Instructions to use phonemetransformers/childes-segmentation-100k-gpt2_lm-model with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use phonemetransformers/childes-segmentation-100k-gpt2_lm-model with Transformers:

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

pipe = pipeline("text-generation", model="phonemetransformers/childes-segmentation-100k-gpt2_lm-model")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("phonemetransformers/childes-segmentation-100k-gpt2_lm-model")
model = AutoModelForCausalLM.from_pretrained("phonemetransformers/childes-segmentation-100k-gpt2_lm-model")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use phonemetransformers/childes-segmentation-100k-gpt2_lm-model with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "phonemetransformers/childes-segmentation-100k-gpt2_lm-model"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "phonemetransformers/childes-segmentation-100k-gpt2_lm-model",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/phonemetransformers/childes-segmentation-100k-gpt2_lm-model

SGLang

How to use phonemetransformers/childes-segmentation-100k-gpt2_lm-model 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 "phonemetransformers/childes-segmentation-100k-gpt2_lm-model" \
    --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": "phonemetransformers/childes-segmentation-100k-gpt2_lm-model",
		"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 "phonemetransformers/childes-segmentation-100k-gpt2_lm-model" \
        --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": "phonemetransformers/childes-segmentation-100k-gpt2_lm-model",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use phonemetransformers/childes-segmentation-100k-gpt2_lm-model with Docker Model Runner:
```
docker model run hf.co/phonemetransformers/childes-segmentation-100k-gpt2_lm-model
```

childes-segmentation-100k-gpt2_lm-model

This model is a fine-tuned version of on an unknown dataset. It achieves the following results on the evaluation set:

epoch: 4000.0
eval_absolute_seg_boundary_fscore_Boundary Prediction: 0.6381
eval_absolute_seg_boundary_fscore_Entropy: 0.4936
eval_absolute_seg_boundary_fscore_Increase in Boundary Prediction: 0.6397
eval_absolute_seg_boundary_fscore_Increase in Entropy: 0.6171
eval_absolute_seg_boundary_fscore_Increase in Loss: 0.6068
eval_absolute_seg_boundary_fscore_Increase in Rank: 0.6806
eval_absolute_seg_boundary_fscore_Loss: 0.5355
eval_absolute_seg_boundary_fscore_Majority Vote Cutoff: 0.7011
eval_absolute_seg_boundary_fscore_Majority Vote Spike: 0.7273
eval_absolute_seg_boundary_fscore_Rank: 0.5571
eval_absolute_seg_type_fscore_Boundary Prediction: 0.1447
eval_absolute_seg_type_fscore_Entropy: 0.2626
eval_absolute_seg_type_fscore_Increase in Boundary Prediction: 0.3233
eval_absolute_seg_type_fscore_Increase in Entropy: 0.3100
eval_absolute_seg_type_fscore_Increase in Loss: 0.2509
eval_absolute_seg_type_fscore_Increase in Rank: 0.4174
eval_absolute_seg_type_fscore_Loss: 0.2412
eval_absolute_seg_type_fscore_Majority Vote Cutoff: 0.4319
eval_absolute_seg_type_fscore_Majority Vote Spike: 0.4164
eval_absolute_seg_type_fscore_Rank: 0.2972
eval_bpc: 4.4805
eval_loss: 3.1056
eval_model_preparation_time: 0.0008
eval_perplexity: 22.3231
eval_runtime: 47.9148
eval_samples_per_second: 2.964
eval_spike_seg_boundary_fscore_Boundary Prediction: 0.7308
eval_spike_seg_boundary_fscore_Entropy: 0.5944
eval_spike_seg_boundary_fscore_Increase in Boundary Prediction: 0.7195
eval_spike_seg_boundary_fscore_Increase in Entropy: 0.6282
eval_spike_seg_boundary_fscore_Increase in Loss: 0.6258
eval_spike_seg_boundary_fscore_Increase in Rank: 0.6735
eval_spike_seg_boundary_fscore_Loss: 0.5578
eval_spike_seg_boundary_fscore_Majority Vote Cutoff: 0.7329
eval_spike_seg_boundary_fscore_Majority Vote Spike: 0.7085
eval_spike_seg_boundary_fscore_Rank: 0.6106
eval_spike_seg_type_fscore_Boundary Prediction: 0.3885
eval_spike_seg_type_fscore_Entropy: 0.2817
eval_spike_seg_type_fscore_Increase in Boundary Prediction: 0.3610
eval_spike_seg_type_fscore_Increase in Entropy: 0.2866
eval_spike_seg_type_fscore_Increase in Loss: 0.3167
eval_spike_seg_type_fscore_Increase in Rank: 0.3724
eval_spike_seg_type_fscore_Loss: 0.2626
eval_spike_seg_type_fscore_Majority Vote Cutoff: 0.4120
eval_spike_seg_type_fscore_Majority Vote Spike: 0.3425
eval_spike_seg_type_fscore_Rank: 0.3228
eval_steps_per_second: 0.104
step: 100000

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: 0.001
train_batch_size: 32
eval_batch_size: 32
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_steps: 30000
training_steps: 100000

Framework versions

Transformers 4.44.2
Pytorch 2.4.0+cu118
Datasets 2.18.0
Tokenizers 0.19.1

Downloads last month: 41

Safetensors

Model size

634k params

Tensor type

F32