Instructions to use houssamDev/WikitextGenModel with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use houssamDev/WikitextGenModel with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="houssamDev/WikitextGenModel")

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("houssamDev/WikitextGenModel")
model = AutoModelForCausalLM.from_pretrained("houssamDev/WikitextGenModel", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use houssamDev/WikitextGenModel with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "houssamDev/WikitextGenModel"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "houssamDev/WikitextGenModel",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/houssamDev/WikitextGenModel

SGLang

How to use houssamDev/WikitextGenModel with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "houssamDev/WikitextGenModel" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "houssamDev/WikitextGenModel",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "houssamDev/WikitextGenModel" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "houssamDev/WikitextGenModel",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use houssamDev/WikitextGenModel with Docker Model Runner:
```
docker model run hf.co/houssamDev/WikitextGenModel
```

houssamDev commited on Jun 17, 2025

Commit

7ec3d53

verified ·

1 Parent(s): b8dc257

Update README.md

Browse files

🧠 Text Summarization Model Evaluation
This project evaluates a sequence-to-sequence Transformer model on the Wikitext-103 dataset using ROUGE metrics. The model was trained to perform abstractive text summarization.

🏋️ Training Performance
Training Loss: 3.4396

This indicates the average model loss during training, showing reasonable convergence.

🧪 Validation Results
Metric Score
ROUGE-1 0.8325
ROUGE-2 0.7163
ROUGE-L 0.8326
ROUGE-Lsum 0.8326

The high ROUGE scores on the validation set demonstrate that the model captures both unigram and bigram overlap effectively, while maintaining structural similarity with the target summaries.

🧾 Test Results
Metric Score
ROUGE-1 0.7806
ROUGE-2 0.6820
ROUGE-L 0.7805
ROUGE-Lsum 0.7805

The model generalizes well to unseen data with a slight drop compared to validation performance, which is expected.

📌 Notes
Model: You can replace this with your specific model name (e.g., t5-base, bart-large, etc.)

Dataset: wikitext-103-raw-v1 from Hugging Face Datasets.

Evaluation Metric: ROUGE – commonly used in summarization tasks to measure the overlap between generated and reference texts.

Files changed (1) hide show

README.md +17 -3

README.md CHANGED Viewed

@@ -1,3 +1,17 @@
----
-license: apache-2.0
----

+---
+license: apache-2.0
+datasets:
+- Salesforce/wikitext
+language:
+- en
+metrics:
+- rouge
+base_model:
+- distilbert/distilgpt2
+pipeline_tag: text-generation
+library_name: transformers
+tags:
+- wikipedia
+- text-generation-inference
+- gbt
+---