salesforce / ctrl

Conditional Transformer Language Model for Controllable Generation

https://arxiv.org/abs/1909.05858

salesforce/ctrl Issues

tips and scripts related to data collection
Updated 2 years ago1
control code
Updated 2 years ago
A transformer decoer-based model or seq2seq model?
Updated 2 years ago
How to finetuning with lower memory fp16 version for p100 GPUs?
Updated 2 years ago5
repeats the last word on AWS
Updated 3 years ago1
Cuda out of memory issue.
Updated 3 years ago
12 layer (huggingface gpt-2 equivalent) ctrl model?
Updated 3 years ago
CTRL model can not work in huggingface transformers
Updated 3 years ago2
Issues with pytorch_generation.py when running the Colab exercise
Updated 3 years ago
Sampling settings used in the paper
Closed 5 years ago2
why set "seq_length = min(args.generate_num, 256)"
Updated 3 years ago1
control code not recognised
Closed 3 years ago2
Fine-tuning on Colab
Updated 4 years ago4
Out of memory when fine-tuning
Updated 4 years ago9
Will BERT+transformer-decoder better than tensor2tensor for text-generation?
Updated 4 years ago
Using ctrl for summarization
Closed 4 years ago2
Altering the tone of the output
Updated 4 years ago
Are control codes required for finetuning?
Updated 4 years ago
Sampling method used for translation
Updated 4 years ago
License for pre-trained model
Updated 4 years ago
training curriculum used
Updated 4 years ago
Source attribution - Cannot replicate results
Updated 4 years ago
How to finetune on TPU v3-8 nodes? It runs without error but does not seem to progress.
Updated 4 years ago5
Is that a way to do "general" generation?
Closed 4 years ago1
TPU configuration - fine tuning
Updated 4 years ago
'tensorflow._api.v1.keras.layers' has no attribute 'LayerNormalization'
Closed 4 years ago2
Penalize/punish some generated words more than others??!
Closed 4 years ago1
Using new top-level domain with Links control code
Closed 4 years ago1
Is it possible to run CTRL (full) on Gradient?
Updated 5 years ago
❓ Question about testing procedure
Updated 5 years ago
Custom model conversion to Hugging Face
Closed 5 years ago1
Line breaks in prompts?
Closed 5 years ago1
Running full model on V100 outputs last word
Closed 5 years ago12
training_utils on new vocab and codes
Closed 5 years ago1
Running the model on TPUs?
Updated 5 years ago2
What's the main practical difference between 256 and 512 models?
Closed 5 years ago1
OutOfMemory in fine-tuning.
Closed 5 years ago2
more than 600 labels
Closed 5 years ago1
How to add new control code into vocabulary?
Closed 5 years ago4
benchmarking with GPT-2
Updated 5 years ago2
Finetuning Errors
Closed 5 years ago5
Enhancement request: How to read prompts from a text file?
Closed 5 years ago1
I can do inference with pretrained model but face error when finetuning.
Closed 5 years ago2
Just for fun: How long would training this model take on a Nvidia 1080Ti GPU (12gb)
Closed 5 years ago1
using pytorch_generation.py: setting --temperature argument to any value causes a failure
Closed 5 years ago2
Update Google Colab Notebook to Correct Model Path Flag
Closed 5 years ago
multiple tags as control code
Closed 5 years ago1
smaller model
Closed 5 years ago1
Question about vocabulary file
Closed 5 years ago1
Can a conditional control code be multiple tags or multiple keywords？
Closed 5 years ago1