report: "ValueError: Batch does not contain any data (None)." when streaming=True

[holyYodu]

System Info

- `Accelerate` version: 0.21.0
- Platform: Linux-3.10.0-1160.119.1.el7.x86_64-x86_64-with-glibc2.17
- Python version: 3.11.0
- Numpy version: 1.26.4
- PyTorch version (GPU?): 2.5.1+cu124 (True)
- PyTorch XPU available: False
- PyTorch NPU available: False
- System RAM: 62.92 GB
- GPU type: NVIDIA A10
- `Accelerate` default config:
        Not found

Information

• The official example scripts
• My own modified scripts

Tasks

• One of the scripts in the examples/ folder of Accelerate or an officially supported no_trainer script in the examples folder of the transformers repo (such as run_no_trainer_glue.py)
• My own task or dataset (give details below)

Reproduction

from datasets import load_dataset
from transformers import AutoTokenizer

from uniem.finetuner import FineTuner
from uniem.model import create_uniem_embedder

dataset = load_dataset('shibing624/nli-zh-all', streaming=True)
dataset = dataset.rename_columns({'text1': 'sentence1', 'text2': 'sentence2'})

# 由于是从头训练，我们需要自己初始化 embedder 和 tokenizer。当然，我们也可以选择新的 pooling 策略。
embedder = create_uniem_embedder('uer/chinese_roberta_L-2_H-128', pooling_strategy='cls')
tokenizer = AutoTokenizer.from_pretrained('uer/chinese_roberta_L-2_H-128')

finetuner = FineTuner(embedder, tokenizer=tokenizer, dataset=dataset)
fintuned_model = finetuner.run(epochs=3, batch_size=32, lr=1e-3)

Expected behavior