Instructions to use kyLELEng/market-gpt-return-token with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use kyLELEng/market-gpt-return-token with Transformers:

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

pipe = pipeline("text-generation", model="kyLELEng/market-gpt-return-token")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("kyLELEng/market-gpt-return-token")
model = AutoModelForCausalLM.from_pretrained("kyLELEng/market-gpt-return-token")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use kyLELEng/market-gpt-return-token with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/kyLELEng/market-gpt-return-token

SGLang

How to use kyLELEng/market-gpt-return-token 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 "kyLELEng/market-gpt-return-token" \
    --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": "kyLELEng/market-gpt-return-token",
		"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 "kyLELEng/market-gpt-return-token" \
        --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": "kyLELEng/market-gpt-return-token",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use kyLELEng/market-gpt-return-token with Docker Model Runner:
```
docker model run hf.co/kyLELEng/market-gpt-return-token
```

market-gpt-return-token / training_config.json

kyLELEng

Train MarketGPT return-token model

c011bf2 verified 17 days ago

raw

history blame contribute delete

1.28 kB

	{
	"model_repo_id": "kyLELEng/market-gpt-return-token-v2",
	"dataset_id": "siddharthmb/stocks-ohlcv",
	"dataset_file": "ohlcv.csv",
	"use_synthetic": false,
	"synthetic_steps": 900,
	"tickers": [
	"AAPL",
	"MSFT",
	"AMZN",
	"GOOGL",
	"NVDA",
	"TSLA",
	"AMD",
	"INTC",
	"JPM",
	"BAC",
	"V",
	"MA"
	],
	"num_assets": 12,
	"num_buckets": 101,
	"bucket_method": "quantile",
	"clip_value": 5.0,
	"vol_span": 60,
	"context_steps": 128,
	"prediction_steps": 32,
	"window_stride": 1,
	"train_fraction": 0.8,
	"validation_fraction": 0.1,
	"n_embd": 384,
	"n_layer": 6,
	"n_head": 6,
	"dropout": 0.1,
	"forecast_only_loss": true,
	"per_device_train_batch_size": 4,
	"per_device_eval_batch_size": 4,
	"gradient_accumulation_steps": 8,
	"learning_rate": 0.0003,
	"weight_decay": 0.1,
	"warmup_ratio": 0.05,
	"num_train_epochs": 5.0,
	"max_steps": -1,
	"logging_steps": 20,
	"eval_strategy": "epoch",
	"save_strategy": "no",
	"bf16": true,
	"fp16": false,
	"random_seed": 7,
	"num_eval_contexts": 12,
	"num_sample_paths": 24,
	"generation_temperature": 1.0,
	"generation_top_p": 0.95,
	"max_csv_chunks": null,
	"csv_chunksize": 500000,
	"min_rows_per_ticker": 1000,
	"output_dir": "market-gpt-return-token"
	}