# Installazione di Transformers
! pip install transformers datasets
# Per installare dalla fonte invece dell'ultima versione rilasciata, commenta il comando sopra e
# rimuovi la modalità commento al comando seguente.
# ! pip install git+https://github.com/huggingface/transformers.git

#@title
from IPython.display import HTML

HTML('<iframe width="560" height="315" src="https://www.youtube.com/embed/tiZFewofSLM?rel=0&amp;controls=0&amp;showinfo=0" frameborder="0" allowfullscreen></iframe>')

from transformers import pipeline

classificatore = pipeline("sentiment-analysis", model="MilaNLProc/feel-it-italian-sentiment")

classificatore("Siamo molto felici di mostrarti la libreria 🤗 Transformers.")

risultati = classificatore(
    ["Siamo molto felici di mostrarti la libreria 🤗 Transformers.", "Speriamo te non la odierai."]
)
for risultato in risultati:
    print(f"etichetta: {risultato['label']}, con punteggio: {round(risultato['score'], 4)}")

import torch
from transformers import pipeline

riconoscitore_vocale = pipeline(
    "automatic-speech-recognition", model="radiogroup-crits/wav2vec2-xls-r-1b-italian-doc4lm-5gram"
)

from datasets import load_dataset, Audio

dataset = load_dataset("PolyAI/minds14", name="it-IT", split="train")

dataset = dataset.cast_column("audio", Audio(sampling_rate=riconoscitore_vocale.feature_extractor.sampling_rate))

risultato = riconoscitore_vocale(dataset[:4]["audio"])
print([d["text"] for d in risultato])

model_name = "nlptown/bert-base-multilingual-uncased-sentiment"

from transformers import AutoTokenizer, AutoModelForSequenceClassification

model = AutoModelForSequenceClassification.from_pretrained(model_name)
tokenizer = AutoTokenizer.from_pretrained(model_name)

from transformers import AutoTokenizer, TFAutoModelForSequenceClassification

model = TFAutoModelForSequenceClassification.from_pretrained(model_name)
tokenizer = AutoTokenizer.from_pretrained(model_name)

classifier = pipeline("sentiment-analysis", model=model, tokenizer=tokenizer)
classifier("Nous sommes très heureux de vous présenter la bibliothèque 🤗 Transformers.")

#@title
from IPython.display import HTML

HTML('<iframe width="560" height="315" src="https://www.youtube.com/embed/AhChOFRegn4?rel=0&amp;controls=0&amp;showinfo=0" frameborder="0" allowfullscreen></iframe>')

from transformers import AutoTokenizer

nome_del_modello = "nlptown/bert-base-multilingual-uncased-sentiment"
tokenizer = AutoTokenizer.from_pretrained(nome_del_modello)

encoding = tokenizer("Siamo molto felici di mostrarti la libreria 🤗 Transformers.")
print(encoding)

pt_batch = tokenizer(
    ["Siamo molto felici di mostrarti la libreria 🤗 Transformers.", "Speriamo te non la odierai."],
    padding=True,
    truncation=True,
    max_length=512,
    return_tensors="pt",
)

tf_batch = tokenizer(
    ["Siamo molto felici di mostrarti la libreria 🤗 Transformers.", "Speriamo te non la odierai."],
    padding=True,
    truncation=True,
    max_length=512,
    return_tensors="tf",
)

from transformers import AutoModelForSequenceClassification

model_name = "nlptown/bert-base-multilingual-uncased-sentiment"
pt_model = AutoModelForSequenceClassification.from_pretrained(model_name)

pt_outputs = pt_model(**pt_batch)

from torch import nn

pt_predictions = nn.functional.softmax(pt_outputs.logits, dim=-1)
print(pt_predictions)

from transformers import TFAutoModelForSequenceClassification

nome_del_modello = "nlptown/bert-base-multilingual-uncased-sentiment"
tf_model = TFAutoModelForSequenceClassification.from_pretrained(nome_del_modello)

tf_outputs = tf_model(tf_batch)

import tensorflow as tf

tf_predictions = tf.nn.softmax(tf_outputs.logits, axis=-1)
tf_predictions

pt_save_directory = "./pt_save_pretrained"
tokenizer.save_pretrained(pt_save_directory)
pt_model.save_pretrained(pt_save_directory)

pt_model = AutoModelForSequenceClassification.from_pretrained("./pt_save_pretrained")

tf_save_directory = "./tf_save_pretrained"
tokenizer.save_pretrained(tf_save_directory)
tf_model.save_pretrained(tf_save_directory)

tf_model = TFAutoModelForSequenceClassification.from_pretrained("./tf_save_pretrained")

from transformers import AutoModel

tokenizer = AutoTokenizer.from_pretrained(tf_save_directory)
pt_model = AutoModelForSequenceClassification.from_pretrained(tf_save_directory, from_tf=True)

from transformers import TFAutoModel

tokenizer = AutoTokenizer.from_pretrained(pt_save_directory)
tf_model = TFAutoModelForSequenceClassification.from_pretrained(pt_save_directory, from_pt=True)