# Auto Train Error

**URL:** <https://discuss.huggingface.co/t/auto-train-error/82839>\
**Category:** 🤗AutoTrain\
**Created:** [April 21, 2024, 12:30am UTC](https://discuss.huggingface.co/t/auto-train-error/82839 "2024-04-21T00:30:40Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![Adilmar](https://sea2.discourse-cdn.com/hellohellohello/user_avatar/discuss.huggingface.co/adilmar/32/52970_2.png) [@Adilmar](https://discuss.huggingface.co/u/Adilmar)\
**Post date:** [April 21, 2024, 12:30am UTC](https://discuss.huggingface.co/t/auto-train-error/82839/1 "2024-04-21T00:30:40Z")

</div>

My training was interrupted today, and now I can’t start a new one, it always gives me the error, since nothing was changed at the base, would it be possible to restart where the training left off?

```
 data_path = dset.prepare()

```

File “/app/env/lib/python3.10/site-packages/autotrain/dataset.py”, line 275, in prepare  
return preprocessor.prepare()  
File “/app/env/lib/python3.10/site-packages/autotrain/preprocessor/text.py”, line 83, in prepare  
train\_df[:, “autotrain\_label”] = train\_df[“autotrain\_label”].astype(str)  
File “/app/env/lib/python3.10/site-packages/pandas/core/frame.py”, line 4311, in **setitem**  
self.\_set\_item(key, value)  
File “/app/env/lib/python3.10/site-packages/pandas/core/frame.py”, line 4527, in \_set\_item  
key in self.columns  
File “/app/env/lib/python3.10/site-packages/pandas/core/indexes/base.py”, line 5358, in **contains**  
hash(key)  
TypeError: nailable type: ‘slice’

---

<div class="post-metadata">

**Author:** ![baptle](https://avatars.discourse-cdn.com/v4/letter/b/4af34b/32.png) [@baptle](https://discuss.huggingface.co/u/baptle)\
**Post date:** [April 22, 2024, 8:09am UTC](https://discuss.huggingface.co/t/auto-train-error/82839/2 "2024-04-22T08:09:10Z")

</div>

Hi,

I have kind of the same error with this message:

> File “/app/env/lib/python3.10/site-packages/autotrain/preprocessor/text.py”, line 83, in prepare  
> train\_df[:, “autotrain\_label”] = train\_df[“autotrain\_label”].astype(str)  
> File “/app/env/lib/python3.10/site-packages/pandas/core/frame.py”, line 4311, in **setitem**  
> self.\_set\_item(key, value)  
> File “/app/env/lib/python3.10/site-packages/pandas/core/frame.py”, line 4527, in \_set\_item  
> key in self.columns  
> File “/app/env/lib/python3.10/site-packages/pandas/core/indexes/base.py”, line 5358, in **contains**  
> hash(key)  
> TypeError: unhashable type: ‘slice’

Have you find a solution?

---

<div class="post-metadata">

**Author:** ![Adilmar](https://sea2.discourse-cdn.com/hellohellohello/user_avatar/discuss.huggingface.co/adilmar/32/52970_2.png) [@Adilmar](https://discuss.huggingface.co/u/Adilmar)\
**Post date:** [April 22, 2024, 2:49pm UTC](https://discuss.huggingface.co/t/auto-train-error/82839/3 "2024-04-22T14:49:21Z")

</div>

Not yet, I’m still having problems.