init

iter_11628.pth/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:80fa8b0a2d18a086edc4427abe3235ce83ac62eec9fba40b4d363f554a87b3b6
+size 19676784

iter_11628.pth/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6fee544a3b533b90133bdb531442b0424b6fc1df8ed938035368397ba5614129
+size 19676912

iter_11628.pth/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d054aaf321ebccde7b7abf24f88f781ad9712d41fdd65b68c4c205251131bccb
+size 19676784

iter_11628.pth/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:935528913d084969a5cc9ea0294e7e34ad48e60dc6c85ab014cf1da9030f8ed7
+size 19676848

iter_11628.pth/mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:260b0e1988cd523cc76ef05f76ba172fd1de395a83d565fc8f2625bb960b153e
+size 13988972

pretrain.py

@@ -0,0 +1,214 @@
+# Copyright (c) OpenMMLab. All rights reserved.
+import torch
+from mmengine.dataset import DefaultSampler
+from mmengine.hooks import (CheckpointHook, DistSamplerSeedHook, IterTimerHook,
+                            LoggerHook, ParamSchedulerHook)
+from mmengine.optim import AmpOptimWrapper, CosineAnnealingLR, LinearLR
+from torch.optim import AdamW
+from transformers import (AutoModelForCausalLM, AutoTokenizer,
+                          BitsAndBytesConfig, SiglipImageProcessor,
+                          SiglipVisionModel)
+
+from xtuner.dataset import LLaVADataset
+from xtuner.dataset.collate_fns import default_collate_fn
+from xtuner.dataset.map_fns import llava_map_fn, template_map_fn_factory
+from xtuner.engine.hooks import DatasetInfoHook, EvaluateChatHook
+from xtuner.engine.runner import TrainLoop
+from xtuner.model import LLaVAModel
+from xtuner.utils import PROMPT_TEMPLATE
+
+#######################################################################
+#                          PART 1  Settings                           #
+#######################################################################
+# Model
+llm_name_or_path = 'internlm/internlm2-1_8b'
+visual_encoder_name_or_path = 'google/siglip-so400m-patch14-384'
+
+# Data
+data_root = './'
+data_path = data_root + 'LLaVA-Pretrain/blip_laion_cc_sbu_558k.json'
+image_folder = data_root + 'LLaVA-Pretrain/images'
+prompt_template = PROMPT_TEMPLATE.internlm2_chat
+max_length = int(2048 - (336 / 14)**2)
+
+# Scheduler & Optimizer
+batch_size = 12  # per_device
+accumulative_counts = 5
+dataloader_num_workers = 12
+max_epochs = 1
+optim_type = AdamW
+lr = 1e-3
+betas = (0.9, 0.999)
+weight_decay = 0
+max_norm = 1  # grad clip
+warmup_ratio = 0.03
+
+# Save
+save_steps = 2000
+save_total_limit = 2  # Maximum checkpoints to keep (-1 means unlimited)
+
+# Evaluate the generation performance during the training
+evaluation_freq = 2000
+SYSTEM = ''
+evaluation_images = 'https://llava-vl.github.io/static/images/view.jpg'
+evaluation_inputs = ['请描述一下这张照片', 'Please describe this picture']
+
+#######################################################################
+#            PART 2  Model & Tokenizer & Image Processor              #
+#######################################################################
+tokenizer = dict(
+    type=AutoTokenizer.from_pretrained,
+    pretrained_model_name_or_path=llm_name_or_path,
+    trust_remote_code=True,
+    padding_side='right')
+
+image_processor = dict(
+    type=SiglipImageProcessor.from_pretrained,
+    pretrained_model_name_or_path=visual_encoder_name_or_path,
+    trust_remote_code=True)
+
+model = dict(
+    type=LLaVAModel,
+    freeze_llm=True,
+    freeze_visual_encoder=True,
+    llm=dict(
+        type=AutoModelForCausalLM.from_pretrained,
+        pretrained_model_name_or_path=llm_name_or_path,
+        trust_remote_code=True,
+        torch_dtype=torch.float16,
+        quantization_config=dict(
+            type=BitsAndBytesConfig,
+            load_in_4bit=True,
+            load_in_8bit=False,
+            llm_int8_threshold=6.0,
+            llm_int8_has_fp16_weight=False,
+            bnb_4bit_compute_dtype=torch.float16,
+            bnb_4bit_use_double_quant=True,
+            bnb_4bit_quant_type='nf4')),
+    visual_encoder=dict(
+        type=SiglipVisionModel.from_pretrained,
+        pretrained_model_name_or_path=visual_encoder_name_or_path))
+
+#######################################################################
+#                      PART 3  Dataset & Dataloader                   #
+#######################################################################
+llava_dataset = dict(
+    type=LLaVADataset,
+    data_path=data_path,
+    image_folder=image_folder,
+    tokenizer=tokenizer,
+    image_processor=image_processor,
+    dataset_map_fn=llava_map_fn,
+    template_map_fn=dict(
+        type=template_map_fn_factory, template=prompt_template),
+    max_length=max_length,
+    pad_image_to_square=False)
+
+train_dataloader = dict(
+    batch_size=batch_size,
+    num_workers=dataloader_num_workers,
+    dataset=llava_dataset,
+    sampler=dict(type=DefaultSampler, shuffle=True),
+    collate_fn=dict(type=default_collate_fn))
+
+#######################################################################
+#                    PART 4  Scheduler & Optimizer                    #
+#######################################################################
+# optimizer
+optim_wrapper = dict(
+    type=AmpOptimWrapper,
+    optimizer=dict(
+        type=optim_type, lr=lr, betas=betas, weight_decay=weight_decay),
+    clip_grad=dict(max_norm=max_norm, error_if_nonfinite=False),
+    accumulative_counts=accumulative_counts,
+    loss_scale='dynamic',
+    dtype='float16')
+
+# learning policy
+# More information: https://github.com/open-mmlab/mmengine/blob/main/docs/en/tutorials/param_scheduler.md  # noqa: E501
+param_scheduler = [
+    dict(
+        type=LinearLR,
+        start_factor=1e-5,
+        by_epoch=True,
+        begin=0,
+        end=warmup_ratio * max_epochs,
+        convert_to_iter_based=True),
+    dict(
+        type=CosineAnnealingLR,
+        eta_min=0.0,
+        by_epoch=True,
+        begin=warmup_ratio * max_epochs,
+        end=max_epochs,
+        convert_to_iter_based=True)
+]
+
+# train, val, test setting
+train_cfg = dict(type=TrainLoop, max_epochs=max_epochs)
+
+#######################################################################
+#                           PART 5  Runtime                           #
+#######################################################################
+# Log the dialogue periodically during the training process, optional
+custom_hooks = [
+    dict(type=DatasetInfoHook, tokenizer=tokenizer),
+    dict(
+        type=EvaluateChatHook,
+        tokenizer=tokenizer,
+        image_processor=image_processor,
+        every_n_iters=evaluation_freq,
+        evaluation_inputs=evaluation_inputs,
+        evaluation_images=evaluation_images,
+        system=SYSTEM,
+        prompt_template=prompt_template)
+]
+
+# configure default hooks
+default_hooks = dict(
+    # record the time of every iteration.
+    timer=dict(type=IterTimerHook),
+    # print log every 10 iterations.
+    logger=dict(type=LoggerHook, log_metric_by_epoch=False, interval=10),
+    # enable the parameter scheduler.
+    param_scheduler=dict(type=ParamSchedulerHook),
+    # save checkpoint per `save_steps`.
+    checkpoint=dict(
+        type=CheckpointHook,
+        by_epoch=False,
+        interval=save_steps,
+        max_keep_ckpts=save_total_limit),
+    # set sampler seed in distributed evrionment.
+    sampler_seed=dict(type=DistSamplerSeedHook),
+)
+
+# configure environment
+env_cfg = dict(
+    # whether to enable cudnn benchmark
+    cudnn_benchmark=False,
+    # set multi process parameters
+    mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0),
+    # set distributed parameters
+    dist_cfg=dict(backend='nccl'),
+)
+
+# set visualizer
+from mmengine.visualization import Visualizer, TensorboardVisBackend
+visualizer = dict(
+    type=Visualizer,
+    vis_backends=[dict(type=TensorboardVisBackend)]
+)
+
+# set log level
+log_level = 'INFO'
+
+# load from which checkpoint
+load_from = None
+
+# whether to resume training from the loaded checkpoint
+resume = False
+
+# Defaults to use random seed and disable `deterministic`
+randomness = dict(seed=None, deterministic=False)
+
+# set log processor
+log_processor = dict(by_epoch=False)