elesun2018
elesun2018
i have the same problem,thanks
i got the key.you should use git clone https://github.com/weiliu89/caffe.git first that maybe different caffe. cp MobileNet-SSD-master to ssd/caffe/examples train.sh : ../../build/tools/caffe train -solver then run good luck
MODEL_TYPE="XrayGLM" MODEL_ARGS="--max_source_length 64 \ --max_target_length 256 \ --lora_rank 10\ --pre_seq_len 4" --lr-decay-style cosine \ --warmup .02 \ --checkpoint-activations \ --save-interval 3000 \ --eval-interval 10000 \ --save "./checkpoints" \ --split 1...
已解决，原因为需要添加/THUDM/chatglm-6b模型文件
VisualGLM-6B# python web_demo.py遇到了相同的问题， OSError: We couldn't connect to 'https://huggingface.co' to load this file
遇到了相同的问题，OSError: We couldn't connect to 'https://huggingface.co' to load this file
同问，谢谢分享
请问微调VisualGLM-6B# bash finetune/finetune_visualglm.sh，出现问题torch/utils/cpp_extension.py raise RuntimeError(message) from e RuntimeError: Error building extension 'fused_adam'是什么原因
同样的问题 服务响应 的路走的路走的路走的路走的路走的路走的路走的路走的路走的路走的路')] error: index 2055 is out of bounds for dimension 0 with size 205 问题未解决
请问解决了吗 VisualGLM-6B# bash finetune/finetune_visualglm.sh NCCL INFO NET/Plugin : No plugin found (libnccl-net.so), using internal implementation misc/ibvwrap.cc:63 NCCL WARN Failed to open libibverbs.so[.1] FAILED: multi_tensor_adam.cuda.o ninja: build stopped: subcommand failed. subprocess.CalledProcessError:...