total_limit 2 \
+    --logging_steps 5 \
+    --max_length 2048 \
+    --output_dir output \
+    --warmup_ratio 0.05 \
+    --dataloader_num_workers 4 \
+    --dataset_num_proc 4 \
+    --num_labels '<num-labels>' \
+    --task_type seq_cls \
+    --use_chat_template false \
+    --problem_type multi_label_classification

examples/train/seq_cls/multi_label/vlm.sh

@@ -0,0 +1,31 @@
+CUDA_VISIBLE_DEVICES=0 \
+IMAGE_MAX_TOKEN_NUM=1024 \
+VIDEO_MAX_TOKEN_NUM=128 \
+FPS_MAX_FRAMES=16 \
+swift sft \
+    --model Qwen/Qwen3-VL-4B-Instruct \
+    --tuner_type lora \
+    --dataset 'clip-benchmark/wds_voc2007_multilabel' \
+    --load_from_cache_file true \
+    --split_dataset_ratio 0.01 \
+    --torch_dtype bfloat16 \
+    --num_train_epochs 2 \
+    --per_device_train_batch_size 16 \
+    --per_device_eval_batch_size 16 \
+    --learning_rate 1e-4 \
+    --lora_rank 8 \
+    --lora_alpha 32 \
+    --target_modules all-linear \
+    --gradient_accumulation_steps 1 \
+    --eval_steps 100 \
+    --save_steps 100 \
+    --save_total_limit 2 \
+    --logging_steps 5 \
+    --max_length 2048 \
+    --output_dir output \
+    --warmup_ratio 0.05 \
+    --dataloader_num_workers 4 \
+    --dataset_num_proc 4 \
+    --num_labels 20 \
+    --task_type seq_cls \
+    --problem_type multi_label_classification