monkeyshichi

The requested URL was not found on the server. If you entered the URL manually please check your spelling and try again.

hi,where is actor-critic code?

如果是YARN 集群，category.spark.masterUrl应该怎么设置，还需要配置其他的吗？

1

如果是YARN 集群，category.spark.masterUrl应该怎么设置，还需要配置其他的吗？

import error

1

from thriftpy2.protocol import TBinaryProtocol, TCompactProtocol from .cybin import TCyBinaryProtocol, TCyBinaryProtocolFactory ImportError: /root/anaconda3/envs/alan_cjf/lib/python3.7/site-packages/thriftpy2/protocol/cybin.cpython-37m-x86_64-linux-gnu.so: undefined symbol: bswap_64 i already installed cython. linux system python 3.7.10 conda happybase 1.2.0 ply 3.11 thriftpy2 0.4.14

fm模型特征处理方法

3

使用fm模型需要将连续性特征值离散化后再输入模型吗？离散型特征值要one-hot转换再输入模型吗？

a = K.permute_dimensions(a, (0, 3, 2, 1))

2

ValueError: Dimension must be 5 but is 4 for 'attention_1/transpose_7' (op: 'Transpose') with input shapes: [?,8,?,8,?], [4]. 请问这个怎么解决 run example https://kexue.fm/archives/4765 代码测试 在Keras上对IMDB进行简单的测试

Supported Models [LLaMA](https://github.com/facebookresearch/llama) (7B/13B/33B/65B) [BLOOM](https://huggingface.co/bigscience/bloom) & [BLOOMZ](https://huggingface.co/bigscience/bloomz) (560M/1.1B/1.7B/3B/7.1B/176B) [baichuan](https://huggingface.co/baichuan-inc/baichuan-7B) (7B) 我看目前支持的模型只有这些，如果不支持需要进行哪些改造以便支持

RuntimeError: unscale_() has already been called on this optimizer since the last update

2

运行命令如下： CUDA_VISIBLE_DEVICES=1 python src/train_pt.py \ --model_name_or_path baichuan-7B \ --do_train \ --lora_target W_pack \ --dataset wiki_demo \ --finetuning_type lora \ --quantization_bit 4 \ --output_dir pretrained \ --overwrite_cache \ --per_device_train_batch_size 4 \...

预训练之后监督微调训练是否需要合并权重并把合并后的权重路径放入--model_name_or_path path_to_your_model

2

预训练之后sft微调训练是否需要合并权重并把合并后的权重路径放入model_name_or_path？ CUDA_VISIBLE_DEVICES=0 python src/train_sft.py \ --model_name_or_path path_to_your_model \ --do_train \ --dataset alpaca_gpt4_en \ --finetuning_type lora \ --output_dir path_to_sft_checkpoint \ --overwrite_cache \ --per_device_train_batch_size 4 \ --gradient_accumulation_steps 4 \ --lr_scheduler_type cosine \...

windows下双击clash翻墙cmd文件没有反应

1

windows下双击clash翻墙cmd文件没有反应，双击其他翻墙文件也没反应