How do I run your first inference pipeline?

# Quickstart with Pipeline API Open Python/Jupyter/Colab and run: ```python from transformers import pipeline # Sentiment Analysis classifier = pipeline("sentiment-analysis") result = classifier("Hugging Face is awesome!") print(result) # [{'label': 'POSITIVE', 'score': 0.9998}] # Text Generation generator = pipeline("text-generation", model="gpt2") print(generator("The future of AI is", max_length=30)) ``` Experiment with tasks: `fill-mask`, `summarization`, `question-answering`. See all at [huggingface.co/tasks](https://huggingface.co/tasks). 47 48

How do I explore and load models from the hub?

# Browse the Model Hub 1. Visit [huggingface.co/models](https://huggingface.co/models?library=transformers) – filter by task/language. 2. Pick a model e.g. `distilbert-base-uncased-finetuned-sst-2-english`. # Load Custom Model ```python from transformers import pipeline pipe = pipeline("sentiment-analysis", model="distilbert/distilbert-base-uncased-finetuned-sst-2-english") print(pipe("This is great!")) # Low-level: Tokenizer + Model from transformers import AutoTokenizer, AutoModelForSequenceClassification tokenizer = AutoTokenizer.from_pretrained("distilbert/distilbert-base-uncased-finetuned-sst-2-english") model = AutoModelForSequenceClassification.from_pretrained("distilbert/distilbert-base-uncased-finetuned-sst-2-english") ``` Login for gated models: `from huggingface_hub import login; login()`. 45 48

How do I fine-tune a model with trainer?

# Fine-Tune Example: Sentiment on Rotten Tomatoes 1. Load dataset/model: ```python from datasets import load_dataset from transformers import AutoTokenizer, AutoModelForSequenceClassification dataset = load_dataset("rotten_tomatoes") model = AutoModelForSequenceClassification.from_pretrained("distilbert/distilbert-base-uncased", num_labels=2) tokenizer = AutoTokenizer.from_pretrained("distilbert/distilbert-base-uncased") ``` 2. Tokenize: ```python def tokenize(ex): return tokenizer(ex["text"], truncation=True) dataset = dataset.map(tokenize, batched=True) ``` 3. Train: ```python from transformers import Trainer, TrainingArguments data_collator = DataCollatorWithPadding(tokenizer=tokenizer) training_args = TrainingArguments(output_dir="my-model", num_train_epochs=2, per_device_train_batch_size=8) trainer = Trainer(model=model, args=training_args, train_dataset=dataset["train"], eval_dataset=dataset["test"], tokenizer=tokenizer, data_collator=data_collator) trainer.train() ``` 48

How do I share model and build a demo space?

# Push to Hub ```python trainer.push_to_hub("my-sentiment-model") ``` # Create Space Demo 1. Go to [huggingface.co/new-space](https://huggingface.co/new-space). 2. Choose Gradio SDK, add code: ```python import gradio as gr from transformers import pipeline pipe = pipeline("sentiment-analysis", "yourusername/my-sentiment-model") def predict(text): return pipe(text) iface = gr.Interface(fn=predict, inputs="text", outputs="text") iface.launch() ``` 3. Commit and share public link. Explore Spaces at [huggingface.co/spaces](https://huggingface.co/spaces). 45

Master Hugging Face Transformers in 5 Steps

Q: How do I set up account and environment?

# Step 1: Create Hugging Face Account 1. Visit [huggingface.co/join](https://huggingface.co/join) and sign up for a free account. 2. Generate a User Access Token: Go to Settings > Access Tokens > New token (read access for now). # Install Python Environment 1. Ensure Python 3.10+ installed. 2. Create virtual env: ```bash python -m venv hf-env source hf-env/bin/activate # Linux/Mac # or hf-env\Scripts\activate on Windows ``` 3. Install Transformers: ```bash pip install "transformers[torch]" pip install datasets evaluate accelerate ``` Use Google Colab for no-install option: Open [colab.research.google.com](https://colab.research.google.com) and run `!pip install transformers[torch]`. 47 48

🧰

Instant Toolkit

2 artifacts

📋
Step-by-Step Guide

1

Step 1: Create Hugging Face Account

Visit huggingface.co/join and sign up for a free account.
Generate a User Access Token: Go to Settings > Access Tokens > New token (read access for now).

Install Python Environment

Ensure Python 3.10+ installed.

Create virtual env:

python -m venv hf-env
source hf-env/bin/activate  # Linux/Mac
# or hf-env\Scripts\activate on Windows

Install Transformers:

pip install "transformers[torch]"
pip install datasets evaluate accelerate

Use Google Colab for no-install option: Open colab.research.google.com and run !pip install transformers[torch].

Why this step matters:

-Establishes access to 1M+ models and datasets on the Hub
-Prepares a reproducible environment for ML experiments

30-60 minutes

Python 3.10+, pip, Google Colab or VS Code, Hugging Face Account

$0

Definition of Done

•HF account created and token generated
•transformers library installed and importable via `python -c 'import transformers'`

Common Mistakes to Avoid

❌ Skipping virtual environment

→Always use venv to avoid package conflicts

❌ Using CPU-only PyTorch on large models

→Install with [torch] extras for GPU support if available

2

Following along, or just reading? 👀

Spin up a personalized “learn Hugging Face” plan you can save, check off, and return to anytime — unlimited on the free trial.

Start free trial →

3

4

5

🎥
Suggested Videos

Hugging Face Tutorial for Beginners

Hugging Face Course Playlist

📚
Learning Resources

📖 Online Courses

NLP Course

beginner

Hugging Face

LLM Course

beginner

Hugging Face

Open Source Models with Hugging Face

intermediate

DeepLearning.AI

📖 Recommended Books

Natural Language Processing with Transformers

intermediate

by Lewis Tunstall, Leandro von Werra, Thomas Wolf

🌐 Helpful Websites

Transformers Documentation

documentation

Official docs for installation, quickstart, and advanced usage

Hugging Face Hub

documentation

Host models, datasets, and Spaces

Hugging Face Learn

courses

Free courses on NLP, LLMs, and more

🛠️ Useful Tools

Transformers Library

ML Library

Core library for models and pipelines

Datasets Library

Data

Load and process datasets easily

Google Colab

IDE

Free GPU notebooks

Master Hugging Face Transformers in 5 Steps