Skip to content

Commit

Permalink
Update README.md
Browse files Browse the repository at this point in the history
  • Loading branch information
rvalenciaaz authored Mar 28, 2024
1 parent 6c0aca8 commit 44e31e4
Showing 1 changed file with 44 additions and 13 deletions.
57 changes: 44 additions & 13 deletions README.md
Original file line number Diff line number Diff line change
@@ -1,18 +1,5 @@
# bope-gpt

To run the code, I'm typically updating a conda/mamba environment that, on the first time, can be installed using the following commands:

`mamba create -n botorch_mar2024 pytorch torchvision torchaudio pytorch-cuda=11.8 python==3.11 -c pytorch -c nvidia`

`mamba install botorch matplotlib seaborn -c pytorch -c gpytorch -c conda-forge`

`mamba update -c conda-forge ffmpeg`

`mamba install -c conda-forge dash`

`pip install keras`

`pip install tensorflow`

![reactor](https://github.com/AC-BO-Hackathon/BOPE-GPT/blob/main/images/reactor_small.jpg) | ![cat](https://github.com/AC-BO-Hackathon/BOPE-GPT/blob/main/images/chemcat_small.jpg)

Expand Down Expand Up @@ -103,3 +90,47 @@ The numbers from the model output are entered to prompt as strings with one deci


# An app to rule them all

**Coming soon to your home for every preferential BO enthusiast**

*Miscellaneous*

**Environment**
To run the code, we're typically updating a conda/mamba environment that, on the first time, can be installed using the following commands:

`mamba create -n botorch_mar2024 pytorch torchvision torchaudio pytorch-cuda=11.8 python==3.11 -c pytorch -c nvidia`

`mamba install botorch matplotlib seaborn -c pytorch -c gpytorch -c conda-forge`

`mamba update -c conda-forge ffmpeg`

`mamba install -c conda-forge dash`

`pip install keras`

`pip install tensorflow`

**Seeding**
Remember to define the seed for random generators when comparing different algorithms:


`#Your chosen seed
your_seed = 42

# Set seed for Python's RNG
random.seed(your_seed)

# Set seed for NumPy RNG
np.random.seed(your_seed)

# Set seed for PyTorch RNGs
torch.manual_seed(your_seed)

# Ensure reproducibility for PyTorch operations (might reduce performance)
torch.backends.cudnn.deterministic = True
torch.backends.cudnn.benchmark = False

# If using CUDA (PyTorch)
torch.cuda.manual_seed(your_seed)
torch.cuda.manual_seed_all(your_seed) # For multi-GPU setups`

0 comments on commit 44e31e4

Please sign in to comment.