Dref360

# Wav2Vec in Baal

from datasets import load_dataset
from transformers import Wav2Vec2Processor, Wav2Vec2ForCTC, TrainingArguments

from baal.active.heuristics import BALD
from baal.bayesian.dropout import patch_module
from baal.transformers_trainer_wrapper import BaalTransformersTrainer

# load model and tokenizer

Dref360

from datasets import load_dataset
from transformers import pipeline, DataCollatorForTokenClassification

from baal.active.active_loop import ActiveLearningLoop
from baal.active.dataset import ActiveLearningDataset
from baal.active.heuristics import BALD
from baal.bayesian.dropout import patch_module
from baal.transformers_trainer_wrapper import BaalTransformersTrainer

dataset = load_dataset("conll2003")

Dref360

import os.path
import shutil
import tempfile

from datasets import Dataset, load_from_disk

PATH = '/tmp/b.arrow'


def overwrite_dataset(ds: Dataset, path) -> Dataset:

Dref360

import numpy as np
import torch
from torchvision.models import vgg16

from baal.bayesian.dropout import MCDropoutModule
from baal.modelwrapper import ModelWrapper

model = vgg16()
wrapper = ModelWrapper(model, None)
input = torch.randn([2, 3, 64, 64])

Dref360

Baal with HuggingFace on Label Studio

Instructions to run Label Studio with Bayesian active learning on Text Classification.

Documentation Github

Environment:

• export LABEL_STUDIO_HOSTNAME=http://localhost:8080
• export LABEL_STUDIO_ML_BACKEND_V2=True

Dref360

from pprint import pprint

import datasets
import numpy as np
import torch
from sklearn.metrics.pairwise import cosine_similarity
from tqdm import tqdm
from transformers import AutoTokenizer, AutoModel

"""

Dref360

import argparse

from datasets import load_dataset
from transformers import AutoModelForSequenceClassification, TrainingArguments, Trainer
from transformers import AutoTokenizer
from transformers import DataCollatorWithPadding

LABEL_COL = "label"

TEXT_COL = "text"

Dref360

import gensim
import nltk
from gensim import corpora, models
from nltk.stem import WordNetLemmatizer, SnowballStemmer
import pandas as pd

nltk.download('wordnet')
nltk.download('omw-1.4')
nltk.download('stopwords')

Dref360

from typing import List, Optional, Callable

import torch
from torch.optim import Adam
from torchvision.datasets.voc import VOCDetection
from torchvision.models.detection.ssd import ssd300_vgg16
from torchvision.transforms import Compose, Resize, ToTensor

from baal import ModelWrapper

Dref360

The code should run as is with the following dependencies:

pip install transformers datasets baal matplotlib tqdm