lewtun

# 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

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

lewtun

lewtun

{"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

lewtun

import json

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

lewtun

def get_grouped_params(model, no_decay=["bias", "LayerNorm.weight"]):
    params_with_wd, params_without_wd = [], []
    for n, p in model.named_parameters():
        if any(nd in n for nd in no_decay): 
            params_without_wd.append(p)
        else: 
            params_with_wd.append(p)
    return [{'params': params_with_wd, 'weight_decay': args.weight_decay},
            {'params': params_without_wd, 'weight_decay': 0.0}]

lewtun

if any(nd in n for nd in no_decay):
    params_without_wd.append(p)
else:
    params_with_wd.append(p)

lewtun

from tqdm import tqdm
import torch

device = "cuda" if torch.cuda.is_available() else "cpu"

def chunks(list_of_elements, batch_size):
    """Yield successive batch-sized chunks from list_of_elements."""
    for i in range(0, len(list_of_elements), batch_size):
        yield list_of_elements[i : i + batch_size]