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

Libraries

How to use huggingtweets/marknorm with Transformers:

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

pipe = pipeline("text-generation", model="huggingtweets/marknorm")

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("huggingtweets/marknorm")
model = AutoModelForCausalLM.from_pretrained("huggingtweets/marknorm", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use huggingtweets/marknorm with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/huggingtweets/marknorm

SGLang

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

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

boris commited on Jan 24, 2021

Commit

aa3ff04

1 Parent(s): c217604

New model from https://wandb.ai/wandb/huggingtweets/runs/3dsmkmlr

Browse files

Files changed (4) hide show

README.md +6 -6
config.json +1 -0
pytorch_model.bin +2 -2
training_args.bin +2 -2

README.md CHANGED Viewed

@@ -51,7 +51,7 @@ The model was trained on [@marknorm's tweets](https://twitter.com/marknorm).
 <tbody style='border-width:0'>
 <tr style='border-width:0 0 1px 0; border-color: #E2E8F0'>
 <td style='border-width:0'>Tweets downloaded</td>
-<td style='border-width:0'>3239</td>
 </tr>
 <tr style='border-width:0 0 1px 0; border-color: #E2E8F0'>
 <td style='border-width:0'>Retweets</td>
@@ -59,24 +59,24 @@ The model was trained on [@marknorm's tweets](https://twitter.com/marknorm).
 </tr>
 <tr style='border-width:0 0 1px 0; border-color: #E2E8F0'>
 <td style='border-width:0'>Short tweets</td>
-<td style='border-width:0'>582</td>
 </tr>
 <tr style='border-width:0'>
 <td style='border-width:0'>Tweets kept</td>
-<td style='border-width:0'>2497</td>
 </tr>
 </tbody>
 </table>
-[Explore the data](https://wandb.ai/wandb/huggingtweets/runs/36it8gtp/artifacts), which is tracked with [W&B artifacts](https://docs.wandb.com/artifacts) at every step of the pipeline.
 ## Training procedure
 The model is based on a pre-trained [GPT-2](https://huggingface.co/gpt2) which is fine-tuned on @marknorm's tweets.
-Hyperparameters and metrics are recorded in the [W&B training run](https://wandb.ai/wandb/huggingtweets/runs/135qj47d) for full transparency and reproducibility.
-At the end of training, [the final model](https://wandb.ai/wandb/huggingtweets/runs/135qj47d/artifacts) is logged and versioned.
 ## Intended uses & limitations

 <tbody style='border-width:0'>
 <tr style='border-width:0 0 1px 0; border-color: #E2E8F0'>
 <td style='border-width:0'>Tweets downloaded</td>
+<td style='border-width:0'>3241</td>
 </tr>
 <tr style='border-width:0 0 1px 0; border-color: #E2E8F0'>
 <td style='border-width:0'>Retweets</td>
 </tr>
 <tr style='border-width:0 0 1px 0; border-color: #E2E8F0'>
 <td style='border-width:0'>Short tweets</td>
+<td style='border-width:0'>578</td>
 </tr>
 <tr style='border-width:0'>
 <td style='border-width:0'>Tweets kept</td>
+<td style='border-width:0'>2503</td>
 </tr>
 </tbody>
 </table>
+[Explore the data](https://wandb.ai/wandb/huggingtweets/runs/1e2qmw0f/artifacts), which is tracked with [W&B artifacts](https://docs.wandb.com/artifacts) at every step of the pipeline.
 ## Training procedure
 The model is based on a pre-trained [GPT-2](https://huggingface.co/gpt2) which is fine-tuned on @marknorm's tweets.
+Hyperparameters and metrics are recorded in the [W&B training run](https://wandb.ai/wandb/huggingtweets/runs/3dsmkmlr) for full transparency and reproducibility.
+At the end of training, [the final model](https://wandb.ai/wandb/huggingtweets/runs/3dsmkmlr/artifacts) is logged and versioned.
 ## Intended uses & limitations

config.json CHANGED Viewed

@@ -34,6 +34,7 @@
       "top_p": 0.95
     }
   },
   "use_cache": true,
   "vocab_size": 50257
 }

       "top_p": 0.95
     }
   },
+  "transformers_version": "4.2.2",
   "use_cache": true,
   "vocab_size": 50257
 }

pytorch_model.bin CHANGED Viewed

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fa9b22357a4e6502a085dad370a3e6eac19d928e65b4252f29f725d4c5279e31
-size 510406558

 version https://git-lfs.github.com/spec/v1
+oid sha256:c184ccf0f84e140691634ff258469c0d497252d073748458048b4278b10c36a1
+size 510406560

training_args.bin CHANGED Viewed

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:17b22856af244d96dce1c87e304bf8122e388ac857a56d7f41f24bb146544fdc
-size 1839

 version https://git-lfs.github.com/spec/v1
+oid sha256:416ffec643d580c00528245e0da82826d91119b36b042fc01230a1a330c986d7
+size 2031