Update README.md

Update instructions for version 1.2.

Author: BBC-Esq

README.md

@@ -19,31 +19,33 @@
   <table>
     <thead>
       <tr>
-        <th>GPU Brand</th>
-        <th>Windows Support</th>
-        <th>Linux Support</th>
+        <th>GPU</th>
+        <th>Windows</th>
+        <th>Linux</th>
+        <th>Requirements</th>
       </tr>
     </thead>
     <tbody>
       <tr>
         <td>Nvidia</td>
-        <td>✅ (requires CUDA 11.7 or 11.8)</td>
-        <td>✅ (requires CUDA 11.7 or 11.8)</td>
+        <td>✅</td>
+        <td>✅</td>
+        <td>CUDA 11.8 or 11.7</td>
       </tr>
       <tr>
         <td>AMD</td>
         <td>❌</td>
-        <td>✅ (requires ROCm 5.4.2)</td>
+        <td>✅</td>
+        <td>ROCm 5.4.2</td>
       </tr>
       <tr>
         <td>Apple/Metal</td>
-        <td colspan="2" align="center"> ✅ (requires "Xcode Command Line Tools")</td>
+        <td colspan="3" align="center"> ✅ ("Xcode Command Line Tools")</td>
       </tr>
     </tbody>
   </table>
 </div>
 
-
 <!-- Table of Contents -->
 
 <div align="center">
@@ -52,77 +54,75 @@
 
 <div align="center">
   <a href="#installation">Installation</a> | 
-  <a href="#usage-guide">Usage</a> | 
+  <a href="#usage">Usage</a> | 
   <a href="#contact">Contact Me</a>
 </div>
 
 ## Installation
 
-* **Step 1**: Install the appropriate software if you intend to use GPU-acceleration:
-
+* **Step 1** ➜ Install the appropriate software if you intend to use GPU-acceleration:
   * **For NVIDIA GPUs** ➜ install [CUDA 11.8](https://developer.nvidia.com/cuda-11-8-0-download-archive) or [CUDA 11.7](https://developer.nvidia.com/cuda-11-7-0-download-archive) for your specific operating system.
   * **For AMD GPUs** ➜ install ROCm version 5.4.2; instructions are [HERE](https://rocmdocs.amd.com/en/latest/deploy/linux/quick_start.html) and [HERE](https://rocmdocs.amd.com/en/latest/deploy/linux/index.html).
     * Unfortunately, gpu-accleration will [only work on Linux systems](https://github.com/RadeonOpenCompute/ROCm/blob/develop/docs/rocm.md).  
   * **For Apple/Metal/MPS** ➜ install [Xcode Command Line Tools](https://www.makeuseof.com/install-xcode-command-line-tools/).
-* **Step 2**: Download or clone this repository to a directory on your computer.
-* **Step 3**: Open a command prompt from within the directory and create a virtual environment:
+* **Step 2** ➜ Download or clone this repository to a directory on your computer.
+* **Step 3** ➜ Open a command prompt from within the directory and create a virtual environment:
 ```
 python -m venv .
 ```
-* **Step 4**: Activate the virtual environment:
+* **Step 4** ➜ Activate the virtual environment:
 ```
 .\Scripts\activate
 ```
-* **Step 5**: Update [PIP](https://pip.pypa.io/en/stable/index.html):
+* **Step 5** ➜ Update [PIP](https://pip.pypa.io/en/stable/index.html):
 ```
 python -m pip install --upgrade pip
 ```
-* **Step 6**: Install PyTorch with the appropriate "build:"
-
-  * **Windows/CUDA 11.8:** ```pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu118```
-  * **Windows/CUDA 11.7:** ```pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu117```
-  * **Windows/CPU-only:** ```pip install torch torchvision torchaudio```
-  * **Linux/CUDA 11.8:** ```pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu118```
-  * **Linux/CUDA 11.7:** ```pip install torch torchvision torchaudio```
-  * **Linux/ROCm 5.4.2:** ```install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/rocm5.4.2```
-  * **Linux/CPU-only:** ```pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cpu```
-  * **Apple/Metal/MPS:** ```pip pip install torch torchvision torchaudio```
+* **Step 6** ➜ Install PyTorch with the appropriate "build:"
+
+  * **Windows/CUDA 11.8** ➜ ```pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu118```
+  * **Windows/CUDA 11.7** ➜ ```pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu117```
+  * **Windows/CPU-only** ➜ ```pip install torch torchvision torchaudio```
+  * **Linux/CUDA 11.8** ➜ ```pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu118```
+  * **Linux/CUDA 11.7** ➜ ```pip install torch torchvision torchaudio```
+  * **Linux/ROCm 5.4.2** ➜ ```install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/rocm5.4.2```
+  * **Linux/CPU-only** ➜ ```pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cpu```
+  * **Apple/Metal/MPS** ➜ ```pip pip install torch torchvision torchaudio```
     * **Metal/MPS** [speedup comparison](https://explosion.ai/blog/metal-performance-shaders) to a 5950x and RTX 3090.
 
-* **Step 7:** Doublecheck that you installed gpu-acceleration properly:
+* **Step 7** ➜ Doublecheck that you installed gpu-acceleration properly:
 ```
 python check_gpu.py
 ```
 
-* **Step 8**: Install the dependencies listed in [requirements.txt](https://github.com/MicrosoftDocs/visualstudio-docs/blob/main/docs/python/managing-required-packages-with-requirements-txt.md):
+* **Step 8** ➜ Install the dependencies listed in [requirements.txt](https://github.com/MicrosoftDocs/visualstudio-docs/blob/main/docs/python/managing-required-packages-with-requirements-txt.md):
 ```
 pip install -r requirements.txt
 ```
-* **Step 9**:
-  * Lastly, you must install the appropriate version of Git (https://git-scm.com/downloads).
+* **Step 9** ➜ Lastly, you must install the appropriate version of Git (https://git-scm.com/downloads).
 
 [Back to top](#top)
 
-## Usage Guide
+## Usage
 
 * **Step 1**: Open a command prompt in the directory of my scripts, activate the virtual environment, and run:
 ```
 python gui.py
 ```
-* **Step 2**: Click "Download Embedding Model" and download a model. The GUI will hang. Wait.  Proceed to the next step.
-* **Step 3**: Click "Select Embedding Model Directory" and select the DIRECTORY containing the model you want to use.
-* **Step 4**: Click "Choose Documents for Database" and choose one or more PDF files to put in the database.
+* **Step 2** ➜ Click "Download Embedding Model" and download a model. The GUI will hang. Wait.  Proceed to the next step.
+* **Step 3** ➜ Click "Select Embedding Model Directory" and select the DIRECTORY containing the model you want to use.
+* **Step 4** ➜ Click "Choose Documents for Database" and choose one or more PDF files to put in the database.
   * Must have had OCR done on them or have text in them already. If you can select the text when viewing it, you're good.
-* **Step 5**: Click "Create Vector Database." The GUI will hang. Watch "CUDA" usage. When CUDA drops to zero, proceed to the next step.
-* **Step 6**: Open up LM Studio and load a model (only Llama2-based models currently work).
-* **Step 7**: Click "Start Server" within LM Studio, enter your question, and click "Submit Question."
+* **Step 5** ➜ Click "Create Vector Database." The GUI will hang. Watch "CUDA" usage. When CUDA drops to zero, proceed to the next step.
+* **Step 6** ➜ Open up LM Studio and load a model (only Llama2-based models currently work).
+* **Step 7** ➜ Click "Start Server" within LM Studio, enter your question, and click "Submit Question."
 
 
 [Back to top](#top)
 
 ## Contact
 
-All suggestions (positive and negative) are welcome.  I can be reached at "bbc@chintellalaw.com" or feel free to message me on the [LM Studio Discord Server](https://discord.gg/aPQfnNkxGC).
+All suggestions (positive and negative) are welcome.  "bbc@chintellalaw.com" or feel free to message me on the [LM Studio Discord Server](https://discord.gg/aPQfnNkxGC).
 
 [Back to top](#top)