# Trainer() issue: AttributeError:\` 'str' object \`has no attribute 'dtype'

**URL:** <https://discuss.huggingface.co/t/trainer-issue-attributeerror-str-object-has-no-attribute-dtype/42835>\
**Category:** 🤗Transformers\
**Created:** [June 11, 2023, 1:42pm UTC](https://discuss.huggingface.co/t/trainer-issue-attributeerror-str-object-has-no-attribute-dtype/42835 "2023-06-11T13:42:09Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![fkov](https://avatars.discourse-cdn.com/v4/letter/f/e274bd/32.png) [@fkov](https://discuss.huggingface.co/u/fkov)\
**Post date:** [June 11, 2023, 1:42pm UTC](https://discuss.huggingface.co/t/trainer-issue-attributeerror-str-object-has-no-attribute-dtype/42835/1 "2023-06-11T13:42:09Z")

</div>

### System Info

- `transformers` version: 4.30.0.dev0
- Platform: Linux-5.4.204-ql-generic-12.0-19-x86\_64-with-glibc2.31
- Python version: 3.11.3
- Huggingface\_hub version: 0.14.1
- Safetensors version: 0.3.1
- PyTorch version (GPU?): 2.0.1 (True)  
Versions of relevant libraries:  
[pip3] numpy==1.23.0  
[pip3] torch==2.0.1  
[pip3] torchaudio==2.0.2  
[pip3] torchvision==0.15.2  
[conda] blas 1.0 mkl  
[conda] ffmpeg 4.3 hf484d3e\_0 pytorch  
[conda] mkl 2023.1.0 h6d00ec8\_46342  
[conda] mkl-service 2.4.0 py311h5eee18b\_1  
[conda] mkl\_fft 1.3.6 py311ha02d727\_1  
[conda] mkl\_random 1.2.2 py311ha02d727\_1  
[conda] numpy 1.23.0 pypi\_0 pypi  
[conda] pytorch 2.0.1 py3.11\_cuda11.8\_cudnn8.7.0\_0 pytorch  
[conda] pytorch-cuda 11.8 h7e8668a\_5 pytorch  
[conda] pytorch-mutex 1.0 cuda pytorch  
[conda] torchaudio 2.0.2 py311\_cu118 pytorch  
[conda] torchtriton 2.0.0 py311 pytorch  
[conda] torchvision 0.15.2 py311\_cu118 pytorch

### Who can help?

@sgugger @sanchit-gandhi

### Information

- The official example scripts
- My own modified scripts

### Tasks

- An officially supported task in the `examples` folder (such as GLUE/SQuAD, …)
- My own task or dataset (give details below)

### Reproduction

## ERROR:

train\_result = `trainer.train`(resume\_from\_checkpoint=checkpoint)  
…  
python3.11/site-packages/transformers/`feature_extraction_sequence_utils.py`", line 220, in pad  
if value.dtype is np.dtype(np.float64):  
^^^^^^^^^^^  
AttributeError:`'str' object`has no attribute ‘dtype’

I am not sure which element of the dataset is read as ‘str’

### 1. OFFICIAL SCRIPT: [transformers/examples/pytorch/audio-classification/run\_audio\_classification.py](https://github.com/huggingface/transformers/blob/main/examples/pytorch/audio-classification/README.md)

### 2. LOADED DATASET:

DatasetDict({  
train: Dataset({  
features: [`'audio',` `'label'`],  
num\_rows: 1280  
})  
validation: Dataset({  
features: [‘audio’, ‘label’],  
num\_rows: 160  
})  
test: Dataset({  
features: [‘audio’, ‘label’],  
num\_rows: 160  
})

### 3. logger.info(raw\_datasets[‘train’][0])

{‘audio’: {`'path'`: ‘/transformers/examples/pytorch/audio-classification/s/data/s/s/train/audio1.wav’, `'array'`: array([0.02072144, 0.02767944, 0.03274536, …, 0.00079346, 0.00088501,  
0.00149536]), ‘sampling\_rate’: 16000}, `'label'`: ‘happy’}

@mariosasko any idea about this?

### Expected behavior

load the dataset to model for training in train\_result = `trainer.train`(resume\_from\_checkpoint=checkpoint)