princeton-nlp/SimCSE

Issues

• 4 x 24 GB OOM for sup-SimCSE-BERT-base

  #286 opened 21 days ago by Anonymous-AI1
  1

• [Question] Pre-trained models for RoBERTa

  #290 opened a month ago by fahos
  0

• Sampling Negatives for Sentiment Analysis

  #288 opened a month ago by konan009
  2

• 3.10及以上的python无法安装simcse0.4版本

  #289 opened a month ago by ganlinganlin
  0

• [Question] The optimizer used for training

  #287 opened 2 months ago by fahos
  4

• Installation Issues

  #284 opened 2 months ago by Diamondterritory
  1

• Cannot reproduce the result for `bert-base-uncased`, `avg_first_last` setting

  #285 opened 2 months ago by kuriyan1204
  4

• The function ‘search’ only returns one result

  #282 opened 4 months ago by MeiNanXue
  2

• when I run evaluation.py, the spearman is unusually low

  #283 opened 4 months ago by UGUESS-lzx
  4

• Question about the comparison of data augmentations's reproduce

  #281 opened 6 months ago by tinaa4869
  2

• couldn't install cimcse

  #280 opened 7 months ago by vincent507cpu
  7

• drpout

  #279 opened 8 months ago by riyajatar37003
  2

• An error when max_seq_length is set too long

  #278 opened 8 months ago by Madilynalisa
  1

• setting an array element with a sequence. The requested array has an inhomogeneous shape after 1 dimensions. The detected shape was (750,) + inhomogeneous part.

  #276 opened 8 months ago by JYru
  2

• 关于两次前向传播

  #274 opened 9 months ago by noob-ctrl
  2

• AttributeError: 'OurTrainingArguments' object has no attribute 'distributed_state'

  #275 opened 9 months ago by zongxindu
  2

• TypeError: object of type 'IndexFlatIP' has no len()

  #268 opened 9 months ago by Maydaytyh
  4

• 关于 Supervised SimCSE 的 GPU Memory Usage

  #277 opened 10 months ago by AnonymXXXXX
  2

• Can I replace the base model with longformer? Is it a simple replacement or does the code also need to be changed? Thank you for your answer.

  #271 opened 10 months ago by Struggle-lsl
  4

• Can I load saved index to GPU?

  #270 opened 10 months ago by Maydaytyh
  2

• ValueError: Mixed precision training with AMP or APEX (`--fp16`) can only be used on CUDA devices.

  #273 opened 10 months ago by LittleZ2022
  0

• Why divide by temp when calculating cosine similarity

  #272 opened a year ago by FinalFlowers
  1

• model = SimCSE("princeton-nlp/sup-simcse-bert-base-uncased")调用

  #269 opened a year ago by ArchimedesLiu
  3

• Difference in models between train and evaluation scripts.

  #267 opened a year ago by seanswyi
  4

• 关于simcse build_index 的速度问题

  #266 opened a year ago by Maydaytyh
  6

• About model format conversion

  #262 opened a year ago by fzxxg
  4

• Problems pretraining simcse on custom dataset

  #260 opened a year ago by mphomokoatle
  6

• [question] Pretrined sentence embeddings model fine tuning

  #265 opened a year ago by nurulchamidah
  2

• why mlp_only_train=True during unsupervised training?

  #264 opened a year ago by lankuohsing
  2

• 　Why add two sentences in prepare_features?

  #263 opened a year ago by FinalFlowers
  3

• no response with training customer dataset

  #261 opened a year ago by BarryC7
  1

• unsup-SimCSE-BERTbase复现不出paper的结果

  #258 opened a year ago by oufangwei
  2

• Support python 3.10 (current Google Colab runtime)

  #248 opened a year ago by mocobeta
  5

• 求roberta中文预训练模型

  #256 opened a year ago by XiangwenNing
  2

• Training dimension error for supervised SIMCSE

  #251 opened a year ago by ko120
  3

• questions related to the comparison with other common data augmentation in Table 1

  #259 opened a year ago by Shuwen27
  2

• Cannot install transformers==4.2.1

  #239 opened 2 years ago by namin-an
  2

• Google Colab error

  #252 opened a year ago by NorahUseringithub
  0

• 请问非对称语义的数据集可否应用在有监督的simCSE模型上呢？

  #253 opened a year ago by zhuzhushiw
  2

• 求一份中文版albert预训练模型

  #255 opened a year ago by XiangwenNing
  0

• 求一份中文roberta预训练模型

  #254 opened a year ago by XiangwenNing
  0

• training code error OSerror

  #250 opened a year ago by shyzzz521
  1

• 你好，请问我用model.encode为什么获取不到句子的长度信息

  #245 opened a year ago by Reset-aa
  1

• small result gap between reproduction and huggingface model

  #246 opened a year ago by kimwongyuda
  1

• RuntimeError: Found no NVIDIA driver on your system.

  #244 opened a year ago by AbdullahUTS
  1

• 如何用于层级多标签数据呢？

  #243 opened a year ago by littttttlebird
  1

• Question About bad results from trained model

  #240 opened 2 years ago by Alison-starbeat
  2

• How do you get the supervised nli dataset?

  #242 opened 2 years ago by leoozy
  2

• How to train your model to better fit SentEval

  #241 opened 2 years ago by ZBWpro
  3

• Number of rows in NLI dataset

  #238 opened 2 years ago by xlpczv
  2