gpt4-torch 0.0.3

GPT4 - Pytorch

GPT4

The open source implementation of the base model behind GPT-4 from OPENAI [Language + Multi-Modal], click here for the Research Paper

Installation

pip install gpt4-torch

Usage

Here's an illustrative code snippet that showcases GPT-3 in action:

import torch
from gpt4 import GPT4

# Generate a random input sequence
x = torch.randint(0, 256, (1, 1024)).cuda()

# Initialize GPT-3 model
model = GPT4()

# Pass the input sequence through the model
output = model(x)

MultiModal Iteration

• Pass in text and and image tensors into GPT4

import torch
from gpt4.gpt4 import GPT4MultiModal

#usage
img = torch.randn(1, 3, 256, 256)
text = torch.randint(0, 20000, (1, 1024))


model = GPT4MultiModal()
output = model(text, img)

📚 Training

from gpt4 import train

train()

For further instructions, refer to the Training SOP.

1. Set the environment variables:

   • ENTITY_NAME: Your wandb project name
   • OUTPUT_DIR: Directory to save the weights (e.g., ./weights)
   • MASTER_ADDR: For distributed training
   • MASTER_PORT For master port distributed training
   • RANK- Number of nodes services
   • WORLD_SIZE Number of gpus

2. Configure the training:

   • Accelerate Config
   • Enable Deepspeed 3
   • Accelerate launch train_distributed_accelerate.py

For more information, refer to the Training SOP.