Text Generation
Transformers
PyTorch
English
MAELM
feature-extraction
audio2text
music2text
musicllm
music foundation model
custom_code
Instructions to use UniMus/OpenJMLA with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use UniMus/OpenJMLA with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="UniMus/OpenJMLA", trust_remote_code=True)# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("UniMus/OpenJMLA", trust_remote_code=True, dtype="auto") - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use UniMus/OpenJMLA with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "UniMus/OpenJMLA" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "UniMus/OpenJMLA", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/UniMus/OpenJMLA
- SGLang
How to use UniMus/OpenJMLA 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 "UniMus/OpenJMLA" \ --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": "UniMus/OpenJMLA", "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 "UniMus/OpenJMLA" \ --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": "UniMus/OpenJMLA", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use UniMus/OpenJMLA with Docker Model Runner:
docker model run hf.co/UniMus/OpenJMLA
sino commited on
Commit ·
6c624ee
1
Parent(s): b4d1dc8
Update configuration_maelm.py
Browse files- configuration_maelm.py +2 -2
configuration_maelm.py
CHANGED
|
@@ -111,13 +111,13 @@ class MAELMConfig(PretrainedConfig):
|
|
| 111 |
per_device_train_batch_size=12,
|
| 112 |
learning_rate=0.00005,
|
| 113 |
lm_lr_ratio=0.1,
|
| 114 |
-
tokenizer_name='Llama-2-7b-hf',
|
| 115 |
resume_from_checkpoint=None,
|
| 116 |
resume_from_pth='epoch_4-step_8639-allstep_60000.pth',
|
| 117 |
backbone={'name': 'MAEViT', 'arch': 'b', 'patch_size': 16, 'mask_ratio': 0.0, 'img_size': [80, 2992], \
|
| 118 |
'ckpt': 'epoch_20.pth'},
|
| 119 |
neck={'name': 'LMDecoder', 'patch_size': 16, 'img_size': [80, 2992], 'in_chans': 3, 'embed_dim': 768, \
|
| 120 |
-
'decoder_embed_dim': 4544, 'freeze_decoder': True, 'decoder_type': 'Llama-2-7b-hf'},
|
| 121 |
wandb={'proj': 'ATRena_cap', 'expname': 'cap_lynx_apmPT_mccaigc1wFT'},
|
| 122 |
**kwargs,
|
| 123 |
):
|
|
|
|
| 111 |
per_device_train_batch_size=12,
|
| 112 |
learning_rate=0.00005,
|
| 113 |
lm_lr_ratio=0.1,
|
| 114 |
+
tokenizer_name='meta-llama/Llama-2-7b-hf',
|
| 115 |
resume_from_checkpoint=None,
|
| 116 |
resume_from_pth='epoch_4-step_8639-allstep_60000.pth',
|
| 117 |
backbone={'name': 'MAEViT', 'arch': 'b', 'patch_size': 16, 'mask_ratio': 0.0, 'img_size': [80, 2992], \
|
| 118 |
'ckpt': 'epoch_20.pth'},
|
| 119 |
neck={'name': 'LMDecoder', 'patch_size': 16, 'img_size': [80, 2992], 'in_chans': 3, 'embed_dim': 768, \
|
| 120 |
+
'decoder_embed_dim': 4544, 'freeze_decoder': True, 'decoder_type': 'meta-llama/Llama-2-7b-hf'},
|
| 121 |
wandb={'proj': 'ATRena_cap', 'expname': 'cap_lynx_apmPT_mccaigc1wFT'},
|
| 122 |
**kwargs,
|
| 123 |
):
|