Instructions to use Mar2Ding/songcomposer_sft with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use Mar2Ding/songcomposer_sft with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="Mar2Ding/songcomposer_sft", trust_remote_code=True)# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("Mar2Ding/songcomposer_sft", trust_remote_code=True, dtype="auto") - Notebooks
- Google Colab
- Kaggle
- Local Apps
- vLLM
How to use Mar2Ding/songcomposer_sft with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "Mar2Ding/songcomposer_sft" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "Mar2Ding/songcomposer_sft", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/Mar2Ding/songcomposer_sft
- SGLang
How to use Mar2Ding/songcomposer_sft 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 "Mar2Ding/songcomposer_sft" \ --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": "Mar2Ding/songcomposer_sft", "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 "Mar2Ding/songcomposer_sft" \ --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": "Mar2Ding/songcomposer_sft", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use Mar2Ding/songcomposer_sft with Docker Model Runner:
docker model run hf.co/Mar2Ding/songcomposer_sft
Update modeling_internlm2.py
Browse files- modeling_internlm2.py +2 -2
modeling_internlm2.py
CHANGED
|
@@ -1163,9 +1163,9 @@ class InternLM2ForCausalLM(InternLM2PreTrainedModel):
|
|
| 1163 |
img_embeds, atts_img, img_target = self.img2emb(torch.zeros(1,3,self.im_size,self.im_size).to(image.device).to(image.dtype))
|
| 1164 |
to_regress_embeds += img_embeds.sum() * 0
|
| 1165 |
im_mask = torch.zeros(to_regress_embeds.shape[:2]).cuda()
|
| 1166 |
-
temp_max_length =
|
| 1167 |
|
| 1168 |
-
temp_max_length =
|
| 1169 |
inputs_embeds = to_regress_embeds[:, :temp_max_length]
|
| 1170 |
attention_mask = attention_mask[:, :temp_max_length]
|
| 1171 |
targets = targets[:, :temp_max_length]
|
|
|
|
| 1163 |
img_embeds, atts_img, img_target = self.img2emb(torch.zeros(1,3,self.im_size,self.im_size).to(image.device).to(image.dtype))
|
| 1164 |
to_regress_embeds += img_embeds.sum() * 0
|
| 1165 |
im_mask = torch.zeros(to_regress_embeds.shape[:2]).cuda()
|
| 1166 |
+
temp_max_length = self.max_length
|
| 1167 |
|
| 1168 |
+
temp_max_length = self.max_length
|
| 1169 |
inputs_embeds = to_regress_embeds[:, :temp_max_length]
|
| 1170 |
attention_mask = attention_mask[:, :temp_max_length]
|
| 1171 |
targets = targets[:, :temp_max_length]
|