lando22

import tensorflow as tf
import tensorflow_datasets as tfds


(ds_train, ds_test), ds_info = tfds.load(
    'mnist',
    split=['train', 'test'],
    shuffle_files=True,
    as_supervised=True,
    with_info=True,

lando22

import torch
import torch.nn as nn
import torch.nn.functional as F
import torch.optim as optim
from torchvision import datasets, transforms
from torch.autograd import Variable

# Training settings
batch_size = 64

lando22

import humingbird

# creation + prediction in one call
prediction = humningbird.Image.predict(,
   image="digit_9.jpg",
   labels=["1", "2", "3", "4", "5", "6", "7", "8", "9"]
)

print(prediction)

lando22

import humingbird

prediction = humingbird.Image.predict(
  image_path="digit_9.jpg",
  labels=["1", "2", "3", "4", "5", "6", "7", "8", "9", "10"]    # added 10 
)

print(prediction)

lando22

import tensorflow as tf
from tensorflow import keras
from tensorflow.keras.datasets import imdb
from tensorflow.keras.preprocessing.sequence import pad_sequences
from tensorflow.keras import Sequential
from tensorflow.keras.layers import Embedding, Flatten, Dense
import numpy as np
import matplotlib.pyplot as plt

# Set the number of words to consider as features

lando22

import os
import glob
import io

from .. import data


class IMDB(data.Dataset):

    urls = ['http://ai.stanford.edu/~amaas/data/sentiment/aclImdb_v1.tar.gz']

lando22

import humingbird

prediction = humingbird.Text.predict(
  text="A example sentence from the dataset",
  labels=["positive", "negative", "neutral"]
)

print(prediction)

lando22

import transformers
from transformers import pipeline

generator = pipeline('text-generation', model='gpt2')

lando22

import humingbird

content_filter = humingbird.Text.predict(
  text="our sample text for Humingbird! This is so easy.",
  labels=["toxic", "not toxic"]
)

print(content_filter)

lando22

import transformers
from transformers import pipeline
import humingbird

# load the model
generator = pipeline('text-generation', model='gpt2')

# generate a sample
generation = (generator("I went for a walk today and saw a", max_length=60))[0]['generated_text']