lewtun’s gists

lewtun / update-label-mappings.py

Created July 15, 2022 18:21

Update label mappings in config.json

	import json

	import datasets
	import transformers
	from datasets import ClassLabel, load_dataset
	from huggingface_hub import (
	HfFolder,
	ModelFilter,
	hf_hub_download,
	list_models,

lewtun / format_spaces_urls.ipynb

Created November 22, 2022 12:30

[HF Course] Format Gradio URLs

Sorry, something went wrong. Reload?

Sorry, we cannot display this file.

Sorry, this file is invalid so it cannot be displayed.

lewtun / metrics.jsonl

Created December 2, 2022 09:47

metric.jsonl

	{"id":"628dfaf7554de818ab126e2d","dataset":{"name":"glue","type":"glue","config":"sst2","split":"validation"},"metric":{"type":"accuracy","value":0.8967889908256881,"name":"Accuracy"}}
	{"id":"628dfaf7554de818ab126e2d","dataset":{"name":"glue","type":"glue","config":"sst2","split":"validation"},"metric":{"type":"precision","value":0.8898678414096917,"name":"Precision"}}
	{"id":"628dfaf7554de818ab126e2d","dataset":{"name":"glue","type":"glue","config":"sst2","split":"validation"},"metric":{"type":"recall","value":0.9099099099099099,"name":"Recall"}}
	{"id":"628dfaf7554de818ab126e2d","dataset":{"name":"glue","type":"glue","config":"sst2","split":"validation"},"metric":{"type":"auc","value":0.9672186789593331,"name":"AUC"}}
	{"id":"628dfaf7554de818ab126e2d","dataset":{"name":"glue","type":"glue","config":"sst2","split":"validation"},"metric":{"type":"f1","value":0.8997772828507795,"name":"F1"}}
	{"id":"628dfaf7554de818ab126e2d","dataset":{"name":"glue","type":"glue","config":"sst2","split":"validation"},"metric":{"ty

lewtun / dataset-sharding.ipynb

Created February 12, 2023 15:22

Sharding dataset subsets

Sorry, something went wrong. Reload?

Sorry, we cannot display this file.

Sorry, this file is invalid so it cannot be displayed.

lewtun / hf-endpoints-inference.ipynb

Created June 10, 2023 07:53

Demo of synchronous and token streaming text generation

Sorry, something went wrong. Reload?

Sorry, we cannot display this file.

Sorry, this file is invalid so it cannot be displayed.

lewtun / m4_inference.py

Created July 4, 2023 15:28

M4 inference

	import torch
	from m4.training.packing import image_attention_mask_for_packed_input_ids, incremental_to_binary_attention_mask
	from m4.training.utils import build_image_transform
	from io import BytesIO
	from PIL import Image
	import requests
	from transformers import AutoTokenizer, AutoModelForCausalLM


	MAX_SEQ_LEN=2048

lewtun / dialogue_template.py

Last active October 6, 2023 12:57

Dialogue template

	# coding=utf-8
	# Copyright 2023 The HuggingFace Team. All rights reserved.
	#
	# Licensed under the Apache License, Version 2.0 (the "License");
	# you may not use this file except in compliance with the License.
	# You may obtain a copy of the License at
	#
	# http://www.apache.org/licenses/LICENSE-2.0
	#
	# Unless required by applicable law or agreed to in writing, software

lewtun / api_demo.ipynb

Created July 12, 2023 14:36

API demo

Sorry, something went wrong. Reload?

Sorry, we cannot display this file.

Sorry, this file is invalid so it cannot be displayed.

lewtun / sentiment_tuning.py

Created August 1, 2023 16:56

TRL Sentiment Tuning with DeepSpeed ZeRO-3

	# coding=utf-8
	# Copyright 2023 The HuggingFace Inc. team. All rights reserved.
	#
	# Licensed under the Apache License, Version 2.0 (the "License");
	# you may not use this file except in compliance with the License.
	# You may obtain a copy of the License at
	#
	# http://www.apache.org/licenses/LICENSE-2.0
	#
	# Unless required by applicable law or agreed to in writing, software

lewtun / sft_trainer.py

Last active April 21, 2025 16:04

Fine-tuning Mistral 7B with TRL & DeepSpeed ZeRO-3

	# This is a modified version of TRL's `SFTTrainer` example (https://github.com/huggingface/trl/blob/main/examples/scripts/sft_trainer.py),
	# adapted to run with DeepSpeed ZeRO-3 and Mistral-7B-V1.0. The settings below were run on 1 node of 8 x A100 (80GB) GPUs.
	#
	# Usage:
	# - Install the latest transformers & accelerate versions: `pip install -U transformers accelerate`
	# - Install deepspeed: `pip install deepspeed==0.9.5`
	# - Install TRL from main: pip install git+https://github.com/huggingface/trl.git
	# - Clone the repo: git clone github.com/huggingface/trl.git
	# - Copy this Gist into trl/examples/scripts
	# - Run from root of trl repo with: accelerate launch --config_file=examples/accelerate_configs/deepspeed_zero3.yaml --gradient_accumulation_steps 8 examples/scripts/sft_trainer.py