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

Libraries

How to use sedrickkeh/checkpoints with Transformers:

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

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

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("sedrickkeh/checkpoints")
model = AutoModelForCausalLM.from_pretrained("sedrickkeh/checkpoints")
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 sedrickkeh/checkpoints with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/sedrickkeh/checkpoints

SGLang

How to use sedrickkeh/checkpoints 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 "sedrickkeh/checkpoints" \
    --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": "sedrickkeh/checkpoints",
		"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 "sedrickkeh/checkpoints" \
        --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": "sedrickkeh/checkpoints",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

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

sedrickkeh commited on Oct 25, 2024

Commit

eb3dcb0

verified ·

1 Parent(s): 996c766

Model save

Browse files

Files changed (4) hide show

README.md +7 -7
model-00001-of-00003.safetensors.sagemaker-uploaded +0 -0
model-00002-of-00003.safetensors.sagemaker-uploaded +0 -0
model-00003-of-00003.safetensors.sagemaker-uploaded +0 -0

README.md CHANGED Viewed

@@ -17,7 +17,7 @@ should probably proofread and complete it, then remove this comment. -->
 This model is a fine-tuned version of [mistralai/Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 1.1202
 ## Model description
@@ -41,10 +41,10 @@ The following hyperparameters were used during training:
 - eval_batch_size: 8
 - seed: 42
 - distributed_type: multi-GPU
-- num_devices: 16
 - gradient_accumulation_steps: 4
-- total_train_batch_size: 1024
-- total_eval_batch_size: 128
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: cosine
 - lr_scheduler_warmup_ratio: 0.1
@@ -52,9 +52,9 @@ The following hyperparameters were used during training:
 ### Training results
-| Training Loss | Epoch  | Step | Validation Loss |
-|:-------------:|:------:|:----:|:---------------:|
-| 1.509         | 0.9565 | 11   | 1.1202          |
 ### Framework versions

 This model is a fine-tuned version of [mistralai/Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1) on an unknown dataset.
 It achieves the following results on the evaluation set:
+- Loss: 1.0612
 ## Model description
 - eval_batch_size: 8
 - seed: 42
 - distributed_type: multi-GPU
+- num_devices: 8
 - gradient_accumulation_steps: 4
+- total_train_batch_size: 512
+- total_eval_batch_size: 64
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: cosine
 - lr_scheduler_warmup_ratio: 0.1
 ### Training results
+| Training Loss | Epoch | Step | Validation Loss |
+|:-------------:|:-----:|:----:|:---------------:|
+| 1.1092        | 1.0   | 23   | 1.0612          |
 ### Framework versions

model-00001-of-00003.safetensors.sagemaker-uploaded ADDED Viewed

File without changes

model-00002-of-00003.safetensors.sagemaker-uploaded ADDED Viewed

File without changes

model-00003-of-00003.safetensors.sagemaker-uploaded ADDED Viewed

File without changes