Instructions to use Scottie201/custom_model_trained with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Scottie201/custom_model_trained with Transformers:

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

pipe = pipeline("text-generation", model="Scottie201/custom_model_trained")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Scottie201/custom_model_trained")
model = AutoModelForCausalLM.from_pretrained("Scottie201/custom_model_trained")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use Scottie201/custom_model_trained with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Scottie201/custom_model_trained"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Scottie201/custom_model_trained",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Scottie201/custom_model_trained

SGLang

How to use Scottie201/custom_model_trained 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 "Scottie201/custom_model_trained" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Scottie201/custom_model_trained",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

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 "Scottie201/custom_model_trained" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Scottie201/custom_model_trained",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use Scottie201/custom_model_trained with Docker Model Runner:
```
docker model run hf.co/Scottie201/custom_model_trained
```

You need to agree to share your contact information to access this model

This repository is publicly accessible, but you have to accept the conditions to access its files and content.

fine-tuned-custom-model

This model is a fine-tuned version of HuggingFaceTB/SmolLM2-1.7B-Instruct on the Scottie201 dataset.
It achieves the following results on the evaluation set:

Loss: 2.3012
Evaluation Runtime: 125.50 seconds
Evaluation Samples per Second: 0.096
Evaluation Steps per Second: 0.048

Model description

This is a fine-tuned version of the HuggingFaceTB/SmolLM2-1.7B-Instruct model, trained to handle custom tasks related to text generation. The model can handle a wide range of text completion, summarization, and question-answering tasks.

Intended uses & limitations

The model can be used for tasks like:

Text generation
Question answering
Summarization
General text-based tasks

Limitations:

May not work well for domain-specific tasks that were not part of the training data.
Performance on unseen data may vary depending on the domain of the query.

Training and evaluation data

The model was trained and evaluated on the Scottie201 dataset, which includes various text-based tasks. The evaluation was carried out after 2 epochs of training.

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

learning_rate: 1e-05
train_batch_size: 2
eval_batch_size: 2
seed: 42
optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 2

Training results

Training Loss	Epoch	Step	Validation Loss
2.4066	1.0	25	2.3138
2.3976	2.0	50	2.3012

Evaluation results

Evaluation Metric	Value
Loss	2.3012
Runtime	125.50 sec
Samples per Second	0.096
Steps per Second	0.048

Framework versions

Transformers: 4.48.3
PyTorch: 2.6.0+cu124
Datasets: 3.2.0
Tokenizers: 0.21.0

This model card now includes detailed information about the evaluation procedure, training results, and evaluation results such as loss, runtime, and throughput. The evaluation metric table provides clarity on how the model performed during evaluation.

Let me know if you'd like any more adjustments or additional details!

Downloads last month: -

Safetensors

Model size

2B params

Tensor type

BF16

Model tree for Scottie201/custom_model_trained

Base model

HuggingFaceTB/SmolLM2-1.7B

Quantized

HuggingFaceTB/SmolLM2-1.7B-Instruct

Finetuned

(138)

this model