> For the complete documentation index, see [llms.txt](https://docs.clore.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.clore.ai/guides/guides_v2-fr/entrainement/huggingface-transformers.md).

# HuggingFace Transformers

Utilisez la bibliothèque Transformers pour le NLP, la vision et l'audio sur GPU.

{% hint style="success" %}
Tous les exemples peuvent être exécutés sur des serveurs GPU loués via [CLORE.AI Marketplace](https://clore.ai/marketplace).
{% endhint %}

## Location sur CLORE.AI

1. Visitez [CLORE.AI Marketplace](https://clore.ai/marketplace)
2. Filtrer par type de GPU, VRAM et prix
3. Choisir **À la demande** (tarif fixe) ou **Spot** (prix d'enchère)
4. Configurez votre commande :
   * Sélectionnez l'image Docker
   * Définissez les ports (TCP pour SSH, HTTP pour les interfaces web)
   * Ajoutez des variables d'environnement si nécessaire
   * Entrez la commande de démarrage
5. Sélectionnez le paiement : **CLORE**, **BTC**, ou **USDT/USDC**
6. Créez la commande et attendez le déploiement

### Accédez à votre serveur

* Trouvez les détails de connexion dans **Mes commandes**
* Interfaces Web : utilisez l'URL du port HTTP
* SSH : `ssh -p <port> root@<adresse-proxy>`

## Qu'est-ce que Transformers ?

Hugging Face Transformers fournit :

* Plus de 100 000 modèles pré-entraînés
* Chargement et inférence de modèles faciles
* Prise en charge du fine-tuning
* Capacités multimodales

## Déploiement rapide

**Image Docker :**

```
pytorch/pytorch:2.5.1-cuda12.4-cudnn9-devel
```

**Ports :**

```
22/tcp
```

**Commande :**

```bash
pip install transformers accelerate datasets huggingface_hub
```

## Installation

```bash
pip install transformers[torch]
pip install accelerate  # Pour les grands modèles
pip install datasets    # Pour les données d'entraînement
```

## Génération de texte

### Génération basique

```python
from transformers import AutoModelForCausalLM, AutoTokenizer
import torch

model_name = "mistralai/Mistral-7B-Instruct-v0.2"
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForCausalLM.from_pretrained(
    model_name,
    torch_dtype=torch.float16,
    device_map="auto"
)

prompt = "Expliquer l'informatique quantique en termes simples :"
inputs = tokenizer(prompt, return_tensors="pt").to("cuda")

outputs = model.generate(
    **inputs,
    max_new_tokens=200,
    temperature=0.7,
    do_sample=True
)

response = tokenizer.decode(outputs[0], skip_special_tokens=True)
print(response)
```

### Modèles de chat

```python
from transformers import pipeline

pipe = pipeline(
    "text-generation",
    model="meta-llama/Llama-2-7b-chat-hf",
    torch_dtype=torch.float16,
    device_map="auto"
)

messages = [
    {"role": "user", "content": "Qu'est-ce que l'apprentissage automatique ?"}
]

outputs = pipe(
    messages,
    max_new_tokens=256,
    do_sample=True,
    temperature=0.7
)

print(outputs[0]["generated_text"][-1]["content"])
```

### Streaming

```python
from transformers import TextStreamer

streamer = TextStreamer(tokenizer)

model.generate(
    **inputs,
    max_new_tokens=200,
    streamer=streamer
)
```

## Quantification

### Quantification 4 bits

```python
from transformers import AutoModelForCausalLM, BitsAndBytesConfig
import torch

bnb_config = BitsAndBytesConfig(
    load_in_4bit=True,
    bnb_4bit_quant_type="nf4",
    bnb_4bit_compute_dtype=torch.float16,
    bnb_4bit_use_double_quant=True
)

model = AutoModelForCausalLM.from_pretrained(
    "meta-llama/Llama-2-13b-hf",
    quantization_config=bnb_config,
    device_map="auto"
)
```

### Quantification 8 bits

```python
model = AutoModelForCausalLM.from_pretrained(
    "meta-llama/Llama-2-7b-hf",
    load_in_8bit=True,
    device_map="auto"
)
```

## Embeddings

```python
from transformers import AutoModel, AutoTokenizer
import torch

model_name = "sentence-transformers/all-MiniLM-L6-v2"
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModel.from_pretrained(model_name).cuda()

def get_embedding(text):
    inputs = tokenizer(text, return_tensors="pt", padding=True, truncation=True).to("cuda")
    with torch.no_grad():
        outputs = model(**inputs)
    # Pooling moyen
    embedding = outputs.last_hidden_state.mean(dim=1)
    return embedding

emb = get_embedding("Hello, world!")
print(f"Embedding shape: {emb.shape}")
```

## Classification d'images

```python
from transformers import pipeline
from PIL import Image

classifier = pipeline("image-classification", model="google/vit-base-patch16-224", device=0)

image = Image.open("cat.jpg")
results = classifier(image)

for result in results:
    print(f"{result['label']}: {result['score']:.4f}")
```

## Détection d'objets

```python
from transformers import pipeline
from PIL import Image

detector = pipeline("object-detection", model="facebook/detr-resnet-50", device=0)

image = Image.open("street.jpg")
results = detector(image)

for result in results:
    print(f"{result['label']}: {result['score']:.4f} à {result['box']}")
```

## Segmentation d'images

```python
from transformers import pipeline
from PIL import Image

segmenter = pipeline("image-segmentation", model="facebook/maskformer-swin-base-ade", device=0)

image = Image.open("scene.jpg")
results = segmenter(image)

for segment in results:
    print(f"{segment['label']}: score {segment['score']:.4f}")
```

## Reconnaissance vocale

```python
from transformers import pipeline

transcriber = pipeline(
    "automatic-speech-recognition",
    model="openai/whisper-large-v3",
    device=0
)

result = transcriber("audio.mp3")
print(result["text"])
```

## Texte en parole

```python
from transformers import pipeline
import scipy

synthesizer = pipeline("text-to-speech", model="microsoft/speecht5_tts", device=0)

speech = synthesizer("Bonjour, ceci est un test de synthèse vocale.")

scipy.io.wavfile.write("output.wav", rate=speech["sampling_rate"], data=speech["audio"])
```

## Fine-tuning

### Préparer le jeu de données

```python
from datasets import load_dataset

dataset = load_dataset("imdb")
train_dataset = dataset["train"].select(range(1000))
eval_dataset = dataset["test"].select(range(200))
```

### Entraînement

```python
from transformers import (
    AutoModelForSequenceClassification,
    AutoTokenizer,
    TrainingArguments,
    Trainer
)

model_name = "bert-base-uncased"
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForSequenceClassification.from_pretrained(model_name, num_labels=2)

def tokenize_function(examples):
    return tokenizer(examples["text"], padding="max_length", truncation=True)

tokenized_train = train_dataset.map(tokenize_function, batched=True)
tokenized_eval = eval_dataset.map(tokenize_function, batched=True)

training_args = TrainingArguments(
    output_dir="./results",
    evaluation_strategy="epoch",
    learning_rate=2e-5,
    per_device_train_batch_size=16,
    per_device_eval_batch_size=16,
    num_train_epochs=3,
    weight_decay=0.01,
    fp16=True,
)

trainer = Trainer(
    model=model,
    args=training_args,
    train_dataset=tokenized_train,
    eval_dataset=tokenized_eval,
)

trainer.train()
```

## Tâches Pipeline

| Tâche                   | Nom du pipeline                |
| ----------------------- | ------------------------------ |
| Génération de texte     | `text-generation`              |
| Remplir le masque       | `fill-mask`                    |
| Résumé                  | `summarization`                |
| Traduction              | `translation`                  |
| Question-Réponse        | `question-answering`           |
| Analyse de sentiment    | `sentiment-analysis`           |
| Classification d'images | `image-classification`         |
| Détection d'objets      | `object-detection`             |
| Reconnaissance vocale   | `automatic-speech-recognition` |
| Texte en parole         | `text-to-speech`               |

## Multi-GPU

```python
from transformers import AutoModelForCausalLM

# Placement automatique des dispositifs
model = AutoModelForCausalLM.from_pretrained(
    "meta-llama/Llama-2-70b-hf",
    device_map="auto",
    torch_dtype=torch.float16
)

# Carte des dispositifs manuelle
device_map = {
    "model.embed_tokens": 0,
    "model.layers.0": 0,
    "model.layers.1": 0,
    # ...
    "model.layers.39": 1,
    "model.norm": 1,
    "lm_head": 1
}

model = AutoModelForCausalLM.from_pretrained(
    "model_name",
    device_map=device_map
)
```

## Optimisation de la mémoire

```python

# Flash Attention 2
model = AutoModelForCausalLM.from_pretrained(
    "meta-llama/Llama-2-7b-hf",
    torch_dtype=torch.float16,
    attn_implementation="flash_attention_2",
    device_map="auto"
)

# Checkpointing de gradient
model.gradient_checkpointing_enable()
```

## Model Hub

### Télécharger des modèles

```python
from huggingface_hub import snapshot_download

snapshot_download(
    repo_id="meta-llama/Llama-2-7b-hf",
    local_dir="./llama-2-7b"
)
```

### Téléverser des modèles

```python
from huggingface_hub import HfApi

api = HfApi()
api.upload_folder(
    folder_path="./my_model",
    repo_id="username/my-model",
    repo_type="model"
)
```

## Conseils de performance

| Astuce                               | Effet                    |
| ------------------------------------ | ------------------------ |
| Utilisez `torch_dtype=torch.float16` | 50 % moins de mémoire    |
| Activer Flash Attention 2            | Attention 2x plus rapide |
| Utiliser la quantification           | 75 % moins de mémoire    |
| Inférence par lot                    | Débit plus élevé         |

## Dépannage

## Estimation des coûts

Tarifs typiques du marché CLORE.AI (à partir de 2024) :

| GPU       | Tarif horaire | Tarif journalier | Session de 4 heures |
| --------- | ------------- | ---------------- | ------------------- |
| RTX 3060  | \~$0.03       | \~$0.70          | \~$0.12             |
| RTX 3090  | \~$0.06       | \~$1.50          | \~$0.25             |
| RTX 4090  | \~$0.10       | \~$2.30          | \~$0.40             |
| A100 40GB | \~$0.17       | \~$4.00          | \~$0.70             |
| A100 80GB | \~$0.25       | \~$6.00          | \~$1.00             |

*Les prix varient selon le fournisseur et la demande. Vérifiez* [*CLORE.AI Marketplace*](https://clore.ai/marketplace) *pour les tarifs actuels.*

**Économisez de l'argent :**

* Utilisez **Spot** market pour les charges de travail flexibles (souvent 30-50 % moins cher)
* Payer avec **CLORE** jetons
* Comparer les prix entre différents fournisseurs

## Prochaines étapes

* vLLM Inference - Mise en production
* [Ajuster des LLMs](/guides/guides_v2-fr/entrainement/finetune-llm.md) - Entraînement LoRA
* [Entraînement DeepSpeed](/guides/guides_v2-fr/entrainement/deepspeed-training.md) - Entraînement distribué


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://docs.clore.ai/guides/guides_v2-fr/entrainement/huggingface-transformers.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.