freedomtowin

{'input_ids': tensor([[ 101, 1045, 2293, 2023, 3185,  999,  102],
        [ 101, 2009, 2001, 3100, 1012,  102,    0]]), 'attention_mask': tensor([[1, 1, 1, 1, 1, 1, 1],
        [1, 1, 1, 1, 1, 1, 0]])}

freedomtowin

from transformers import AutoTokenizer, AutoModelForSequenceClassification
import torch

tokenizer = AutoTokenizer.from_pretrained("distilbert-base-uncased-finetuned-sst-2-english")
model = AutoModelForSequenceClassification.from_pretrained("distilbert-base-uncased-finetuned-sst-2-english")

sentences = ["I love this movie!", "It was okay."]

# Tokenize and pad
encoded_input = tokenizer(sentences, padding=True, return_tensors="pt")

freedomtowin

Tokens: ['i', "'", 've', 'been', 'waiting', 'for', 'a', 'hugging', '##face', 'course', 'my', 'whole', 'life',           '.']
Input IDs: tensor([[ 1045,  1005,  2310,  2042,  3403,  2005,  1037, 17662, 12172,  2607,
        2026,  2878,  2166,  1012]])
Final Inputs: {'input_ids': [101, 1045, 1005, 2310, 2042, 3403, 2005, 1037, 17662, 12172, 2607, 2026, 2878, 2166,       1012, 102], 'attention_mask': [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]}
Decoded Inputs: [CLS] i've been waiting for a huggingface course my whole life. [SEP]

freedomtowin

import torch
from transformers import AutoTokenizer

checkpoint = "distilbert-base-uncased-finetuned-sst-2-english"
tokenizer = AutoTokenizer.from_pretrained(checkpoint)

sequence = "I've been waiting for a HuggingFace course my whole life."

tokens = tokenizer.tokenize(sequence)
print("Tokens:", tokens)

freedomtowin

BertConfig {
  "_attn_implementation_autoset": true,
  "architectures": [
    "BertForMaskedLM"
  ],
  "attention_probs_dropout_prob": 0.1,
  "classifier_dropout": null,
  "gradient_checkpointing": false,
  "hidden_act": "gelu",
  "hidden_dropout_prob": 0.1,

freedomtowin

from transformers import BertConfig, BertModel 
bert_config = BertConfig.from_pretrained("bert-base-cased" , num_hidden_Iayers=10) 
bert_model = BertModel(bert_config)

#### Training Code ####

bert_model.save_pretrained("my-bert-model")
bert_model = BertModel.from_pretrained("my-bert-model")

freedomtowin

Inputs: {'input_ids': tensor([[  101,  1045,  1005,  2310,  2042,  3403,  2005,  1037, 17662, 12172,
          2607,  2026,  2878,  2166,  1012,   102],
        [  101,  1045,  5223,  2023,  2061,  2172,   999,   102,     0,     0,
             0,     0,     0,     0,     0,     0]]), 
        'attention_mask': tensor([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1],
        [1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0]])
        }
Logits: tensor([[-1.5607,  1.6123],
                [ 4.1692, -3.3464]], grad_fn=\<AddmmBackward0>)
Predictions: tensor([[4.0195e-02, 9.5980e-01],

freedomtowin

import torch
from transformers import AutoTokenizer, AutoModelForSequenceClassification#, AutoModel

checkpoint = "distilbert-base-uncased-finetuned-sst-2-english"
tokenizer = AutoTokenizer.from_pretrained(checkpoint)
# model = AutoModel.from_pretrained(checkpoint)
model_seq_classification = AutoModelForSequenceClassification.from_pretrained(checkpoint)

raw_inputs = [
    "I've been waiting for a HuggingFace course my whole life.",

freedomtowin

graph TD
Model-->Task-A-large-dataset
Task-A-large-dataset-.->Model
Model-->Task-B-small-dataset
Task-B-small-dataset-->Finetuned-Model

freedomtowin

[{'translation_text': 'This course is produced by Hugging Face.'}]