Spaces:

JLouisBiz
/

GNU-LLM-Integration

Running

App Files Files Community

Jean Louis commited on Mar 23

Commit

01a937e

1 Parent(s): 0c3183e

Updated HTML files and CSS

Browse files

Files changed (1) hide show

01-prepare-python.html +74 -62

01-prepare-python.html CHANGED Viewed

@@ -11,105 +11,121 @@
 <body>
 <h1>Prepare Python environment to download Hugging Face models</h1>
-<p>This guide will help you install the necessary <a href="https://huggingface.co">Hugging
-Face</a> packages and tools to download and use
-models from the Hugging Face Hub.</p>
 <p>Follow this guide, or <a href="index.html">return to the main page</a> if needed.</p>
 <hr />
-<h2>1. Install <code>transformers</code> Package</h2>
-<p>The <code>transformers</code> package is used to load and use Hugging Face models in Python.</p>
-<h3>Installation</h3>
-<p>Run the following command in your terminal or command prompt:</p>
 <p><code>bash
-pip install transformers
 </code></p>
-<h3>Verify Installation</h3>
-<p>To confirm the installation was successful, run:</p>
 <p><code>bash
-python -c "from transformers import pipeline; print('Installation successful!')"
 </code></p>
 <hr />
-<h2>2. Install <code>huggingface_hub</code> Package</h2>
-<p>The <code>huggingface_hub</code> package provides the <code>huggingface-cli</code> tool for interacting with the Hugging Face Hub (e.g., downloading models, uploading files, etc.).</p>
-<h3>Installation</h3>
-<p>Run the following command:</p>
 <p><code>bash
-pip install huggingface_hub
 </code></p>
 <h3>Verify Installation</h3>
-<p>After installation, check if the <code>huggingface-cli</code> is available:</p>
 <p><code>bash
-huggingface-cli --help
 </code></p>
 <hr />
-<h2>3. Using <code>huggingface-cli</code></h2>
-<p>The <code>huggingface-cli</code> tool allows you to interact with the Hugging Face Hub directly from the command line.</p>
-<h3>Common Commands</h3>
-<h3>Log in to Hugging Face</h3>
-<p>To log in to your Hugging Face account:</p>
-<p><code>
-huggingface-cli login
 </code></p>
-<h3>Download a Model</h3>
-<p>To download a model (e.g., <code>gpt2</code>):</p>
-<p><code>
-huggingface-cli download gpt2
-</code></p>
-<h3>List Available Commands</h3>
-<p>To see all available commands and options:</p>
-<p><code>
-huggingface-cli --help
-</code></p>
-<hr />
-<h2>4. Example: Download and Use a Model</h2>
-<p>Here’s an example of downloading and using a model in Python:</p>
-<pre>
-from transformers import pipeline
-# Download and load a model
-generator = pipeline("text-generation", model="gpt2")
-# Generate text
-output = generator("Hello, how are you?", max_length=50)
-print(output)
-</pre>
 <hr />
@@ -124,24 +140,20 @@ print(output)
 </thead>
 <tbody>
 <tr>
-<td> <code>pip install transformers</code>       </td>
-<td> Install the <code>transformers</code> package.             </td>
-</tr>
-<tr>
-<td> <code>pip install huggingface_hub</code>    </td>
-<td> Install the <code>huggingface_hub</code> package.          </td>
 </tr>
 <tr>
-<td> <code>huggingface-cli --help</code>         </td>
-<td> List all available <code>huggingface-cli</code> commands.  </td>
 </tr>
 <tr>
-<td> <code>huggingface-cli login</code>          </td>
-<td> Log in to your Hugging Face account.            </td>
 </tr>
 <tr>
-<td> <code>huggingface-cli download gpt2</code>  </td>
-<td> Download the <code>gpt2</code> model.                      </td>
 </tr>
 </tbody>
 </table>
@@ -149,7 +161,7 @@ print(output)
 <hr />
-<p>Now you’re ready to use Hugging Face models and tools in Python! 🚀</p>
 <h1>Proceed to next step</h1>

 <body>
 <h1>Prepare Python environment to download Hugging Face models</h1>
+<p>This guide will walk you through setting up a Python environment named
+<strong><code>empower</code></strong>, installing the necessary Hugging Face packages, and
+downloading and using the <strong><code>microsoft/Phi-4-mini-instruct</code></strong> model on
+a GNU/Linux system.</p>
 <p>Follow this guide, or <a href="index.html">return to the main page</a> if needed.</p>
 <hr />
+<h2>1. Set Up Python Environment</h2>
+<h3>Check Python Installation</h3>
+<p>Ensure Python 3 is installed by running:</p>
 <p><code>bash
+python3 --version
 </code></p>
+<p>If Python is not installed, install it using your package manager. For example:</p>
+<ul>
+<li><p>On Debian/Ubuntu:</p>
+<p><code>bash
+sudo apt update
+sudo apt install python3 python3-venv
+</code></p></li>
+<li><p>On Fedora:</p>
+<p><code>bash
+sudo dnf install python3
+</code></p></li>
+</ul>
+<h3>Create a Virtual Environment</h3>
+<p>Create a virtual environment named <strong><code>empower</code></strong>:</p>
 <p><code>bash
+python3 -m venv empower
+</code></p>
+<p>Activate the virtual environment:
+<code>bash
+source empower/bin/activate
 </code></p>
 <hr />
+<h2>2. Install Hugging Face Packages</h2>
+<p>Install the necessary Hugging Face packages to interact with models and the Hugging Face Hub.</p>
+<h3>Install <code>transformers</code> and <code>huggingface_hub</code></h3>
+<p>Run the following command to install both packages:</p>
 <p><code>bash
+pip install transformers huggingface_hub
 </code></p>
 <h3>Verify Installation</h3>
+<p>Check if the packages are installed correctly:</p>
 <p><code>bash
+python3 -c "from transformers import pipeline; print('Transformers installed successfully!')"
+python3 -c "from huggingface_hub import HfApi; print('Hugging Face Hub installed successfully!')"
 </code></p>
 <hr />
+<h2>3. Download the <code>microsoft/Phi-4-mini-instruct</code> Model</h2>
+<p>To download and use the <strong><code>microsoft/Phi-4-mini-instruct</code></strong> model, follow these steps.</p>
+<h3>Using <code>huggingface-cli</code> to Download the Model</h3>
+<p>Run the following command to download the model:</p>
+<p><code>bash
+huggingface-cli download microsoft/Phi-4-mini-instruct
 </code></p>
+<p>This will download the model files to your current directory.</p>
+<hr />
+<h2>4. Load and Use the Model in Python</h2>
+<p>Once the model is downloaded, you can load and use it in your Python code.</p>
+<h3>Example Code</h3>
+<p>```python
+from transformers import AutoModelForCausalLM, AutoTokenizer</p>
+<h1>Load the model and tokenizer</h1>
+<p>model_name = &ldquo;microsoft/Phi-4-mini-instruct&rdquo;
+tokenizer = AutoTokenizer.from_pretrained(model_name)
+model = AutoModelForCausalLM.from_pretrained(model_name)</p>
+<h1>Generate text</h1>
+<p>input_text = &ldquo;What is the capital of France?&rdquo;
+inputs = tokenizer(input_text, return_tensors=&ldquo;pt&rdquo;)
+outputs = model.generate(**inputs, max_length=50)</p>
+<h1>Decode and print the output</h1>
+<p>print(tokenizer.decode(outputs[0], skip_special_tokens=True))
+```</p>
 <hr />
 </thead>
 <tbody>
 <tr>
+<td> <code>python3 -m venv empower</code>        </td>
+<td> Create a virtual environment named <code>empower</code>.   </td>
 </tr>
 <tr>
+<td> <code>source empower/bin/activate</code>    </td>
+<td> Activate the <code>empower</code> environment.             </td>
 </tr>
 <tr>
+<td> <code>pip install transformers huggingface_hub</code> </td>
+<td> Install Hugging Face packages.          </td>
 </tr>
 <tr>
+<td> <code>huggingface-cli download microsoft/Phi-4-mini-instruct</code> </td>
+<td> Download the model.          </td>
 </tr>
 </tbody>
 </table>
 <hr />
+<p>Now you’re ready to use the <strong><code>microsoft/Phi-4-mini-instruct</code></strong> model in your Python projects on GNU/Linux! 🚀</p>
 <h1>Proceed to next step</h1>