Upload folder using huggingface_hub
Browse filesThis view is limited to 50 files because it contains too many changes.
See raw diff
- .devcontainer/Dockerfile +53 -0
- .devcontainer/devcontainer.env +2 -0
- .devcontainer/devcontainer.json +71 -0
- .devcontainer/postCreateCommand.sh +45 -0
- .editorconfig +18 -0
- .gitattributes +96 -35
- .github/ISSUE_TEMPLATE/1-usage.yaml +31 -0
- .github/ISSUE_TEMPLATE/2-feature-request.yaml +13 -0
- .github/ISSUE_TEMPLATE/3-question.yaml +13 -0
- .github/ISSUE_TEMPLATE/4-discussion.yaml +13 -0
- LICENSE +201 -0
- README.md +460 -0
- accuracy_scores.txt +140 -0
- checkpoints_ft/.gitattributes +55 -0
- checkpoints_ft/coco_gen_200k/llava-v1.5-13b-pretrain/config.json +46 -0
- checkpoints_ft/coco_gen_200k/llava-v1.5-13b-pretrain/mm_projector.bin +3 -0
- checkpoints_ft/coco_gen_200k/llava-v1.5-13b-pretrain/trainer_state.json +0 -0
- checkpoints_ft/coco_gen_558k/llava-v1.5-13b-pretrain/config.json +46 -0
- checkpoints_ft/coco_gen_558k/llava-v1.5-13b-pretrain/mm_projector.bin +3 -0
- checkpoints_ft/coco_gen_558k/llava-v1.5-13b-pretrain/trainer_state.json +0 -0
- checkpoints_ft/coco_raw_200k/llava-v1.5-13b-pretrain/config.json +46 -0
- checkpoints_ft/coco_raw_200k/llava-v1.5-13b-pretrain/mm_projector.bin +3 -0
- checkpoints_ft/coco_raw_200k/llava-v1.5-13b-pretrain/trainer_state.json +0 -0
- checkpoints_ft/coco_raw_558k/llava-v1.5-13b-pretrain/config.json +46 -0
- checkpoints_ft/coco_raw_558k/llava-v1.5-13b-pretrain/mm_projector.bin +3 -0
- checkpoints_ft/coco_raw_558k/llava-v1.5-13b-pretrain/trainer_state.json +0 -0
- checkpoints_ft/coco_select_200k/llava-v1.5-13b-pretrain/config.json +46 -0
- checkpoints_ft/coco_select_200k/llava-v1.5-13b-pretrain/mm_projector.bin +3 -0
- checkpoints_ft/coco_select_200k/llava-v1.5-13b-pretrain/trainer_state.json +0 -0
- checkpoints_ft/llava_coco_gen_758k/llava-v1.5-13b-pretrain/config.json +46 -0
- checkpoints_ft/llava_coco_gen_758k/llava-v1.5-13b-pretrain/mm_projector.bin +3 -0
- checkpoints_ft/llava_coco_gen_758k/llava-v1.5-13b-pretrain/trainer_state.json +0 -0
- checkpoints_ft/llava_coco_raw_758k/llava-v1.5-13b-pretrain/config.json +46 -0
- checkpoints_ft/llava_coco_raw_758k/llava-v1.5-13b-pretrain/mm_projector.bin +3 -0
- checkpoints_ft/llava_coco_raw_758k/llava-v1.5-13b-pretrain/trainer_state.json +0 -0
- checkpoints_ft/llava_gen_200k/llava-v1.5-13b-pretrain/config.json +46 -0
- checkpoints_ft/llava_gen_200k/llava-v1.5-13b-pretrain/mm_projector.bin +3 -0
- checkpoints_ft/llava_gen_200k/llava-v1.5-13b-pretrain/trainer_state.json +0 -0
- checkpoints_ft/llava_gen_200k/llava-v1.5-7b-pretrain/config.json +45 -0
- checkpoints_ft/llava_gen_200k/llava-v1.5-7b-pretrain/mm_projector.bin +3 -0
- checkpoints_ft/llava_gen_200k/llava-v1.5-7b-pretrain/trainer_state.json +0 -0
- checkpoints_ft/llava_gen_200k/llava-v1.5-7b/config.json +44 -0
- checkpoints_ft/llava_gen_200k/llava-v1.5-7b/trainer_state.json +0 -0
- checkpoints_ft/llava_gen_558k/llava-v1.5-13b-pretrain/config.json +46 -0
- checkpoints_ft/llava_gen_558k/llava-v1.5-13b-pretrain/mm_projector.bin +3 -0
- checkpoints_ft/llava_gen_558k/llava-v1.5-13b-pretrain/trainer_state.json +0 -0
- checkpoints_ft/llava_raw_200k/llava-v1.5-13b-pretrain/config.json +46 -0
- checkpoints_ft/llava_raw_200k/llava-v1.5-13b-pretrain/mm_projector.bin +3 -0
- checkpoints_ft/llava_raw_200k/llava-v1.5-13b-pretrain/trainer_state.json +0 -0
- checkpoints_ft/llava_raw_558k/llava-v1.5-13b-pretrain/config.json +46 -0
.devcontainer/Dockerfile
ADDED
|
@@ -0,0 +1,53 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
FROM mcr.microsoft.com/devcontainers/base:ubuntu-20.04
|
| 2 |
+
|
| 3 |
+
SHELL [ "bash", "-c" ]
|
| 4 |
+
|
| 5 |
+
# update apt and install packages
|
| 6 |
+
RUN apt update && \
|
| 7 |
+
apt install -yq \
|
| 8 |
+
ffmpeg \
|
| 9 |
+
dkms \
|
| 10 |
+
build-essential
|
| 11 |
+
|
| 12 |
+
# add user tools
|
| 13 |
+
RUN sudo apt install -yq \
|
| 14 |
+
jq \
|
| 15 |
+
jp \
|
| 16 |
+
tree \
|
| 17 |
+
tldr
|
| 18 |
+
|
| 19 |
+
# add git-lfs and install
|
| 20 |
+
RUN curl -s https://packagecloud.io/install/repositories/github/git-lfs/script.deb.sh | sudo bash && \
|
| 21 |
+
sudo apt-get install -yq git-lfs && \
|
| 22 |
+
git lfs install
|
| 23 |
+
|
| 24 |
+
############################################
|
| 25 |
+
# Setup user
|
| 26 |
+
############################################
|
| 27 |
+
|
| 28 |
+
USER vscode
|
| 29 |
+
|
| 30 |
+
# install azcopy, a tool to copy to/from blob storage
|
| 31 |
+
# for more info: https://learn.microsoft.com/en-us/azure/storage/common/storage-use-azcopy-blobs-upload#upload-a-file
|
| 32 |
+
RUN cd /tmp && \
|
| 33 |
+
wget https://azcopyvnext.azureedge.net/release20230123/azcopy_linux_amd64_10.17.0.tar.gz && \
|
| 34 |
+
tar xvf azcopy_linux_amd64_10.17.0.tar.gz && \
|
| 35 |
+
mkdir -p ~/.local/bin && \
|
| 36 |
+
mv azcopy_linux_amd64_10.17.0/azcopy ~/.local/bin && \
|
| 37 |
+
chmod +x ~/.local/bin/azcopy && \
|
| 38 |
+
rm -rf azcopy_linux_amd64*
|
| 39 |
+
|
| 40 |
+
# Setup conda
|
| 41 |
+
RUN cd /tmp && \
|
| 42 |
+
wget https://repo.anaconda.com/miniconda/Miniconda3-latest-Linux-x86_64.sh && \
|
| 43 |
+
bash ./Miniconda3-latest-Linux-x86_64.sh -b && \
|
| 44 |
+
rm ./Miniconda3-latest-Linux-x86_64.sh
|
| 45 |
+
|
| 46 |
+
# Install dotnet
|
| 47 |
+
RUN cd /tmp && \
|
| 48 |
+
wget https://dot.net/v1/dotnet-install.sh && \
|
| 49 |
+
chmod +x dotnet-install.sh && \
|
| 50 |
+
./dotnet-install.sh --channel 7.0 && \
|
| 51 |
+
./dotnet-install.sh --channel 3.1 && \
|
| 52 |
+
rm ./dotnet-install.sh
|
| 53 |
+
|
.devcontainer/devcontainer.env
ADDED
|
@@ -0,0 +1,2 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
SAMPLE_ENV_VAR1="Sample Value"
|
| 2 |
+
SAMPLE_ENV_VAR2=332431bf-68bf
|
.devcontainer/devcontainer.json
ADDED
|
@@ -0,0 +1,71 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"name": "LLaVA",
|
| 3 |
+
"build": {
|
| 4 |
+
"dockerfile": "Dockerfile",
|
| 5 |
+
"context": "..",
|
| 6 |
+
"args": {}
|
| 7 |
+
},
|
| 8 |
+
"features": {
|
| 9 |
+
"ghcr.io/devcontainers/features/docker-in-docker:2": {},
|
| 10 |
+
"ghcr.io/devcontainers/features/azure-cli:1": {},
|
| 11 |
+
"ghcr.io/azure/azure-dev/azd:0": {},
|
| 12 |
+
"ghcr.io/devcontainers/features/powershell:1": {},
|
| 13 |
+
"ghcr.io/devcontainers/features/common-utils:2": {},
|
| 14 |
+
"ghcr.io/devcontainers-contrib/features/zsh-plugins:0": {},
|
| 15 |
+
},
|
| 16 |
+
// "forwardPorts": [],
|
| 17 |
+
"postCreateCommand": "bash ./.devcontainer/postCreateCommand.sh",
|
| 18 |
+
"customizations": {
|
| 19 |
+
"vscode": {
|
| 20 |
+
"settings": {
|
| 21 |
+
"python.analysis.autoImportCompletions": true,
|
| 22 |
+
"python.analysis.autoImportUserSymbols": true,
|
| 23 |
+
"python.defaultInterpreterPath": "~/miniconda3/envs/llava/bin/python",
|
| 24 |
+
"python.formatting.provider": "yapf",
|
| 25 |
+
"python.linting.enabled": true,
|
| 26 |
+
"python.linting.flake8Enabled": true,
|
| 27 |
+
"isort.check": true,
|
| 28 |
+
"dev.containers.copyGitConfig": true,
|
| 29 |
+
"terminal.integrated.defaultProfile.linux": "zsh",
|
| 30 |
+
"terminal.integrated.profiles.linux": {
|
| 31 |
+
"zsh": {
|
| 32 |
+
"path": "/usr/bin/zsh"
|
| 33 |
+
},
|
| 34 |
+
}
|
| 35 |
+
},
|
| 36 |
+
"extensions": [
|
| 37 |
+
"aaron-bond.better-comments",
|
| 38 |
+
"eamodio.gitlens",
|
| 39 |
+
"EditorConfig.EditorConfig",
|
| 40 |
+
"foxundermoon.shell-format",
|
| 41 |
+
"GitHub.copilot-chat",
|
| 42 |
+
"GitHub.copilot-labs",
|
| 43 |
+
"GitHub.copilot",
|
| 44 |
+
"lehoanganh298.json-lines-viewer",
|
| 45 |
+
"mhutchie.git-graph",
|
| 46 |
+
"ms-azuretools.vscode-docker",
|
| 47 |
+
"ms-dotnettools.dotnet-interactive-vscode",
|
| 48 |
+
"ms-python.flake8",
|
| 49 |
+
"ms-python.isort",
|
| 50 |
+
"ms-python.python",
|
| 51 |
+
"ms-python.vscode-pylance",
|
| 52 |
+
"njpwerner.autodocstring",
|
| 53 |
+
"redhat.vscode-yaml",
|
| 54 |
+
"stkb.rewrap",
|
| 55 |
+
"yzhang.markdown-all-in-one",
|
| 56 |
+
]
|
| 57 |
+
}
|
| 58 |
+
},
|
| 59 |
+
"mounts": [],
|
| 60 |
+
"runArgs": [
|
| 61 |
+
"--gpus",
|
| 62 |
+
"all",
|
| 63 |
+
// "--ipc",
|
| 64 |
+
// "host",
|
| 65 |
+
"--ulimit",
|
| 66 |
+
"memlock=-1",
|
| 67 |
+
"--env-file",
|
| 68 |
+
".devcontainer/devcontainer.env"
|
| 69 |
+
],
|
| 70 |
+
// "remoteUser": "root"
|
| 71 |
+
}
|
.devcontainer/postCreateCommand.sh
ADDED
|
@@ -0,0 +1,45 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
git config --global safe.directory '*'
|
| 2 |
+
git config --global core.editor "code --wait"
|
| 3 |
+
git config --global pager.branch false
|
| 4 |
+
|
| 5 |
+
# Set AZCOPY concurrency to auto
|
| 6 |
+
echo "export AZCOPY_CONCURRENCY_VALUE=AUTO" >> ~/.zshrc
|
| 7 |
+
echo "export AZCOPY_CONCURRENCY_VALUE=AUTO" >> ~/.bashrc
|
| 8 |
+
|
| 9 |
+
# Activate conda by default
|
| 10 |
+
echo ". /home/vscode/miniconda3/bin/activate" >> ~/.zshrc
|
| 11 |
+
echo ". /home/vscode/miniconda3/bin/activate" >> ~/.bashrc
|
| 12 |
+
|
| 13 |
+
# Use llava environment by default
|
| 14 |
+
echo "conda activate llava" >> ~/.zshrc
|
| 15 |
+
echo "conda activate llava" >> ~/.bashrc
|
| 16 |
+
|
| 17 |
+
# Add dotnet to PATH
|
| 18 |
+
echo 'export PATH="$PATH:$HOME/.dotnet"' >> ~/.bashrc
|
| 19 |
+
echo 'export PATH="$PATH:$HOME/.dotnet"' >> ~/.zshrc
|
| 20 |
+
|
| 21 |
+
# Create and activate llava environment
|
| 22 |
+
source /home/vscode/miniconda3/bin/activate
|
| 23 |
+
conda create -y -q -n llava python=3.10
|
| 24 |
+
conda activate llava
|
| 25 |
+
|
| 26 |
+
# Install Nvidia Cuda Compiler
|
| 27 |
+
conda install -y -c nvidia cuda-compiler
|
| 28 |
+
|
| 29 |
+
pip install pre-commit==3.0.2
|
| 30 |
+
|
| 31 |
+
# Install package locally
|
| 32 |
+
pip install --upgrade pip # enable PEP 660 support
|
| 33 |
+
pip install -e .
|
| 34 |
+
|
| 35 |
+
# Install additional packages for training
|
| 36 |
+
pip install -e ".[train]"
|
| 37 |
+
pip install flash-attn --no-build-isolation
|
| 38 |
+
|
| 39 |
+
# Download checkpoints to location outside of the repo
|
| 40 |
+
git clone https://huggingface.co/liuhaotian/llava-v1.5-7b ~/llava-v1.5-7b
|
| 41 |
+
|
| 42 |
+
# Commented because it is unlikely for users to have enough local GPU memory to load the model
|
| 43 |
+
# git clone https://huggingface.co/liuhaotian/llava-v1.5-13b ~/llava-v1.5-13b
|
| 44 |
+
|
| 45 |
+
echo "postCreateCommand.sh COMPLETE!"
|
.editorconfig
ADDED
|
@@ -0,0 +1,18 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
root = true
|
| 2 |
+
|
| 3 |
+
# Unix-style newlines with a newline ending every file
|
| 4 |
+
[*]
|
| 5 |
+
end_of_line = lf
|
| 6 |
+
insert_final_newline = true
|
| 7 |
+
trim_trailing_whitespace = true
|
| 8 |
+
charset = utf-8
|
| 9 |
+
|
| 10 |
+
# 4 space indentation
|
| 11 |
+
[*.{py,json}]
|
| 12 |
+
indent_style = space
|
| 13 |
+
indent_size = 4
|
| 14 |
+
|
| 15 |
+
# 2 space indentation
|
| 16 |
+
[*.{md,sh,yaml,yml}]
|
| 17 |
+
indent_style = space
|
| 18 |
+
indent_size = 2
|
.gitattributes
CHANGED
|
@@ -1,35 +1,96 @@
|
|
| 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 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
# https://git-scm.com/docs/gitattributes
|
| 2 |
+
|
| 3 |
+
# Set the default behavior, in case people don't have core.autocrlf set.
|
| 4 |
+
# https://git-scm.com/docs/gitattributes#_end_of_line_conversion
|
| 5 |
+
* text=auto
|
| 6 |
+
|
| 7 |
+
# common python attributes, taken from https://github.com/alexkaratarakis/gitattributes/blob/710900479a2bedeec7003d381719521ffbb18bf8/Python.gitattributes
|
| 8 |
+
# Source files
|
| 9 |
+
# ============
|
| 10 |
+
*.pxd text diff=python
|
| 11 |
+
*.py text diff=python
|
| 12 |
+
*.py3 text diff=python
|
| 13 |
+
*.pyw text diff=python
|
| 14 |
+
*.pyx text diff=python
|
| 15 |
+
*.pyz text diff=python
|
| 16 |
+
*.pyi text diff=python
|
| 17 |
+
|
| 18 |
+
# Binary files
|
| 19 |
+
# ============
|
| 20 |
+
*.db binary
|
| 21 |
+
*.p binary
|
| 22 |
+
*.pkl binary
|
| 23 |
+
*.pickle binary
|
| 24 |
+
*.pyc binary export-ignore
|
| 25 |
+
*.pyo binary export-ignore
|
| 26 |
+
*.pyd binary
|
| 27 |
+
|
| 28 |
+
# Jupyter notebook
|
| 29 |
+
*.ipynb text eol=lf
|
| 30 |
+
checkpoints_ft/coco_gen_200k/llava-v1.5-13b-pretrain/mm_projector.bin filter=lfs diff=lfs merge=lfs -text
|
| 31 |
+
checkpoints_ft/coco_gen_558k/llava-v1.5-13b-pretrain/mm_projector.bin filter=lfs diff=lfs merge=lfs -text
|
| 32 |
+
checkpoints_ft/coco_raw_200k/llava-v1.5-13b-pretrain/mm_projector.bin filter=lfs diff=lfs merge=lfs -text
|
| 33 |
+
checkpoints_ft/coco_raw_558k/llava-v1.5-13b-pretrain/mm_projector.bin filter=lfs diff=lfs merge=lfs -text
|
| 34 |
+
checkpoints_ft/coco_select_200k/llava-v1.5-13b-pretrain/mm_projector.bin filter=lfs diff=lfs merge=lfs -text
|
| 35 |
+
checkpoints_ft/llava_coco_gen_758k/llava-v1.5-13b-pretrain/mm_projector.bin filter=lfs diff=lfs merge=lfs -text
|
| 36 |
+
checkpoints_ft/llava_coco_raw_758k/llava-v1.5-13b-pretrain/mm_projector.bin filter=lfs diff=lfs merge=lfs -text
|
| 37 |
+
checkpoints_ft/llava_gen_200k/llava-v1.5-13b-pretrain/mm_projector.bin filter=lfs diff=lfs merge=lfs -text
|
| 38 |
+
checkpoints_ft/llava_gen_200k/llava-v1.5-7b-pretrain/mm_projector.bin filter=lfs diff=lfs merge=lfs -text
|
| 39 |
+
checkpoints_ft/llava_gen_558k/llava-v1.5-13b-pretrain/mm_projector.bin filter=lfs diff=lfs merge=lfs -text
|
| 40 |
+
checkpoints_ft/llava_raw_200k/llava-v1.5-13b-pretrain/mm_projector.bin filter=lfs diff=lfs merge=lfs -text
|
| 41 |
+
checkpoints_ft/llava_raw_558k/llava-v1.5-13b-pretrain/mm_projector.bin filter=lfs diff=lfs merge=lfs -text
|
| 42 |
+
checkpoints_ft/llava_raw_558k/llava-v1.5-7b-pretrain/mm_projector.bin filter=lfs diff=lfs merge=lfs -text
|
| 43 |
+
checkpoints_ft/pure_gen_200k/llava-v1.5-13b-pretrain/mm_projector.bin filter=lfs diff=lfs merge=lfs -text
|
| 44 |
+
checkpoints_ft/pure_gen_558k/llava-v1.5-13b-pretrain/mm_projector.bin filter=lfs diff=lfs merge=lfs -text
|
| 45 |
+
checkpoints_ft/random/llava-v1.5-13b-pretrain/mm_projector.bin filter=lfs diff=lfs merge=lfs -text
|
| 46 |
+
checkpoints_ft/select_gen_100k/llava-v1.5-13b-pretrain/mm_projector.bin filter=lfs diff=lfs merge=lfs -text
|
| 47 |
+
checkpoints_ft/select_gen_100k/llava-v1.5-7b-pretrain/mm_projector.bin filter=lfs diff=lfs merge=lfs -text
|
| 48 |
+
checkpoints_ft/select_gen_200k/llava-v1.5-13b-pretrain/mm_projector.bin filter=lfs diff=lfs merge=lfs -text
|
| 49 |
+
checkpoints_ft/select_raw_100k/llava-v1.5-13b-pretrain/mm_projector.bin filter=lfs diff=lfs merge=lfs -text
|
| 50 |
+
checkpoints_ft/select_raw_100k/llava-v1.5-7b-pretrain/mm_projector.bin filter=lfs diff=lfs merge=lfs -text
|
| 51 |
+
checkpoints_ft/select_raw_200k/llava-v1.5-13b-pretrain/mm_projector.bin filter=lfs diff=lfs merge=lfs -text
|
| 52 |
+
checkpoints_ft/sharegpt4v_100k/llava-v1.5-7b-pretrain/mm_projector.bin filter=lfs diff=lfs merge=lfs -text
|
| 53 |
+
checkpoints_pt/coco_gen_200k/llava-v1.5-13b-pretrain/mm_projector.bin filter=lfs diff=lfs merge=lfs -text
|
| 54 |
+
checkpoints_pt/coco_gen_558k/llava-v1.5-13b-pretrain/mm_projector.bin filter=lfs diff=lfs merge=lfs -text
|
| 55 |
+
checkpoints_pt/coco_raw_200k/llava-v1.5-13b-pretrain/mm_projector.bin filter=lfs diff=lfs merge=lfs -text
|
| 56 |
+
checkpoints_pt/coco_raw_558k/llava-v1.5-13b-pretrain/mm_projector.bin filter=lfs diff=lfs merge=lfs -text
|
| 57 |
+
checkpoints_pt/coco_select_200k/llava-v1.5-13b-pretrain/mm_projector.bin filter=lfs diff=lfs merge=lfs -text
|
| 58 |
+
checkpoints_pt/llava_coco_gen_758k/llava-v1.5-13b-pretrain/mm_projector.bin filter=lfs diff=lfs merge=lfs -text
|
| 59 |
+
checkpoints_pt/llava_coco_raw_758k/llava-v1.5-13b-pretrain/mm_projector.bin filter=lfs diff=lfs merge=lfs -text
|
| 60 |
+
checkpoints_pt/llava_gen_200k/llava-v1.5-13b-pretrain/mm_projector.bin filter=lfs diff=lfs merge=lfs -text
|
| 61 |
+
checkpoints_pt/llava_raw_200k/llava-v1.5-13b-pretrain/mm_projector.bin filter=lfs diff=lfs merge=lfs -text
|
| 62 |
+
checkpoints_pt/llava_raw_558k/llava-v1.5-13b-pretrain/mm_projector.bin filter=lfs diff=lfs merge=lfs -text
|
| 63 |
+
images/demo_cli.gif filter=lfs diff=lfs merge=lfs -text
|
| 64 |
+
playground/data/LLaVA-Pretrain/coco_annotations_500k.json filter=lfs diff=lfs merge=lfs -text
|
| 65 |
+
playground/data/LLaVA-Pretrain/coco_gen_200k.json filter=lfs diff=lfs merge=lfs -text
|
| 66 |
+
playground/data/LLaVA-Pretrain/coco_gen_200k_2.json filter=lfs diff=lfs merge=lfs -text
|
| 67 |
+
playground/data/LLaVA-Pretrain/coco_raw_200k.json filter=lfs diff=lfs merge=lfs -text
|
| 68 |
+
playground/data/LLaVA-Pretrain/llava_coco_gen_758k.json filter=lfs diff=lfs merge=lfs -text
|
| 69 |
+
playground/data/LLaVA-Pretrain/llava_coco_raw_758k.json filter=lfs diff=lfs merge=lfs -text
|
| 70 |
+
playground/data/LLaVA-Pretrain/llava_gen_200k.json filter=lfs diff=lfs merge=lfs -text
|
| 71 |
+
playground/data/LLaVA-Pretrain/llava_gen_558k.json filter=lfs diff=lfs merge=lfs -text
|
| 72 |
+
playground/data/LLaVA-Pretrain/llava_raw_200k.json filter=lfs diff=lfs merge=lfs -text
|
| 73 |
+
playground/data/LLaVA-Pretrain/llava_raw_558k.json filter=lfs diff=lfs merge=lfs -text
|
| 74 |
+
playground/data/LLaVA-Pretrain/mixed_200k.json filter=lfs diff=lfs merge=lfs -text
|
| 75 |
+
playground/data/LLaVA-Pretrain/select_gen_100k.json filter=lfs diff=lfs merge=lfs -text
|
| 76 |
+
playground/data/LLaVA-Pretrain/select_gen_200k.json filter=lfs diff=lfs merge=lfs -text
|
| 77 |
+
playground/data/LLaVA-Pretrain/select_mscoco_200k.json filter=lfs diff=lfs merge=lfs -text
|
| 78 |
+
playground/data/LLaVA-Pretrain/select_raw_100k.json filter=lfs diff=lfs merge=lfs -text
|
| 79 |
+
playground/data/LLaVA-Pretrain/select_raw_200k.json filter=lfs diff=lfs merge=lfs -text
|
| 80 |
+
playground/data/LLaVA-Pretrain/sharegpt4v_100k.json filter=lfs diff=lfs merge=lfs -text
|
| 81 |
+
playground/data/LLaVA-Pretrain/sharegpt4v_instruct_gpt4-vision_cap100k.json filter=lfs diff=lfs merge=lfs -text
|
| 82 |
+
playground/data/eval/gqa/data/challenge_all_questions.json filter=lfs diff=lfs merge=lfs -text
|
| 83 |
+
playground/data/eval/gqa/data/submission_all_questions.json filter=lfs diff=lfs merge=lfs -text
|
| 84 |
+
playground/data/eval/gqa/data/test_all_questions.json filter=lfs diff=lfs merge=lfs -text
|
| 85 |
+
playground/data/eval/gqa/data/test_balanced_questions.json filter=lfs diff=lfs merge=lfs -text
|
| 86 |
+
playground/data/eval/gqa/data/testdev_all_questions.json filter=lfs diff=lfs merge=lfs -text
|
| 87 |
+
playground/data/eval/gqa/data/testdev_balanced_questions.json filter=lfs diff=lfs merge=lfs -text
|
| 88 |
+
playground/data/eval/gqa/data/train_balanced_questions.json filter=lfs diff=lfs merge=lfs -text
|
| 89 |
+
playground/data/eval/gqa/data/val_all_questions.json filter=lfs diff=lfs merge=lfs -text
|
| 90 |
+
playground/data/eval/gqa/data/val_balanced_questions.json filter=lfs diff=lfs merge=lfs -text
|
| 91 |
+
playground/data/eval/gqa/data/val_choices.json filter=lfs diff=lfs merge=lfs -text
|
| 92 |
+
playground/data/llava_v1_5_mix200k.json filter=lfs diff=lfs merge=lfs -text
|
| 93 |
+
playground/data/llava_v1_5_mix665k.json filter=lfs diff=lfs merge=lfs -text
|
| 94 |
+
playground/data/llava_v1_5_mix865k.json filter=lfs diff=lfs merge=lfs -text
|
| 95 |
+
playground/data/llava_v1_5_mix865k_2.json filter=lfs diff=lfs merge=lfs -text
|
| 96 |
+
playground/data/mscoco_200k.json filter=lfs diff=lfs merge=lfs -text
|
.github/ISSUE_TEMPLATE/1-usage.yaml
ADDED
|
@@ -0,0 +1,31 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
name: Usage issues
|
| 2 |
+
description: Report issues in usage.
|
| 3 |
+
title: "[Usage] "
|
| 4 |
+
body:
|
| 5 |
+
- type: markdown
|
| 6 |
+
attributes:
|
| 7 |
+
value: |
|
| 8 |
+
Thanks for taking the time to fill out this form. Please give as detailed description as possible for us to better assist with the issue :)
|
| 9 |
+
- type: textarea
|
| 10 |
+
id: what-happened
|
| 11 |
+
attributes:
|
| 12 |
+
label: Describe the issue
|
| 13 |
+
description: Please give as detailed description as possible for us to better assist with the issue. Please paste the **FULL** error log here, so that we can better understand the issue. Wrap the log with ``` for better readability in GitHub.
|
| 14 |
+
placeholder: Issue
|
| 15 |
+
value: |
|
| 16 |
+
Issue:
|
| 17 |
+
|
| 18 |
+
Command:
|
| 19 |
+
```
|
| 20 |
+
PASTE THE COMMANDS HERE.
|
| 21 |
+
```
|
| 22 |
+
|
| 23 |
+
Log:
|
| 24 |
+
```
|
| 25 |
+
PASTE THE LOGS HERE.
|
| 26 |
+
```
|
| 27 |
+
|
| 28 |
+
Screenshots:
|
| 29 |
+
You may attach screenshots if it better explains the issue.
|
| 30 |
+
validations:
|
| 31 |
+
required: true
|
.github/ISSUE_TEMPLATE/2-feature-request.yaml
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
name: Feature Request
|
| 2 |
+
description: Request for a new feature
|
| 3 |
+
title: "[Feature request] "
|
| 4 |
+
body:
|
| 5 |
+
- type: markdown
|
| 6 |
+
attributes:
|
| 7 |
+
value: |
|
| 8 |
+
Thanks for your interest in our work. Please share your thoughts of the new features below.
|
| 9 |
+
- type: textarea
|
| 10 |
+
id: feature
|
| 11 |
+
attributes:
|
| 12 |
+
label: feature
|
| 13 |
+
placeholder: Start your thoughts here...
|
.github/ISSUE_TEMPLATE/3-question.yaml
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
name: Questions
|
| 2 |
+
description: General questions about the work
|
| 3 |
+
title: "[Question] "
|
| 4 |
+
body:
|
| 5 |
+
- type: markdown
|
| 6 |
+
attributes:
|
| 7 |
+
value: |
|
| 8 |
+
Thanks for your interest in our work. For this type of question, it may be more suitable to go to [discussion](https://github.com/haotian-liu/LLaVA/discussions) sections. If you believe an issue would be better for your request, please continue your post below :)
|
| 9 |
+
- type: textarea
|
| 10 |
+
id: question
|
| 11 |
+
attributes:
|
| 12 |
+
label: Question
|
| 13 |
+
placeholder: Start question here...
|
.github/ISSUE_TEMPLATE/4-discussion.yaml
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
name: Discussions
|
| 2 |
+
description: General discussions about the work
|
| 3 |
+
title: "[Discussion] "
|
| 4 |
+
body:
|
| 5 |
+
- type: markdown
|
| 6 |
+
attributes:
|
| 7 |
+
value: |
|
| 8 |
+
Thanks for your interest in our work. For this type of question, it may be more suitable to go to [discussion](https://github.com/haotian-liu/LLaVA/discussions) sections. If you believe an issue would be better for your request, please continue your post below :)
|
| 9 |
+
- type: textarea
|
| 10 |
+
id: discussion
|
| 11 |
+
attributes:
|
| 12 |
+
label: Discussion
|
| 13 |
+
placeholder: Start discussion here...
|
LICENSE
ADDED
|
@@ -0,0 +1,201 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
Apache License
|
| 2 |
+
Version 2.0, January 2004
|
| 3 |
+
http://www.apache.org/licenses/
|
| 4 |
+
|
| 5 |
+
TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
|
| 6 |
+
|
| 7 |
+
1. Definitions.
|
| 8 |
+
|
| 9 |
+
"License" shall mean the terms and conditions for use, reproduction,
|
| 10 |
+
and distribution as defined by Sections 1 through 9 of this document.
|
| 11 |
+
|
| 12 |
+
"Licensor" shall mean the copyright owner or entity authorized by
|
| 13 |
+
the copyright owner that is granting the License.
|
| 14 |
+
|
| 15 |
+
"Legal Entity" shall mean the union of the acting entity and all
|
| 16 |
+
other entities that control, are controlled by, or are under common
|
| 17 |
+
control with that entity. For the purposes of this definition,
|
| 18 |
+
"control" means (i) the power, direct or indirect, to cause the
|
| 19 |
+
direction or management of such entity, whether by contract or
|
| 20 |
+
otherwise, or (ii) ownership of fifty percent (50%) or more of the
|
| 21 |
+
outstanding shares, or (iii) beneficial ownership of such entity.
|
| 22 |
+
|
| 23 |
+
"You" (or "Your") shall mean an individual or Legal Entity
|
| 24 |
+
exercising permissions granted by this License.
|
| 25 |
+
|
| 26 |
+
"Source" form shall mean the preferred form for making modifications,
|
| 27 |
+
including but not limited to software source code, documentation
|
| 28 |
+
source, and configuration files.
|
| 29 |
+
|
| 30 |
+
"Object" form shall mean any form resulting from mechanical
|
| 31 |
+
transformation or translation of a Source form, including but
|
| 32 |
+
not limited to compiled object code, generated documentation,
|
| 33 |
+
and conversions to other media types.
|
| 34 |
+
|
| 35 |
+
"Work" shall mean the work of authorship, whether in Source or
|
| 36 |
+
Object form, made available under the License, as indicated by a
|
| 37 |
+
copyright notice that is included in or attached to the work
|
| 38 |
+
(an example is provided in the Appendix below).
|
| 39 |
+
|
| 40 |
+
"Derivative Works" shall mean any work, whether in Source or Object
|
| 41 |
+
form, that is based on (or derived from) the Work and for which the
|
| 42 |
+
editorial revisions, annotations, elaborations, or other modifications
|
| 43 |
+
represent, as a whole, an original work of authorship. For the purposes
|
| 44 |
+
of this License, Derivative Works shall not include works that remain
|
| 45 |
+
separable from, or merely link (or bind by name) to the interfaces of,
|
| 46 |
+
the Work and Derivative Works thereof.
|
| 47 |
+
|
| 48 |
+
"Contribution" shall mean any work of authorship, including
|
| 49 |
+
the original version of the Work and any modifications or additions
|
| 50 |
+
to that Work or Derivative Works thereof, that is intentionally
|
| 51 |
+
submitted to Licensor for inclusion in the Work by the copyright owner
|
| 52 |
+
or by an individual or Legal Entity authorized to submit on behalf of
|
| 53 |
+
the copyright owner. For the purposes of this definition, "submitted"
|
| 54 |
+
means any form of electronic, verbal, or written communication sent
|
| 55 |
+
to the Licensor or its representatives, including but not limited to
|
| 56 |
+
communication on electronic mailing lists, source code control systems,
|
| 57 |
+
and issue tracking systems that are managed by, or on behalf of, the
|
| 58 |
+
Licensor for the purpose of discussing and improving the Work, but
|
| 59 |
+
excluding communication that is conspicuously marked or otherwise
|
| 60 |
+
designated in writing by the copyright owner as "Not a Contribution."
|
| 61 |
+
|
| 62 |
+
"Contributor" shall mean Licensor and any individual or Legal Entity
|
| 63 |
+
on behalf of whom a Contribution has been received by Licensor and
|
| 64 |
+
subsequently incorporated within the Work.
|
| 65 |
+
|
| 66 |
+
2. Grant of Copyright License. Subject to the terms and conditions of
|
| 67 |
+
this License, each Contributor hereby grants to You a perpetual,
|
| 68 |
+
worldwide, non-exclusive, no-charge, royalty-free, irrevocable
|
| 69 |
+
copyright license to reproduce, prepare Derivative Works of,
|
| 70 |
+
publicly display, publicly perform, sublicense, and distribute the
|
| 71 |
+
Work and such Derivative Works in Source or Object form.
|
| 72 |
+
|
| 73 |
+
3. Grant of Patent License. Subject to the terms and conditions of
|
| 74 |
+
this License, each Contributor hereby grants to You a perpetual,
|
| 75 |
+
worldwide, non-exclusive, no-charge, royalty-free, irrevocable
|
| 76 |
+
(except as stated in this section) patent license to make, have made,
|
| 77 |
+
use, offer to sell, sell, import, and otherwise transfer the Work,
|
| 78 |
+
where such license applies only to those patent claims licensable
|
| 79 |
+
by such Contributor that are necessarily infringed by their
|
| 80 |
+
Contribution(s) alone or by combination of their Contribution(s)
|
| 81 |
+
with the Work to which such Contribution(s) was submitted. If You
|
| 82 |
+
institute patent litigation against any entity (including a
|
| 83 |
+
cross-claim or counterclaim in a lawsuit) alleging that the Work
|
| 84 |
+
or a Contribution incorporated within the Work constitutes direct
|
| 85 |
+
or contributory patent infringement, then any patent licenses
|
| 86 |
+
granted to You under this License for that Work shall terminate
|
| 87 |
+
as of the date such litigation is filed.
|
| 88 |
+
|
| 89 |
+
4. Redistribution. You may reproduce and distribute copies of the
|
| 90 |
+
Work or Derivative Works thereof in any medium, with or without
|
| 91 |
+
modifications, and in Source or Object form, provided that You
|
| 92 |
+
meet the following conditions:
|
| 93 |
+
|
| 94 |
+
(a) You must give any other recipients of the Work or
|
| 95 |
+
Derivative Works a copy of this License; and
|
| 96 |
+
|
| 97 |
+
(b) You must cause any modified files to carry prominent notices
|
| 98 |
+
stating that You changed the files; and
|
| 99 |
+
|
| 100 |
+
(c) You must retain, in the Source form of any Derivative Works
|
| 101 |
+
that You distribute, all copyright, patent, trademark, and
|
| 102 |
+
attribution notices from the Source form of the Work,
|
| 103 |
+
excluding those notices that do not pertain to any part of
|
| 104 |
+
the Derivative Works; and
|
| 105 |
+
|
| 106 |
+
(d) If the Work includes a "NOTICE" text file as part of its
|
| 107 |
+
distribution, then any Derivative Works that You distribute must
|
| 108 |
+
include a readable copy of the attribution notices contained
|
| 109 |
+
within such NOTICE file, excluding those notices that do not
|
| 110 |
+
pertain to any part of the Derivative Works, in at least one
|
| 111 |
+
of the following places: within a NOTICE text file distributed
|
| 112 |
+
as part of the Derivative Works; within the Source form or
|
| 113 |
+
documentation, if provided along with the Derivative Works; or,
|
| 114 |
+
within a display generated by the Derivative Works, if and
|
| 115 |
+
wherever such third-party notices normally appear. The contents
|
| 116 |
+
of the NOTICE file are for informational purposes only and
|
| 117 |
+
do not modify the License. You may add Your own attribution
|
| 118 |
+
notices within Derivative Works that You distribute, alongside
|
| 119 |
+
or as an addendum to the NOTICE text from the Work, provided
|
| 120 |
+
that such additional attribution notices cannot be construed
|
| 121 |
+
as modifying the License.
|
| 122 |
+
|
| 123 |
+
You may add Your own copyright statement to Your modifications and
|
| 124 |
+
may provide additional or different license terms and conditions
|
| 125 |
+
for use, reproduction, or distribution of Your modifications, or
|
| 126 |
+
for any such Derivative Works as a whole, provided Your use,
|
| 127 |
+
reproduction, and distribution of the Work otherwise complies with
|
| 128 |
+
the conditions stated in this License.
|
| 129 |
+
|
| 130 |
+
5. Submission of Contributions. Unless You explicitly state otherwise,
|
| 131 |
+
any Contribution intentionally submitted for inclusion in the Work
|
| 132 |
+
by You to the Licensor shall be under the terms and conditions of
|
| 133 |
+
this License, without any additional terms or conditions.
|
| 134 |
+
Notwithstanding the above, nothing herein shall supersede or modify
|
| 135 |
+
the terms of any separate license agreement you may have executed
|
| 136 |
+
with Licensor regarding such Contributions.
|
| 137 |
+
|
| 138 |
+
6. Trademarks. This License does not grant permission to use the trade
|
| 139 |
+
names, trademarks, service marks, or product names of the Licensor,
|
| 140 |
+
except as required for reasonable and customary use in describing the
|
| 141 |
+
origin of the Work and reproducing the content of the NOTICE file.
|
| 142 |
+
|
| 143 |
+
7. Disclaimer of Warranty. Unless required by applicable law or
|
| 144 |
+
agreed to in writing, Licensor provides the Work (and each
|
| 145 |
+
Contributor provides its Contributions) on an "AS IS" BASIS,
|
| 146 |
+
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
|
| 147 |
+
implied, including, without limitation, any warranties or conditions
|
| 148 |
+
of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
|
| 149 |
+
PARTICULAR PURPOSE. You are solely responsible for determining the
|
| 150 |
+
appropriateness of using or redistributing the Work and assume any
|
| 151 |
+
risks associated with Your exercise of permissions under this License.
|
| 152 |
+
|
| 153 |
+
8. Limitation of Liability. In no event and under no legal theory,
|
| 154 |
+
whether in tort (including negligence), contract, or otherwise,
|
| 155 |
+
unless required by applicable law (such as deliberate and grossly
|
| 156 |
+
negligent acts) or agreed to in writing, shall any Contributor be
|
| 157 |
+
liable to You for damages, including any direct, indirect, special,
|
| 158 |
+
incidental, or consequential damages of any character arising as a
|
| 159 |
+
result of this License or out of the use or inability to use the
|
| 160 |
+
Work (including but not limited to damages for loss of goodwill,
|
| 161 |
+
work stoppage, computer failure or malfunction, or any and all
|
| 162 |
+
other commercial damages or losses), even if such Contributor
|
| 163 |
+
has been advised of the possibility of such damages.
|
| 164 |
+
|
| 165 |
+
9. Accepting Warranty or Additional Liability. While redistributing
|
| 166 |
+
the Work or Derivative Works thereof, You may choose to offer,
|
| 167 |
+
and charge a fee for, acceptance of support, warranty, indemnity,
|
| 168 |
+
or other liability obligations and/or rights consistent with this
|
| 169 |
+
License. However, in accepting such obligations, You may act only
|
| 170 |
+
on Your own behalf and on Your sole responsibility, not on behalf
|
| 171 |
+
of any other Contributor, and only if You agree to indemnify,
|
| 172 |
+
defend, and hold each Contributor harmless for any liability
|
| 173 |
+
incurred by, or claims asserted against, such Contributor by reason
|
| 174 |
+
of your accepting any such warranty or additional liability.
|
| 175 |
+
|
| 176 |
+
END OF TERMS AND CONDITIONS
|
| 177 |
+
|
| 178 |
+
APPENDIX: How to apply the Apache License to your work.
|
| 179 |
+
|
| 180 |
+
To apply the Apache License to your work, attach the following
|
| 181 |
+
boilerplate notice, with the fields enclosed by brackets "[]"
|
| 182 |
+
replaced with your own identifying information. (Don't include
|
| 183 |
+
the brackets!) The text should be enclosed in the appropriate
|
| 184 |
+
comment syntax for the file format. We also recommend that a
|
| 185 |
+
file or class name and description of purpose be included on the
|
| 186 |
+
same "printed page" as the copyright notice for easier
|
| 187 |
+
identification within third-party archives.
|
| 188 |
+
|
| 189 |
+
Copyright [yyyy] [name of copyright owner]
|
| 190 |
+
|
| 191 |
+
Licensed under the Apache License, Version 2.0 (the "License");
|
| 192 |
+
you may not use this file except in compliance with the License.
|
| 193 |
+
You may obtain a copy of the License at
|
| 194 |
+
|
| 195 |
+
http://www.apache.org/licenses/LICENSE-2.0
|
| 196 |
+
|
| 197 |
+
Unless required by applicable law or agreed to in writing, software
|
| 198 |
+
distributed under the License is distributed on an "AS IS" BASIS,
|
| 199 |
+
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
| 200 |
+
See the License for the specific language governing permissions and
|
| 201 |
+
limitations under the License.
|
README.md
ADDED
|
@@ -0,0 +1,460 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
# 🌋 LLaVA: Large Language and Vision Assistant
|
| 2 |
+
|
| 3 |
+
*Visual instruction tuning towards large language and vision models with GPT-4 level capabilities.*
|
| 4 |
+
|
| 5 |
+
[📢 [LLaVA-NeXT Blog](https://llava-vl.github.io/blog/2024-01-30-llava-next/)] [[Project Page](https://llava-vl.github.io/)] [[Demo](https://llava.hliu.cc/)] [[Data](https://github.com/haotian-liu/LLaVA/blob/main/docs/Data.md)] [[Model Zoo](https://github.com/haotian-liu/LLaVA/blob/main/docs/MODEL_ZOO.md)]
|
| 6 |
+
|
| 7 |
+
🤝Community Contributions: [[llama.cpp](https://github.com/ggerganov/llama.cpp/pull/3436)] [[Colab](https://github.com/camenduru/LLaVA-colab)] [[🤗Space](https://huggingface.co/spaces/badayvedat/LLaVA)] [[Replicate](https://replicate.com/yorickvp/llava-13b)] [[AutoGen](https://github.com/microsoft/autogen/blob/main/notebook/agentchat_lmm_llava.ipynb)] [[BakLLaVA](https://github.com/SkunkworksAI/BakLLaVA)]
|
| 8 |
+
|
| 9 |
+
**Improved Baselines with Visual Instruction Tuning** [[Paper](https://arxiv.org/abs/2310.03744)] [[HF](https://huggingface.co/papers/2310.03744)] <br>
|
| 10 |
+
[Haotian Liu](https://hliu.cc), [Chunyuan Li](https://chunyuan.li/), [Yuheng Li](https://yuheng-li.github.io/), [Yong Jae Lee](https://pages.cs.wisc.edu/~yongjaelee/)
|
| 11 |
+
|
| 12 |
+
**Visual Instruction Tuning** (NeurIPS 2023, **Oral**) [[Paper](https://arxiv.org/abs/2304.08485)] [[HF](https://huggingface.co/papers/2304.08485)] <br>
|
| 13 |
+
[Haotian Liu*](https://hliu.cc), [Chunyuan Li*](https://chunyuan.li/), [Qingyang Wu](https://scholar.google.ca/citations?user=HDiw-TsAAAAJ&hl=en/), [Yong Jae Lee](https://pages.cs.wisc.edu/~yongjaelee/) (*Equal Contribution)
|
| 14 |
+
|
| 15 |
+
<!--p align="center">
|
| 16 |
+
<a href="https://llava.hliu.cc/"><img src="images/llava_logo.png" width="50%"></a> <br>
|
| 17 |
+
Generated by <a href="https://gligen.github.io/">GLIGEN</a> via "a cute lava llama with glasses" and box prompt
|
| 18 |
+
</p-->
|
| 19 |
+
|
| 20 |
+
|
| 21 |
+
## Release
|
| 22 |
+
- [03/10] Releasing **LMMs-Eval**, a highly efficient evaluation pipeline we used when developing LLaVA-NeXT. It supports the evaluation of LMMs on dozens of public datasets and allows new dataset onboarding, making the dev of new LMMs much faster. [[Blog](https://lmms-lab.github.io/lmms-eval-blog/lmms-eval-0.1/)] [[Codebase](https://github.com/EvolvingLMMs-Lab/lmms-eval)]
|
| 23 |
+
- [1/30] 🔥 LLaVA-NeXT (LLaVA-1.6) is out! With additional scaling to LLaVA-1.5, LLaVA-NeXT-34B outperforms Gemini Pro on some benchmarks. It can now process 4x more pixels and perform more tasks/applications than before. Check out the [blog post](https://llava-vl.github.io/blog/2024-01-30-llava-next/), and explore the [demo](https://llava.hliu.cc/)! Models are available in [Model Zoo](https://github.com/haotian-liu/LLaVA/blob/main/docs/MODEL_ZOO.md). Training/eval data and scripts coming soon.
|
| 24 |
+
- [11/10] [LLaVA-Plus](https://llava-vl.github.io/llava-plus/) is released: Learning to Use Tools for Creating Multimodal Agents, with LLaVA-Plus (LLaVA that Plug and Learn to Use Skills). [[Project Page](https://llava-vl.github.io/llava-plus/)] [[Demo](https://llavaplus.ngrok.io/)] [[Code](https://github.com/LLaVA-VL/LLaVA-Plus-Codebase)] [[Paper](https://arxiv.org/abs/2311.05437)]
|
| 25 |
+
- [11/2] [LLaVA-Interactive](https://llava-vl.github.io/llava-interactive/) is released: Experience the future of human-AI multimodal interaction with an all-in-one demo for Image Chat, Segmentation, Generation and Editing. [[Project Page](https://llava-vl.github.io/llava-interactive/)] [[Demo](https://llavainteractive.ngrok.io/)] [[Code](https://github.com/LLaVA-VL/LLaVA-Interactive-Demo)] [[Paper](https://arxiv.org/abs/2311.00571)]
|
| 26 |
+
- [10/26] 🔥 LLaVA-1.5 with LoRA achieves comparable performance as full-model finetuning, with a reduced GPU RAM requirement ([ckpts](https://github.com/haotian-liu/LLaVA/blob/main/docs/MODEL_ZOO.md#llava-v15), [script](https://github.com/haotian-liu/LLaVA#train)). We also provide a [doc](https://github.com/haotian-liu/LLaVA/blob/main/docs/Finetune_Custom_Data.md) on how to finetune LLaVA-1.5 on your own dataset with LoRA.
|
| 27 |
+
- [10/12] Check out the Korean LLaVA (Ko-LLaVA), created by ETRI, who has generously supported our research! [[🤗 Demo](https://huggingface.co/spaces/etri-vilab/Ko-LLaVA)]
|
| 28 |
+
- [10/5] 🔥 LLaVA-1.5 is out! Achieving SoTA on 11 benchmarks, with just simple modifications to the original LLaVA, utilizes all public data, completes training in ~1 day on a single 8-A100 node, and surpasses methods like Qwen-VL-Chat that use billion-scale data. Check out the [technical report](https://arxiv.org/abs/2310.03744), and explore the [demo](https://llava.hliu.cc/)! Models are available in [Model Zoo](https://github.com/haotian-liu/LLaVA/blob/main/docs/MODEL_ZOO.md). The training data and scripts of LLaVA-1.5 are released [here](https://github.com/haotian-liu/LLaVA#train), and evaluation scripts are released [here](https://github.com/haotian-liu/LLaVA/blob/main/docs/Evaluation.md)!
|
| 29 |
+
- [9/26] LLaVA is improved with reinforcement learning from human feedback (RLHF) to improve fact grounding and reduce hallucination. Check out the new SFT and RLHF checkpoints at project [[LLavA-RLHF]](https://llava-rlhf.github.io/)
|
| 30 |
+
- [9/22] [LLaVA](https://arxiv.org/abs/2304.08485) is accepted by NeurIPS 2023 as **oral presentation**, and [LLaVA-Med](https://arxiv.org/abs/2306.00890) is accepted by NeurIPS 2023 Datasets and Benchmarks Track as **spotlight presentation**.
|
| 31 |
+
|
| 32 |
+
<details>
|
| 33 |
+
<summary>More</summary>
|
| 34 |
+
|
| 35 |
+
- [11/6] Support **Intel** dGPU and CPU platforms. [More details here.](https://github.com/haotian-liu/LLaVA/tree/intel/docs/intel)
|
| 36 |
+
- [10/12] LLaVA is now supported in [llama.cpp](https://github.com/ggerganov/llama.cpp/pull/3436) with 4-bit / 5-bit quantization support!
|
| 37 |
+
- [10/11] The training data and scripts of LLaVA-1.5 are released [here](https://github.com/haotian-liu/LLaVA#train), and evaluation scripts are released [here](https://github.com/haotian-liu/LLaVA/blob/main/docs/Evaluation.md)!
|
| 38 |
+
- [10/10] [Roboflow Deep Dive](https://blog.roboflow.com/first-impressions-with-llava-1-5/): First Impressions with LLaVA-1.5.
|
| 39 |
+
- [9/20] We summarize our empirical study of training 33B and 65B LLaVA models in a [note](https://arxiv.org/abs/2309.09958). Further, if you are interested in the comprehensive review, evolution and trend of multimodal foundation models, please check out our recent survey paper [``Multimodal Foundation Models: From Specialists to General-Purpose Assistants''.](https://arxiv.org/abs/2309.10020)
|
| 40 |
+
<p align="center">
|
| 41 |
+
<img src="https://github.com/Computer-Vision-in-the-Wild/CVinW_Readings/blob/main/images/mfm_evolution.jpeg?raw=true" width=50%/>
|
| 42 |
+
</p>
|
| 43 |
+
|
| 44 |
+
- [7/19] 🔥 We release a major upgrade, including support for LLaMA-2, LoRA training, 4-/8-bit inference, higher resolution (336x336), and a lot more. We release [LLaVA Bench](https://github.com/haotian-liu/LLaVA/blob/main/docs/LLaVA_Bench.md) for benchmarking open-ended visual chat with results from Bard and Bing-Chat. We also support and verify training with RTX 3090 and RTX A6000. Check out [LLaVA-from-LLaMA-2](https://github.com/haotian-liu/LLaVA/blob/main/docs/LLaVA_from_LLaMA2.md), and our [model zoo](https://github.com/haotian-liu/LLaVA/blob/main/docs/MODEL_ZOO.md)!
|
| 45 |
+
- [6/26] [CVPR 2023 Tutorial](https://vlp-tutorial.github.io/) on **Large Multimodal Models: Towards Building and Surpassing Multimodal GPT-4**! Please check out [[Slides](https://datarelease.blob.core.windows.net/tutorial/vision_foundation_models_2023/slides/Chunyuan_cvpr2023_tutorial_lmm.pdf)] [[Notes](https://arxiv.org/abs/2306.14895)] [[YouTube](https://youtu.be/mkI7EPD1vp8)] [[Bilibli](https://www.bilibili.com/video/BV1Ng4y1T7v3/)].
|
| 46 |
+
- [6/11] We released the preview for the most requested feature: DeepSpeed and LoRA support! Please see documentations [here](./docs/LoRA.md).
|
| 47 |
+
- [6/1] We released **LLaVA-Med: Large Language and Vision Assistant for Biomedicine**, a step towards building biomedical domain large language and vision models with GPT-4 level capabilities. Checkout the [paper](https://arxiv.org/abs/2306.00890) and [page](https://github.com/microsoft/LLaVA-Med).
|
| 48 |
+
- [5/6] We are releasing [LLaVA-Lighting-MPT-7B-preview](https://huggingface.co/liuhaotian/LLaVA-Lightning-MPT-7B-preview), based on MPT-7B-Chat! See [here](#LLaVA-MPT-7b) for more details.
|
| 49 |
+
- [5/2] 🔥 We are releasing LLaVA-Lighting! Train a lite, multimodal GPT-4 with just $40 in 3 hours! See [here](#train-llava-lightning) for more details.
|
| 50 |
+
- [4/27] Thanks to the community effort, LLaVA-13B with 4-bit quantization allows you to run on a GPU with as few as 12GB VRAM! Try it out [here](https://github.com/oobabooga/text-generation-webui/tree/main/extensions/llava).
|
| 51 |
+
- [4/17] 🔥 We released **LLaVA: Large Language and Vision Assistant**. We propose visual instruction tuning, towards building large language and vision models with GPT-4 level capabilities. Checkout the [paper](https://arxiv.org/abs/2304.08485) and [demo](https://llava.hliu.cc/).
|
| 52 |
+
|
| 53 |
+
</details>
|
| 54 |
+
|
| 55 |
+
<!-- <a href="https://llava.hliu.cc/"><img src="assets/demo.gif" width="70%"></a> -->
|
| 56 |
+
|
| 57 |
+
[](https://github.com/tatsu-lab/stanford_alpaca/blob/main/LICENSE)
|
| 58 |
+
**Usage and License Notices**: This project utilizes certain datasets and checkpoints that are subject to their respective original licenses. Users must comply with all terms and conditions of these original licenses, including but not limited to the [OpenAI Terms of Use](https://openai.com/policies/terms-of-use) for the dataset and the specific licenses for base language models for checkpoints trained using the dataset (e.g. [Llama community license](https://ai.meta.com/llama/license/) for LLaMA-2 and Vicuna-v1.5). This project does not impose any additional constraints beyond those stipulated in the original licenses. Furthermore, users are reminded to ensure that their use of the dataset and checkpoints is in compliance with all applicable laws and regulations.
|
| 59 |
+
|
| 60 |
+
|
| 61 |
+
## Contents
|
| 62 |
+
- [Install](#install)
|
| 63 |
+
- [LLaVA Weights](#llava-weights)
|
| 64 |
+
- [Demo](#Demo)
|
| 65 |
+
- [Model Zoo](https://github.com/haotian-liu/LLaVA/blob/main/docs/MODEL_ZOO.md)
|
| 66 |
+
- [Dataset](https://github.com/haotian-liu/LLaVA/blob/main/docs/Data.md)
|
| 67 |
+
- [Train](#train)
|
| 68 |
+
- [Evaluation](#evaluation)
|
| 69 |
+
|
| 70 |
+
## Install
|
| 71 |
+
|
| 72 |
+
If you are not using Linux, do *NOT* proceed, see instructions for [macOS](https://github.com/haotian-liu/LLaVA/blob/main/docs/macOS.md) and [Windows](https://github.com/haotian-liu/LLaVA/blob/main/docs/Windows.md).
|
| 73 |
+
|
| 74 |
+
1. Clone this repository and navigate to LLaVA folder
|
| 75 |
+
```bash
|
| 76 |
+
git clone https://github.com/haotian-liu/LLaVA.git
|
| 77 |
+
cd LLaVA
|
| 78 |
+
```
|
| 79 |
+
|
| 80 |
+
2. Install Package
|
| 81 |
+
```Shell
|
| 82 |
+
conda create -n llava python=3.10 -y
|
| 83 |
+
conda activate llava
|
| 84 |
+
pip install --upgrade pip # enable PEP 660 support
|
| 85 |
+
pip install -e .
|
| 86 |
+
```
|
| 87 |
+
|
| 88 |
+
3. Install additional packages for training cases
|
| 89 |
+
```
|
| 90 |
+
pip install -e ".[train]"
|
| 91 |
+
pip install flash-attn --no-build-isolation
|
| 92 |
+
```
|
| 93 |
+
|
| 94 |
+
### Upgrade to latest code base
|
| 95 |
+
|
| 96 |
+
```Shell
|
| 97 |
+
git pull
|
| 98 |
+
pip install -e .
|
| 99 |
+
|
| 100 |
+
# if you see some import errors when you upgrade,
|
| 101 |
+
# please try running the command below (without #)
|
| 102 |
+
# pip install flash-attn --no-build-isolation --no-cache-dir
|
| 103 |
+
```
|
| 104 |
+
|
| 105 |
+
### Quick Start With HuggingFace
|
| 106 |
+
|
| 107 |
+
<details>
|
| 108 |
+
<summary>Example Code</summary>
|
| 109 |
+
|
| 110 |
+
```Python
|
| 111 |
+
from llava.model.builder import load_pretrained_model
|
| 112 |
+
from llava.mm_utils import get_model_name_from_path
|
| 113 |
+
from llava.eval.run_llava import eval_model
|
| 114 |
+
|
| 115 |
+
model_path = "liuhaotian/llava-v1.5-7b"
|
| 116 |
+
|
| 117 |
+
tokenizer, model, image_processor, context_len = load_pretrained_model(
|
| 118 |
+
model_path=model_path,
|
| 119 |
+
model_base=None,
|
| 120 |
+
model_name=get_model_name_from_path(model_path)
|
| 121 |
+
)
|
| 122 |
+
```
|
| 123 |
+
|
| 124 |
+
Check out the details wth the `load_pretrained_model` function in `llava/model/builder.py`.
|
| 125 |
+
|
| 126 |
+
You can also use the `eval_model` function in `llava/eval/run_llava.py` to get the output easily. By doing so, you can use this code on Colab directly after downloading this repository.
|
| 127 |
+
|
| 128 |
+
``` python
|
| 129 |
+
model_path = "liuhaotian/llava-v1.5-7b"
|
| 130 |
+
prompt = "What are the things I should be cautious about when I visit here?"
|
| 131 |
+
image_file = "https://llava-vl.github.io/static/images/view.jpg"
|
| 132 |
+
|
| 133 |
+
args = type('Args', (), {
|
| 134 |
+
"model_path": model_path,
|
| 135 |
+
"model_base": None,
|
| 136 |
+
"model_name": get_model_name_from_path(model_path),
|
| 137 |
+
"query": prompt,
|
| 138 |
+
"conv_mode": None,
|
| 139 |
+
"image_file": image_file,
|
| 140 |
+
"sep": ",",
|
| 141 |
+
"temperature": 0,
|
| 142 |
+
"top_p": None,
|
| 143 |
+
"num_beams": 1,
|
| 144 |
+
"max_new_tokens": 512
|
| 145 |
+
})()
|
| 146 |
+
|
| 147 |
+
eval_model(args)
|
| 148 |
+
```
|
| 149 |
+
</details>
|
| 150 |
+
|
| 151 |
+
## LLaVA Weights
|
| 152 |
+
Please check out our [Model Zoo](https://github.com/haotian-liu/LLaVA/blob/main/docs/MODEL_ZOO.md) for all public LLaVA checkpoints, and the instructions of how to use the weights.
|
| 153 |
+
|
| 154 |
+
## Demo
|
| 155 |
+
|
| 156 |
+
### Gradio Web UI
|
| 157 |
+
|
| 158 |
+
To launch a Gradio demo locally, please run the following commands one by one. If you plan to launch multiple model workers to compare between different checkpoints, you only need to launch the controller and the web server *ONCE*.
|
| 159 |
+
|
| 160 |
+
```mermaid
|
| 161 |
+
flowchart BT
|
| 162 |
+
%% Declare Nodes
|
| 163 |
+
gws("Gradio (UI Server)")
|
| 164 |
+
c("Controller (API Server):<br/>PORT: 10000")
|
| 165 |
+
mw7b("Model Worker:<br/>llava-v1.5-7b<br/>PORT: 40000")
|
| 166 |
+
mw13b("Model Worker:<br/>llava-v1.5-13b<br/>PORT: 40001")
|
| 167 |
+
sglw13b("SGLang Backend:<br/>llava-v1.6-34b<br/>http://localhost:30000")
|
| 168 |
+
lsglw13b("SGLang Worker:<br/>llava-v1.6-34b<br/>PORT: 40002")
|
| 169 |
+
|
| 170 |
+
%% Declare Styles
|
| 171 |
+
classDef data fill:#3af,stroke:#48a,stroke-width:2px,color:#444
|
| 172 |
+
classDef success fill:#8f8,stroke:#0a0,stroke-width:2px,color:#444
|
| 173 |
+
classDef failure fill:#f88,stroke:#f00,stroke-width:2px,color:#444
|
| 174 |
+
|
| 175 |
+
%% Assign Styles
|
| 176 |
+
class id,od data;
|
| 177 |
+
class cimg,cs_s,scsim_s success;
|
| 178 |
+
class ncimg,cs_f,scsim_f failure;
|
| 179 |
+
|
| 180 |
+
subgraph Demo Connections
|
| 181 |
+
direction BT
|
| 182 |
+
c<-->gws
|
| 183 |
+
|
| 184 |
+
mw7b<-->c
|
| 185 |
+
mw13b<-->c
|
| 186 |
+
lsglw13b<-->c
|
| 187 |
+
sglw13b<-->lsglw13b
|
| 188 |
+
end
|
| 189 |
+
```
|
| 190 |
+
|
| 191 |
+
#### Launch a controller
|
| 192 |
+
```Shell
|
| 193 |
+
python -m llava.serve.controller --host 0.0.0.0 --port 10000
|
| 194 |
+
```
|
| 195 |
+
|
| 196 |
+
#### Launch a gradio web server.
|
| 197 |
+
```Shell
|
| 198 |
+
python -m llava.serve.gradio_web_server --controller http://localhost:10000 --model-list-mode reload
|
| 199 |
+
```
|
| 200 |
+
You just launched the Gradio web interface. Now, you can open the web interface with the URL printed on the screen. You may notice that there is no model in the model list. Do not worry, as we have not launched any model worker yet. It will be automatically updated when you launch a model worker.
|
| 201 |
+
|
| 202 |
+
#### Launch a SGLang worker
|
| 203 |
+
|
| 204 |
+
This is the recommended way to serve LLaVA model with high throughput, and you need to install SGLang first. Note that currently `4-bit` quantization is not supported yet on SGLang-LLaVA, and if you have limited GPU VRAM, please check out model worker with [quantization](https://github.com/haotian-liu/LLaVA?tab=readme-ov-file#launch-a-model-worker-4-bit-8-bit-inference-quantized).
|
| 205 |
+
|
| 206 |
+
```Shell
|
| 207 |
+
pip install "sglang[all]"
|
| 208 |
+
```
|
| 209 |
+
|
| 210 |
+
You'll first launch a SGLang backend worker which will execute the models on GPUs. Remember the `--port` you've set and you'll use that later.
|
| 211 |
+
|
| 212 |
+
```Shell
|
| 213 |
+
# Single GPU
|
| 214 |
+
CUDA_VISIBLE_DEVICES=0 python3 -m sglang.launch_server --model-path liuhaotian/llava-v1.5-7b --tokenizer-path llava-hf/llava-1.5-7b-hf --port 30000
|
| 215 |
+
|
| 216 |
+
# Multiple GPUs with tensor parallel
|
| 217 |
+
CUDA_VISIBLE_DEVICES=0,1 python3 -m sglang.launch_server --model-path liuhaotian/llava-v1.5-13b --tokenizer-path llava-hf/llava-1.5-13b-hf --port 30000 --tp 2
|
| 218 |
+
```
|
| 219 |
+
|
| 220 |
+
Tokenizers (temporary): `llava-hf/llava-1.5-7b-hf`, `llava-hf/llava-1.5-13b-hf`, `liuhaotian/llava-v1.6-34b-tokenizer`.
|
| 221 |
+
|
| 222 |
+
You'll then launch a LLaVA-SGLang worker that will communicate between LLaVA controller and SGLang backend to route the requests. Set `--sgl-endpoint` to `http://127.0.0.1:port` where `port` is the one you just set (default: 30000).
|
| 223 |
+
|
| 224 |
+
```Shell
|
| 225 |
+
python -m llava.serve.sglang_worker --host 0.0.0.0 --controller http://localhost:10000 --port 40000 --worker http://localhost:40000 --sgl-endpoint http://127.0.0.1:30000
|
| 226 |
+
```
|
| 227 |
+
|
| 228 |
+
#### Launch a model worker
|
| 229 |
+
|
| 230 |
+
This is the actual *worker* that performs the inference on the GPU. Each worker is responsible for a single model specified in `--model-path`.
|
| 231 |
+
|
| 232 |
+
```Shell
|
| 233 |
+
python -m llava.serve.model_worker --host 0.0.0.0 --controller http://localhost:10000 --port 40000 --worker http://localhost:40000 --model-path liuhaotian/llava-v1.5-13b
|
| 234 |
+
```
|
| 235 |
+
Wait until the process finishes loading the model and you see "Uvicorn running on ...". Now, refresh your Gradio web UI, and you will see the model you just launched in the model list.
|
| 236 |
+
|
| 237 |
+
You can launch as many workers as you want, and compare between different model checkpoints in the same Gradio interface. Please keep the `--controller` the same, and modify the `--port` and `--worker` to a different port number for each worker.
|
| 238 |
+
```Shell
|
| 239 |
+
python -m llava.serve.model_worker --host 0.0.0.0 --controller http://localhost:10000 --port <different from 40000, say 40001> --worker http://localhost:<change accordingly, i.e. 40001> --model-path <ckpt2>
|
| 240 |
+
```
|
| 241 |
+
|
| 242 |
+
If you are using an Apple device with an M1 or M2 chip, you can specify the mps device by using the `--device` flag: `--device mps`.
|
| 243 |
+
|
| 244 |
+
#### Launch a model worker (Multiple GPUs, when GPU VRAM <= 24GB)
|
| 245 |
+
|
| 246 |
+
If the VRAM of your GPU is less than 24GB (e.g., RTX 3090, RTX 4090, etc.), you may try running it with multiple GPUs. Our latest code base will automatically try to use multiple GPUs if you have more than one GPU. You can specify which GPUs to use with `CUDA_VISIBLE_DEVICES`. Below is an example of running with the first two GPUs.
|
| 247 |
+
|
| 248 |
+
```Shell
|
| 249 |
+
CUDA_VISIBLE_DEVICES=0,1 python -m llava.serve.model_worker --host 0.0.0.0 --controller http://localhost:10000 --port 40000 --worker http://localhost:40000 --model-path liuhaotian/llava-v1.5-13b
|
| 250 |
+
```
|
| 251 |
+
|
| 252 |
+
#### Launch a model worker (4-bit, 8-bit inference, quantized)
|
| 253 |
+
|
| 254 |
+
You can launch the model worker with quantized bits (4-bit, 8-bit), which allows you to run the inference with reduced GPU memory footprint, potentially allowing you to run on a GPU with as few as 12GB VRAM. Note that inference with quantized bits may not be as accurate as the full-precision model. Simply append `--load-4bit` or `--load-8bit` to the **model worker** command that you are executing. Below is an example of running with 4-bit quantization.
|
| 255 |
+
|
| 256 |
+
```Shell
|
| 257 |
+
python -m llava.serve.model_worker --host 0.0.0.0 --controller http://localhost:10000 --port 40000 --worker http://localhost:40000 --model-path liuhaotian/llava-v1.5-13b --load-4bit
|
| 258 |
+
```
|
| 259 |
+
|
| 260 |
+
#### Launch a model worker (LoRA weights, unmerged)
|
| 261 |
+
|
| 262 |
+
You can launch the model worker with LoRA weights, without merging them with the base checkpoint, to save disk space. There will be additional loading time, while the inference speed is the same as the merged checkpoints. Unmerged LoRA checkpoints do not have `lora-merge` in the model name, and are usually much smaller (less than 1GB) than the merged checkpoints (13G for 7B, and 25G for 13B).
|
| 263 |
+
|
| 264 |
+
To load unmerged LoRA weights, you simply need to pass an additional argument `--model-base`, which is the base LLM that is used to train the LoRA weights. You can check the base LLM of each LoRA weights in the [model zoo](https://github.com/haotian-liu/LLaVA/blob/main/docs/MODEL_ZOO.md).
|
| 265 |
+
|
| 266 |
+
```Shell
|
| 267 |
+
python -m llava.serve.model_worker --host 0.0.0.0 --controller http://localhost:10000 --port 40000 --worker http://localhost:40000 --model-path liuhaotian/llava-v1-0719-336px-lora-vicuna-13b-v1.3 --model-base lmsys/vicuna-13b-v1.3
|
| 268 |
+
```
|
| 269 |
+
|
| 270 |
+
### CLI Inference
|
| 271 |
+
|
| 272 |
+
Chat about images using LLaVA without the need of Gradio interface. It also supports multiple GPUs, 4-bit and 8-bit quantized inference. With 4-bit quantization, for our LLaVA-1.5-7B, it uses less than 8GB VRAM on a single GPU.
|
| 273 |
+
|
| 274 |
+
```Shell
|
| 275 |
+
python -m llava.serve.cli \
|
| 276 |
+
--model-path liuhaotian/llava-v1.5-7b \
|
| 277 |
+
--image-file "https://llava-vl.github.io/static/images/view.jpg" \
|
| 278 |
+
--load-4bit
|
| 279 |
+
```
|
| 280 |
+
|
| 281 |
+
<img src="images/demo_cli.gif" width="70%">
|
| 282 |
+
|
| 283 |
+
## Train
|
| 284 |
+
|
| 285 |
+
*Below is the latest training configuration for LLaVA v1.5. For legacy models, please refer to README of [this](https://github.com/haotian-liu/LLaVA/tree/v1.0.1) version for now. We'll add them in a separate doc later.*
|
| 286 |
+
|
| 287 |
+
LLaVA training consists of two stages: (1) feature alignment stage: use our 558K subset of the LAION-CC-SBU dataset to connect a *frozen pretrained* vision encoder to a *frozen LLM*; (2) visual instruction tuning stage: use 150K GPT-generated multimodal instruction-following data, plus around 515K VQA data from academic-oriented tasks, to teach the model to follow multimodal instructions.
|
| 288 |
+
|
| 289 |
+
LLaVA is trained on 8 A100 GPUs with 80GB memory. To train on fewer GPUs, you can reduce the `per_device_train_batch_size` and increase the `gradient_accumulation_steps` accordingly. Always keep the global batch size the same: `per_device_train_batch_size` x `gradient_accumulation_steps` x `num_gpus`.
|
| 290 |
+
|
| 291 |
+
### Hyperparameters
|
| 292 |
+
We use a similar set of hyperparameters as Vicuna in finetuning. Both hyperparameters used in pretraining and finetuning are provided below.
|
| 293 |
+
|
| 294 |
+
1. Pretraining
|
| 295 |
+
|
| 296 |
+
| Hyperparameter | Global Batch Size | Learning rate | Epochs | Max length | Weight decay |
|
| 297 |
+
| --- | ---: | ---: | ---: | ---: | ---: |
|
| 298 |
+
| LLaVA-v1.5-13B | 256 | 1e-3 | 1 | 2048 | 0 |
|
| 299 |
+
|
| 300 |
+
2. Finetuning
|
| 301 |
+
|
| 302 |
+
| Hyperparameter | Global Batch Size | Learning rate | Epochs | Max length | Weight decay |
|
| 303 |
+
| --- | ---: | ---: | ---: | ---: | ---: |
|
| 304 |
+
| LLaVA-v1.5-13B | 128 | 2e-5 | 1 | 2048 | 0 |
|
| 305 |
+
|
| 306 |
+
### Download Vicuna checkpoints (automatically)
|
| 307 |
+
|
| 308 |
+
Our base model Vicuna v1.5, which is an instruction-tuned chatbot, will be downloaded automatically when you run our provided training scripts. No action is needed.
|
| 309 |
+
|
| 310 |
+
### Pretrain (feature alignment)
|
| 311 |
+
|
| 312 |
+
Please download the 558K subset of the LAION-CC-SBU dataset with BLIP captions we use in the paper [here](https://huggingface.co/datasets/liuhaotian/LLaVA-Pretrain).
|
| 313 |
+
|
| 314 |
+
Pretrain takes around 5.5 hours for LLaVA-v1.5-13B on 8x A100 (80G), due to the increased resolution to 336px. It takes around 3.5 hours for LLaVA-v1.5-7B.
|
| 315 |
+
|
| 316 |
+
Training script with DeepSpeed ZeRO-2: [`pretrain.sh`](https://github.com/haotian-liu/LLaVA/blob/main/scripts/v1_5/pretrain.sh).
|
| 317 |
+
|
| 318 |
+
- `--mm_projector_type mlp2x_gelu`: the two-layer MLP vision-language connector.
|
| 319 |
+
- `--vision_tower openai/clip-vit-large-patch14-336`: CLIP ViT-L/14 336px.
|
| 320 |
+
|
| 321 |
+
<details>
|
| 322 |
+
<summary>Pretrain takes around 20 hours for LLaVA-7B on 8x V100 (32G)</summary>
|
| 323 |
+
|
| 324 |
+
We provide training script with DeepSpeed [here](https://github.com/haotian-liu/LLaVA/blob/main/scripts/pretrain_xformers.sh).
|
| 325 |
+
Tips:
|
| 326 |
+
- If you are using V100 which is not supported by FlashAttention, you can use the [memory-efficient attention](https://arxiv.org/abs/2112.05682) implemented in [xFormers](https://github.com/facebookresearch/xformers). Install xformers and replace `llava/train/train_mem.py` above with [llava/train/train_xformers.py](llava/train/train_xformers.py).
|
| 327 |
+
</details>
|
| 328 |
+
|
| 329 |
+
### Visual Instruction Tuning
|
| 330 |
+
|
| 331 |
+
1. Prepare data
|
| 332 |
+
|
| 333 |
+
Please download the annotation of the final mixture our instruction tuning data [llava_v1_5_mix665k.json](https://huggingface.co/datasets/liuhaotian/LLaVA-Instruct-150K/blob/main/llava_v1_5_mix665k.json), and download the images from constituting datasets:
|
| 334 |
+
|
| 335 |
+
- COCO: [train2017](http://images.cocodataset.org/zips/train2017.zip)
|
| 336 |
+
- GQA: [images](https://downloads.cs.stanford.edu/nlp/data/gqa/images.zip)
|
| 337 |
+
- OCR-VQA: [download script](https://drive.google.com/drive/folders/1_GYPY5UkUy7HIcR0zq3ZCFgeZN7BAfm_?usp=sharing), **we save all files as `.jpg`**
|
| 338 |
+
- TextVQA: [train_val_images](https://dl.fbaipublicfiles.com/textvqa/images/train_val_images.zip)
|
| 339 |
+
- VisualGenome: [part1](https://cs.stanford.edu/people/rak248/VG_100K_2/images.zip), [part2](https://cs.stanford.edu/people/rak248/VG_100K_2/images2.zip)
|
| 340 |
+
|
| 341 |
+
After downloading all of them, organize the data as follows in `./playground/data`,
|
| 342 |
+
|
| 343 |
+
```
|
| 344 |
+
├── coco
|
| 345 |
+
│ └── train2017
|
| 346 |
+
├── gqa
|
| 347 |
+
│ └── images
|
| 348 |
+
├── ocr_vqa
|
| 349 |
+
│ └── images
|
| 350 |
+
├── textvqa
|
| 351 |
+
│ └── train_images
|
| 352 |
+
└── vg
|
| 353 |
+
├── VG_100K
|
| 354 |
+
└── VG_100K_2
|
| 355 |
+
```
|
| 356 |
+
|
| 357 |
+
2. Start training!
|
| 358 |
+
|
| 359 |
+
You may download our pretrained projectors in [Model Zoo](https://github.com/haotian-liu/LLaVA/blob/main/docs/MODEL_ZOO.md). It is not recommended to use legacy projectors, as they may be trained with a different version of the codebase, and if any option is off, the model will not function/train as we expected.
|
| 360 |
+
|
| 361 |
+
Visual instruction tuning takes around 20 hours for LLaVA-v1.5-13B on 8x A100 (80G), due to the increased resolution to 336px. It takes around 10 hours for LLaVA-v1.5-7B on 8x A100 (40G).
|
| 362 |
+
|
| 363 |
+
Training script with DeepSpeed ZeRO-3: [`finetune.sh`](https://github.com/haotian-liu/LLaVA/blob/main/scripts/v1_5/finetune.sh).
|
| 364 |
+
|
| 365 |
+
If you are do not have enough GPU memory:
|
| 366 |
+
|
| 367 |
+
- Use LoRA: [`finetune_lora.sh`](https://github.com/haotian-liu/LLaVA/blob/main/scripts/v1_5/finetune_lora.sh). We are able to fit 13B training in 8-A100-40G/8-A6000, and 7B training in 8-RTX3090. Make sure `per_device_train_batch_size*gradient_accumulation_steps` is the same as the provided script for best reproducibility.
|
| 368 |
+
- Replace `zero3.json` with `zero3_offload.json` which offloads some parameters to CPU RAM. This slows down the training speed.
|
| 369 |
+
|
| 370 |
+
If you are interested in finetuning LLaVA model to your own task/data, please check out [`Finetune_Custom_Data.md`](https://github.com/haotian-liu/LLaVA/blob/main/docs/Finetune_Custom_Data.md)。
|
| 371 |
+
|
| 372 |
+
New options to note:
|
| 373 |
+
|
| 374 |
+
- `--mm_projector_type mlp2x_gelu`: the two-layer MLP vision-language connector.
|
| 375 |
+
- `--vision_tower openai/clip-vit-large-patch14-336`: CLIP ViT-L/14 336px.
|
| 376 |
+
- `--image_aspect_ratio pad`: this pads the non-square images to square, instead of cropping them; it slightly reduces hallucination.
|
| 377 |
+
- `--group_by_modality_length True`: this should only be used when your instruction tuning dataset contains both language (e.g. ShareGPT) and multimodal (e.g. LLaVA-Instruct). It makes the training sampler only sample a single modality (either image or language) during training, which we observe to speed up training by ~25%, and does not affect the final outcome.
|
| 378 |
+
|
| 379 |
+
## Evaluation
|
| 380 |
+
|
| 381 |
+
In LLaVA-1.5, we evaluate models on a diverse set of 12 benchmarks. To ensure the reproducibility, we evaluate the models with greedy decoding. We do not evaluate using beam search to make the inference process consistent with the chat demo of real-time outputs.
|
| 382 |
+
|
| 383 |
+
See [Evaluation.md](https://github.com/haotian-liu/LLaVA/blob/main/docs/Evaluation.md).
|
| 384 |
+
|
| 385 |
+
### GPT-assisted Evaluation
|
| 386 |
+
|
| 387 |
+
Our GPT-assisted evaluation pipeline for multimodal modeling is provided for a comprehensive understanding of the capabilities of vision-language models. Please see our paper for more details.
|
| 388 |
+
|
| 389 |
+
1. Generate LLaVA responses
|
| 390 |
+
|
| 391 |
+
```Shell
|
| 392 |
+
python model_vqa.py \
|
| 393 |
+
--model-path ./checkpoints/LLaVA-13B-v0 \
|
| 394 |
+
--question-file \
|
| 395 |
+
playground/data/coco2014_val_qa_eval/qa90_questions.jsonl \
|
| 396 |
+
--image-folder \
|
| 397 |
+
/path/to/coco2014_val \
|
| 398 |
+
--answers-file \
|
| 399 |
+
/path/to/answer-file-our.jsonl
|
| 400 |
+
```
|
| 401 |
+
|
| 402 |
+
2. Evaluate the generated responses. In our case, [`answer-file-ref.jsonl`](./playground/data/coco2014_val_qa_eval/qa90_gpt4_answer.jsonl) is the response generated by text-only GPT-4 (0314), with the context captions/boxes provided.
|
| 403 |
+
|
| 404 |
+
```Shell
|
| 405 |
+
OPENAI_API_KEY="sk-***********************************" python llava/eval/eval_gpt_review_visual.py \
|
| 406 |
+
--question playground/data/coco2014_val_qa_eval/qa90_questions.jsonl \
|
| 407 |
+
--context llava/eval/table/caps_boxes_coco2014_val_80.jsonl \
|
| 408 |
+
--answer-list \
|
| 409 |
+
/path/to/answer-file-ref.jsonl \
|
| 410 |
+
/path/to/answer-file-our.jsonl \
|
| 411 |
+
--rule llava/eval/table/rule.json \
|
| 412 |
+
--output /path/to/review.json
|
| 413 |
+
```
|
| 414 |
+
|
| 415 |
+
3. Summarize the evaluation results
|
| 416 |
+
|
| 417 |
+
```Shell
|
| 418 |
+
python summarize_gpt_review.py
|
| 419 |
+
```
|
| 420 |
+
|
| 421 |
+
## Citation
|
| 422 |
+
|
| 423 |
+
If you find LLaVA useful for your research and applications, please cite using this BibTeX:
|
| 424 |
+
```bibtex
|
| 425 |
+
@misc{liu2024llavanext,
|
| 426 |
+
title={LLaVA-NeXT: Improved reasoning, OCR, and world knowledge},
|
| 427 |
+
url={https://llava-vl.github.io/blog/2024-01-30-llava-next/},
|
| 428 |
+
author={Liu, Haotian and Li, Chunyuan and Li, Yuheng and Li, Bo and Zhang, Yuanhan and Shen, Sheng and Lee, Yong Jae},
|
| 429 |
+
month={January},
|
| 430 |
+
year={2024}
|
| 431 |
+
}
|
| 432 |
+
|
| 433 |
+
@misc{liu2023improvedllava,
|
| 434 |
+
title={Improved Baselines with Visual Instruction Tuning},
|
| 435 |
+
author={Liu, Haotian and Li, Chunyuan and Li, Yuheng and Lee, Yong Jae},
|
| 436 |
+
publisher={arXiv:2310.03744},
|
| 437 |
+
year={2023},
|
| 438 |
+
}
|
| 439 |
+
|
| 440 |
+
@misc{liu2023llava,
|
| 441 |
+
title={Visual Instruction Tuning},
|
| 442 |
+
author={Liu, Haotian and Li, Chunyuan and Wu, Qingyang and Lee, Yong Jae},
|
| 443 |
+
publisher={NeurIPS},
|
| 444 |
+
year={2023},
|
| 445 |
+
}
|
| 446 |
+
```
|
| 447 |
+
|
| 448 |
+
## Acknowledgement
|
| 449 |
+
|
| 450 |
+
- [Vicuna](https://github.com/lm-sys/FastChat): the codebase we built upon, and our base model Vicuna-13B that has the amazing language capabilities!
|
| 451 |
+
|
| 452 |
+
## Related Projects
|
| 453 |
+
|
| 454 |
+
- [Instruction Tuning with GPT-4](https://github.com/Instruction-Tuning-with-GPT-4/GPT-4-LLM)
|
| 455 |
+
- [LLaVA-Med: Training a Large Language-and-Vision Assistant for Biomedicine in One Day](https://github.com/microsoft/LLaVA-Med)
|
| 456 |
+
- [Otter: In-Context Multi-Modal Instruction Tuning](https://github.com/Luodian/Otter)
|
| 457 |
+
|
| 458 |
+
For future project ideas, please check out:
|
| 459 |
+
- [SEEM: Segment Everything Everywhere All at Once](https://github.com/UX-Decoder/Segment-Everything-Everywhere-All-At-Once)
|
| 460 |
+
- [Grounded-Segment-Anything](https://github.com/IDEA-Research/Grounded-Segment-Anything) to detect, segment, and generate anything by marrying [Grounding DINO](https://github.com/IDEA-Research/GroundingDINO) and [Segment-Anything](https://github.com/facebookresearch/segment-anything).
|
accuracy_scores.txt
ADDED
|
@@ -0,0 +1,140 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
Model: /mnt/petrelfs/zhuchenglin/LLaVA/checkpoints_ft/coco_gen_558k/llava-v1.5-13b
|
| 2 |
+
Average accuracy: 0.477
|
| 3 |
+
--- End of model ---
|
| 4 |
+
|
| 5 |
+
Model: /mnt/petrelfs/zhuchenglin/LLaVA/checkpoints_ft/llava_raw_558k/llava-v1.5-13b
|
| 6 |
+
Average accuracy: 0.494
|
| 7 |
+
--- End of model ---
|
| 8 |
+
|
| 9 |
+
Model: /mnt/petrelfs/zhuchenglin/LLaVA/checkpoints_ft/llava_coco_gen_758k/llava-v1.5-13b
|
| 10 |
+
Average accuracy: 0.513
|
| 11 |
+
--- End of model ---
|
| 12 |
+
|
| 13 |
+
Model: /mnt/petrelfs/zhuchenglin/LLaVA/checkpoints_ft/llava_coco_raw_758k/llava-v1.5-13b
|
| 14 |
+
Average accuracy: 0.505
|
| 15 |
+
--- End of model ---
|
| 16 |
+
|
| 17 |
+
Model: /mnt/petrelfs/zhuchenglin/LLaVA/checkpoints_ft/coco_gen_200k/llava-v1.5-13b
|
| 18 |
+
Average accuracy: 0.501
|
| 19 |
+
--- End of model ---
|
| 20 |
+
|
| 21 |
+
Model: /mnt/petrelfs/zhuchenglin/LLaVA/checkpoints_ft/llava_gen_200k/llava-v1.5-13b
|
| 22 |
+
Average accuracy: 0.531
|
| 23 |
+
--- End of model ---
|
| 24 |
+
|
| 25 |
+
Model: /mnt/petrelfs/zhuchenglin/LLaVA/checkpoints_ft/coco_raw_200k/llava-v1.5-13b
|
| 26 |
+
Average accuracy: 0.508
|
| 27 |
+
--- End of model ---
|
| 28 |
+
|
| 29 |
+
Model: /mnt/petrelfs/zhuchenglin/LLaVA/checkpoints_ft/coco_raw_200k/llava-v1.5-13b
|
| 30 |
+
Average accuracy: 0.508
|
| 31 |
+
--- End of model ---
|
| 32 |
+
|
| 33 |
+
Model: /mnt/petrelfs/zhuchenglin/LLaVA/checkpoints_ft/coco_raw_558k/llava-v1.5-13b
|
| 34 |
+
Average accuracy: 0.520
|
| 35 |
+
--- End of model ---
|
| 36 |
+
|
| 37 |
+
Model: /mnt/petrelfs/zhuchenglin/LLaVA/checkpoints_ft/llava_raw_558k/llava-v1.5-13b
|
| 38 |
+
Average accuracy: 0.524
|
| 39 |
+
--- End of model ---
|
| 40 |
+
|
| 41 |
+
Model: /mnt/petrelfs/zhuchenglin/LLaVA/checkpoints_ft/coco_raw_558k/llava-v1.5-13b
|
| 42 |
+
Average accuracy: 0.520
|
| 43 |
+
--- End of model ---
|
| 44 |
+
|
| 45 |
+
Model: /mnt/petrelfs/zhuchenglin/LLaVA/checkpoints_ft/coco_gen_200k/llava-v1.5-13b
|
| 46 |
+
Average accuracy: 0.529
|
| 47 |
+
--- End of model ---
|
| 48 |
+
|
| 49 |
+
Model: /mnt/petrelfs/zhuchenglin/LLaVA/checkpoints_ft/coco_gen_558k/llava-v1.5-13b
|
| 50 |
+
Average accuracy: 0.515
|
| 51 |
+
--- End of model ---
|
| 52 |
+
|
| 53 |
+
Model: /mnt/petrelfs/zhuchenglin/LLaVA/checkpoints_ft/llava_coco_gen_758k/llava-v1.5-13b
|
| 54 |
+
Average accuracy: 0.537
|
| 55 |
+
--- End of model ---
|
| 56 |
+
|
| 57 |
+
Model: /mnt/petrelfs/zhuchenglin/LLaVA/checkpoints_ft/llava_gen_200k/llava-v1.5-13b
|
| 58 |
+
Average accuracy: 0.537
|
| 59 |
+
--- End of model ---
|
| 60 |
+
|
| 61 |
+
Model: /mnt/petrelfs/zhuchenglin/LLaVA/checkpoints_ft/coco_raw_200k/llava-v1.5-13b
|
| 62 |
+
Average accuracy: 0.529
|
| 63 |
+
--- End of model ---
|
| 64 |
+
|
| 65 |
+
Model: /mnt/petrelfs/zhuchenglin/LLaVA/checkpoints_ft/llava_raw_558k/llava-v1.5-13b
|
| 66 |
+
Average accuracy: 0.524
|
| 67 |
+
--- End of model ---
|
| 68 |
+
|
| 69 |
+
Model: /mnt/petrelfs/zhuchenglin/LLaVA/checkpoints_ft/llava_coco_raw_758k/llava-v1.5-13b
|
| 70 |
+
Average accuracy: 0.521
|
| 71 |
+
--- End of model ---
|
| 72 |
+
|
| 73 |
+
Model: /mnt/petrelfs/zhuchenglin/LLaVA/checkpoints_ft/coco_raw_558k/llava-v1.5-13b
|
| 74 |
+
Average accuracy: 0.539
|
| 75 |
+
--- End of model ---
|
| 76 |
+
|
| 77 |
+
Model: /mnt/petrelfs/zhuchenglin/LLaVA/checkpoints_ft/coco_gen_200k/llava-v1.5-13b-pretrain
|
| 78 |
+
Average accuracy: 0.531
|
| 79 |
+
--- End of model ---
|
| 80 |
+
|
| 81 |
+
Model: /mnt/petrelfs/zhuchenglin/LLaVA/checkpoints_ft/coco_gen_558k/llava-v1.5-13b-pretrain
|
| 82 |
+
Average accuracy: 0.531
|
| 83 |
+
--- End of model ---
|
| 84 |
+
|
| 85 |
+
Model: /mnt/petrelfs/zhuchenglin/LLaVA/checkpoints_ft/coco_raw_200k/llava-v1.5-13b-pretrain
|
| 86 |
+
Average accuracy: 0.531
|
| 87 |
+
--- End of model ---
|
| 88 |
+
|
| 89 |
+
Model: /mnt/petrelfs/zhuchenglin/LLaVA/checkpoints_ft/coco_raw_558k/llava-v1.5-13b-pretrain
|
| 90 |
+
Average accuracy: 0.531
|
| 91 |
+
--- End of model ---
|
| 92 |
+
|
| 93 |
+
Model: /mnt/petrelfs/zhuchenglin/LLaVA/checkpoints_ft/llava_coco_raw_758k/llava-v1.5-13b-pretrain
|
| 94 |
+
Average accuracy: 0.531
|
| 95 |
+
--- End of model ---
|
| 96 |
+
|
| 97 |
+
Model: /mnt/petrelfs/zhuchenglin/LLaVA/checkpoints_ft/llava_coco_gen_758k/llava-v1.5-13b-pretrain
|
| 98 |
+
Average accuracy: 0.531
|
| 99 |
+
--- End of model ---
|
| 100 |
+
|
| 101 |
+
Model: /mnt/petrelfs/zhuchenglin/LLaVA/checkpoints_ft/llava_raw_558k/llava-v1.5-13b-pretrain
|
| 102 |
+
Average accuracy: 0.531
|
| 103 |
+
--- End of model ---
|
| 104 |
+
|
| 105 |
+
Model: /mnt/petrelfs/zhuchenglin/LLaVA/checkpoints_ft/llava_gen_200k/llava-v1.5-13b-pretrain
|
| 106 |
+
Average accuracy: 0.531
|
| 107 |
+
--- End of model ---
|
| 108 |
+
|
| 109 |
+
Model: /mnt/petrelfs/zhuchenglin/LLaVA/checkpoints_ft/pure_gen_558k/llava-v1.5-13b
|
| 110 |
+
Average accuracy: 0.533
|
| 111 |
+
--- End of model ---
|
| 112 |
+
|
| 113 |
+
Model: /mnt/petrelfs/zhuchenglin/LLaVA/checkpoints_ft/pure_gen_200k/llava-v1.5-13b
|
| 114 |
+
Average accuracy: 0.534
|
| 115 |
+
--- End of model ---
|
| 116 |
+
|
| 117 |
+
Model: /mnt/petrelfs/zhuchenglin/LLaVA/checkpoints_ft/select_gen_100k/llava-v1.5-13b
|
| 118 |
+
Average accuracy: 0.546
|
| 119 |
+
--- End of model ---
|
| 120 |
+
|
| 121 |
+
Model: /mnt/petrelfs/zhuchenglin/LLaVA/checkpoints_ft/select_gen_100k/llava-v1.5-13b
|
| 122 |
+
Average accuracy: 0.542
|
| 123 |
+
--- End of model ---
|
| 124 |
+
|
| 125 |
+
Model: /mnt/petrelfs/zhuchenglin/LLaVA/checkpoints_ft/select_raw_100k/llava-v1.5-13b
|
| 126 |
+
Average accuracy: 0.536
|
| 127 |
+
--- End of model ---
|
| 128 |
+
|
| 129 |
+
Model: /mnt/petrelfs/zhuchenglin/LLaVA/checkpoints_ft/select_gen_200k/llava-v1.5-13b
|
| 130 |
+
Average accuracy: 0.535
|
| 131 |
+
--- End of model ---
|
| 132 |
+
|
| 133 |
+
Model: /mnt/petrelfs/zhuchenglin/LLaVA/checkpoints_ft/select_raw_200k/llava-v1.5-13b
|
| 134 |
+
Average accuracy: 0.537
|
| 135 |
+
--- End of model ---
|
| 136 |
+
|
| 137 |
+
Model: /mnt/petrelfs/zhuchenglin/LLaVA/checkpoints_ft/mixed_200k/llava-v1.5-13b
|
| 138 |
+
Average accuracy: 0.535
|
| 139 |
+
--- End of model ---
|
| 140 |
+
|
checkpoints_ft/.gitattributes
ADDED
|
@@ -0,0 +1,55 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
*.7z filter=lfs diff=lfs merge=lfs -text
|
| 2 |
+
*.arrow filter=lfs diff=lfs merge=lfs -text
|
| 3 |
+
*.bin filter=lfs diff=lfs merge=lfs -text
|
| 4 |
+
*.bz2 filter=lfs diff=lfs merge=lfs -text
|
| 5 |
+
*.ckpt filter=lfs diff=lfs merge=lfs -text
|
| 6 |
+
*.ftz filter=lfs diff=lfs merge=lfs -text
|
| 7 |
+
*.gz filter=lfs diff=lfs merge=lfs -text
|
| 8 |
+
*.h5 filter=lfs diff=lfs merge=lfs -text
|
| 9 |
+
*.joblib filter=lfs diff=lfs merge=lfs -text
|
| 10 |
+
*.lfs.* filter=lfs diff=lfs merge=lfs -text
|
| 11 |
+
*.lz4 filter=lfs diff=lfs merge=lfs -text
|
| 12 |
+
*.mlmodel filter=lfs diff=lfs merge=lfs -text
|
| 13 |
+
*.model filter=lfs diff=lfs merge=lfs -text
|
| 14 |
+
*.msgpack filter=lfs diff=lfs merge=lfs -text
|
| 15 |
+
*.npy filter=lfs diff=lfs merge=lfs -text
|
| 16 |
+
*.npz filter=lfs diff=lfs merge=lfs -text
|
| 17 |
+
*.onnx filter=lfs diff=lfs merge=lfs -text
|
| 18 |
+
*.ot filter=lfs diff=lfs merge=lfs -text
|
| 19 |
+
*.parquet filter=lfs diff=lfs merge=lfs -text
|
| 20 |
+
*.pb filter=lfs diff=lfs merge=lfs -text
|
| 21 |
+
*.pickle filter=lfs diff=lfs merge=lfs -text
|
| 22 |
+
*.pkl filter=lfs diff=lfs merge=lfs -text
|
| 23 |
+
*.pt filter=lfs diff=lfs merge=lfs -text
|
| 24 |
+
*.pth filter=lfs diff=lfs merge=lfs -text
|
| 25 |
+
*.rar filter=lfs diff=lfs merge=lfs -text
|
| 26 |
+
*.safetensors filter=lfs diff=lfs merge=lfs -text
|
| 27 |
+
saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
| 28 |
+
*.tar.* filter=lfs diff=lfs merge=lfs -text
|
| 29 |
+
*.tar filter=lfs diff=lfs merge=lfs -text
|
| 30 |
+
*.tflite filter=lfs diff=lfs merge=lfs -text
|
| 31 |
+
*.tgz filter=lfs diff=lfs merge=lfs -text
|
| 32 |
+
*.wasm filter=lfs diff=lfs merge=lfs -text
|
| 33 |
+
*.xz filter=lfs diff=lfs merge=lfs -text
|
| 34 |
+
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 35 |
+
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 36 |
+
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
| 37 |
+
# Audio files - uncompressed
|
| 38 |
+
*.pcm filter=lfs diff=lfs merge=lfs -text
|
| 39 |
+
*.sam filter=lfs diff=lfs merge=lfs -text
|
| 40 |
+
*.raw filter=lfs diff=lfs merge=lfs -text
|
| 41 |
+
# Audio files - compressed
|
| 42 |
+
*.aac filter=lfs diff=lfs merge=lfs -text
|
| 43 |
+
*.flac filter=lfs diff=lfs merge=lfs -text
|
| 44 |
+
*.mp3 filter=lfs diff=lfs merge=lfs -text
|
| 45 |
+
*.ogg filter=lfs diff=lfs merge=lfs -text
|
| 46 |
+
*.wav filter=lfs diff=lfs merge=lfs -text
|
| 47 |
+
# Image files - uncompressed
|
| 48 |
+
*.bmp filter=lfs diff=lfs merge=lfs -text
|
| 49 |
+
*.gif filter=lfs diff=lfs merge=lfs -text
|
| 50 |
+
*.png filter=lfs diff=lfs merge=lfs -text
|
| 51 |
+
*.tiff filter=lfs diff=lfs merge=lfs -text
|
| 52 |
+
# Image files - compressed
|
| 53 |
+
*.jpg filter=lfs diff=lfs merge=lfs -text
|
| 54 |
+
*.jpeg filter=lfs diff=lfs merge=lfs -text
|
| 55 |
+
*.webp filter=lfs diff=lfs merge=lfs -text
|
checkpoints_ft/coco_gen_200k/llava-v1.5-13b-pretrain/config.json
ADDED
|
@@ -0,0 +1,46 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "lmsys/vicuna-13b-v1.5",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"LlamaForCausalLM"
|
| 5 |
+
],
|
| 6 |
+
"attention_bias": false,
|
| 7 |
+
"attention_dropout": 0.0,
|
| 8 |
+
"bos_token_id": 1,
|
| 9 |
+
"eos_token_id": 2,
|
| 10 |
+
"freeze_mm_mlp_adapter": false,
|
| 11 |
+
"hidden_act": "silu",
|
| 12 |
+
"hidden_size": 5120,
|
| 13 |
+
"image_aspect_ratio": "square",
|
| 14 |
+
"initializer_range": 0.02,
|
| 15 |
+
"intermediate_size": 13824,
|
| 16 |
+
"max_length": 4096,
|
| 17 |
+
"max_position_embeddings": 4096,
|
| 18 |
+
"mm_hidden_size": 1024,
|
| 19 |
+
"mm_patch_merge_type": "flat",
|
| 20 |
+
"mm_projector_lr": null,
|
| 21 |
+
"mm_projector_type": "mlp2x_gelu",
|
| 22 |
+
"mm_use_im_patch_token": false,
|
| 23 |
+
"mm_use_im_start_end": false,
|
| 24 |
+
"mm_vision_select_feature": "patch",
|
| 25 |
+
"mm_vision_select_layer": -2,
|
| 26 |
+
"mm_vision_tower": "openai/clip-vit-large-patch14-336",
|
| 27 |
+
"model_type": "llava_llama",
|
| 28 |
+
"num_attention_heads": 40,
|
| 29 |
+
"num_hidden_layers": 40,
|
| 30 |
+
"num_key_value_heads": 40,
|
| 31 |
+
"pad_token_id": 0,
|
| 32 |
+
"pretraining_tp": 1,
|
| 33 |
+
"rms_norm_eps": 1e-05,
|
| 34 |
+
"rope_scaling": null,
|
| 35 |
+
"rope_theta": 10000.0,
|
| 36 |
+
"tie_word_embeddings": false,
|
| 37 |
+
"tokenizer_model_max_length": 2048,
|
| 38 |
+
"tokenizer_padding_side": "right",
|
| 39 |
+
"torch_dtype": "float16",
|
| 40 |
+
"transformers_version": "4.37.2",
|
| 41 |
+
"tune_mm_mlp_adapter": true,
|
| 42 |
+
"use_cache": true,
|
| 43 |
+
"use_mm_proj": true,
|
| 44 |
+
"vocab_size": 32000,
|
| 45 |
+
"do_sample": true
|
| 46 |
+
}
|
checkpoints_ft/coco_gen_200k/llava-v1.5-13b-pretrain/mm_projector.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:71fab2dc54b692793d5f008f274b55ab8e6f67c093489128fde9f01aae2553d5
|
| 3 |
+
size 62937144
|
checkpoints_ft/coco_gen_200k/llava-v1.5-13b-pretrain/trainer_state.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoints_ft/coco_gen_558k/llava-v1.5-13b-pretrain/config.json
ADDED
|
@@ -0,0 +1,46 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "/root/LLaVA/vicuna-13b-v1.5",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"LlamaForCausalLM"
|
| 5 |
+
],
|
| 6 |
+
"attention_bias": false,
|
| 7 |
+
"attention_dropout": 0.0,
|
| 8 |
+
"bos_token_id": 1,
|
| 9 |
+
"eos_token_id": 2,
|
| 10 |
+
"freeze_mm_mlp_adapter": false,
|
| 11 |
+
"hidden_act": "silu",
|
| 12 |
+
"hidden_size": 5120,
|
| 13 |
+
"image_aspect_ratio": "square",
|
| 14 |
+
"initializer_range": 0.02,
|
| 15 |
+
"intermediate_size": 13824,
|
| 16 |
+
"max_length": 4096,
|
| 17 |
+
"max_position_embeddings": 4096,
|
| 18 |
+
"mm_hidden_size": 1024,
|
| 19 |
+
"mm_patch_merge_type": "flat",
|
| 20 |
+
"mm_projector_lr": null,
|
| 21 |
+
"mm_projector_type": "mlp2x_gelu",
|
| 22 |
+
"mm_use_im_patch_token": false,
|
| 23 |
+
"mm_use_im_start_end": false,
|
| 24 |
+
"mm_vision_select_feature": "patch",
|
| 25 |
+
"mm_vision_select_layer": -2,
|
| 26 |
+
"mm_vision_tower": "openai/clip-vit-large-patch14-336",
|
| 27 |
+
"model_type": "llava_llama",
|
| 28 |
+
"num_attention_heads": 40,
|
| 29 |
+
"num_hidden_layers": 40,
|
| 30 |
+
"num_key_value_heads": 40,
|
| 31 |
+
"pad_token_id": 0,
|
| 32 |
+
"pretraining_tp": 1,
|
| 33 |
+
"rms_norm_eps": 1e-05,
|
| 34 |
+
"rope_scaling": null,
|
| 35 |
+
"rope_theta": 10000.0,
|
| 36 |
+
"tie_word_embeddings": false,
|
| 37 |
+
"tokenizer_model_max_length": 2048,
|
| 38 |
+
"tokenizer_padding_side": "right",
|
| 39 |
+
"torch_dtype": "float16",
|
| 40 |
+
"transformers_version": "4.37.2",
|
| 41 |
+
"tune_mm_mlp_adapter": true,
|
| 42 |
+
"use_cache": true,
|
| 43 |
+
"use_mm_proj": true,
|
| 44 |
+
"vocab_size": 32000,
|
| 45 |
+
"do_sample": true
|
| 46 |
+
}
|
checkpoints_ft/coco_gen_558k/llava-v1.5-13b-pretrain/mm_projector.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b0e7165ff7ac58d8e24ee9c6b6fed5c12731e6a005d45c2de626b72167d0bd05
|
| 3 |
+
size 62937144
|
checkpoints_ft/coco_gen_558k/llava-v1.5-13b-pretrain/trainer_state.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoints_ft/coco_raw_200k/llava-v1.5-13b-pretrain/config.json
ADDED
|
@@ -0,0 +1,46 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "lmsys/vicuna-13b-v1.5",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"LlamaForCausalLM"
|
| 5 |
+
],
|
| 6 |
+
"attention_bias": false,
|
| 7 |
+
"attention_dropout": 0.0,
|
| 8 |
+
"bos_token_id": 1,
|
| 9 |
+
"eos_token_id": 2,
|
| 10 |
+
"freeze_mm_mlp_adapter": false,
|
| 11 |
+
"hidden_act": "silu",
|
| 12 |
+
"hidden_size": 5120,
|
| 13 |
+
"image_aspect_ratio": "square",
|
| 14 |
+
"initializer_range": 0.02,
|
| 15 |
+
"intermediate_size": 13824,
|
| 16 |
+
"max_length": 4096,
|
| 17 |
+
"max_position_embeddings": 4096,
|
| 18 |
+
"mm_hidden_size": 1024,
|
| 19 |
+
"mm_patch_merge_type": "flat",
|
| 20 |
+
"mm_projector_lr": null,
|
| 21 |
+
"mm_projector_type": "mlp2x_gelu",
|
| 22 |
+
"mm_use_im_patch_token": false,
|
| 23 |
+
"mm_use_im_start_end": false,
|
| 24 |
+
"mm_vision_select_feature": "patch",
|
| 25 |
+
"mm_vision_select_layer": -2,
|
| 26 |
+
"mm_vision_tower": "openai/clip-vit-large-patch14-336",
|
| 27 |
+
"model_type": "llava_llama",
|
| 28 |
+
"num_attention_heads": 40,
|
| 29 |
+
"num_hidden_layers": 40,
|
| 30 |
+
"num_key_value_heads": 40,
|
| 31 |
+
"pad_token_id": 0,
|
| 32 |
+
"pretraining_tp": 1,
|
| 33 |
+
"rms_norm_eps": 1e-05,
|
| 34 |
+
"rope_scaling": null,
|
| 35 |
+
"rope_theta": 10000.0,
|
| 36 |
+
"tie_word_embeddings": false,
|
| 37 |
+
"tokenizer_model_max_length": 2048,
|
| 38 |
+
"tokenizer_padding_side": "right",
|
| 39 |
+
"torch_dtype": "float16",
|
| 40 |
+
"transformers_version": "4.37.2",
|
| 41 |
+
"tune_mm_mlp_adapter": true,
|
| 42 |
+
"use_cache": true,
|
| 43 |
+
"use_mm_proj": true,
|
| 44 |
+
"vocab_size": 32000,
|
| 45 |
+
"do_sample": true
|
| 46 |
+
}
|
checkpoints_ft/coco_raw_200k/llava-v1.5-13b-pretrain/mm_projector.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2180785973804f4a52e0a7443cb5684f848ae499169d91101c99a83afef4c9b9
|
| 3 |
+
size 62937144
|
checkpoints_ft/coco_raw_200k/llava-v1.5-13b-pretrain/trainer_state.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoints_ft/coco_raw_558k/llava-v1.5-13b-pretrain/config.json
ADDED
|
@@ -0,0 +1,46 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "lmsys/vicuna-13b-v1.5",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"LlamaForCausalLM"
|
| 5 |
+
],
|
| 6 |
+
"attention_bias": false,
|
| 7 |
+
"attention_dropout": 0.0,
|
| 8 |
+
"bos_token_id": 1,
|
| 9 |
+
"eos_token_id": 2,
|
| 10 |
+
"freeze_mm_mlp_adapter": false,
|
| 11 |
+
"hidden_act": "silu",
|
| 12 |
+
"hidden_size": 5120,
|
| 13 |
+
"image_aspect_ratio": "square",
|
| 14 |
+
"initializer_range": 0.02,
|
| 15 |
+
"intermediate_size": 13824,
|
| 16 |
+
"max_length": 4096,
|
| 17 |
+
"max_position_embeddings": 4096,
|
| 18 |
+
"mm_hidden_size": 1024,
|
| 19 |
+
"mm_patch_merge_type": "flat",
|
| 20 |
+
"mm_projector_lr": null,
|
| 21 |
+
"mm_projector_type": "mlp2x_gelu",
|
| 22 |
+
"mm_use_im_patch_token": false,
|
| 23 |
+
"mm_use_im_start_end": false,
|
| 24 |
+
"mm_vision_select_feature": "patch",
|
| 25 |
+
"mm_vision_select_layer": -2,
|
| 26 |
+
"mm_vision_tower": "openai/clip-vit-large-patch14-336",
|
| 27 |
+
"model_type": "llava_llama",
|
| 28 |
+
"num_attention_heads": 40,
|
| 29 |
+
"num_hidden_layers": 40,
|
| 30 |
+
"num_key_value_heads": 40,
|
| 31 |
+
"pad_token_id": 0,
|
| 32 |
+
"pretraining_tp": 1,
|
| 33 |
+
"rms_norm_eps": 1e-05,
|
| 34 |
+
"rope_scaling": null,
|
| 35 |
+
"rope_theta": 10000.0,
|
| 36 |
+
"tie_word_embeddings": false,
|
| 37 |
+
"tokenizer_model_max_length": 2048,
|
| 38 |
+
"tokenizer_padding_side": "right",
|
| 39 |
+
"torch_dtype": "float16",
|
| 40 |
+
"transformers_version": "4.37.2",
|
| 41 |
+
"tune_mm_mlp_adapter": true,
|
| 42 |
+
"use_cache": true,
|
| 43 |
+
"use_mm_proj": true,
|
| 44 |
+
"vocab_size": 32000,
|
| 45 |
+
"do_sample": true
|
| 46 |
+
}
|
checkpoints_ft/coco_raw_558k/llava-v1.5-13b-pretrain/mm_projector.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:108db710ea90f798e0db72c3c3a785e3adadb4739bbafde1331c2cd5f276e05b
|
| 3 |
+
size 62937144
|
checkpoints_ft/coco_raw_558k/llava-v1.5-13b-pretrain/trainer_state.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoints_ft/coco_select_200k/llava-v1.5-13b-pretrain/config.json
ADDED
|
@@ -0,0 +1,46 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "lmsys/vicuna-13b-v1.5",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"LlamaForCausalLM"
|
| 5 |
+
],
|
| 6 |
+
"attention_bias": false,
|
| 7 |
+
"attention_dropout": 0.0,
|
| 8 |
+
"bos_token_id": 1,
|
| 9 |
+
"eos_token_id": 2,
|
| 10 |
+
"freeze_mm_mlp_adapter": false,
|
| 11 |
+
"hidden_act": "silu",
|
| 12 |
+
"hidden_size": 5120,
|
| 13 |
+
"image_aspect_ratio": "square",
|
| 14 |
+
"initializer_range": 0.02,
|
| 15 |
+
"intermediate_size": 13824,
|
| 16 |
+
"max_length": 4096,
|
| 17 |
+
"max_position_embeddings": 4096,
|
| 18 |
+
"mm_hidden_size": 1024,
|
| 19 |
+
"mm_patch_merge_type": "flat",
|
| 20 |
+
"mm_projector_lr": null,
|
| 21 |
+
"mm_projector_type": "mlp2x_gelu",
|
| 22 |
+
"mm_use_im_patch_token": false,
|
| 23 |
+
"mm_use_im_start_end": false,
|
| 24 |
+
"mm_vision_select_feature": "patch",
|
| 25 |
+
"mm_vision_select_layer": -2,
|
| 26 |
+
"mm_vision_tower": "openai/clip-vit-large-patch14-336",
|
| 27 |
+
"model_type": "llava_llama",
|
| 28 |
+
"num_attention_heads": 40,
|
| 29 |
+
"num_hidden_layers": 40,
|
| 30 |
+
"num_key_value_heads": 40,
|
| 31 |
+
"pad_token_id": 0,
|
| 32 |
+
"pretraining_tp": 1,
|
| 33 |
+
"rms_norm_eps": 1e-05,
|
| 34 |
+
"rope_scaling": null,
|
| 35 |
+
"rope_theta": 10000.0,
|
| 36 |
+
"tie_word_embeddings": false,
|
| 37 |
+
"tokenizer_model_max_length": 2048,
|
| 38 |
+
"tokenizer_padding_side": "right",
|
| 39 |
+
"torch_dtype": "float16",
|
| 40 |
+
"transformers_version": "4.37.2",
|
| 41 |
+
"tune_mm_mlp_adapter": true,
|
| 42 |
+
"use_cache": true,
|
| 43 |
+
"use_mm_proj": true,
|
| 44 |
+
"vocab_size": 32000,
|
| 45 |
+
"do_sample": true
|
| 46 |
+
}
|
checkpoints_ft/coco_select_200k/llava-v1.5-13b-pretrain/mm_projector.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:32f539ec44bea3f9901ed557fd705c33ff4ce872f32ab4aa272baa227235f99c
|
| 3 |
+
size 62937144
|
checkpoints_ft/coco_select_200k/llava-v1.5-13b-pretrain/trainer_state.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoints_ft/llava_coco_gen_758k/llava-v1.5-13b-pretrain/config.json
ADDED
|
@@ -0,0 +1,46 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "lmsys/vicuna-13b-v1.5",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"LlamaForCausalLM"
|
| 5 |
+
],
|
| 6 |
+
"attention_bias": false,
|
| 7 |
+
"attention_dropout": 0.0,
|
| 8 |
+
"bos_token_id": 1,
|
| 9 |
+
"eos_token_id": 2,
|
| 10 |
+
"freeze_mm_mlp_adapter": false,
|
| 11 |
+
"hidden_act": "silu",
|
| 12 |
+
"hidden_size": 5120,
|
| 13 |
+
"image_aspect_ratio": "square",
|
| 14 |
+
"initializer_range": 0.02,
|
| 15 |
+
"intermediate_size": 13824,
|
| 16 |
+
"max_length": 4096,
|
| 17 |
+
"max_position_embeddings": 4096,
|
| 18 |
+
"mm_hidden_size": 1024,
|
| 19 |
+
"mm_patch_merge_type": "flat",
|
| 20 |
+
"mm_projector_lr": null,
|
| 21 |
+
"mm_projector_type": "mlp2x_gelu",
|
| 22 |
+
"mm_use_im_patch_token": false,
|
| 23 |
+
"mm_use_im_start_end": false,
|
| 24 |
+
"mm_vision_select_feature": "patch",
|
| 25 |
+
"mm_vision_select_layer": -2,
|
| 26 |
+
"mm_vision_tower": "openai/clip-vit-large-patch14-336",
|
| 27 |
+
"model_type": "llava_llama",
|
| 28 |
+
"num_attention_heads": 40,
|
| 29 |
+
"num_hidden_layers": 40,
|
| 30 |
+
"num_key_value_heads": 40,
|
| 31 |
+
"pad_token_id": 0,
|
| 32 |
+
"pretraining_tp": 1,
|
| 33 |
+
"rms_norm_eps": 1e-05,
|
| 34 |
+
"rope_scaling": null,
|
| 35 |
+
"rope_theta": 10000.0,
|
| 36 |
+
"tie_word_embeddings": false,
|
| 37 |
+
"tokenizer_model_max_length": 2048,
|
| 38 |
+
"tokenizer_padding_side": "right",
|
| 39 |
+
"torch_dtype": "float16",
|
| 40 |
+
"transformers_version": "4.37.2",
|
| 41 |
+
"tune_mm_mlp_adapter": true,
|
| 42 |
+
"use_cache": true,
|
| 43 |
+
"use_mm_proj": true,
|
| 44 |
+
"vocab_size": 32000,
|
| 45 |
+
"do_sample": true
|
| 46 |
+
}
|
checkpoints_ft/llava_coco_gen_758k/llava-v1.5-13b-pretrain/mm_projector.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f1879a902ae20cf78bb8f66a1d849adf1e88d0f8e81c53285a21e2625497e646
|
| 3 |
+
size 62937144
|
checkpoints_ft/llava_coco_gen_758k/llava-v1.5-13b-pretrain/trainer_state.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoints_ft/llava_coco_raw_758k/llava-v1.5-13b-pretrain/config.json
ADDED
|
@@ -0,0 +1,46 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "lmsys/vicuna-13b-v1.5",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"LlamaForCausalLM"
|
| 5 |
+
],
|
| 6 |
+
"attention_bias": false,
|
| 7 |
+
"attention_dropout": 0.0,
|
| 8 |
+
"bos_token_id": 1,
|
| 9 |
+
"eos_token_id": 2,
|
| 10 |
+
"freeze_mm_mlp_adapter": false,
|
| 11 |
+
"hidden_act": "silu",
|
| 12 |
+
"hidden_size": 5120,
|
| 13 |
+
"image_aspect_ratio": "square",
|
| 14 |
+
"initializer_range": 0.02,
|
| 15 |
+
"intermediate_size": 13824,
|
| 16 |
+
"max_length": 4096,
|
| 17 |
+
"max_position_embeddings": 4096,
|
| 18 |
+
"mm_hidden_size": 1024,
|
| 19 |
+
"mm_patch_merge_type": "flat",
|
| 20 |
+
"mm_projector_lr": null,
|
| 21 |
+
"mm_projector_type": "mlp2x_gelu",
|
| 22 |
+
"mm_use_im_patch_token": false,
|
| 23 |
+
"mm_use_im_start_end": false,
|
| 24 |
+
"mm_vision_select_feature": "patch",
|
| 25 |
+
"mm_vision_select_layer": -2,
|
| 26 |
+
"mm_vision_tower": "openai/clip-vit-large-patch14-336",
|
| 27 |
+
"model_type": "llava_llama",
|
| 28 |
+
"num_attention_heads": 40,
|
| 29 |
+
"num_hidden_layers": 40,
|
| 30 |
+
"num_key_value_heads": 40,
|
| 31 |
+
"pad_token_id": 0,
|
| 32 |
+
"pretraining_tp": 1,
|
| 33 |
+
"rms_norm_eps": 1e-05,
|
| 34 |
+
"rope_scaling": null,
|
| 35 |
+
"rope_theta": 10000.0,
|
| 36 |
+
"tie_word_embeddings": false,
|
| 37 |
+
"tokenizer_model_max_length": 2048,
|
| 38 |
+
"tokenizer_padding_side": "right",
|
| 39 |
+
"torch_dtype": "float16",
|
| 40 |
+
"transformers_version": "4.37.2",
|
| 41 |
+
"tune_mm_mlp_adapter": true,
|
| 42 |
+
"use_cache": true,
|
| 43 |
+
"use_mm_proj": true,
|
| 44 |
+
"vocab_size": 32000,
|
| 45 |
+
"do_sample": true
|
| 46 |
+
}
|
checkpoints_ft/llava_coco_raw_758k/llava-v1.5-13b-pretrain/mm_projector.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4031291543d0a4b64ec39f4560655bcf5d78d5a76281abd642d17b15b679857d
|
| 3 |
+
size 62937144
|
checkpoints_ft/llava_coco_raw_758k/llava-v1.5-13b-pretrain/trainer_state.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoints_ft/llava_gen_200k/llava-v1.5-13b-pretrain/config.json
ADDED
|
@@ -0,0 +1,46 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "lmsys/vicuna-13b-v1.5",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"LlamaForCausalLM"
|
| 5 |
+
],
|
| 6 |
+
"attention_bias": false,
|
| 7 |
+
"attention_dropout": 0.0,
|
| 8 |
+
"bos_token_id": 1,
|
| 9 |
+
"eos_token_id": 2,
|
| 10 |
+
"freeze_mm_mlp_adapter": false,
|
| 11 |
+
"hidden_act": "silu",
|
| 12 |
+
"hidden_size": 5120,
|
| 13 |
+
"image_aspect_ratio": "square",
|
| 14 |
+
"initializer_range": 0.02,
|
| 15 |
+
"intermediate_size": 13824,
|
| 16 |
+
"max_length": 4096,
|
| 17 |
+
"max_position_embeddings": 4096,
|
| 18 |
+
"mm_hidden_size": 1024,
|
| 19 |
+
"mm_patch_merge_type": "flat",
|
| 20 |
+
"mm_projector_lr": null,
|
| 21 |
+
"mm_projector_type": "mlp2x_gelu",
|
| 22 |
+
"mm_use_im_patch_token": false,
|
| 23 |
+
"mm_use_im_start_end": false,
|
| 24 |
+
"mm_vision_select_feature": "patch",
|
| 25 |
+
"mm_vision_select_layer": -2,
|
| 26 |
+
"mm_vision_tower": "openai/clip-vit-large-patch14-336",
|
| 27 |
+
"model_type": "llava_llama",
|
| 28 |
+
"num_attention_heads": 40,
|
| 29 |
+
"num_hidden_layers": 40,
|
| 30 |
+
"num_key_value_heads": 40,
|
| 31 |
+
"pad_token_id": 0,
|
| 32 |
+
"pretraining_tp": 1,
|
| 33 |
+
"rms_norm_eps": 1e-05,
|
| 34 |
+
"rope_scaling": null,
|
| 35 |
+
"rope_theta": 10000.0,
|
| 36 |
+
"tie_word_embeddings": false,
|
| 37 |
+
"tokenizer_model_max_length": 2048,
|
| 38 |
+
"tokenizer_padding_side": "right",
|
| 39 |
+
"torch_dtype": "float16",
|
| 40 |
+
"transformers_version": "4.37.2",
|
| 41 |
+
"tune_mm_mlp_adapter": true,
|
| 42 |
+
"use_cache": true,
|
| 43 |
+
"use_mm_proj": true,
|
| 44 |
+
"vocab_size": 32000,
|
| 45 |
+
"do_sample": true
|
| 46 |
+
}
|
checkpoints_ft/llava_gen_200k/llava-v1.5-13b-pretrain/mm_projector.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6d16c9fcd3548427b2ad57e5305c6e5a5d7b4f804df8558c16fcd2f21c188b99
|
| 3 |
+
size 62937144
|
checkpoints_ft/llava_gen_200k/llava-v1.5-13b-pretrain/trainer_state.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoints_ft/llava_gen_200k/llava-v1.5-7b-pretrain/config.json
ADDED
|
@@ -0,0 +1,45 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "lmsys/vicuna-7b-v1.5",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"LlamaForCausalLM"
|
| 5 |
+
],
|
| 6 |
+
"attention_bias": false,
|
| 7 |
+
"attention_dropout": 0.0,
|
| 8 |
+
"bos_token_id": 1,
|
| 9 |
+
"eos_token_id": 2,
|
| 10 |
+
"freeze_mm_mlp_adapter": false,
|
| 11 |
+
"hidden_act": "silu",
|
| 12 |
+
"hidden_size": 4096,
|
| 13 |
+
"image_aspect_ratio": "square",
|
| 14 |
+
"initializer_range": 0.02,
|
| 15 |
+
"intermediate_size": 11008,
|
| 16 |
+
"max_position_embeddings": 4096,
|
| 17 |
+
"mm_hidden_size": 1024,
|
| 18 |
+
"mm_patch_merge_type": "flat",
|
| 19 |
+
"mm_projector_lr": null,
|
| 20 |
+
"mm_projector_type": "mlp2x_gelu",
|
| 21 |
+
"mm_use_im_patch_token": false,
|
| 22 |
+
"mm_use_im_start_end": false,
|
| 23 |
+
"mm_vision_select_feature": "patch",
|
| 24 |
+
"mm_vision_select_layer": -2,
|
| 25 |
+
"mm_vision_tower": "openai/clip-vit-large-patch14-336",
|
| 26 |
+
"model_type": "llava_llama",
|
| 27 |
+
"num_attention_heads": 32,
|
| 28 |
+
"num_hidden_layers": 32,
|
| 29 |
+
"num_key_value_heads": 32,
|
| 30 |
+
"pad_token_id": 0,
|
| 31 |
+
"pretraining_tp": 1,
|
| 32 |
+
"rms_norm_eps": 1e-05,
|
| 33 |
+
"rope_scaling": null,
|
| 34 |
+
"rope_theta": 10000.0,
|
| 35 |
+
"tie_word_embeddings": false,
|
| 36 |
+
"tokenizer_model_max_length": 2048,
|
| 37 |
+
"tokenizer_padding_side": "right",
|
| 38 |
+
"torch_dtype": "float16",
|
| 39 |
+
"transformers_version": "4.37.2",
|
| 40 |
+
"tune_mm_mlp_adapter": true,
|
| 41 |
+
"use_cache": true,
|
| 42 |
+
"use_mm_proj": true,
|
| 43 |
+
"vocab_size": 32000,
|
| 44 |
+
"do_sample": true
|
| 45 |
+
}
|
checkpoints_ft/llava_gen_200k/llava-v1.5-7b-pretrain/mm_projector.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:bcf3c664a5b88df6e87a1b9e4add5b95133daa6e405ca0f320d92b40e976d43a
|
| 3 |
+
size 41961528
|
checkpoints_ft/llava_gen_200k/llava-v1.5-7b-pretrain/trainer_state.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoints_ft/llava_gen_200k/llava-v1.5-7b/config.json
ADDED
|
@@ -0,0 +1,44 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "lmsys/vicuna-7b-v1.5",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"LlavaLlamaForCausalLM"
|
| 5 |
+
],
|
| 6 |
+
"attention_bias": false,
|
| 7 |
+
"attention_dropout": 0.0,
|
| 8 |
+
"bos_token_id": 1,
|
| 9 |
+
"eos_token_id": 2,
|
| 10 |
+
"freeze_mm_mlp_adapter": false,
|
| 11 |
+
"hidden_act": "silu",
|
| 12 |
+
"hidden_size": 4096,
|
| 13 |
+
"image_aspect_ratio": "pad",
|
| 14 |
+
"initializer_range": 0.02,
|
| 15 |
+
"intermediate_size": 11008,
|
| 16 |
+
"max_position_embeddings": 4096,
|
| 17 |
+
"mm_hidden_size": 1024,
|
| 18 |
+
"mm_patch_merge_type": "flat",
|
| 19 |
+
"mm_projector_lr": null,
|
| 20 |
+
"mm_projector_type": "mlp2x_gelu",
|
| 21 |
+
"mm_use_im_patch_token": false,
|
| 22 |
+
"mm_use_im_start_end": false,
|
| 23 |
+
"mm_vision_select_feature": "patch",
|
| 24 |
+
"mm_vision_select_layer": -2,
|
| 25 |
+
"mm_vision_tower": "openai/clip-vit-large-patch14-336",
|
| 26 |
+
"model_type": "llava_llama",
|
| 27 |
+
"num_attention_heads": 32,
|
| 28 |
+
"num_hidden_layers": 32,
|
| 29 |
+
"num_key_value_heads": 32,
|
| 30 |
+
"pad_token_id": 0,
|
| 31 |
+
"pretraining_tp": 1,
|
| 32 |
+
"rms_norm_eps": 1e-05,
|
| 33 |
+
"rope_scaling": null,
|
| 34 |
+
"rope_theta": 10000.0,
|
| 35 |
+
"tie_word_embeddings": false,
|
| 36 |
+
"tokenizer_model_max_length": 2048,
|
| 37 |
+
"tokenizer_padding_side": "right",
|
| 38 |
+
"torch_dtype": "bfloat16",
|
| 39 |
+
"transformers_version": "4.37.2",
|
| 40 |
+
"tune_mm_mlp_adapter": false,
|
| 41 |
+
"use_cache": true,
|
| 42 |
+
"use_mm_proj": true,
|
| 43 |
+
"vocab_size": 32000
|
| 44 |
+
}
|
checkpoints_ft/llava_gen_200k/llava-v1.5-7b/trainer_state.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoints_ft/llava_gen_558k/llava-v1.5-13b-pretrain/config.json
ADDED
|
@@ -0,0 +1,46 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "lmsys/vicuna-13b-v1.5",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"LlamaForCausalLM"
|
| 5 |
+
],
|
| 6 |
+
"attention_bias": false,
|
| 7 |
+
"attention_dropout": 0.0,
|
| 8 |
+
"bos_token_id": 1,
|
| 9 |
+
"eos_token_id": 2,
|
| 10 |
+
"freeze_mm_mlp_adapter": false,
|
| 11 |
+
"hidden_act": "silu",
|
| 12 |
+
"hidden_size": 5120,
|
| 13 |
+
"image_aspect_ratio": "square",
|
| 14 |
+
"initializer_range": 0.02,
|
| 15 |
+
"intermediate_size": 13824,
|
| 16 |
+
"max_length": 4096,
|
| 17 |
+
"max_position_embeddings": 4096,
|
| 18 |
+
"mm_hidden_size": 1024,
|
| 19 |
+
"mm_patch_merge_type": "flat",
|
| 20 |
+
"mm_projector_lr": null,
|
| 21 |
+
"mm_projector_type": "mlp2x_gelu",
|
| 22 |
+
"mm_use_im_patch_token": false,
|
| 23 |
+
"mm_use_im_start_end": false,
|
| 24 |
+
"mm_vision_select_feature": "patch",
|
| 25 |
+
"mm_vision_select_layer": -2,
|
| 26 |
+
"mm_vision_tower": "openai/clip-vit-large-patch14-336",
|
| 27 |
+
"model_type": "llava_llama",
|
| 28 |
+
"num_attention_heads": 40,
|
| 29 |
+
"num_hidden_layers": 40,
|
| 30 |
+
"num_key_value_heads": 40,
|
| 31 |
+
"pad_token_id": 0,
|
| 32 |
+
"pretraining_tp": 1,
|
| 33 |
+
"rms_norm_eps": 1e-05,
|
| 34 |
+
"rope_scaling": null,
|
| 35 |
+
"rope_theta": 10000.0,
|
| 36 |
+
"tie_word_embeddings": false,
|
| 37 |
+
"tokenizer_model_max_length": 2048,
|
| 38 |
+
"tokenizer_padding_side": "right",
|
| 39 |
+
"torch_dtype": "float16",
|
| 40 |
+
"transformers_version": "4.37.2",
|
| 41 |
+
"tune_mm_mlp_adapter": true,
|
| 42 |
+
"use_cache": true,
|
| 43 |
+
"use_mm_proj": true,
|
| 44 |
+
"vocab_size": 32000,
|
| 45 |
+
"do_sample": true
|
| 46 |
+
}
|
checkpoints_ft/llava_gen_558k/llava-v1.5-13b-pretrain/mm_projector.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cd31ac0099d5e5bad89b180ed632ec1658dfe18a0dceedbab7e0fb976073b7ae
|
| 3 |
+
size 62937144
|
checkpoints_ft/llava_gen_558k/llava-v1.5-13b-pretrain/trainer_state.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoints_ft/llava_raw_200k/llava-v1.5-13b-pretrain/config.json
ADDED
|
@@ -0,0 +1,46 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "lmsys/vicuna-13b-v1.5",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"LlamaForCausalLM"
|
| 5 |
+
],
|
| 6 |
+
"attention_bias": false,
|
| 7 |
+
"attention_dropout": 0.0,
|
| 8 |
+
"bos_token_id": 1,
|
| 9 |
+
"eos_token_id": 2,
|
| 10 |
+
"freeze_mm_mlp_adapter": false,
|
| 11 |
+
"hidden_act": "silu",
|
| 12 |
+
"hidden_size": 5120,
|
| 13 |
+
"image_aspect_ratio": "square",
|
| 14 |
+
"initializer_range": 0.02,
|
| 15 |
+
"intermediate_size": 13824,
|
| 16 |
+
"max_length": 4096,
|
| 17 |
+
"max_position_embeddings": 4096,
|
| 18 |
+
"mm_hidden_size": 1024,
|
| 19 |
+
"mm_patch_merge_type": "flat",
|
| 20 |
+
"mm_projector_lr": null,
|
| 21 |
+
"mm_projector_type": "mlp2x_gelu",
|
| 22 |
+
"mm_use_im_patch_token": false,
|
| 23 |
+
"mm_use_im_start_end": false,
|
| 24 |
+
"mm_vision_select_feature": "patch",
|
| 25 |
+
"mm_vision_select_layer": -2,
|
| 26 |
+
"mm_vision_tower": "openai/clip-vit-large-patch14-336",
|
| 27 |
+
"model_type": "llava_llama",
|
| 28 |
+
"num_attention_heads": 40,
|
| 29 |
+
"num_hidden_layers": 40,
|
| 30 |
+
"num_key_value_heads": 40,
|
| 31 |
+
"pad_token_id": 0,
|
| 32 |
+
"pretraining_tp": 1,
|
| 33 |
+
"rms_norm_eps": 1e-05,
|
| 34 |
+
"rope_scaling": null,
|
| 35 |
+
"rope_theta": 10000.0,
|
| 36 |
+
"tie_word_embeddings": false,
|
| 37 |
+
"tokenizer_model_max_length": 2048,
|
| 38 |
+
"tokenizer_padding_side": "right",
|
| 39 |
+
"torch_dtype": "float16",
|
| 40 |
+
"transformers_version": "4.37.2",
|
| 41 |
+
"tune_mm_mlp_adapter": true,
|
| 42 |
+
"use_cache": true,
|
| 43 |
+
"use_mm_proj": true,
|
| 44 |
+
"vocab_size": 32000,
|
| 45 |
+
"do_sample": true
|
| 46 |
+
}
|
checkpoints_ft/llava_raw_200k/llava-v1.5-13b-pretrain/mm_projector.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cfbaba7ebebbcc1f153df144e3c8b48476d000b948f51e5ed1c78e66d832f65e
|
| 3 |
+
size 62937144
|
checkpoints_ft/llava_raw_200k/llava-v1.5-13b-pretrain/trainer_state.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoints_ft/llava_raw_558k/llava-v1.5-13b-pretrain/config.json
ADDED
|
@@ -0,0 +1,46 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "lmsys/vicuna-13b-v1.5",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"LlamaForCausalLM"
|
| 5 |
+
],
|
| 6 |
+
"attention_bias": false,
|
| 7 |
+
"attention_dropout": 0.0,
|
| 8 |
+
"bos_token_id": 1,
|
| 9 |
+
"eos_token_id": 2,
|
| 10 |
+
"freeze_mm_mlp_adapter": false,
|
| 11 |
+
"hidden_act": "silu",
|
| 12 |
+
"hidden_size": 5120,
|
| 13 |
+
"image_aspect_ratio": "square",
|
| 14 |
+
"initializer_range": 0.02,
|
| 15 |
+
"intermediate_size": 13824,
|
| 16 |
+
"max_length": 4096,
|
| 17 |
+
"max_position_embeddings": 4096,
|
| 18 |
+
"mm_hidden_size": 1024,
|
| 19 |
+
"mm_patch_merge_type": "flat",
|
| 20 |
+
"mm_projector_lr": null,
|
| 21 |
+
"mm_projector_type": "mlp2x_gelu",
|
| 22 |
+
"mm_use_im_patch_token": false,
|
| 23 |
+
"mm_use_im_start_end": false,
|
| 24 |
+
"mm_vision_select_feature": "patch",
|
| 25 |
+
"mm_vision_select_layer": -2,
|
| 26 |
+
"mm_vision_tower": "openai/clip-vit-large-patch14-336",
|
| 27 |
+
"model_type": "llava_llama",
|
| 28 |
+
"num_attention_heads": 40,
|
| 29 |
+
"num_hidden_layers": 40,
|
| 30 |
+
"num_key_value_heads": 40,
|
| 31 |
+
"pad_token_id": 0,
|
| 32 |
+
"pretraining_tp": 1,
|
| 33 |
+
"rms_norm_eps": 1e-05,
|
| 34 |
+
"rope_scaling": null,
|
| 35 |
+
"rope_theta": 10000.0,
|
| 36 |
+
"tie_word_embeddings": false,
|
| 37 |
+
"tokenizer_model_max_length": 2048,
|
| 38 |
+
"tokenizer_padding_side": "right",
|
| 39 |
+
"torch_dtype": "float16",
|
| 40 |
+
"transformers_version": "4.37.2",
|
| 41 |
+
"tune_mm_mlp_adapter": true,
|
| 42 |
+
"use_cache": true,
|
| 43 |
+
"use_mm_proj": true,
|
| 44 |
+
"vocab_size": 32000,
|
| 45 |
+
"do_sample": true
|
| 46 |
+
}
|