Instructions to use dparra19/dparcon with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use dparra19/dparcon with PEFT:

from peft import PeftModel
from transformers import AutoModelForCausalLM

base_model = AutoModelForCausalLM.from_pretrained("gplsi/Aitana-2B-S-base")
model = PeftModel.from_pretrained(base_model, "dparra19/dparcon")

Transformers

How to use dparra19/dparcon with Transformers:

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

pipe = pipeline("text-generation", model="dparra19/dparcon")

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("dparra19/dparcon", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use dparra19/dparcon with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "dparra19/dparcon"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "dparra19/dparcon",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/dparra19/dparcon

SGLang

How to use dparra19/dparcon 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 "dparra19/dparcon" \
    --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": "dparra19/dparcon",
		"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 "dparra19/dparcon" \
        --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": "dparra19/dparcon",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use dparra19/dparcon with Docker Model Runner:
```
docker model run hf.co/dparra19/dparcon
```

dparcon

This model is a fine-tuned version of gplsi/Aitana-2B-S-base on the dataset showed in my account (descriptions about fruits and vegetables). It achieves the following results on the evaluation set:

Loss: 1.8984

Model description

The gplsi/Aitana-2B-S-base consists on a generative language model on multilingual data (Spanish, Valencian and English). In this case, we have fine-tuning this model to answer information about certain fruits and vegetables.

Intended uses & limitations

More information needed

Training and evaluation data

We have used the following dataset

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

learning_rate: 0.0002
train_batch_size: 4
eval_batch_size: 4
seed: 42
gradient_accumulation_steps: 4
total_train_batch_size: 16
optimizer: Use OptimizerNames.PAGED_ADAMW_8BIT with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: linear
lr_scheduler_warmup_steps: 2
num_epochs: 10
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
2.1405	1.0	3	2.1311
2.0658	2.0	6	2.0559
1.9534	3.0	9	2.0014
1.8923	4.0	12	1.9646
1.8391	5.0	15	1.9432
1.7374	6.0	18	1.9290
1.7135	7.0	21	1.9172
1.6700	8.0	24	1.9076
1.6643	9.0	27	1.9014
1.6647	10.0	30	1.8984

Framework versions

PEFT 0.19.1
Transformers 5.14.1
Pytorch 2.11.0+cu128
Datasets 5.0.0
Tokenizers 0.22.2

Downloads last month: 8

Model tree for dparra19/dparcon

Base model

BSC-LT/salamandra-2b

Quantized

gplsi/Aitana-2B-S-base

Adapter

(1)

this model

dparra19
/

dparcon