transfer-learning-conv-ai
transfer-learning-conv-ai copied to clipboard
🦄 State-of-the-Art Conversational AI with Transfer Learning
Hello so i just found this and wanted to explore with my own language (Indonesian Language). I'm aware that in Installation I need to download the English version for spacy:...
Hi, Can anyone explain why the code snippet ([interact.py](https://github.com/huggingface/transfer-learning-conv-ai/blob/d4c760731e8f062abd3d61ccdc8fa29ccb5e3485/interact.py#L72)) (underlined in red) was necessary?  As far as I know the logits returned by **OpenAIGPTLMHeadModel** is of...
I just figured it out. You need both "train" and "valid" examples, following the same structure as:
I just figured it out. You need both "train" and "valid" examples, following the same structure as: ``` { "train": [ "your data here" ], "valid": [ "your data here"...
from transformers import cached_path ImportError: cannot import name 'cached_path' from 'transformers' (/root/anaconda3/envs/transformers/lib/python3.10/site-packages/transformers/__init__.py)
Bumps [transformers](https://github.com/huggingface/transformers) from 2.5.1 to 4.30.0. Release notes Sourced from transformers's releases. v4.30.0: 100k, Agents improvements, Safetensors core dependency, Swiftformer, Autoformer, MobileViTv2, timm-as-a-backbone 100k Transformers has just reached 100k stars...
I cannot find how to input a particular persona (i.e. .json file) to narrow down and/or fine-tune the model. The various attempts at editing the utils and/or interact files have...
metrics:
for utterance in dialog["utterances"]: history = utterance["history"][-(2*args.max_history+1):] for j, candidate in enumerate(utterance["candidates"][-num_candidates:]): lm_labels = bool(j == num_candidates-1) instance = build_input_from_segments(persona, history, candidate, tokenizer, lm_labels) Hello, I am new to transfer,...
I don't understand what you mean by command line parameter `--num_candidates` and `--personality_permutations`.(even though read help) Should these be changed depending on the content of the training data? For example,...
I wish to install this project but find no instructions about which Python version I should use in a virtual environment to install the packages mentioned in requirements.txt.