From 8e44363ec9255fbbe8c26e2eb4a514f788008c93 Mon Sep 17 00:00:00 2001
From: Abhinav <60320192+blacksmithop@users.noreply.github.com>
Date: Mon, 29 Jan 2024 22:11:29 +0530
Subject: [PATCH] langchain_community: Update documentation for installing
 llama-cpp-python on windows (#16666)

**Description** : This PR updates the documentation for installing
llama-cpp-python on Windows.

- Updates install command to support pyproject.toml
- Makes CPU/GPU install instructions clearer
- Adds reinstall with GPU support command

**Issue**: Existing
[documentation](https://python.langchain.com/docs/integrations/llms/llamacpp#compiling-and-installing)
lists the following commands for installing llama-cpp-python
```
python setup.py clean
python setup.py install
````
The current version of the repo does not include a `setup.py` and uses a
`pyproject.toml` instead.
This can be replaced with
```
python -m pip install -e .
```
As explained in
https://github.com/abetlen/llama-cpp-python/issues/965#issuecomment-1837268339
**Dependencies**: None
**Twitter handle**: None

---------

Co-authored-by: blacksmithop <angstycoder101@gmaii.com>
---
 docs/docs/integrations/llms/llamacpp.ipynb | 26 +++++++++++++++++-----
 1 file changed, 21 insertions(+), 5 deletions(-)

diff --git a/docs/docs/integrations/llms/llamacpp.ipynb b/docs/docs/integrations/llms/llamacpp.ipynb
index 58bb7f38d8..a3a22acb7b 100644
--- a/docs/docs/integrations/llms/llamacpp.ipynb
+++ b/docs/docs/integrations/llms/llamacpp.ipynb
@@ -144,24 +144,40 @@
     "git clone --recursive -j8 https://github.com/abetlen/llama-cpp-python.git\n",
     "```\n",
     "\n",
-    "2. Open up command Prompt (or anaconda prompt if you have it installed), set up environment variables to install. Follow this if you do not have a GPU, you must set both of the following variables.\n",
+    "2. Open up a command Prompt and set the following environment variables.\n",
+    "\n",
     "\n",
     "```\n",
     "set FORCE_CMAKE=1\n",
     "set CMAKE_ARGS=-DLLAMA_CUBLAS=OFF\n",
     "```\n",
-    "You can ignore the second environment variable if you have an NVIDIA GPU.\n",
+    "If you have an NVIDIA GPU make sure `DLLAMA_CUBLAS` is set to `ON`\n",
     "\n",
     "#### Compiling and installing\n",
     "\n",
-    "In the same command prompt (anaconda prompt) you set the variables, you can `cd` into `llama-cpp-python` directory and run the following commands.\n",
+    "Now you can `cd` into the `llama-cpp-python` directory and install the package\n",
     "\n",
     "```\n",
-    "python setup.py clean\n",
-    "python setup.py install\n",
+    "python -m pip install -e .\n",
     "```"
    ]
   },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "**IMPORTANT**: If you have already installed a cpu only version of the package, you need to reinstall it from scratch: consider the following command: "
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "!python -m pip install -e . --force-reinstall --no-cache-dir"
+   ]
+  },
   {
    "cell_type": "markdown",
    "metadata": {},