Text Generation
Transformers
Safetensors
Spanish
English
gemma
finance
conversational
text-generation-inference
Instructions to use NickyNicky/Financial_graph with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use NickyNicky/Financial_graph with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="NickyNicky/Financial_graph") messages = [ {"role": "user", "content": "Who are you?"}, ] pipe(messages)# Load model directly from transformers import AutoTokenizer, AutoModelForCausalLM tokenizer = AutoTokenizer.from_pretrained("NickyNicky/Financial_graph") model = AutoModelForCausalLM.from_pretrained("NickyNicky/Financial_graph", device_map="auto") messages = [ {"role": "user", "content": "Who are you?"}, ] inputs = tokenizer.apply_chat_template( messages, add_generation_prompt=True, tokenize=True, return_dict=True, return_tensors="pt", ).to(model.device) outputs = model.generate(**inputs, max_new_tokens=40) print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:])) - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use NickyNicky/Financial_graph with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "NickyNicky/Financial_graph" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "NickyNicky/Financial_graph", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'Use Docker
docker model run hf.co/NickyNicky/Financial_graph
- SGLang
How to use NickyNicky/Financial_graph 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 "NickyNicky/Financial_graph" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "NickyNicky/Financial_graph", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'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 "NickyNicky/Financial_graph" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "NickyNicky/Financial_graph", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }' - Docker Model Runner
How to use NickyNicky/Financial_graph with Docker Model Runner:
docker model run hf.co/NickyNicky/Financial_graph
| language: | |
| - es | |
| - en | |
| library_name: transformers | |
| pipeline_tag: text-generation | |
| tags: | |
| - finance | |
|  | |
| ## Descripcion. | |
| Este modelo de IA es competente en la comprensión y análisis de noticias financieras y de la bolsa de valores. Puede interpretar y procesar información relacionada con los mercados financieros, incluyendo tendencias bursátiles, informes económicos y análisis de inversiones. Además, tiene la capacidad de comprender y comunicarse de manera efectiva tanto en español como en inglés, lo que le permite interactuar con usuarios y fuentes de información en ambos idiomas de manera fluida. Asimismo, el modelo puede trabajar con grafos en formato JSON, permitiendo la visualización y análisis de estructuras de datos. | |
| ## Code. | |
| ```py | |
| import torch,gc | |
| import transformers | |
| from transformers import AutoModelForCausalLM, AutoTokenizer, TrainingArguments, BitsAndBytesConfig | |
| from datasets import load_dataset | |
| from peft import LoraConfig, PeftModel, get_peft_model, prepare_model_for_kbit_training | |
| from transformers import ( | |
| AutoModelForCausalLM, | |
| AutoTokenizer, | |
| BitsAndBytesConfig, | |
| HfArgumentParser, | |
| TrainingArguments, | |
| pipeline, | |
| logging, | |
| GenerationConfig, | |
| TextIteratorStreamer, | |
| ) | |
| from transformers import StoppingCriteria, StoppingCriteriaList | |
| import os | |
| model_id="NickyNicky/Financial_graph" | |
| max_seq_length=4048 | |
| tokenizer = AutoTokenizer.from_pretrained(model_id, | |
| token=access_token, | |
| max_length=max_seq_length) | |
| model = AutoModelForCausalLM.from_pretrained(model_id, | |
| # quantization_config=bnb_config, | |
| device_map={"":0}, | |
| token=access_token, | |
| attn_implementation="flash_attention_2", | |
| torch_dtype=torch.bfloat16 | |
| ).eval() | |
| # dpo_trainer.model.config.use_cache = True # silence the warnings. Please re-enable for inference! | |
| # tokenizer.padding_side='left' | |
| class ListOfTokensStoppingCriteria(StoppingCriteria): | |
| """ | |
| Clase para definir un criterio de parada basado en una lista de tokens específicos. | |
| """ | |
| def __init__(self, tokenizer, stop_tokens): | |
| self.tokenizer = tokenizer | |
| # Codifica cada token de parada y guarda sus IDs en una lista | |
| self.stop_token_ids_list = [tokenizer.encode(stop_token, add_special_tokens=False) for stop_token in stop_tokens] | |
| def __call__(self, input_ids, scores, **kwargs): | |
| # Verifica si los últimos tokens generados coinciden con alguno de los conjuntos de tokens de parada | |
| for stop_token_ids in self.stop_token_ids_list: | |
| len_stop_tokens = len(stop_token_ids) | |
| if len(input_ids[0]) >= len_stop_tokens: | |
| if input_ids[0, -len_stop_tokens:].tolist() == stop_token_ids: | |
| return True | |
| return False | |
| # Uso del criterio de parada personalizado | |
| stop_tokens = ["<end_of_turn>"] # Lista de tokens de parada | |
| # Inicializa tu criterio de parada con el tokenizer y la lista de tokens de parada | |
| stopping_criteria = ListOfTokensStoppingCriteria(tokenizer, stop_tokens) | |
| # Añade tu criterio de parada a una StoppingCriteriaList | |
| stopping_criteria_list = StoppingCriteriaList([stopping_criteria]) | |
| prompt= """<bos><start_of_turn>system | |
| Eres un agente experto en finanzas multilenguaje.<end_of_turn> | |
| <start_of_turn>user | |
| **news:** | |
| In a sudden decision, investment firm 21Shares submitted a filing to the Securities and Exchange Commission in the late afternoon of Friday, May 31, to rename its proposed spot Ether | |
| ETH | |
| tickers down | |
| $3,686 | |
| exchange-traded fund (ETF) and to remove ARK Invest from the application. | |
| A spokesperson for ARK Invest later confirmed the firm decided not to move forward with the crypto product, citing a need to reassess its investment strategy. The move raised concerns among the crypto community concerning the short-term viability of the newly approved ETFs. | |
| During an interview on June 5, United States Securities and Exchange Commission (SEC) Chair Gary Gensler suggested a delay in final approvals for asset managers. ETFs will “take some time,” said Gensler. The SEC has yet to sign off on S-1 registration statements from applicants. | |
| ARK Invest and 21Shares will remain partners on their spot Bitcoin | |
| BTC | |
| tickers down | |
| $69,346 | |
| ETF, launched in January. | |
| This week’s Crypto Biz also explores Galaxy Digital’s tokenized loan for Animoca Brands, Avail’s fundraising, the Toposware acquisition, and Bitcoin miners’ first reports since the halving. | |
| Galaxy Digital uses historic violin NFT to secure loan | |
| Michael Novogratz’s Galaxy Digital and Animoca Brands co-founder Yat Siu have tokenized a Stradivarius violin from 1708 to use as collateral for a multimillion-dollar loan. On June 4, Galaxy reportedly lent an undisclosed amount of funds to Siu, who used the 316-year-old violin he owned as collateral. The digital assets firm turned the violin into a nonfungible token (NFT) and will hold the NFT and the physical version until Siu settles the loan. The violin once belonged to the Russian Empress Catherine the Great. | |
| genera un grafo de razonamiento que contenga nodes: 10 y edges: 20<end_of_turn> | |
| <start_of_turn>model | |
| """ | |
| input= tokenizer(prompt, | |
| return_tensors="pt", | |
| add_special_tokens=False).to(model.device) | |
| max_new_tokens=1500 | |
| generation_config = GenerationConfig( | |
| max_new_tokens = max_new_tokens, | |
| temperature = .3, | |
| # top_p=0.55, | |
| # top_k = 50, | |
| # repetition_penalty = 1.1, | |
| do_sample=True, | |
| ) | |
| outputs = model.generate(**input, | |
| generation_config=generation_config, | |
| stopping_criteria=stopping_criteria_list, | |
| ) | |
| print(tokenizer.decode(outputs[0], skip_special_tokens=False) ) | |
| ##output: | |
| # {'nodes': [{'entity': 'ARK Invest', 'type': 'Organization'}, {'entity': '21Shares', 'type': 'Organization'}, {'entity': 'SEC', 'type': 'Organization'}, {'entity': 'Gary Gensler', 'type': 'Person'}, {'entity': 'United States Securities and Exchange Commission', 'type': 'Organization'}, {'entity': 'Galaxy Digital', 'type': 'Organization'}, {'entity': 'Animoca Brands', 'type': 'Organization'}, {'entity': 'Toposware', 'type': 'Organization'}, {'entity': 'Bitcoin miners', 'type': 'Organization'}, {'entity': 'S-1 registration statements', 'type': 'Document'}], 'edges': [{'from': 'ARK Invest', 'to': '21Shares', 'relation': 'ASSOCIATED_WITH'}, {'from': 'ARK Invest', 'to': 'SEC', 'relation': 'ASSOCIATED_WITH'}, {'from': 'ARK Invest', 'to': 'Gary Gensler', 'relation': 'ASSOCIATED_WITH'}, {'from': '21Shares', 'to': 'SEC', 'relation': 'ASSOCIATED_WITH'}, {'from': '21Shares', 'to': 'Gary Gensler', 'relation': 'ASSOCIATED_WITH'}, {'from': 'SEC', 'to': 'Gary Gensler', 'relation': 'WORKED_AT'}, {'from': 'SEC', 'to': 'United States Securities and Exchange Commission', 'relation': 'PART_OF'}, {'from': 'Galaxy Digital', 'to': 'Animoca Brands', 'relation': 'ASSOCIATED_WITH'}, {'from': 'Galaxy Digital', 'to': 'Toposware', 'relation': 'ASSOCIATED_WITH'}, {'from': 'Galaxy Digital', 'to': 'Bitcoin miners', 'relation': 'ASSOCIATED_WITH'}, {'from': 'Galaxy Digital', 'to': 'S-1 registration statements', 'relation': 'ASSOCIATED_WITH'}, {'from': 'Animoca Brands', 'to': 'Toposware', 'relation': 'ASSOCIATED_WITH'}, {'from': 'Toposware', 'to': 'Bitcoin miners', 'relation': 'ASSOCIATED_WITH'}, {'from': 'Bitcoin miners', 'to': 'S-1 registration statements', 'relation': 'ASSOCIATED_WITH'}, {'from': 'S-1 registration statements', 'to': 'SEC', 'relation': 'ASSOCIATED_WITH'}, {'from': 'S-1 registration statements', 'to': 'Galaxy Digital', 'relation': 'ASSOCIATED_WITH'}, {'from': 'S-1 registration statements', 'to': 'Animoca Brands', 'relation': 'ASSOCIATED_WITH'}, {'from': 'S-1 registration statements', 'to': 'Toposware', 'relation': 'ASSOCIATED_WITH'}, {'from': 'S-1 registration statements', 'to': 'Bitcoin miners', 'relation': 'ASSOCIATED_WITH'}, {'from': 'S-1 registration statements', 'to': '21Shares', 'relation': 'ASSOCIATED_WITH'}]} | |
| ``` | |
|  | |
|  | |
| ## https://huggingface.co/NickyNicky/Financial_graph/blob/main/3d_graph%20(1).html | |
|  | |
| ``` | |
| better than BloombergGPT :V | |
| ``` | |
| ## colab | |
| ``` | |
| https://colab.research.google.com/drive/1-D7kgig6ODqhjCPaQBrZ8QG2OWsS29ML?usp=sharing | |
| ``` |