innit

Author: ChenDRAG

LlamaGen/.gitignore

@@ -0,0 +1,165 @@
+# Byte-compiled / optimized / DLL files
+__pycache__/
+*.py[cod]
+*$py.class
+
+# C extensions
+*.so
+
+# Distribution / packaging
+.Python
+build/
+develop-eggs/
+dist/
+downloads/
+eggs/
+.eggs/
+lib/
+lib64/
+evaluations/
+results/
+results2/
+parts/
+sdist/
+var/
+wheels/
+share/python-wheels/
+*.egg-info/
+.installed.cfg
+*.egg
+MANIFEST
+
+# PyInstaller
+#  Usually these files are written by a python script from a template
+#  before PyInstaller builds the exe, so as to inject date/other infos into it.
+*.manifest
+*.spec
+
+# Installer logs
+pip-log.txt
+pip-delete-this-directory.txt
+
+# Unit test / coverage reports
+htmlcov/
+.tox/
+.nox/
+.coverage
+.coverage.*
+.cache
+nosetests.xml
+coverage.xml
+*.cover
+*.py,cover
+.hypothesis/
+.pytest_cache/
+cover/
+
+# Translations
+*.mo
+*.pot
+
+# Django stuff:
+*.log
+local_settings.py
+db.sqlite3
+db.sqlite3-journal
+
+# Flask stuff:
+instance/
+.webassets-cache
+
+# Scrapy stuff:
+.scrapy
+
+# Sphinx documentation
+docs/_build/
+
+# PyBuilder
+.pybuilder/
+target/
+
+# Jupyter Notebook
+.ipynb_checkpoints
+
+# IPython
+profile_default/
+ipython_config.py
+
+# pyenv
+#   For a library or package, you might want to ignore these files since the code is
+#   intended to run in multiple environments; otherwise, check them in:
+# .python-version
+
+# pipenv
+#   According to pypa/pipenv#598, it is recommended to include Pipfile.lock in version control.
+#   However, in case of collaboration, if having platform-specific dependencies or dependencies
+#   having no cross-platform support, pipenv may install dependencies that don't work, or not
+#   install all needed dependencies.
+#Pipfile.lock
+
+# poetry
+#   Similar to Pipfile.lock, it is generally recommended to include poetry.lock in version control.
+#   This is especially recommended for binary packages to ensure reproducibility, and is more
+#   commonly ignored for libraries.
+#   https://python-poetry.org/docs/basic-usage/#commit-your-poetrylock-file-to-version-control
+#poetry.lock
+
+# pdm
+#   Similar to Pipfile.lock, it is generally recommended to include pdm.lock in version control.
+#pdm.lock
+#   pdm stores project-wide configurations in .pdm.toml, but it is recommended to not include it
+#   in version control.
+#   https://pdm.fming.dev/latest/usage/project/#working-with-version-control
+.pdm.toml
+.pdm-python
+.pdm-build/
+
+# PEP 582; used by e.g. github.com/David-OConnor/pyflow and github.com/pdm-project/pdm
+__pypackages__/
+
+# Celery stuff
+celerybeat-schedule
+celerybeat.pid
+
+# SageMath parsed files
+*.sage.py
+
+# Environments
+.env
+.venv
+env/
+venv/
+ENV/
+env.bak/
+venv.bak/
+
+# Spyder project settings
+.spyderproject
+.spyproject
+
+# Rope project settings
+.ropeproject
+
+# mkdocs documentation
+/site
+
+# mypy
+.mypy_cache/
+.dmypy.json
+dmypy.json
+
+# Pyre type checker
+.pyre/
+
+# pytype static type analyzer
+.pytype/
+
+# Cython debug symbols
+cython_debug/
+
+# PyCharm
+#  JetBrains specific template is maintained in a separate JetBrains.gitignore that can
+#  be found at https://github.com/github/gitignore/blob/main/Global/JetBrains.gitignore
+#  and can be added to the global gitignore or merged into this file.  For a more nuclear
+#  option (not recommended) you can uncomment the following to ignore the entire idea folder.
+#.idea/

LlamaGen/GETTING_STARTED.md

@@ -0,0 +1,61 @@
+## Getting Started
+### Requirements
+- Linux with Python ≥ 3.7
+- PyTorch ≥ 2.1
+- A100 GPUs
+
+### Train VQVAE models
+```
+bash scripts/tokenizer/train_vq.sh --cloud-save-path /path/to/cloud_disk --data-path /path/to/imagenet/train --image-size 256 --vq-model VQ-16
+```
+
+
+### Pre-extract discrete codes of training images
+```
+bash scripts/autoregressive/extract_codes_c2i.sh --vq-ckpt ./pretrained_models/vq_ds16_c2i.pt --data-path /path/to/imagenet/train --code-path /path/to/imagenet_code_c2i_flip_ten_crop --ten-crop --crop-range 1.1 --image-size 384
+```
+and/or
+``` 
+bash scripts/autoregressive/extract_codes_c2i.sh --vq-ckpt ./pretrained_models/vq_ds16_c2i.pt --data-path /path/to/imagenet/train --code-path /path/to/imagenet_code_c2i_flip_ten_crop_105 --ten-crop --crop-range 1.05 --image-size 384
+```
+
+
+### Train AR models with DDP
+Before running, please change `nnodes, nproc_per_node, node_rank, master_addr, master_port` in `.sh`
+```
+bash scripts/autoregressive/train_c2i.sh --cloud-save-path /path/to/cloud_disk --code-path /path/to/imagenet_code_c2i_flip_ten_crop --image-size 384 --gpt-model GPT-B
+
+bash scripts/autoregressive/train_c2i.sh --cloud-save-path /path/to/cloud_disk --code-path /path/to/imagenet_code_c2i_flip_ten_crop --image-size 384 --gpt-model GPT-L
+
+bash scripts/autoregressive/train_c2i.sh --cloud-save-path /path/to/cloud_disk --code-path /path/to/imagenet_code_c2i_flip_ten_crop --image-size 384 --gpt-model GPT-XL
+```
+
+
+### Train AR models with FSDP
+Before running, please change `nnodes, nproc_per_node, node_rank, master_addr, master_port` in `.sh`
+```
+bash scripts/autoregressive/train_c2i_fsdp.sh --cloud-save-path /path/to/cloud_disk --code-path /path/to/imagenet_code_c2i_flip_ten_crop --image-size 384 --gpt-model GPT-XXL
+
+bash scripts/autoregressive/train_c2i_fsdp.sh --cloud-save-path /path/to/cloud_disk --code-path /path/to/imagenet_code_c2i_flip_ten_crop --image-size 384 --gpt-model GPT-3B
+```
+
+
+### Sampling
+```
+bash scripts/autoregressive/sample_c2i.sh --vq-ckpt ./pretrained_models/vq_ds16_c2i.pt --gpt-ckpt ./pretrained_models/c2i_B.pt --gpt-model GPT-B --image-size 384 --image-size-eval 256 --cfg-scale 2.0
+
+bash scripts/autoregressive/sample_c2i.sh --vq-ckpt ./pretrained_models/vq_ds16_c2i.pt --gpt-ckpt ./pretrained_models/c2i_L.pt --gpt-model GPT-L --image-size 384 --image-size-eval 256 --cfg-scale 2.0
+
+bash scripts/autoregressive/sample_c2i.sh --vq-ckpt ./pretrained_models/vq_ds16_c2i.pt --gpt-ckpt ./pretrained_models/c2i_XL.pt --gpt-model GPT-XL --image-size 384 --image-size-eval 256 --cfg-scale 1.75
+
+bash scripts/autoregressive/sample_c2i.sh --vq-ckpt ./pretrained_models/vq_ds16_c2i.pt --gpt-ckpt ./pretrained_models/c2i_XXL.pt --gpt-model GPT-XXL --from-fsdp --image-size 384 --image-size-eval 256 --cfg-scale 1.75
+
+bash scripts/autoregressive/sample_c2i.sh --vq-ckpt ./pretrained_models/vq_ds16_c2i.pt --gpt-ckpt ./pretrained_models/c2i_3B.pt --gpt-model GPT-3B --from-fsdp --image-size 384 --image-size-eval 256 --cfg-scale 1.65
+```
+
+
+### Evaluation
+Before evaluation, please refer [evaluation readme](evaluations/c2i/README.md) to install required packages. 
+```
+python3 evaluations/c2i/evaluator.py VIRTUAL_imagenet256_labeled.npz samples/GPT-B-c2i_B-size-384-size-256-VQ-16-topk-0-topp-1.0-temperature-1.0-cfg-2.0-seed-0.npz
+```

LlamaGen/GFT.sh

@@ -0,0 +1,26 @@
+set -x
+port=12345
+
+torchrun --nproc_per_node=8 --nnodes=1 --node_rank=0 --master_addr="127.0.0.1" --master_port=$port autoregressive/train/finetune_c2i_fsdp.py \
+    --global-batch-size 256  --epochs=15 --ckpt-every=25000 --gradient-accumulation-steps=2 \
+    --lr=2e-4 --expid="3843B_beta-1_lr24_ep15_maskbeta_detach_betainput2_refeval_cosinelr" --beta=-1.0 \
+    --loss_type="detachu" \
+    --beta_input_type=2 \
+    --ref_eval=1 \
+    --cosinelr=1 \
+    --ref_ckpt="/data/home/chenhuayu/LlamaGen/c2i_3B_384.pt" \
+    --code-path="/data/home/chenhuayu/imagenet384_train_code_c2i_flip_ten_crop/" \
+    --results-dir="results2" \
+    --image-size=384 --gpt-model="GPT-3B"
+port=$((port + 1))
+
+
+# torchrun --nproc_per_node=8 --nnodes=1 --node_rank=0 --master_addr="127.0.0.1" --master_port=$port autoregressive/train/finetune_c2i.py \
+#     --global-batch-size 768 --gradient-accumulation-step 2 --epochs=300 --ckpt-every=100000 \
+#     --lr=1e-4 --expid="384L_beta-1_lr14_bz768_ep1_maskbeta_detach_betainput2_fromscratch_cosinelr_fix" --beta=-1 --cosinelr=1 \
+#     --loss_type="detachu" --beta_input_type=2 \
+#     --ref_eval=1 \
+#     --code-path="/data/home/chenhuayu/imagenet384_train_code_c2i_flip_ten_crop/" \
+#     --results-dir="results2" \
+#     --image-size=384 --gpt-model="GPT-L"
+# port=$((port + 1))

LlamaGen/LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2024 FoundationVision
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.