| { | |
| "global": { | |
| "check_for_updates_on_startup": true, | |
| "show_in_menu_bar": true, | |
| "show_profile_name_in_menu_bar": false | |
| }, | |
| "profiles": [ | |
| { | |
| "complex_modifications": { | |
| "parameters": { |
Sometimes you want to have a subdirectory on the master branch be the root directory of a repository’s gh-pages branch. This is useful for things like sites developed with Yeoman, or if you have a Jekyll site contained in the master branch alongside the rest of your code.
For the sake of this example, let’s pretend the subfolder containing your site is named dist.
Remove the dist directory from the project’s .gitignore file (it’s ignored by default by Yeoman).
| from transformers import AutoTokenizer, TextGenerationPipeline | |
| from auto_gptq import AutoGPTQForCausalLM, BaseQuantizeConfig | |
| import logging | |
| logging.basicConfig( | |
| format="%(asctime)s %(levelname)s [%(name)s] %(message)s", level=logging.INFO, datefmt="%Y-%m-%d %H:%M:%S" | |
| ) | |
| """ | |
| Download https://huggingface.co/liuhaotian/llava-llama-2-13b-chat-lightning-preview to local |
| # Evaluation code for GQA. | |
| # Computes a suite of metrics such as accuracy, consistency, plausibility and scores per question type and length. | |
| # Visit https://gqadataset.org/ for all information about the dataset, including examples, visualizations, paper and slides. | |
| # | |
| # | |
| # Metrics: | |
| # - Accuracy: Standard accuracy, computed over the balanced version of the dataset, which is more robust against | |
| # cheating by making educated guesses. For each question-answer pair (q,a), we give 1 point if the | |
| # predicted answer p matches a and 0 otherwise, and average over all questions in the dataset. | |
| # |