DataAnalysist

关于创建数据处理函数问题

3

如题，如果我有多个任务：A、B、C、D那是不是需要创建多个类似A_cls这样的装饰器函数，如下所示： @preprocessing_fn def A_cls(params, mode): # get data # your work return input_list, target_list 能否根据具体任务动态的创建数据处理函数呢

0.3.4版本，执行notebook示例：Run Self Defined Problem，出现以下错误： InvalidArgumentError: 2 root error(s) found. (0) Invalid argument: `cycle_length` must be > 0 [[node ExperimentalParallelInterleaveDataset (defined at /home/appadmin/anaconda3/envs/multi_task/lib/python3.6/site-packages/bert_multitask_learning-0.3.4-py3.6.egg/bert_multitask_learning/read_write_tfrecord.py:520) ]] [[MultiDeviceIteratorToStringHandle/_8865]] (1) Invalid argument: `cycle_length` must be...

预训练模型可否添加网盘地址

1

rt

### Feature request i find this version can only support faiss-gpu, does it support faiss-gpu. maybe we can add something like annoy, hnswlib and so on ### Motivation add more...

Feature Description An optional configuration is provided, which allows users to freely choose to use faiss local library(or annoy,hnswlib， jina) remote/distributed vector database as vector storage. Problem Solved When the...

### Reminder - [X] I have read the README and searched the existing issues. ### Reproduction 最新的微调方法lisa可否集成进去，类似lora，qlora，galore微调方法 ### Expected behavior none ### System Info _No response_ ### Others _No response_

rt，使用多卡训练时，若不设置ddp_find_unused_parameters=false，则会出现如下错误： RuntimeError: Expected to mark a variable ready only once. This error is caused by one of the following reasons: 1) Use of a module parameter outside the `forward` function....

Mixture-of-Depths Finetune，IndexError: too many indices for tensor of dimension 0

1

### Reminder - [X] I have read the README and searched the existing issues. ### Reproduction MASTER_PORT=$(shuf -n 1 -i 10000-65535) DEEPSPEED_PATH=../config/ds_config_sft_z2_offload.json MODEL_PATH=/your path/Meta-Llama-3-8B OUTPUT_PATH=../output/llama3-8b-mod-sft LOG_PATH=../logs/result_mod_sft.log nohup deepspeed --num_gpus=4 --master_port...

can't load reranker-v2-minicpm-layerwise model

7

when uses code like this： ``` from FlagEmbedding import LayerWiseFlagLLMReranker reranker = LayerWiseFlagLLMReranker('/path/bge-reranker-v2-minicpm-layerwise', use_fp16=True) score = reranker.compute_score(['query', 'passage'], cutoff_layers=[28]) # Adjusting 'cutoff_layers' to pick which layers are used for computing...

MODPO: Multi-Objective Direct Preference Optimization

1

[https://arxiv.org/pdf/2310.03708](url) MODPO folds language modeling directly into reward modeling, training LMs as implicit collective reward models (cRMs) that combine all objectives with specific weightings. While theoretically guaranteed to produce the...