AssertionError: config.json missing

我在执行stage1.sh的时候，出现报错

Traceback (most recent call last):
  File "graphgpt/train/train_mem.py", line 20, in <module>
    train()
  File "/root/autodl-tmp/GraphGPT/graphgpt/train/train_graph.py", line 871, in train
    model_graph_dict = model.get_model().initialize_graph_modules(
  File "/root/autodl-tmp/GraphGPT/graphgpt/model/GraphLlama.py", line 139, in initialize_graph_modules
    clip_graph, args= load_model_pretrained(CLIP, self.config.pretrain_graph_model_path) 
  File "/root/autodl-tmp/GraphGPT/graphgpt/model/GraphLlama.py", line 54, in load_model_pretrained
    assert osp.exists(osp.join(pretrain_model_path, 'config.json')), 'config.json missing'
AssertionError: config.json missing

我参考其他issue，已经修改了vicuna里面的config.json如下：

  "_name_or_path": "vicuna-7b-v1.5-16k",
  "architectures": [
    "LlamaForCausalLM"
  ],
  "bos_token_id": 1,
  "eos_token_id": 2,
  "hidden_act": "silu",
  "hidden_size": 4096,
  "initializer_range": 0.02,
  "intermediate_size": 11008,
  "max_sequence_length": 16384,
  "max_position_embeddings": 4096,
  "model_type": "llama",
  "num_attention_heads": 32,
  "num_hidden_layers": 32,
  "num_key_value_heads": 32,
  "pad_token_id": 0,
  "pretraining_tp": 1,
  "rms_norm_eps": 1e-05,
  "rope_scaling": {
    "factor": 4.0,
    "type": "linear"
  },
  "tie_word_embeddings": false,
  "torch_dtype": "float16",
  "transformers_version": "4.31.0",
  "use_cache": true,
  "vocab_size": 32000, 
  "graph_hidden_size": 128, 
  "pretrain_graph_model_path": "/root/autodl-tmp/GraphGPT/Arxiv-PubMed-GraphCLIP-GT/"
}

sh文件如下：


model_path=./vicuna-7b-v1.5-16k
instruct_ds=./data/graph_matching.json
graph_data_path=./graph_data/all_graph_data.pt
pretra_gnn=clip_gt_arxiv
output_model=./stage_1
wandb offline
python3 -m  torch.distributed.run  --nnodes=1 --nproc_per_node=1 --master_port=20001 \
    graphgpt/train/train_mem.py \
    --model_name_or_path ${model_path} \
    --version v1 \
    --data_path ${instruct_ds} \
    --graph_content ./arxiv_ti_ab.json \
    --graph_data_path ${graph_data_path} \
    --graph_tower ${pretra_gnn} \
    --tune_graph_mlp_adapter True \
    --graph_select_layer -2 \
    --use_graph_start_end \
    --bf16 True \
    --output_dir ${output_model} \
    --num_train_epochs 3 \
    --per_device_train_batch_size 2 \
    --per_device_eval_batch_size 2 \
    --gradient_accumulation_steps 1 \
    --evaluation_strategy "no" \
    --save_strategy "steps" \
    --save_steps 2400 \
    --save_total_limit 1 \
    --learning_rate 2e-3 \
    --weight_decay 0. \
    --warmup_ratio 0.03 \
    --lr_scheduler_type "cosine" \
    --logging_steps 1 \
    --tf32 True \
    --model_max_length 2048 \
    --gradient_checkpointing True \
    --lazy_preprocess True \
    --report_to wandb```

HKUDS / GraphGPT

AssertionError: config.json missing #57