Spaces:
Paused
Paused
File size: 2,632 Bytes
5b6bc45 4c79a74 81395fc b01335d 5b6bc45 8867e8a 4ffc0ce 5189ab0 c30f436 81395fc bdec0c5 bf9669d 237d9d2 bdec0c5 bf9669d bdec0c5 bf9669d bdec0c5 81395fc 7f45f98 5b6bc45 535aa50 81395fc 237d9d2 81395fc af69c70 81395fc bf9669d 81395fc d992640 237d9d2 b01335d |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 |
from transformers import AutoConfig, AutoTokenizer, AutoModelForSeq2SeqLM, AutoModelForCausalLM, MistralForCausalLM
from peft import PeftModel, PeftConfig
import torch
import gradio as gr
# Use the base model's ID
base_model_id = "mistralai/Mistral-7B-v0.1"
model_directory = "Tonic/mistralmed"
# Instantiate the Models
tokenizer = AutoTokenizer.from_pretrained("mistralai/Mistral-7B-v0.1", trust_remote_code=True)
#tokenizer.pad_token = tokenizer.eos_token
#tokenizer.padding_side = 'left'
# Specify the configuration class for the model
#model_config = AutoConfig.from_pretrained(base_model_id)
# Load the PEFT model with the specified configuration
#peft_model = AutoModelForCausalLM.from_pretrained(base_model_id, config=model_config)
# Load the PEFT model
peft_config = PeftConfig.from_pretrained("Tonic/mistralmed", token="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")
peft_model = MistralForCausalLM.from_pretrained("mistralai/Mistral-7B-v0.1", trust_remote_code=True)
peft_model = PeftModel.from_pretrained(peft_model, "Tonic/mistralmed", token="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")
class ChatBot:
def __init__(self):
self.history = []
def predict(self, input):
# Encode user input
user_input_ids = tokenizer.encode(input + tokenizer.eos_token, return_tensors="pt")
# Concatenate the user input with chat history
if self.history:
chat_history_ids = torch.cat([self.history, user_input_ids], dim=-1)
else:
chat_history_ids = user_input_ids
# Generate a response using the PEFT model
# response = peft_model.generate(chat_history_ids, max_length=512, pad_token_id=tokenizer.eos_token_id)
# response = peft_model.generate(chat_history_ids)
response = peft_model.generate(input_ids=chat_history_ids, max_length=512, pad_token_id=tokenizer.eos_token_id)
# Update chat history
self.history = chat_history_ids
# Decode and return the response
response_text = tokenizer.decode(response[0], skip_special_tokens=True)
return response_text
bot = ChatBot()
title = "👋🏻Welcome to Tonic's MistralMed Chat🚀"
description = "You can use this Space to test out the current model (MistralMed) or duplicate this Space and use it for any other model on 🤗HuggingFace. Join me on Discord to build together."
examples = [["What is the boiling point of nitrogen"]]
iface = gr.Interface(
fn=bot.predict,
title=title,
description=description,
examples=examples,
inputs="text",
outputs="text",
theme="ParityError/Anime"
)
iface.launch()
|