No module named numpy.strings in chapter 2

[marcusround]

https://colab.research.google.com/github/HandsOnLLM/Hands-On-Large-Language-Models/blob/main/chapter02/Chapter%202%20-%20Tokens%20and%20Token%20Embeddings.ipynb

Simply running the first two cells results in this error.

[MaartenGr]

Could you try installing the dependencies from the requirements.txt instead? There might be a dependency issue here.

[zbilgeozkan]

I encountered a similar issue and resolved it by redesigning the initial part of the code as follows:

%%capture
!pip install transformers>=4.40.1 accelerate>=0.27.2

from transformers import AutoModelForCausalLM, AutoTokenizer

# Load the model and tokenizer
model = AutoModelForCausalLM.from_pretrained(
    "microsoft/Phi-3-mini-4k-instruct",
    device_map="cuda",        # or "auto"
    torch_dtype="auto",
    trust_remote_code=False,  # set to True if you trust the model repo
)

tokenizer = AutoTokenizer.from_pretrained("microsoft/Phi-3-mini-4k-instruct")

The core issue was that newer versions of transformers depend on features introduced in numpy 2.x, which are not yet fully supported by other dependencies and can lead to errors like ModuleNotFoundError: No module named 'numpy.strings'.

Hopefully, this helps you too! :)

[arulk]

I encounter the same issue. I tried the fix about suggested and issue still persists.

I use google collab to open the chapter 2 notebook

[MaartenGr]

@arulk I just tried it out and it seems that the following should work:

!pip install --upgrade transformers==4.41.2 sentence-transformers>=3.0.1 gensim>=4.3.2 scikit-learn>=1.5.0 accelerate==0.31.0

You can also fix everything:

!pip install --upgrade transformers==4.41.2 sentence-transformers==3.0.1 gensim==4.3.2 scikit-learn==1.5.0 accelerate==0.31.0

The latter is more likely to prevent conflicting issues. Plus, if all else fails, the requirements.txt should work.

Let me know what works for you so I can update the notebook!

[xindi-dumbledore]

I have a similar but different issue with gensim on Google colab

ModuleNotFoundError                       Traceback (most recent call last)
[<ipython-input-9-7ff9ea44eb5b>](https://9p0hqhor15n-496ff2e9c6d22116-0-colab.googleusercontent.com/outputframe.html?vrz=colab_20250603-060055_RC00_766582280#) in <cell line: 0>()
----> 1 import gensim.downloader as api
      2 
      3 model = api.load("glove-wiki-gigaword-50")

9 frames
[/usr/local/lib/python3.11/dist-packages/numpy/__init__.py](https://9p0hqhor15n-496ff2e9c6d22116-0-colab.googleusercontent.com/outputframe.html?vrz=colab_20250603-060055_RC00_766582280#) in __getattr__(attr)
    365                 raise AssertionError()
    366         except AssertionError:
--> 367             msg = ("The current Numpy installation ({!r}) fails to "
    368                    "pass simple sanity checks. This can be caused for example "
    369                    "by incorrect BLAS library being linked in, or by mixing "

ModuleNotFoundError: No module named 'numpy.char'

I think it might be something related to numpy version?

[xindi-dumbledore]

ah tried this and it works

!pip install --upgrade transformers==4.41.2 sentence-transformers==3.0.1 gensim==4.3.2 scikit-learn==1.5.0 accelerate==0.31.0 scipy==1.10.1

[MaartenGr]

@xindi-dumbledore Thanks, it seems that the requirements need to be updated. The following is working for me in Chapter 2 and I'll make sure to fix that soon!

!pip install --upgrade transformers==4.41.2 sentence-transformers==3.0.1 gensim==4.3.2 scikit-learn==1.5.0 accelerate==0.31.0 peft==0.11.1 scipy==1.10.1 numpy==1.26.4

Also, the notebook needs to be restarted after this otherwise it doesn't work.

I'm also going to test out uv for a bit, but that will take a bit longer considering I might want to have chapter exclusive dependencies in the pyproject.toml and I have to test that for each chapter.