huggingface/transfer-learning-conv-ai
🦄 State-of-the-Art Conversational AI with Transfer Learning
PythonMIT
Issues
- 3
cannot import name 'cached_path' from 'transformers'
#132 opened by ncorbuk - 0
How to input personality?
#133 opened by Fshrink - 1
metrics:
#131 opened by steveluo01 - 0
the meaning of code
#130 opened by steveluo01 - 1
Logical error in train.py
#81 opened by monisha08041998 - 0
How should the value of --num_candidates and --personality_permutations be determined?
#129 opened by kassy11 - 0
- 0
Questions about segment embedding
#127 opened by YunahJang - 0
- 0
Questions about interact.py
#124 opened by GlitchBox - 1
I just figured it out. You need both "train" and "valid" examples, following the same structure as:
#123 opened by TangoIndiaMango - 0
Load model checkpoint
#122 opened by raihan0824 - 0
Batchwise padding dataset
#121 opened by mrghofrani - 1
Reason for parameter 'personality_permutations'
#119 opened by Fengpei-Yuan - 0
Question about "token_type_ids"
#120 opened by Fengpei-Yuan - 1
Difference in fine-tuned models linked in README and used in interact.py script
#82 opened by AmanTiwari1503 - 1
Custom vocabulary
#110 opened by harrystuart - 1
Saved model after training?
#104 opened by CiaraG98 - 0
Train.py not working
#115 opened by HadiAskari - 0
- 0
GPT2 chat-bot single interaction… Attribute Error: 'NoneType' object has no attribute 'multiprocessing_chunksize' issue:2
#113 opened by chkshahin - 2
- 1
Issue when running interact script
#109 opened by harrystuart - 0
pip version problems - updated requirements.txt
#108 opened by jnehring - 0
Another language training
#95 opened by pourdabiri - 0
Docker image build fails on AWS EC2
#107 opened by Raydrian - 0
Question on Next Utterance Classification task
#106 opened by wsp317 - 3
Online demo is not available duo to CORS
#105 opened by imageslr - 2
"Cuda Out of Memory" when running train.py
#103 opened by CiaraG98 - 1
- 1
- 2
Readme doesn't give python version
#97 opened by MikeyBeez - 1
No module named 'numpy' during docker build
#102 opened by moon-home - 1
Error During Installation
#93 opened by 19hunter19 - 1
Fine-tuning the model
#96 opened by ShivanshuPurohit - 0
- 1
dataset tokenization returns None values
#99 opened by dnns92 - 2
TypeError: tuple indices must be integers or slices, not tuple when running interact.py
#98 opened by alyonavyshnevska - 3
- 2
Multilingual chat-bot
#79 opened by VikasRajashekar - 1
Evaluation in other languages
#94 opened by VikasRajashekar - 0
I can't replicate the training
#92 opened by albusdemens - 0
Out of memory XLM-roberta-large
#91 opened by hadifar - 1
Error while saving models
#87 opened by Rasipuram - 2
- 0
pytorch-ignite issue:AttributeError: type object 'torch.distributed.ReduceOp' has no attribute 'BAND'
#86 opened by VikasRajashekar - 1
personalities = [dialog["personality"] for dataset in dataset.values() for dialog in dataset], TypeError: list indices must be integers or slices, not str
#84 opened by RisabBiswas - 3
How are the distractors made in the dataset?
#77 opened by Cakeszy - 1
Building docker fails
#73 opened by FBosler - 1
Output of tokenizer.encode has None for space
#75 opened by ziweiji