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

Libraries

How to use AhmedSSoliman/MarianCausalLM with Transformers:

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

pipe = pipeline("text-generation", model="AhmedSSoliman/MarianCausalLM")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("AhmedSSoliman/MarianCausalLM")
model = AutoModelForCausalLM.from_pretrained("AhmedSSoliman/MarianCausalLM", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use AhmedSSoliman/MarianCausalLM with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/AhmedSSoliman/MarianCausalLM

SGLang

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

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

AhmedSSoliman commited on Jul 16, 2023

Commit

fc4a287

1 Parent(s): e63aa78

Upload tokenizer

Browse files

Files changed (5) hide show

source.spm +0 -0
special_tokens_map.json +5 -0
target.spm +0 -0
tokenizer_config.json +12 -0
vocab.json +0 -0

source.spm ADDED Viewed

Binary file (790 kB). View file

special_tokens_map.json ADDED Viewed

	@@ -0,0 +1,5 @@

+{
+  "eos_token": "</s>",
+  "pad_token": "<pad>",
+  "unk_token": "<unk>"
+}

target.spm ADDED Viewed

Binary file (814 kB). View file

tokenizer_config.json ADDED Viewed

	@@ -0,0 +1,12 @@

+{
+  "clean_up_tokenization_spaces": true,
+  "eos_token": "</s>",
+  "model_max_length": 512,
+  "pad_token": "<pad>",
+  "separate_vocabs": false,
+  "source_lang": "en",
+  "sp_model_kwargs": {},
+  "target_lang": "nl",
+  "tokenizer_class": "MarianTokenizer",
+  "unk_token": "<unk>"
+}

vocab.json ADDED Viewed

The diff for this file is too large to render. See raw diff