minimaxir/aitextgen

Issues

• Empty Text Generation?

  #235 opened 6 months ago by theSoberSobber
  0

• Error while using the Google collab : cannot import name 'DeepSpeedPlugin' from 'pytorch_lightning.plugins'

  #215 opened 2 years ago by Fqlox
  24

• ImportError: cannot import name '_compare_version' from 'torchmetrics.utilities.imports' (/usr/local/lib/python3.10/dist-packages/torchmetrics/utilities/imports.py)

  #234 opened a year ago by Kiyuma
  0

• getting error while importing aitextgen

  #233 opened a year ago by sabajhn
  2

• Can't use aitextgen to train arabic module

  #232 opened a year ago by ibraheem-abusnineh
  1

• No module named 'transformers.models.cpmant.configuration_cpmant'

  #231 opened a year ago by huyremy
  0

• Proposed code to add Apple MPS support to aitextgen

  #229 opened a year ago by tony352
  1

• ImportError: cannot import name '_TPU_AVAILABLE' from 'pytorch_lightning.utilities' (C:\Users\sambi\anaconda3\lib\site-packages\pytorch_lightning\utilities\__init__.py)

  #228 opened a year ago by SambitSahuIT
  0

• Encoding 1.5GB corpus file runs out of RAM

  #213 opened 2 years ago by enchantinggg4
  2

• Gpt-J

  #223 opened a year ago by robitec97
  1

• Is it possible to set a prompt for training generations?

  #226 opened a year ago by xzuyn
  0

• How do you continue training from the same place in the training set?

  #225 opened a year ago by xzuyn
  0

• The colab notebook is broken again

  #222 opened a year ago by Blazeolmo
  3

• still getting the same error.

  #218 opened 2 years ago by koutkout
  4

• The provided lr scheduler `LambdaLR` doesn't follow PyTorch's LRScheduler API.

  #219 opened 2 years ago by benninkcorien
  5

• RuntimeError when finetuning.

  #193 opened 2 years ago by wisplite
  5

• KeyError: 'biogpt'

  #210 opened 2 years ago by breadbrowser
  2

• ImportError: cannot import name 'ProgressBarBase' from 'pytorch_lightning.callbacks.progress'

  #214 opened 2 years ago by dumpsters
  2

• Key error "loss" issue while training custom dataset

  #217 opened 2 years ago by vjarora1978
  0

• Suppress Generate config GenerationConfig

  #212 opened 2 years ago by francescoboc
  2

• ImportError: cannot import name '_TPU_AVAILABLE' from 'pytorch_lightning.utilities'

  #200 opened 2 years ago by ProfessorQu
  13

• Stop generating text after period

  #211 opened 2 years ago by rmallick6806
  1

• padding_side

  #205 opened 2 years ago by g-i-o-r-g-i-o
  1

• error while trying import aitextgen

  #208 opened 2 years ago by dogukan33
  2

• ImportError: cannot import name '_TPU_AVAILABLE' from 'pytorch_lightning.utilities'

  #207 opened 2 years ago by ZT00OK53
  1

• Error when trying to import

  #209 opened 2 years ago by rashidkisejjere0784
  0

• gradient_accumulation_steps + warmup_steps completely breaks the model's ability to learn

  #206 opened 2 years ago by Vectorrent
  1

• A decoder-only architecture is being used, but right-padding was detected! For correct generation results, please set `padding_side='left'` when initializing the tokenizer.

  #197 opened 2 years ago by Blazeolmo
  3

• AttributeError: can't set attribute

  #204 opened 2 years ago by jatinarora96
  1

• TokenDataset merge_datasets outputting incorrect format of list

  #196 opened 2 years ago by Benw8888
  1

• How do you specify model with CLI interface?

  #201 opened 2 years ago by rmc135
  0

• Aitextgen colab notebook fp16 training

  #199 opened 2 years ago by Blazeolmo
  1

• Warmup steps

  #195 opened 2 years ago by TheGullahanMaster
  4

• FP16 training requires NVIDIA Apex

  #198 opened 2 years ago by Blazeolmo
  0

• Documentation expansion - Trained set produces segments of sentences from the model

  #194 opened 2 years ago by ephemeralfuture
  2

• Gpu vram runs out on colab

  #178 opened 3 years ago by Blazeolmo
  4

• Error with fp16: Original ImportError was: ModuleNotFoundError("No module named 'amp_C'")

  #192 opened 2 years ago by wisplite
  1

• TypeError: __init__() got an unexpected keyword argument 'checkpoint_callback'

  #189 opened 2 years ago by kageraaron
  6

• Missing parameter `amp_backend=apex` with when using `fp16=True`

  #180 opened 2 years ago by tientr
  2

• [WinError 1455] The paging file is too small for this operation to complete.

  #188 opened 2 years ago by GucciFlipFlops1917
  1

• please to putting lies into the demo section README.md.

  #187 opened 2 years ago by breadbrowser
  1

• ValueError: Unfeasable length constraints: the minimum length (50) is larger than the maximum length (32)

  #186 opened 2 years ago by breadbrowser
  0

• Continuing training for a pre-trained model doesn't seem to work

  #183 opened 2 years ago by accountForIssues
  1

• Colab Train a GPT-2 (or GPT Neo) Text-Generating Model w GPU: datascience 0.10.6 requires folium==0.2.1, but you have folium 0.8.3 which is incompatible.

  #174 opened 3 years ago by RobertoPuglieseFusco
  1

• Can I finetune custom trained model?

  #173 opened 3 years ago by varna9000
  1

• How to load already trained models for more training or fine tuning?

  #177 opened 2 years ago by msglm
  1

• Documentation on train() and train_tokenizer() arguments.

  #181 opened 2 years ago by coworo
  0

• Apple Silicon support possible?

  #176 opened 3 years ago by swcrazyfan
  1

• gradient_accumulation_steps argument is not taken into account by ATGProgressBar

  #175 opened 3 years ago by jtourille
  0

• integrate with Lightning ecosystem CI

  #172 opened 3 years ago by Borda
  0