Instructions to use EshAhm/phi-2-LoRa-Python with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use EshAhm/phi-2-LoRa-Python with PEFT:

from peft import PeftModel
from transformers import AutoModelForCausalLM

base_model = AutoModelForCausalLM.from_pretrained("microsoft/phi-2")
model = PeftModel.from_pretrained(base_model, "EshAhm/phi-2-LoRa-Python")

Transformers

How to use EshAhm/phi-2-LoRa-Python with Transformers:

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

pipe = pipeline("text-generation", model="EshAhm/phi-2-LoRa-Python")

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("EshAhm/phi-2-LoRa-Python", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use EshAhm/phi-2-LoRa-Python with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "EshAhm/phi-2-LoRa-Python"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "EshAhm/phi-2-LoRa-Python",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/EshAhm/phi-2-LoRa-Python

SGLang

How to use EshAhm/phi-2-LoRa-Python 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 "EshAhm/phi-2-LoRa-Python" \
    --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": "EshAhm/phi-2-LoRa-Python",
		"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 "EshAhm/phi-2-LoRa-Python" \
        --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": "EshAhm/phi-2-LoRa-Python",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use EshAhm/phi-2-LoRa-Python with Docker Model Runner:
```
docker model run hf.co/EshAhm/phi-2-LoRa-Python
```

phi-2-LoRa-Python / README.md

EshAhm

EshAhm/LoRa-Python2

fffce84 verified about 1 year ago

preview code

Raw History Blame Contribute Delete

2.64 kB

---
library_name: peft
license: mit
base_model: microsoft/phi-2
tags:
- base_model:adapter:microsoft/phi-2
- lora
- transformers
pipeline_tag: text-generation
model-index:
- name: phi-2-LoRa-Python
  results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->

# phi-2-LoRa-Python

This model is a fine-tuned version of [microsoft/phi-2](https://huggingface.co/microsoft/phi-2) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 4.4864

## Model description

More information needed

## Intended uses & limitations

More information needed

## Training and evaluation data

More information needed

## Training procedure

### Training hyperparameters

The following hyperparameters were used during training:
- learning_rate: 0.0002
- train_batch_size: 2
- eval_batch_size: 2
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 4
- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED 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: 100
- num_epochs: 3
### Training results
| Training Loss | Epoch  | Step | Validation Loss |
|:-------------:|:------:|:----:|:---------------:|
| 4.5433        | 0.1333 | 100  | 4.4747          |
| 4.4162        | 0.2667 | 200  | 4.4774          |
| 4.2759        | 0.4    | 300  | 4.4755          |
| 4.4021        | 0.5333 | 400  | 4.4764          |
| 4.3925        | 0.6667 | 500  | 4.4775          |
| 4.6016        | 0.8    | 600  | 4.5976          |
| 4.2527        | 0.9333 | 700  | 4.5484          |
| 4.5823        | 1.0667 | 800  | 4.5037          |
| 4.5826        | 1.2    | 900  | 4.5032          |
| 4.0654        | 1.3333 | 1000 | 4.5073          |
| 4.5386        | 1.4667 | 1100 | 4.4961          |
| 4.1046        | 1.6    | 1200 | 4.4957          |
| 4.1705        | 1.7333 | 1300 | 4.4943          |
| 4.4922        | 1.8667 | 1400 | 4.4918          |
| 4.7333        | 2.0    | 1500 | 4.4901          |
| 3.997         | 2.1333 | 1600 | 4.4889          |
| 4.5407        | 2.2667 | 1700 | 4.4880          |
| 3.8457        | 2.4    | 1800 | 4.4877          |
| 4.6653        | 2.5333 | 1900 | 4.4873          |
| 4.4524        | 2.6667 | 2000 | 4.4865          |
| 4.3138        | 2.8    | 2100 | 4.4870          |
| 4.3203        | 2.9333 | 2200 | 4.4865          |
### Framework versions
- PEFT 0.17.1
- Transformers 4.56.1
- Pytorch 2.8.0+cu126
- Datasets 4.0.0
- Tokenizers 0.22.0

	---
	library_name: peft
	license: mit
	base_model: microsoft/phi-2
	tags:
	- base_model:adapter:microsoft/phi-2
	- lora
	- transformers
	pipeline_tag: text-generation
	model-index:
	- name: phi-2-LoRa-Python
	results: []
	---

	<!-- This model card has been generated automatically according to the information the Trainer had access to. You
	should probably proofread and complete it, then remove this comment. -->

	# phi-2-LoRa-Python

	This model is a fine-tuned version of [microsoft/phi-2](https://huggingface.co/microsoft/phi-2) on the None dataset.
	It achieves the following results on the evaluation set:
	- Loss: 4.4864

	## Model description

	More information needed

	## Intended uses & limitations

	More information needed

	## Training and evaluation data

	More information needed

	## Training procedure

	### Training hyperparameters

	The following hyperparameters were used during training:
	- learning_rate: 0.0002
	- train_batch_size: 2
	- eval_batch_size: 2
	- seed: 42
	- gradient_accumulation_steps: 2
	- total_train_batch_size: 4
	- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED 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: 100
	- num_epochs: 3

	### Training results

	\| Training Loss \| Epoch \| Step \| Validation Loss \|
	\|:-------------:\|:------:\|:----:\|:---------------:\|
	\| 4.5433 \| 0.1333 \| 100 \| 4.4747 \|
	\| 4.4162 \| 0.2667 \| 200 \| 4.4774 \|
	\| 4.2759 \| 0.4 \| 300 \| 4.4755 \|
	\| 4.4021 \| 0.5333 \| 400 \| 4.4764 \|
	\| 4.3925 \| 0.6667 \| 500 \| 4.4775 \|
	\| 4.6016 \| 0.8 \| 600 \| 4.5976 \|
	\| 4.2527 \| 0.9333 \| 700 \| 4.5484 \|
	\| 4.5823 \| 1.0667 \| 800 \| 4.5037 \|
	\| 4.5826 \| 1.2 \| 900 \| 4.5032 \|
	\| 4.0654 \| 1.3333 \| 1000 \| 4.5073 \|
	\| 4.5386 \| 1.4667 \| 1100 \| 4.4961 \|
	\| 4.1046 \| 1.6 \| 1200 \| 4.4957 \|
	\| 4.1705 \| 1.7333 \| 1300 \| 4.4943 \|
	\| 4.4922 \| 1.8667 \| 1400 \| 4.4918 \|
	\| 4.7333 \| 2.0 \| 1500 \| 4.4901 \|
	\| 3.997 \| 2.1333 \| 1600 \| 4.4889 \|
	\| 4.5407 \| 2.2667 \| 1700 \| 4.4880 \|
	\| 3.8457 \| 2.4 \| 1800 \| 4.4877 \|
	\| 4.6653 \| 2.5333 \| 1900 \| 4.4873 \|
	\| 4.4524 \| 2.6667 \| 2000 \| 4.4865 \|
	\| 4.3138 \| 2.8 \| 2100 \| 4.4870 \|
	\| 4.3203 \| 2.9333 \| 2200 \| 4.4865 \|


	### Framework versions

	- PEFT 0.17.1
	- Transformers 4.56.1
	- Pytorch 2.8.0+cu126
	- Datasets 4.0.0
	- Tokenizers 0.22.0