Upload folder using huggingface_hub
Browse files- .gitattributes +2 -0
- debug-internal.log +7 -7
- debug.log +26 -24
- offline-run-20250328_203038-ni3gydc1/files/output.log +0 -0
- offline-run-20250328_203038-ni3gydc1/files/requirements.txt +167 -0
- offline-run-20250328_203038-ni3gydc1/files/wandb-metadata.json +105 -0
- offline-run-20250328_203038-ni3gydc1/logs/debug-core.log +7 -0
- offline-run-20250328_203038-ni3gydc1/logs/debug-internal.log +7 -0
- offline-run-20250328_203038-ni3gydc1/logs/debug.log +26 -0
- offline-run-20250328_203038-ni3gydc1/run-ni3gydc1.wandb +3 -0
- offline-run-20250328_203844-gc2qytpj/files/output.log +0 -0
- offline-run-20250328_203844-gc2qytpj/files/requirements.txt +167 -0
- offline-run-20250328_203844-gc2qytpj/files/wandb-metadata.json +105 -0
- offline-run-20250328_203844-gc2qytpj/logs/debug-core.log +7 -0
- offline-run-20250328_203844-gc2qytpj/logs/debug-internal.log +7 -0
- offline-run-20250328_203844-gc2qytpj/logs/debug.log +26 -0
- offline-run-20250328_203844-gc2qytpj/run-gc2qytpj.wandb +3 -0
.gitattributes
CHANGED
|
@@ -35,3 +35,5 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
| 36 |
run-20250329_010934-3x35hjks/run-3x35hjks.wandb filter=lfs diff=lfs merge=lfs -text
|
| 37 |
run-20250329_012205-co1ecmky/run-co1ecmky.wandb filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
|
|
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
| 36 |
run-20250329_010934-3x35hjks/run-3x35hjks.wandb filter=lfs diff=lfs merge=lfs -text
|
| 37 |
run-20250329_012205-co1ecmky/run-co1ecmky.wandb filter=lfs diff=lfs merge=lfs -text
|
| 38 |
+
offline-run-20250328_203038-ni3gydc1/run-ni3gydc1.wandb filter=lfs diff=lfs merge=lfs -text
|
| 39 |
+
offline-run-20250328_203844-gc2qytpj/run-gc2qytpj.wandb filter=lfs diff=lfs merge=lfs -text
|
debug-internal.log
CHANGED
|
@@ -1,7 +1,7 @@
|
|
| 1 |
-
{"time":"2025-03-
|
| 2 |
-
{"time":"2025-03-
|
| 3 |
-
{"time":"2025-03-
|
| 4 |
-
{"time":"2025-03-
|
| 5 |
-
{"time":"2025-03-
|
| 6 |
-
{"time":"2025-03-
|
| 7 |
-
{"time":"2025-03-
|
|
|
|
| 1 |
+
{"time":"2025-03-28T20:38:44.692258573+08:00","level":"INFO","msg":"stream: starting","core version":"0.19.8","symlink path":"../outputs/test_13B/wandb/offline-run-20250328_203844-gc2qytpj/logs/debug-core.log"}
|
| 2 |
+
{"time":"2025-03-28T20:38:44.90653769+08:00","level":"INFO","msg":"created new stream","id":"gc2qytpj"}
|
| 3 |
+
{"time":"2025-03-28T20:38:44.90660149+08:00","level":"INFO","msg":"stream: started","id":"gc2qytpj"}
|
| 4 |
+
{"time":"2025-03-28T20:38:44.906615038+08:00","level":"INFO","msg":"handler: started","stream_id":"gc2qytpj"}
|
| 5 |
+
{"time":"2025-03-28T20:38:44.906633311+08:00","level":"INFO","msg":"writer: Do: started","stream_id":"gc2qytpj"}
|
| 6 |
+
{"time":"2025-03-28T20:38:44.90663431+08:00","level":"INFO","msg":"sender: started","stream_id":"gc2qytpj"}
|
| 7 |
+
{"time":"2025-03-28T20:38:44.910439284+08:00","level":"INFO","msg":"Starting system monitor"}
|
debug.log
CHANGED
|
@@ -1,24 +1,26 @@
|
|
| 1 |
-
2025-03-
|
| 2 |
-
2025-03-
|
| 3 |
-
2025-03-
|
| 4 |
-
2025-03-
|
| 5 |
-
2025-03-
|
| 6 |
-
2025-03-
|
| 7 |
-
2025-03-
|
| 8 |
-
2025-03-
|
| 9 |
-
2025-03-
|
| 10 |
-
config: {'train_cfgs': {'save_checkpoint':
|
| 11 |
-
2025-03-
|
| 12 |
-
2025-03-
|
| 13 |
-
2025-03-
|
| 14 |
-
2025-03-
|
| 15 |
-
2025-03-
|
| 16 |
-
2025-03-
|
| 17 |
-
2025-03-
|
| 18 |
-
2025-03-
|
| 19 |
-
2025-03-
|
| 20 |
-
2025-03-
|
| 21 |
-
2025-03-
|
| 22 |
-
2025-03-
|
| 23 |
-
2025-03-
|
| 24 |
-
2025-03-
|
|
|
|
|
|
|
|
|
| 1 |
+
2025-03-28 20:38:44,447 INFO MainThread:3871311 [wandb_setup.py:_flush():67] Current SDK version is 0.19.8
|
| 2 |
+
2025-03-28 20:38:44,447 INFO MainThread:3871311 [wandb_setup.py:_flush():67] Configure stats pid to 3871311
|
| 3 |
+
2025-03-28 20:38:44,447 INFO MainThread:3871311 [wandb_setup.py:_flush():67] Loading settings from /home/yangyaodong/.config/wandb/settings
|
| 4 |
+
2025-03-28 20:38:44,447 INFO MainThread:3871311 [wandb_setup.py:_flush():67] Loading settings from /aifs4su/yaodong/hantao/align-anything/scripts/wandb/settings
|
| 5 |
+
2025-03-28 20:38:44,447 INFO MainThread:3871311 [wandb_setup.py:_flush():67] Loading settings from environment variables
|
| 6 |
+
2025-03-28 20:38:44,447 INFO MainThread:3871311 [wandb_init.py:setup_run_log_directory():647] Logging user logs to ../outputs/test_13B/wandb/offline-run-20250328_203844-gc2qytpj/logs/debug.log
|
| 7 |
+
2025-03-28 20:38:44,447 INFO MainThread:3871311 [wandb_init.py:setup_run_log_directory():648] Logging internal logs to ../outputs/test_13B/wandb/offline-run-20250328_203844-gc2qytpj/logs/debug-internal.log
|
| 8 |
+
2025-03-28 20:38:44,447 INFO MainThread:3871311 [wandb_init.py:init():761] calling init triggers
|
| 9 |
+
2025-03-28 20:38:44,447 INFO MainThread:3871311 [wandb_init.py:init():766] wandb.init called with sweep_config: {}
|
| 10 |
+
config: {'train_cfgs': {'save_checkpoint': True, 'load_checkpoint': False, 'ds_cfgs': 'ds_z3_config.json', 'epochs': 3, 'seed': 42, 'per_device_train_batch_size': 1, 'per_device_eval_batch_size': 1, 'gradient_accumulation_steps': 1, 'gradient_checkpointing': True, 'learning_rate': 1e-06, 'lr_scheduler_type': 'cosine', 'lr_warmup_ratio': 0.03, 'weight_decay': 0.0, 'adam_betas': [0.9, 0.95], 'bf16': True, 'fp16': False, 'eval_strategy': 'epoch', 'eval_interval': 10, 'regularization': 0.001, 'scale_coeff': 0.1, 'freeze_mm_proj': False, 'freeze_vision_tower': True, 'freeze_language_model': False}, 'data_cfgs': {'train_datasets': '/aifs4su/yaodong/hantao/datasets/AA_preference_vicuna-13b_cosi_cut/merged/top1-10', 'train_template': 'AA_TI2T_LLAVA', 'train_size': {}, 'train_split': 'train', 'train_name': 'text-image-to-text', 'train_data_files': {}, 'train_optional_args': [], 'eval_datasets': {}, 'eval_template': {}, 'eval_size': {}, 'eval_split': {}, 'eval_subset': {}, 'eval_data_files': {}, 'eval_optional_args': []}, 'logger_cfgs': {'log_type': 'wandb', 'log_project': 'align-anything', 'log_run_name': 'dpo', 'output_dir': '../outputs/test_13B', 'cache_dir': {}, 'save_total_limit': 6}, 'model_cfgs': {'model_name_or_path': '/aifs4su/yaodong/hantao/models/llava-v1.6-vicuna-13b-hf', 'trust_remote_code': True, 'model_max_length': 2048}, 'special_tokens': {}, '_wandb': {}}
|
| 11 |
+
2025-03-28 20:38:44,447 INFO MainThread:3871311 [wandb_init.py:init():784] starting backend
|
| 12 |
+
2025-03-28 20:38:44,657 INFO MainThread:3871311 [wandb_init.py:init():788] sending inform_init request
|
| 13 |
+
2025-03-28 20:38:44,660 INFO MainThread:3871311 [backend.py:_multiprocessing_setup():101] multiprocessing start_methods=fork,spawn,forkserver, using: spawn
|
| 14 |
+
2025-03-28 20:38:44,660 INFO MainThread:3871311 [wandb_init.py:init():798] backend started and connected
|
| 15 |
+
2025-03-28 20:38:44,662 INFO MainThread:3871311 [wandb_init.py:init():891] updated telemetry
|
| 16 |
+
2025-03-28 20:38:44,672 INFO MainThread:3871311 [wandb_init.py:init():915] communicating run to backend with 90.0 second timeout
|
| 17 |
+
2025-03-28 20:38:44,908 INFO MainThread:3871311 [wandb_init.py:init():990] starting run threads in backend
|
| 18 |
+
2025-03-28 20:38:45,071 INFO MainThread:3871311 [wandb_run.py:_console_start():2375] atexit reg
|
| 19 |
+
2025-03-28 20:38:45,071 INFO MainThread:3871311 [wandb_run.py:_redirect():2227] redirect: wrap_raw
|
| 20 |
+
2025-03-28 20:38:45,071 INFO MainThread:3871311 [wandb_run.py:_redirect():2292] Wrapping output streams.
|
| 21 |
+
2025-03-28 20:38:45,071 INFO MainThread:3871311 [wandb_run.py:_redirect():2315] Redirects installed.
|
| 22 |
+
2025-03-28 20:38:45,074 INFO MainThread:3871311 [wandb_init.py:init():1032] run started, returning control to user process
|
| 23 |
+
2025-03-28 20:47:40,230 INFO MainThread:3871311 [wandb_run.py:_finish():2112] finishing run align-anything/gc2qytpj
|
| 24 |
+
2025-03-28 20:47:40,230 INFO MainThread:3871311 [wandb_run.py:_atexit_cleanup():2340] got exitcode: 0
|
| 25 |
+
2025-03-28 20:47:40,230 INFO MainThread:3871311 [wandb_run.py:_restore():2322] restore
|
| 26 |
+
2025-03-28 20:47:40,230 INFO MainThread:3871311 [wandb_run.py:_restore():2328] restore done
|
offline-run-20250328_203038-ni3gydc1/files/output.log
ADDED
|
File without changes
|
offline-run-20250328_203038-ni3gydc1/files/requirements.txt
ADDED
|
@@ -0,0 +1,167 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
maskrcnn_benchmark==0.0.0
|
| 2 |
+
webdataset==0.2.111
|
| 3 |
+
websockets==15.0.1
|
| 4 |
+
typer==0.15.2
|
| 5 |
+
blobfile==3.0.0
|
| 6 |
+
pooch==1.8.2
|
| 7 |
+
python-dateutil==2.9.0.post0
|
| 8 |
+
gmpy2==2.2.1
|
| 9 |
+
httpcore==1.0.7
|
| 10 |
+
charset-normalizer==3.3.2
|
| 11 |
+
torchlibrosa==0.1.0
|
| 12 |
+
multiprocess==0.70.16
|
| 13 |
+
Werkzeug==3.1.3
|
| 14 |
+
aiofiles==23.2.1
|
| 15 |
+
six==1.17.0
|
| 16 |
+
typing_extensions==4.12.2
|
| 17 |
+
psutil==7.0.0
|
| 18 |
+
frozenlist==1.5.0
|
| 19 |
+
einops==0.8.1
|
| 20 |
+
flash_attn==2.7.4.post1
|
| 21 |
+
PySocks==1.7.1
|
| 22 |
+
regex==2024.11.6
|
| 23 |
+
markdown-it-py==3.0.0
|
| 24 |
+
ruff==0.11.2
|
| 25 |
+
docker-pycreds==0.4.0
|
| 26 |
+
protobuf==5.29.4
|
| 27 |
+
resampy==0.4.3
|
| 28 |
+
aiohappyeyeballs==2.6.1
|
| 29 |
+
httpx==0.28.1
|
| 30 |
+
encodec==0.1.1
|
| 31 |
+
ffmpy==0.5.0
|
| 32 |
+
mkl_random==1.2.8
|
| 33 |
+
soxr==0.5.0.post1
|
| 34 |
+
absl-py==2.2.1
|
| 35 |
+
networkx==3.4.2
|
| 36 |
+
h5py==3.13.0
|
| 37 |
+
hjson==3.1.0
|
| 38 |
+
tensorboard==2.19.0
|
| 39 |
+
aiosignal==1.3.2
|
| 40 |
+
pip==25.0
|
| 41 |
+
triton==3.1.0
|
| 42 |
+
zipp==3.21.0
|
| 43 |
+
ftfy==6.3.1
|
| 44 |
+
attrs==25.3.0
|
| 45 |
+
requests==2.32.3
|
| 46 |
+
progressbar==2.5
|
| 47 |
+
sniffio==1.3.1
|
| 48 |
+
lxml==5.3.1
|
| 49 |
+
starlette==0.46.1
|
| 50 |
+
Markdown==3.7
|
| 51 |
+
mdurl==0.1.2
|
| 52 |
+
torchaudio==2.5.1
|
| 53 |
+
safetensors==0.5.3
|
| 54 |
+
opencv-python==4.6.0.66
|
| 55 |
+
torchvision==0.20.1
|
| 56 |
+
shellingham==1.5.4
|
| 57 |
+
gradio==5.23.1
|
| 58 |
+
timm==1.0.15
|
| 59 |
+
multidict==6.2.0
|
| 60 |
+
semantic-version==2.10.0
|
| 61 |
+
numba==0.60.0
|
| 62 |
+
gradio_client==1.8.0
|
| 63 |
+
pydantic_core==2.33.0
|
| 64 |
+
dill==0.3.8
|
| 65 |
+
msgpack==1.1.0
|
| 66 |
+
sentry-sdk==2.24.1
|
| 67 |
+
grpcio==1.71.0
|
| 68 |
+
cffi==1.17.1
|
| 69 |
+
PyYAML==6.0.2
|
| 70 |
+
tensorboard-data-server==0.7.2
|
| 71 |
+
fastapi==0.115.12
|
| 72 |
+
lazy_loader==0.4
|
| 73 |
+
mkl_fft==1.3.11
|
| 74 |
+
annotated-types==0.7.0
|
| 75 |
+
scikit-learn==1.6.1
|
| 76 |
+
wget==3.2
|
| 77 |
+
setuptools==75.8.0
|
| 78 |
+
certifi==2025.1.31
|
| 79 |
+
click==8.1.8
|
| 80 |
+
laion_clap==1.1.5
|
| 81 |
+
Pygments==2.19.1
|
| 82 |
+
tomlkit==0.13.2
|
| 83 |
+
idna==3.7
|
| 84 |
+
propcache==0.3.1
|
| 85 |
+
platformdirs==4.3.7
|
| 86 |
+
align-anything==0.0.1.dev0
|
| 87 |
+
deepspeed==0.16.5
|
| 88 |
+
smmap==5.0.2
|
| 89 |
+
pillow==11.1.0
|
| 90 |
+
typing-inspection==0.4.0
|
| 91 |
+
braceexpand==0.1.7
|
| 92 |
+
decorator==5.2.1
|
| 93 |
+
pandas==2.2.3
|
| 94 |
+
huggingface-hub==0.29.3
|
| 95 |
+
pyarrow==19.0.1
|
| 96 |
+
tokenizers==0.21.1
|
| 97 |
+
GitPython==3.1.44
|
| 98 |
+
xxhash==3.5.0
|
| 99 |
+
packaging==24.2
|
| 100 |
+
numpy==1.23.4
|
| 101 |
+
setproctitle==1.3.5
|
| 102 |
+
llvmlite==0.43.0
|
| 103 |
+
tiktoken==0.9.0
|
| 104 |
+
mpmath==1.3.0
|
| 105 |
+
nvidia-ml-py==12.570.86
|
| 106 |
+
pydantic==2.11.0
|
| 107 |
+
datasets==3.5.0
|
| 108 |
+
librosa==0.11.0
|
| 109 |
+
frechet_audio_distance==0.3.1
|
| 110 |
+
sympy==1.13.1
|
| 111 |
+
safehttpx==0.1.6
|
| 112 |
+
Jinja2==3.1.6
|
| 113 |
+
h11==0.14.0
|
| 114 |
+
aiohttp==3.11.14
|
| 115 |
+
diffusers==0.32.2
|
| 116 |
+
tqdm==4.67.1
|
| 117 |
+
filelock==3.13.1
|
| 118 |
+
transformers==4.51.0.dev0
|
| 119 |
+
scipy==1.10.1
|
| 120 |
+
audioread==3.0.1
|
| 121 |
+
sentencepiece==0.2.0
|
| 122 |
+
pytz==2025.2
|
| 123 |
+
tzdata==2025.2
|
| 124 |
+
python-multipart==0.0.20
|
| 125 |
+
urllib3==2.3.0
|
| 126 |
+
pycryptodomex==3.22.0
|
| 127 |
+
yarl==1.18.3
|
| 128 |
+
pydub==0.25.1
|
| 129 |
+
pycparser==2.22
|
| 130 |
+
soundfile==0.13.1
|
| 131 |
+
wcwidth==0.2.13
|
| 132 |
+
groovy==0.1.2
|
| 133 |
+
torch==2.5.1
|
| 134 |
+
anyio==4.9.0
|
| 135 |
+
wandb==0.19.8
|
| 136 |
+
joblib==1.4.2
|
| 137 |
+
fsspec==2024.12.0
|
| 138 |
+
peft==0.15.1
|
| 139 |
+
accelerate==1.5.2
|
| 140 |
+
py-cpuinfo==9.0.0
|
| 141 |
+
uvicorn==0.34.0
|
| 142 |
+
orjson==3.10.16
|
| 143 |
+
Brotli==1.0.9
|
| 144 |
+
rich==13.9.4
|
| 145 |
+
importlib_metadata==8.6.1
|
| 146 |
+
ninja==1.11.1.4
|
| 147 |
+
wheel==0.45.1
|
| 148 |
+
MarkupSafe==3.0.2
|
| 149 |
+
threadpoolctl==3.6.0
|
| 150 |
+
gitdb==4.0.12
|
| 151 |
+
mkl-service==2.4.0
|
| 152 |
+
typing_extensions==4.12.2
|
| 153 |
+
tomli==2.0.1
|
| 154 |
+
zipp==3.19.2
|
| 155 |
+
wheel==0.43.0
|
| 156 |
+
jaraco.text==3.12.1
|
| 157 |
+
packaging==24.2
|
| 158 |
+
autocommand==2.2.2
|
| 159 |
+
jaraco.functools==4.0.1
|
| 160 |
+
jaraco.collections==5.1.0
|
| 161 |
+
platformdirs==4.2.2
|
| 162 |
+
more-itertools==10.3.0
|
| 163 |
+
inflect==7.3.1
|
| 164 |
+
jaraco.context==5.3.0
|
| 165 |
+
typeguard==4.3.0
|
| 166 |
+
backports.tarfile==1.2.0
|
| 167 |
+
importlib_metadata==8.0.0
|
offline-run-20250328_203038-ni3gydc1/files/wandb-metadata.json
ADDED
|
@@ -0,0 +1,105 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"os": "Linux-5.15.0-1040-nvidia-x86_64-with-glibc2.35",
|
| 3 |
+
"python": "CPython 3.11.11",
|
| 4 |
+
"startedAt": "2025-03-28T12:30:39.058990Z",
|
| 5 |
+
"args": [
|
| 6 |
+
"--local_rank=0",
|
| 7 |
+
"--model_name_or_path",
|
| 8 |
+
"/aifs4su/yaodong/hantao/models/llava-v1.6-vicuna-13b-hf",
|
| 9 |
+
"--train_datasets",
|
| 10 |
+
"/aifs4su/yaodong/hantao/datasets/AA_preference_vicuna-13b_cosi_cut/merged/top1-10",
|
| 11 |
+
"--train_template",
|
| 12 |
+
"AA_TI2T_LLAVA",
|
| 13 |
+
"--train_split",
|
| 14 |
+
"train",
|
| 15 |
+
"--train_name",
|
| 16 |
+
"text-image-to-text",
|
| 17 |
+
"--output_dir",
|
| 18 |
+
"../outputs/test_13B",
|
| 19 |
+
"--save_total_limit",
|
| 20 |
+
"3",
|
| 21 |
+
"--train_batch_size",
|
| 22 |
+
"1",
|
| 23 |
+
"--epochs",
|
| 24 |
+
"2"
|
| 25 |
+
],
|
| 26 |
+
"program": "-m align_anything.trainers.text_image_to_text.dpo",
|
| 27 |
+
"git": {
|
| 28 |
+
"remote": "git@github.com-hantao:PKU-Alignment/align-anything.git",
|
| 29 |
+
"commit": "106588f9802757a3283c1aff1f33ea9afd737f31"
|
| 30 |
+
},
|
| 31 |
+
"root": "../outputs/test_13B",
|
| 32 |
+
"host": "dgx-092",
|
| 33 |
+
"executable": "/aifs4su/yaodong/miniconda3/envs/hantao_llama/bin/python",
|
| 34 |
+
"cpu_count": 112,
|
| 35 |
+
"cpu_count_logical": 224,
|
| 36 |
+
"gpu": "NVIDIA H800",
|
| 37 |
+
"gpu_count": 8,
|
| 38 |
+
"disk": {
|
| 39 |
+
"/": {
|
| 40 |
+
"total": "1888556142592",
|
| 41 |
+
"used": "148592041984"
|
| 42 |
+
}
|
| 43 |
+
},
|
| 44 |
+
"memory": {
|
| 45 |
+
"total": "2164195454976"
|
| 46 |
+
},
|
| 47 |
+
"cpu": {
|
| 48 |
+
"count": 112,
|
| 49 |
+
"countLogical": 224
|
| 50 |
+
},
|
| 51 |
+
"gpu_nvidia": [
|
| 52 |
+
{
|
| 53 |
+
"name": "NVIDIA H800",
|
| 54 |
+
"memoryTotal": "85520809984",
|
| 55 |
+
"cudaCores": 16896,
|
| 56 |
+
"architecture": "Hopper"
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"name": "NVIDIA H800",
|
| 60 |
+
"memoryTotal": "85520809984",
|
| 61 |
+
"cudaCores": 16896,
|
| 62 |
+
"architecture": "Hopper"
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"name": "NVIDIA H800",
|
| 66 |
+
"memoryTotal": "85520809984",
|
| 67 |
+
"cudaCores": 16896,
|
| 68 |
+
"architecture": "Hopper"
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"name": "NVIDIA H800",
|
| 72 |
+
"memoryTotal": "85520809984",
|
| 73 |
+
"cudaCores": 16896,
|
| 74 |
+
"architecture": "Hopper"
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"name": "NVIDIA H800",
|
| 78 |
+
"memoryTotal": "85520809984",
|
| 79 |
+
"cudaCores": 16896,
|
| 80 |
+
"architecture": "Hopper"
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"name": "NVIDIA H800",
|
| 84 |
+
"memoryTotal": "85520809984",
|
| 85 |
+
"cudaCores": 16896,
|
| 86 |
+
"architecture": "Hopper"
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"name": "NVIDIA H800",
|
| 90 |
+
"memoryTotal": "85520809984",
|
| 91 |
+
"cudaCores": 16896,
|
| 92 |
+
"architecture": "Hopper"
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"name": "NVIDIA H800",
|
| 96 |
+
"memoryTotal": "85520809984",
|
| 97 |
+
"cudaCores": 16896,
|
| 98 |
+
"architecture": "Hopper"
|
| 99 |
+
}
|
| 100 |
+
],
|
| 101 |
+
"slurm": {
|
| 102 |
+
"conf": "/cm/shared/apps/slurm/var/etc/slurm/slurm.conf"
|
| 103 |
+
},
|
| 104 |
+
"cudaVersion": "12.2"
|
| 105 |
+
}
|
offline-run-20250328_203038-ni3gydc1/logs/debug-core.log
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{"time":"2025-03-28T20:30:38.872129247+08:00","level":"INFO","msg":"main: starting server","port-filename":"/tmp/tmpmsjq2qrj/port-3856880.txt","pid":3856880,"log-level":0,"disable-analytics":false,"shutdown-on-parent-exit":false}
|
| 2 |
+
{"time":"2025-03-28T20:30:38.873015217+08:00","level":"INFO","msg":"Will exit if parent process dies.","ppid":3856880}
|
| 3 |
+
{"time":"2025-03-28T20:30:38.873005488+08:00","level":"INFO","msg":"server is running","addr":{"IP":"127.0.0.1","Port":44091,"Zone":""}}
|
| 4 |
+
{"time":"2025-03-28T20:30:39.052962944+08:00","level":"INFO","msg":"connection: ManageConnectionData: new connection created","id":"127.0.0.1:34304"}
|
| 5 |
+
{"time":"2025-03-28T20:30:39.059989776+08:00","level":"INFO","msg":"handleInformInit: received","streamId":"ni3gydc1","id":"127.0.0.1:34304"}
|
| 6 |
+
{"time":"2025-03-28T20:30:39.456558723+08:00","level":"INFO","msg":"handleInformInit: stream started","streamId":"ni3gydc1","id":"127.0.0.1:34304"}
|
| 7 |
+
{"time":"2025-03-28T20:36:33.911468496+08:00","level":"INFO","msg":"received shutdown signal","signal":15}
|
offline-run-20250328_203038-ni3gydc1/logs/debug-internal.log
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{"time":"2025-03-28T20:30:39.242295984+08:00","level":"INFO","msg":"stream: starting","core version":"0.19.8","symlink path":"../outputs/test_13B/wandb/offline-run-20250328_203038-ni3gydc1/logs/debug-core.log"}
|
| 2 |
+
{"time":"2025-03-28T20:30:39.456487825+08:00","level":"INFO","msg":"created new stream","id":"ni3gydc1"}
|
| 3 |
+
{"time":"2025-03-28T20:30:39.456549716+08:00","level":"INFO","msg":"stream: started","id":"ni3gydc1"}
|
| 4 |
+
{"time":"2025-03-28T20:30:39.456572806+08:00","level":"INFO","msg":"writer: Do: started","stream_id":"ni3gydc1"}
|
| 5 |
+
{"time":"2025-03-28T20:30:39.456577964+08:00","level":"INFO","msg":"handler: started","stream_id":"ni3gydc1"}
|
| 6 |
+
{"time":"2025-03-28T20:30:39.456615028+08:00","level":"INFO","msg":"sender: started","stream_id":"ni3gydc1"}
|
| 7 |
+
{"time":"2025-03-28T20:30:39.462148507+08:00","level":"INFO","msg":"Starting system monitor"}
|
offline-run-20250328_203038-ni3gydc1/logs/debug.log
ADDED
|
@@ -0,0 +1,26 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
2025-03-28 20:30:38,838 INFO MainThread:3856880 [wandb_setup.py:_flush():67] Current SDK version is 0.19.8
|
| 2 |
+
2025-03-28 20:30:38,839 INFO MainThread:3856880 [wandb_setup.py:_flush():67] Configure stats pid to 3856880
|
| 3 |
+
2025-03-28 20:30:38,839 INFO MainThread:3856880 [wandb_setup.py:_flush():67] Loading settings from /home/yangyaodong/.config/wandb/settings
|
| 4 |
+
2025-03-28 20:30:38,839 INFO MainThread:3856880 [wandb_setup.py:_flush():67] Loading settings from /aifs4su/yaodong/hantao/align-anything/scripts/wandb/settings
|
| 5 |
+
2025-03-28 20:30:38,839 INFO MainThread:3856880 [wandb_setup.py:_flush():67] Loading settings from environment variables
|
| 6 |
+
2025-03-28 20:30:38,839 INFO MainThread:3856880 [wandb_init.py:setup_run_log_directory():647] Logging user logs to ../outputs/test_13B/wandb/offline-run-20250328_203038-ni3gydc1/logs/debug.log
|
| 7 |
+
2025-03-28 20:30:38,839 INFO MainThread:3856880 [wandb_init.py:setup_run_log_directory():648] Logging internal logs to ../outputs/test_13B/wandb/offline-run-20250328_203038-ni3gydc1/logs/debug-internal.log
|
| 8 |
+
2025-03-28 20:30:38,839 INFO MainThread:3856880 [wandb_init.py:init():761] calling init triggers
|
| 9 |
+
2025-03-28 20:30:38,839 INFO MainThread:3856880 [wandb_init.py:init():766] wandb.init called with sweep_config: {}
|
| 10 |
+
config: {'train_cfgs': {'save_checkpoint': True, 'load_checkpoint': False, 'ds_cfgs': 'ds_z3_config.json', 'epochs': 2, 'seed': 42, 'per_device_train_batch_size': 1, 'per_device_eval_batch_size': 1, 'gradient_accumulation_steps': 1, 'gradient_checkpointing': True, 'learning_rate': 1e-06, 'lr_scheduler_type': 'cosine', 'lr_warmup_ratio': 0.03, 'weight_decay': 0.0, 'adam_betas': [0.9, 0.95], 'bf16': True, 'fp16': False, 'eval_strategy': 'epoch', 'eval_interval': 10, 'regularization': 0.001, 'scale_coeff': 0.1, 'freeze_mm_proj': False, 'freeze_vision_tower': True, 'freeze_language_model': False}, 'data_cfgs': {'train_datasets': '/aifs4su/yaodong/hantao/datasets/AA_preference_vicuna-13b_cosi_cut/merged/top1-10', 'train_template': 'AA_TI2T_LLAVA', 'train_size': {}, 'train_split': 'train', 'train_name': 'text-image-to-text', 'train_data_files': {}, 'train_optional_args': [], 'eval_datasets': {}, 'eval_template': {}, 'eval_size': {}, 'eval_split': {}, 'eval_subset': {}, 'eval_data_files': {}, 'eval_optional_args': []}, 'logger_cfgs': {'log_type': 'wandb', 'log_project': 'align-anything', 'log_run_name': 'dpo', 'output_dir': '../outputs/test_13B', 'cache_dir': {}, 'save_total_limit': 3}, 'model_cfgs': {'model_name_or_path': '/aifs4su/yaodong/hantao/models/llava-v1.6-vicuna-13b-hf', 'trust_remote_code': True, 'model_max_length': 2048}, 'special_tokens': {}, '_wandb': {}}
|
| 11 |
+
2025-03-28 20:30:38,839 INFO MainThread:3856880 [wandb_init.py:init():784] starting backend
|
| 12 |
+
2025-03-28 20:30:39,053 INFO MainThread:3856880 [wandb_init.py:init():788] sending inform_init request
|
| 13 |
+
2025-03-28 20:30:39,058 INFO MainThread:3856880 [backend.py:_multiprocessing_setup():101] multiprocessing start_methods=fork,spawn,forkserver, using: spawn
|
| 14 |
+
2025-03-28 20:30:39,058 INFO MainThread:3856880 [wandb_init.py:init():798] backend started and connected
|
| 15 |
+
2025-03-28 20:30:39,060 INFO MainThread:3856880 [wandb_init.py:init():891] updated telemetry
|
| 16 |
+
2025-03-28 20:30:39,069 INFO MainThread:3856880 [wandb_init.py:init():915] communicating run to backend with 90.0 second timeout
|
| 17 |
+
2025-03-28 20:30:39,458 INFO MainThread:3856880 [wandb_init.py:init():990] starting run threads in backend
|
| 18 |
+
2025-03-28 20:30:39,624 INFO MainThread:3856880 [wandb_run.py:_console_start():2375] atexit reg
|
| 19 |
+
2025-03-28 20:30:39,624 INFO MainThread:3856880 [wandb_run.py:_redirect():2227] redirect: wrap_raw
|
| 20 |
+
2025-03-28 20:30:39,624 INFO MainThread:3856880 [wandb_run.py:_redirect():2292] Wrapping output streams.
|
| 21 |
+
2025-03-28 20:30:39,624 INFO MainThread:3856880 [wandb_run.py:_redirect():2315] Redirects installed.
|
| 22 |
+
2025-03-28 20:30:39,627 INFO MainThread:3856880 [wandb_init.py:init():1032] run started, returning control to user process
|
| 23 |
+
2025-03-28 20:36:33,901 INFO MainThread:3856880 [wandb_run.py:_finish():2112] finishing run align-anything/ni3gydc1
|
| 24 |
+
2025-03-28 20:36:33,902 INFO MainThread:3856880 [wandb_run.py:_atexit_cleanup():2340] got exitcode: 0
|
| 25 |
+
2025-03-28 20:36:33,902 INFO MainThread:3856880 [wandb_run.py:_restore():2322] restore
|
| 26 |
+
2025-03-28 20:36:33,902 INFO MainThread:3856880 [wandb_run.py:_restore():2328] restore done
|
offline-run-20250328_203038-ni3gydc1/run-ni3gydc1.wandb
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7551a30274cb4766f54360f45c34e659bd36728d49e2f3ed7d96d44362f0c1ca
|
| 3 |
+
size 229376
|
offline-run-20250328_203844-gc2qytpj/files/output.log
ADDED
|
File without changes
|
offline-run-20250328_203844-gc2qytpj/files/requirements.txt
ADDED
|
@@ -0,0 +1,167 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
maskrcnn_benchmark==0.0.0
|
| 2 |
+
webdataset==0.2.111
|
| 3 |
+
websockets==15.0.1
|
| 4 |
+
typer==0.15.2
|
| 5 |
+
blobfile==3.0.0
|
| 6 |
+
pooch==1.8.2
|
| 7 |
+
python-dateutil==2.9.0.post0
|
| 8 |
+
gmpy2==2.2.1
|
| 9 |
+
httpcore==1.0.7
|
| 10 |
+
charset-normalizer==3.3.2
|
| 11 |
+
torchlibrosa==0.1.0
|
| 12 |
+
multiprocess==0.70.16
|
| 13 |
+
Werkzeug==3.1.3
|
| 14 |
+
aiofiles==23.2.1
|
| 15 |
+
six==1.17.0
|
| 16 |
+
typing_extensions==4.12.2
|
| 17 |
+
psutil==7.0.0
|
| 18 |
+
frozenlist==1.5.0
|
| 19 |
+
einops==0.8.1
|
| 20 |
+
flash_attn==2.7.4.post1
|
| 21 |
+
PySocks==1.7.1
|
| 22 |
+
regex==2024.11.6
|
| 23 |
+
markdown-it-py==3.0.0
|
| 24 |
+
ruff==0.11.2
|
| 25 |
+
docker-pycreds==0.4.0
|
| 26 |
+
protobuf==5.29.4
|
| 27 |
+
resampy==0.4.3
|
| 28 |
+
aiohappyeyeballs==2.6.1
|
| 29 |
+
httpx==0.28.1
|
| 30 |
+
encodec==0.1.1
|
| 31 |
+
ffmpy==0.5.0
|
| 32 |
+
mkl_random==1.2.8
|
| 33 |
+
soxr==0.5.0.post1
|
| 34 |
+
absl-py==2.2.1
|
| 35 |
+
networkx==3.4.2
|
| 36 |
+
h5py==3.13.0
|
| 37 |
+
hjson==3.1.0
|
| 38 |
+
tensorboard==2.19.0
|
| 39 |
+
aiosignal==1.3.2
|
| 40 |
+
pip==25.0
|
| 41 |
+
triton==3.1.0
|
| 42 |
+
zipp==3.21.0
|
| 43 |
+
ftfy==6.3.1
|
| 44 |
+
attrs==25.3.0
|
| 45 |
+
requests==2.32.3
|
| 46 |
+
progressbar==2.5
|
| 47 |
+
sniffio==1.3.1
|
| 48 |
+
lxml==5.3.1
|
| 49 |
+
starlette==0.46.1
|
| 50 |
+
Markdown==3.7
|
| 51 |
+
mdurl==0.1.2
|
| 52 |
+
torchaudio==2.5.1
|
| 53 |
+
safetensors==0.5.3
|
| 54 |
+
opencv-python==4.6.0.66
|
| 55 |
+
torchvision==0.20.1
|
| 56 |
+
shellingham==1.5.4
|
| 57 |
+
gradio==5.23.1
|
| 58 |
+
timm==1.0.15
|
| 59 |
+
multidict==6.2.0
|
| 60 |
+
semantic-version==2.10.0
|
| 61 |
+
numba==0.60.0
|
| 62 |
+
gradio_client==1.8.0
|
| 63 |
+
pydantic_core==2.33.0
|
| 64 |
+
dill==0.3.8
|
| 65 |
+
msgpack==1.1.0
|
| 66 |
+
sentry-sdk==2.24.1
|
| 67 |
+
grpcio==1.71.0
|
| 68 |
+
cffi==1.17.1
|
| 69 |
+
PyYAML==6.0.2
|
| 70 |
+
tensorboard-data-server==0.7.2
|
| 71 |
+
fastapi==0.115.12
|
| 72 |
+
lazy_loader==0.4
|
| 73 |
+
mkl_fft==1.3.11
|
| 74 |
+
annotated-types==0.7.0
|
| 75 |
+
scikit-learn==1.6.1
|
| 76 |
+
wget==3.2
|
| 77 |
+
setuptools==75.8.0
|
| 78 |
+
certifi==2025.1.31
|
| 79 |
+
click==8.1.8
|
| 80 |
+
laion_clap==1.1.5
|
| 81 |
+
Pygments==2.19.1
|
| 82 |
+
tomlkit==0.13.2
|
| 83 |
+
idna==3.7
|
| 84 |
+
propcache==0.3.1
|
| 85 |
+
platformdirs==4.3.7
|
| 86 |
+
align-anything==0.0.1.dev0
|
| 87 |
+
deepspeed==0.16.5
|
| 88 |
+
smmap==5.0.2
|
| 89 |
+
pillow==11.1.0
|
| 90 |
+
typing-inspection==0.4.0
|
| 91 |
+
braceexpand==0.1.7
|
| 92 |
+
decorator==5.2.1
|
| 93 |
+
pandas==2.2.3
|
| 94 |
+
huggingface-hub==0.29.3
|
| 95 |
+
pyarrow==19.0.1
|
| 96 |
+
tokenizers==0.21.1
|
| 97 |
+
GitPython==3.1.44
|
| 98 |
+
xxhash==3.5.0
|
| 99 |
+
packaging==24.2
|
| 100 |
+
numpy==1.23.4
|
| 101 |
+
setproctitle==1.3.5
|
| 102 |
+
llvmlite==0.43.0
|
| 103 |
+
tiktoken==0.9.0
|
| 104 |
+
mpmath==1.3.0
|
| 105 |
+
nvidia-ml-py==12.570.86
|
| 106 |
+
pydantic==2.11.0
|
| 107 |
+
datasets==3.5.0
|
| 108 |
+
librosa==0.11.0
|
| 109 |
+
frechet_audio_distance==0.3.1
|
| 110 |
+
sympy==1.13.1
|
| 111 |
+
safehttpx==0.1.6
|
| 112 |
+
Jinja2==3.1.6
|
| 113 |
+
h11==0.14.0
|
| 114 |
+
aiohttp==3.11.14
|
| 115 |
+
diffusers==0.32.2
|
| 116 |
+
tqdm==4.67.1
|
| 117 |
+
filelock==3.13.1
|
| 118 |
+
transformers==4.51.0.dev0
|
| 119 |
+
scipy==1.10.1
|
| 120 |
+
audioread==3.0.1
|
| 121 |
+
sentencepiece==0.2.0
|
| 122 |
+
pytz==2025.2
|
| 123 |
+
tzdata==2025.2
|
| 124 |
+
python-multipart==0.0.20
|
| 125 |
+
urllib3==2.3.0
|
| 126 |
+
pycryptodomex==3.22.0
|
| 127 |
+
yarl==1.18.3
|
| 128 |
+
pydub==0.25.1
|
| 129 |
+
pycparser==2.22
|
| 130 |
+
soundfile==0.13.1
|
| 131 |
+
wcwidth==0.2.13
|
| 132 |
+
groovy==0.1.2
|
| 133 |
+
torch==2.5.1
|
| 134 |
+
anyio==4.9.0
|
| 135 |
+
wandb==0.19.8
|
| 136 |
+
joblib==1.4.2
|
| 137 |
+
fsspec==2024.12.0
|
| 138 |
+
peft==0.15.1
|
| 139 |
+
accelerate==1.5.2
|
| 140 |
+
py-cpuinfo==9.0.0
|
| 141 |
+
uvicorn==0.34.0
|
| 142 |
+
orjson==3.10.16
|
| 143 |
+
Brotli==1.0.9
|
| 144 |
+
rich==13.9.4
|
| 145 |
+
importlib_metadata==8.6.1
|
| 146 |
+
ninja==1.11.1.4
|
| 147 |
+
wheel==0.45.1
|
| 148 |
+
MarkupSafe==3.0.2
|
| 149 |
+
threadpoolctl==3.6.0
|
| 150 |
+
gitdb==4.0.12
|
| 151 |
+
mkl-service==2.4.0
|
| 152 |
+
typing_extensions==4.12.2
|
| 153 |
+
tomli==2.0.1
|
| 154 |
+
zipp==3.19.2
|
| 155 |
+
wheel==0.43.0
|
| 156 |
+
jaraco.text==3.12.1
|
| 157 |
+
packaging==24.2
|
| 158 |
+
autocommand==2.2.2
|
| 159 |
+
jaraco.functools==4.0.1
|
| 160 |
+
jaraco.collections==5.1.0
|
| 161 |
+
platformdirs==4.2.2
|
| 162 |
+
more-itertools==10.3.0
|
| 163 |
+
inflect==7.3.1
|
| 164 |
+
jaraco.context==5.3.0
|
| 165 |
+
typeguard==4.3.0
|
| 166 |
+
backports.tarfile==1.2.0
|
| 167 |
+
importlib_metadata==8.0.0
|
offline-run-20250328_203844-gc2qytpj/files/wandb-metadata.json
ADDED
|
@@ -0,0 +1,105 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"os": "Linux-5.15.0-1040-nvidia-x86_64-with-glibc2.35",
|
| 3 |
+
"python": "CPython 3.11.11",
|
| 4 |
+
"startedAt": "2025-03-28T12:38:44.660819Z",
|
| 5 |
+
"args": [
|
| 6 |
+
"--local_rank=0",
|
| 7 |
+
"--model_name_or_path",
|
| 8 |
+
"/aifs4su/yaodong/hantao/models/llava-v1.6-vicuna-13b-hf",
|
| 9 |
+
"--train_datasets",
|
| 10 |
+
"/aifs4su/yaodong/hantao/datasets/AA_preference_vicuna-13b_cosi_cut/merged/top1-10",
|
| 11 |
+
"--train_template",
|
| 12 |
+
"AA_TI2T_LLAVA",
|
| 13 |
+
"--train_split",
|
| 14 |
+
"train",
|
| 15 |
+
"--train_name",
|
| 16 |
+
"text-image-to-text",
|
| 17 |
+
"--output_dir",
|
| 18 |
+
"../outputs/test_13B",
|
| 19 |
+
"--save_total_limit",
|
| 20 |
+
"6",
|
| 21 |
+
"--train_batch_size",
|
| 22 |
+
"2",
|
| 23 |
+
"--epochs",
|
| 24 |
+
"3"
|
| 25 |
+
],
|
| 26 |
+
"program": "-m align_anything.trainers.text_image_to_text.dpo",
|
| 27 |
+
"git": {
|
| 28 |
+
"remote": "git@github.com-hantao:PKU-Alignment/align-anything.git",
|
| 29 |
+
"commit": "106588f9802757a3283c1aff1f33ea9afd737f31"
|
| 30 |
+
},
|
| 31 |
+
"root": "../outputs/test_13B",
|
| 32 |
+
"host": "dgx-092",
|
| 33 |
+
"executable": "/aifs4su/yaodong/miniconda3/envs/hantao_llama/bin/python",
|
| 34 |
+
"cpu_count": 112,
|
| 35 |
+
"cpu_count_logical": 224,
|
| 36 |
+
"gpu": "NVIDIA H800",
|
| 37 |
+
"gpu_count": 8,
|
| 38 |
+
"disk": {
|
| 39 |
+
"/": {
|
| 40 |
+
"total": "1888556142592",
|
| 41 |
+
"used": "148592906240"
|
| 42 |
+
}
|
| 43 |
+
},
|
| 44 |
+
"memory": {
|
| 45 |
+
"total": "2164195454976"
|
| 46 |
+
},
|
| 47 |
+
"cpu": {
|
| 48 |
+
"count": 112,
|
| 49 |
+
"countLogical": 224
|
| 50 |
+
},
|
| 51 |
+
"gpu_nvidia": [
|
| 52 |
+
{
|
| 53 |
+
"name": "NVIDIA H800",
|
| 54 |
+
"memoryTotal": "85520809984",
|
| 55 |
+
"cudaCores": 16896,
|
| 56 |
+
"architecture": "Hopper"
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"name": "NVIDIA H800",
|
| 60 |
+
"memoryTotal": "85520809984",
|
| 61 |
+
"cudaCores": 16896,
|
| 62 |
+
"architecture": "Hopper"
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"name": "NVIDIA H800",
|
| 66 |
+
"memoryTotal": "85520809984",
|
| 67 |
+
"cudaCores": 16896,
|
| 68 |
+
"architecture": "Hopper"
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"name": "NVIDIA H800",
|
| 72 |
+
"memoryTotal": "85520809984",
|
| 73 |
+
"cudaCores": 16896,
|
| 74 |
+
"architecture": "Hopper"
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"name": "NVIDIA H800",
|
| 78 |
+
"memoryTotal": "85520809984",
|
| 79 |
+
"cudaCores": 16896,
|
| 80 |
+
"architecture": "Hopper"
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"name": "NVIDIA H800",
|
| 84 |
+
"memoryTotal": "85520809984",
|
| 85 |
+
"cudaCores": 16896,
|
| 86 |
+
"architecture": "Hopper"
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"name": "NVIDIA H800",
|
| 90 |
+
"memoryTotal": "85520809984",
|
| 91 |
+
"cudaCores": 16896,
|
| 92 |
+
"architecture": "Hopper"
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"name": "NVIDIA H800",
|
| 96 |
+
"memoryTotal": "85520809984",
|
| 97 |
+
"cudaCores": 16896,
|
| 98 |
+
"architecture": "Hopper"
|
| 99 |
+
}
|
| 100 |
+
],
|
| 101 |
+
"slurm": {
|
| 102 |
+
"conf": "/cm/shared/apps/slurm/var/etc/slurm/slurm.conf"
|
| 103 |
+
},
|
| 104 |
+
"cudaVersion": "12.2"
|
| 105 |
+
}
|
offline-run-20250328_203844-gc2qytpj/logs/debug-core.log
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{"time":"2025-03-28T20:38:44.475652335+08:00","level":"INFO","msg":"main: starting server","port-filename":"/tmp/tmpe73e3keo/port-3871311.txt","pid":3871311,"log-level":0,"disable-analytics":false,"shutdown-on-parent-exit":false}
|
| 2 |
+
{"time":"2025-03-28T20:38:44.476536664+08:00","level":"INFO","msg":"Will exit if parent process dies.","ppid":3871311}
|
| 3 |
+
{"time":"2025-03-28T20:38:44.476519981+08:00","level":"INFO","msg":"server is running","addr":{"IP":"127.0.0.1","Port":37967,"Zone":""}}
|
| 4 |
+
{"time":"2025-03-28T20:38:44.657114013+08:00","level":"INFO","msg":"connection: ManageConnectionData: new connection created","id":"127.0.0.1:34698"}
|
| 5 |
+
{"time":"2025-03-28T20:38:44.662194892+08:00","level":"INFO","msg":"handleInformInit: received","streamId":"gc2qytpj","id":"127.0.0.1:34698"}
|
| 6 |
+
{"time":"2025-03-28T20:38:44.906611127+08:00","level":"INFO","msg":"handleInformInit: stream started","streamId":"gc2qytpj","id":"127.0.0.1:34698"}
|
| 7 |
+
{"time":"2025-03-28T20:47:40.250303335+08:00","level":"INFO","msg":"received shutdown signal","signal":15}
|
offline-run-20250328_203844-gc2qytpj/logs/debug-internal.log
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{"time":"2025-03-28T20:38:44.692258573+08:00","level":"INFO","msg":"stream: starting","core version":"0.19.8","symlink path":"../outputs/test_13B/wandb/offline-run-20250328_203844-gc2qytpj/logs/debug-core.log"}
|
| 2 |
+
{"time":"2025-03-28T20:38:44.90653769+08:00","level":"INFO","msg":"created new stream","id":"gc2qytpj"}
|
| 3 |
+
{"time":"2025-03-28T20:38:44.90660149+08:00","level":"INFO","msg":"stream: started","id":"gc2qytpj"}
|
| 4 |
+
{"time":"2025-03-28T20:38:44.906615038+08:00","level":"INFO","msg":"handler: started","stream_id":"gc2qytpj"}
|
| 5 |
+
{"time":"2025-03-28T20:38:44.906633311+08:00","level":"INFO","msg":"writer: Do: started","stream_id":"gc2qytpj"}
|
| 6 |
+
{"time":"2025-03-28T20:38:44.90663431+08:00","level":"INFO","msg":"sender: started","stream_id":"gc2qytpj"}
|
| 7 |
+
{"time":"2025-03-28T20:38:44.910439284+08:00","level":"INFO","msg":"Starting system monitor"}
|
offline-run-20250328_203844-gc2qytpj/logs/debug.log
ADDED
|
@@ -0,0 +1,26 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
2025-03-28 20:38:44,447 INFO MainThread:3871311 [wandb_setup.py:_flush():67] Current SDK version is 0.19.8
|
| 2 |
+
2025-03-28 20:38:44,447 INFO MainThread:3871311 [wandb_setup.py:_flush():67] Configure stats pid to 3871311
|
| 3 |
+
2025-03-28 20:38:44,447 INFO MainThread:3871311 [wandb_setup.py:_flush():67] Loading settings from /home/yangyaodong/.config/wandb/settings
|
| 4 |
+
2025-03-28 20:38:44,447 INFO MainThread:3871311 [wandb_setup.py:_flush():67] Loading settings from /aifs4su/yaodong/hantao/align-anything/scripts/wandb/settings
|
| 5 |
+
2025-03-28 20:38:44,447 INFO MainThread:3871311 [wandb_setup.py:_flush():67] Loading settings from environment variables
|
| 6 |
+
2025-03-28 20:38:44,447 INFO MainThread:3871311 [wandb_init.py:setup_run_log_directory():647] Logging user logs to ../outputs/test_13B/wandb/offline-run-20250328_203844-gc2qytpj/logs/debug.log
|
| 7 |
+
2025-03-28 20:38:44,447 INFO MainThread:3871311 [wandb_init.py:setup_run_log_directory():648] Logging internal logs to ../outputs/test_13B/wandb/offline-run-20250328_203844-gc2qytpj/logs/debug-internal.log
|
| 8 |
+
2025-03-28 20:38:44,447 INFO MainThread:3871311 [wandb_init.py:init():761] calling init triggers
|
| 9 |
+
2025-03-28 20:38:44,447 INFO MainThread:3871311 [wandb_init.py:init():766] wandb.init called with sweep_config: {}
|
| 10 |
+
config: {'train_cfgs': {'save_checkpoint': True, 'load_checkpoint': False, 'ds_cfgs': 'ds_z3_config.json', 'epochs': 3, 'seed': 42, 'per_device_train_batch_size': 1, 'per_device_eval_batch_size': 1, 'gradient_accumulation_steps': 1, 'gradient_checkpointing': True, 'learning_rate': 1e-06, 'lr_scheduler_type': 'cosine', 'lr_warmup_ratio': 0.03, 'weight_decay': 0.0, 'adam_betas': [0.9, 0.95], 'bf16': True, 'fp16': False, 'eval_strategy': 'epoch', 'eval_interval': 10, 'regularization': 0.001, 'scale_coeff': 0.1, 'freeze_mm_proj': False, 'freeze_vision_tower': True, 'freeze_language_model': False}, 'data_cfgs': {'train_datasets': '/aifs4su/yaodong/hantao/datasets/AA_preference_vicuna-13b_cosi_cut/merged/top1-10', 'train_template': 'AA_TI2T_LLAVA', 'train_size': {}, 'train_split': 'train', 'train_name': 'text-image-to-text', 'train_data_files': {}, 'train_optional_args': [], 'eval_datasets': {}, 'eval_template': {}, 'eval_size': {}, 'eval_split': {}, 'eval_subset': {}, 'eval_data_files': {}, 'eval_optional_args': []}, 'logger_cfgs': {'log_type': 'wandb', 'log_project': 'align-anything', 'log_run_name': 'dpo', 'output_dir': '../outputs/test_13B', 'cache_dir': {}, 'save_total_limit': 6}, 'model_cfgs': {'model_name_or_path': '/aifs4su/yaodong/hantao/models/llava-v1.6-vicuna-13b-hf', 'trust_remote_code': True, 'model_max_length': 2048}, 'special_tokens': {}, '_wandb': {}}
|
| 11 |
+
2025-03-28 20:38:44,447 INFO MainThread:3871311 [wandb_init.py:init():784] starting backend
|
| 12 |
+
2025-03-28 20:38:44,657 INFO MainThread:3871311 [wandb_init.py:init():788] sending inform_init request
|
| 13 |
+
2025-03-28 20:38:44,660 INFO MainThread:3871311 [backend.py:_multiprocessing_setup():101] multiprocessing start_methods=fork,spawn,forkserver, using: spawn
|
| 14 |
+
2025-03-28 20:38:44,660 INFO MainThread:3871311 [wandb_init.py:init():798] backend started and connected
|
| 15 |
+
2025-03-28 20:38:44,662 INFO MainThread:3871311 [wandb_init.py:init():891] updated telemetry
|
| 16 |
+
2025-03-28 20:38:44,672 INFO MainThread:3871311 [wandb_init.py:init():915] communicating run to backend with 90.0 second timeout
|
| 17 |
+
2025-03-28 20:38:44,908 INFO MainThread:3871311 [wandb_init.py:init():990] starting run threads in backend
|
| 18 |
+
2025-03-28 20:38:45,071 INFO MainThread:3871311 [wandb_run.py:_console_start():2375] atexit reg
|
| 19 |
+
2025-03-28 20:38:45,071 INFO MainThread:3871311 [wandb_run.py:_redirect():2227] redirect: wrap_raw
|
| 20 |
+
2025-03-28 20:38:45,071 INFO MainThread:3871311 [wandb_run.py:_redirect():2292] Wrapping output streams.
|
| 21 |
+
2025-03-28 20:38:45,071 INFO MainThread:3871311 [wandb_run.py:_redirect():2315] Redirects installed.
|
| 22 |
+
2025-03-28 20:38:45,074 INFO MainThread:3871311 [wandb_init.py:init():1032] run started, returning control to user process
|
| 23 |
+
2025-03-28 20:47:40,230 INFO MainThread:3871311 [wandb_run.py:_finish():2112] finishing run align-anything/gc2qytpj
|
| 24 |
+
2025-03-28 20:47:40,230 INFO MainThread:3871311 [wandb_run.py:_atexit_cleanup():2340] got exitcode: 0
|
| 25 |
+
2025-03-28 20:47:40,230 INFO MainThread:3871311 [wandb_run.py:_restore():2322] restore
|
| 26 |
+
2025-03-28 20:47:40,230 INFO MainThread:3871311 [wandb_run.py:_restore():2328] restore done
|
offline-run-20250328_203844-gc2qytpj/run-gc2qytpj.wandb
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8aa9b355d934125304017d0d212405953c1aba2e8d426c9e82a0b3097e7e3c89
|
| 3 |
+
size 393216
|