fix Python tutorial examples for accuracy

### Exercise 3: Create a Micromamba environment

1. Create an environment called `nlp-env` with Python 3.11, transformers, and datasets

2. Activate it and verify transformers is installed

1. Create an environment called `sci-env` with Python 3.11, numpy, and scipy

2. Activate it and verify scipy is installed

3. Export the environment to `environment.yml`

{{% alert title="Check your work" color="info" %}}

```shell-session

$ micromamba activate nlp-env

(nlp-env) $ python -c "import transformers; print(transformers.__version__)"

4.47.1

$ micromamba activate sci-env

(sci-env) $ python -c "import scipy; print(scipy.__version__)"

1.14.1

$ cat environment.yml | head -5

name: nlp-env

$ head -5 environment.yml

name: sci-env

channels:

  - conda-forge

dependencies:

  - python=3.11

  - python=3.11.*

```

{{% /alert %}}

PyTorch can't find CUDA or wrong version.

**Solution**: Match PyTorch CUDA version to loaded modules:

**Solution**: Match PyTorch CUDA version to the host driver. Check driver version:

```shell-session

$ module load cuda/12.9

$ uv add torch --index https://download.pytorch.org/whl/cu124

$ nvidia-smi | grep "Driver Version"

Driver Version: 550.54.15    CUDA Version: 12.4

```

Then install matching PyTorch:

```shell-session

$ uv add torch --index https://download.pytorch.org/whl/cu124  # for CUDA 12.4

```

### Slow package installation

$ git add pixi.lock pixi.toml     # For Pixi

```

### Exercise 5: Debug a broken environment

### Exercise 5: Restore from lockfile

1. Create a UV project

2. Add packages, then manually delete `.venv/`

3. Run `uv sync` to restore the environment

1. Create a UV project and add packages

2. Delete `.venv/` to simulate a fresh clone

3. Run `uv sync` to restore the exact environment

4. Verify packages work

{{% alert title="Check your work" color="info" %}}

```shell-session

$ rm -rf .venv/

$ uv run python -c "import torch"

error: No virtual environment found

$ ls

pyproject.toml  uv.lock  src/

$ uv sync

Resolved 15 packages in 12ms

Installed 15 packages in 0.8s

$ uv run python -c "import torch; print('Fixed!')"

Fixed!

$ uv run python -c "import torch; print('Restored!')"

Restored!

```

The lockfile (`uv.lock`) ensures the exact same versions are installed.

{{% /alert %}}

---