bert of thesus模型实战，采用transformers直接调用

from transformers import AutoTokenizer, AutoModel

tokenizer = AutoTokenizer.from_pretrained("canwenxu/BERT-of-Theseus-MNLI")

model = AutoModel.from_pretrained("canwenxu/BERT-of-Theseus-MNLI")

Downloading:   0%|          | 0.00/151 [00:00

#coding=utf-8
import numpy as np
import pandas as pd
import os
import matplotlib.pyplot as plt
from sklearn.model_selection import train_test_split
from sklearn.metrics import accuracy_score
import torch
import transformers
import random
from tqdm import tqdm

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

torch.cuda.empty_cache()

import os
import datetime

#打印时间
def printbar():
    nowtime = datetime.datetime.now().strftime('%Y-%m-%d %H:%M:%S')
    print("\n"+"=========="*8 + "%s"%nowtime)

#mac系统上pytorch和matplotlib在jupyter中同时跑需要更改环境变量
os.environ["KMP_DUPLICATE_LIB_OK"]="TRUE" 

def read_sst2_split(split_dir):
#     split_dir = Path(split_dir)
    texts = []
    labels = []
    spilt_data=pd.read_csv(split_dir,sep='\t'
                       )
    spilt_data.columns=['text','label']
    texts=list(spilt_data.text)
    labels=list(spilt_data.label)
    return texts, labels

path='D://Dataset//GLUE//SST//SST-2//'

train_texts, train_labels = read_sst2_split(path+'train.tsv')
test_texts, test_labels = read_sst2_split(path+'dev.tsv')

数据处理

random_seed = 0
numpy_seed = 1
torch_seed = 2
cuda_seed = 3

random.seed(random_seed)
np.random.seed(numpy_seed)
torch.manual_seed(torch_seed)
torch.cuda.manual_seed_all(cuda_seed)

#数据处理
from sklearn.model_selection import train_test_split
train_texts, val_texts, train_labels, val_labels = train_test_split(train_texts,
                                                                    train_labels, test_size=.4,random_state=random_seed)

model_path="canwenxu/BERT-of-Theseus-MNLI"

from transformers import BertForSequenceClassification,BertTokenizer

tokenizer = BertTokenizer.from_pretrained(model_path)

##文本向量化
# train_encodings = tokenizer(train_texts[:100], truncation=True, padding=True)
# val_encodings = tokenizer(val_texts[:50], truncation=True, padding=True)
# test_encodings = tokenizer(test_texts[:25], truncation=True, padding=True)

## 文本向量化版本2
maxlen=20

##文本向量化
train_encodings = tokenizer(train_texts[:128],
                    padding='max_length',  # Pad to max_length
                    truncation=True,       # Truncate to max_length
                    max_length=maxlen,  
                    return_tensors='pt')  # Return torch.Tensor objects
val_encodings = tokenizer(val_texts[:64],
                    padding='max_length',  # Pad to max_length
                    truncation=True,       # Truncate to max_length
                    max_length=maxlen,  
                    return_tensors='pt')  # Return torch.Tensor objects
test_encodings = tokenizer(test_texts[:32],
                    padding='max_length',  # Pad to max_length
                    truncation=True,       # Truncate to max_length
                    max_length=maxlen,  
                    return_tensors='pt')  # Return torch.Tensor objects

import gc
gc.collect()

1324

class IMDbDataset(torch.utils.data.Dataset):
    def __init__(self, encodings, labels):
        self.encodings = encodings
        self.labels = labels
 
    def __getitem__(self, idx):
        item = {key: torch.tensor(val[idx]) for key, val in self.encodings.items()}
        item['labels'] = torch.tensor(self.labels[idx])
        return item
 
    def __len__(self):
        return len(self.labels)

# train_dataset = IMDbDataset(train_encodings, train_labels)
# val_dataset = IMDbDataset(val_encodings, val_labels)
# test_dataset = IMDbDataset(test_encodings, test_labels)
train_dataset = IMDbDataset(train_encodings, train_labels[:128])
val_dataset = IMDbDataset(val_encodings, val_labels[:64])
test_dataset = IMDbDataset(test_encodings, test_labels[:32])

from torch.utils.data import DataLoader
# from openprompt import DataLoader
train_data_loader = DataLoader(
    dataset = train_dataset,
#     tokenizer = tokenizer, 
#     tokenizer = tokenizer, 
#     template = template,
    batch_size=32,
    shuffle=True
)
val_data_loader = DataLoader(
    dataset = val_dataset,
#     tokenizer = tokenizer, 
#     tokenizer = tokenizer, 
#     template = template, 
    batch_size=32,
    shuffle=True
)
test_data_loader = DataLoader(
    dataset = val_dataset,
#     tokenizer = tokenizer, 
#     tokenizer = tokenizer, 
#     template = template, 
    batch_size=32,
    shuffle=True
)

model = BertForSequenceClassification.from_pretrained(model_path,num_labels=2)

Some weights of BertForSequenceClassification were not initialized from the model checkpoint at canwenxu/BERT-of-Theseus-MNLI and are newly initialized: ['classifier.bias', 'classifier.weight']
You should probably TRAIN this model on a down-stream task to be able to use it for predictions and inference.

model

BertForSequenceClassification(
  (bert): BertModel(
    (embeddings): BertEmbeddings(
      (word_embeddings): Embedding(30522, 768, padding_idx=0)
      (position_embeddings): Embedding(512, 768)
      (token_type_embeddings): Embedding(2, 768)
      (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
      (dropout): Dropout(p=0.1, inplace=False)
    )
    (encoder): BertEncoder(
      (layer): ModuleList(
        (0): BertLayer(
          (attention): BertAttention(
            (self): BertSelfAttention(
              (query): Linear(in_features=768, out_features=768, bias=True)
              (key): Linear(in_features=768, out_features=768, bias=True)
              (value): Linear(in_features=768, out_features=768, bias=True)
              (dropout): Dropout(p=0.1, inplace=False)
            )
            (output): BertSelfOutput(
              (dense): Linear(in_features=768, out_features=768, bias=True)
              (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
              (dropout): Dropout(p=0.1, inplace=False)
            )
          )
          (intermediate): BertIntermediate(
            (dense): Linear(in_features=768, out_features=3072, bias=True)
          )
          (output): BertOutput(
            (dense): Linear(in_features=3072, out_features=768, bias=True)
            (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
            (dropout): Dropout(p=0.1, inplace=False)
          )
        )
        (1): BertLayer(
          (attention): BertAttention(
            (self): BertSelfAttention(
              (query): Linear(in_features=768, out_features=768, bias=True)
              (key): Linear(in_features=768, out_features=768, bias=True)
              (value): Linear(in_features=768, out_features=768, bias=True)
              (dropout): Dropout(p=0.1, inplace=False)
            )
            (output): BertSelfOutput(
              (dense): Linear(in_features=768, out_features=768, bias=True)
              (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
              (dropout): Dropout(p=0.1, inplace=False)
            )
          )
          (intermediate): BertIntermediate(
            (dense): Linear(in_features=768, out_features=3072, bias=True)
          )
          (output): BertOutput(
            (dense): Linear(in_features=3072, out_features=768, bias=True)
            (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
            (dropout): Dropout(p=0.1, inplace=False)
          )
        )
        (2): BertLayer(
          (attention): BertAttention(
            (self): BertSelfAttention(
              (query): Linear(in_features=768, out_features=768, bias=True)
              (key): Linear(in_features=768, out_features=768, bias=True)
              (value): Linear(in_features=768, out_features=768, bias=True)
              (dropout): Dropout(p=0.1, inplace=False)
            )
            (output): BertSelfOutput(
              (dense): Linear(in_features=768, out_features=768, bias=True)
              (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
              (dropout): Dropout(p=0.1, inplace=False)
            )
          )
          (intermediate): BertIntermediate(
            (dense): Linear(in_features=768, out_features=3072, bias=True)
          )
          (output): BertOutput(
            (dense): Linear(in_features=3072, out_features=768, bias=True)
            (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
            (dropout): Dropout(p=0.1, inplace=False)
          )
        )
        (3): BertLayer(
          (attention): BertAttention(
            (self): BertSelfAttention(
              (query): Linear(in_features=768, out_features=768, bias=True)
              (key): Linear(in_features=768, out_features=768, bias=True)
              (value): Linear(in_features=768, out_features=768, bias=True)
              (dropout): Dropout(p=0.1, inplace=False)
            )
            (output): BertSelfOutput(
              (dense): Linear(in_features=768, out_features=768, bias=True)
              (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
              (dropout): Dropout(p=0.1, inplace=False)
            )
          )
          (intermediate): BertIntermediate(
            (dense): Linear(in_features=768, out_features=3072, bias=True)
          )
          (output): BertOutput(
            (dense): Linear(in_features=3072, out_features=768, bias=True)
            (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
            (dropout): Dropout(p=0.1, inplace=False)
          )
        )
        (4): BertLayer(
          (attention): BertAttention(
            (self): BertSelfAttention(
              (query): Linear(in_features=768, out_features=768, bias=True)
              (key): Linear(in_features=768, out_features=768, bias=True)
              (value): Linear(in_features=768, out_features=768, bias=True)
              (dropout): Dropout(p=0.1, inplace=False)
            )
            (output): BertSelfOutput(
              (dense): Linear(in_features=768, out_features=768, bias=True)
              (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
              (dropout): Dropout(p=0.1, inplace=False)
            )
          )
          (intermediate): BertIntermediate(
            (dense): Linear(in_features=768, out_features=3072, bias=True)
          )
          (output): BertOutput(
            (dense): Linear(in_features=3072, out_features=768, bias=True)
            (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
            (dropout): Dropout(p=0.1, inplace=False)
          )
        )
        (5): BertLayer(
          (attention): BertAttention(
            (self): BertSelfAttention(
              (query): Linear(in_features=768, out_features=768, bias=True)
              (key): Linear(in_features=768, out_features=768, bias=True)
              (value): Linear(in_features=768, out_features=768, bias=True)
              (dropout): Dropout(p=0.1, inplace=False)
            )
            (output): BertSelfOutput(
              (dense): Linear(in_features=768, out_features=768, bias=True)
              (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
              (dropout): Dropout(p=0.1, inplace=False)
            )
          )
          (intermediate): BertIntermediate(
            (dense): Linear(in_features=768, out_features=3072, bias=True)
          )
          (output): BertOutput(
            (dense): Linear(in_features=3072, out_features=768, bias=True)
            (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
            (dropout): Dropout(p=0.1, inplace=False)
          )
        )
      )
    )
    (pooler): BertPooler(
      (dense): Linear(in_features=768, out_features=768, bias=True)
      (activation): Tanh()
    )
  )
  (dropout): Dropout(p=0.1, inplace=False)
  (classifier): Linear(in_features=768, out_features=2, bias=True)
)

del model

import gc
gc.collect()

2367

from sklearn.metrics import accuracy_score
from sklearn.metrics import f1_score, precision_score, recall_score, accuracy_score

loss_func = torch.nn.CrossEntropyLoss()
# from transformers import AdamW
# tmp_optimizer=AdamW(model.parameters(),lr=0.001)
optimizer = torch.optim.Adam(params=model.parameters(),lr = 0.01)
metric_func = accuracy_score
metric_name = "accuracy"

def train_epoch(tmp_model,dataloader,tmp_op,epoch):
    loss_function=torch.nn.CrossEntropyLoss()
    tmp_model.train()
    tmp_model.zero_grad()
    total_loss = 0.0
    total_metric=0.0
    #########
    #这里计算
    preds = []
    labels = []
    ###########
    for step,batch in enumerate(dataloader,1):
        tmp_op.zero_grad()
        input_ids = batch['input_ids'].to(device)
        attention_mask = batch['attention_mask'].to(device)
        label = batch['labels'].to(device)
        outputs = model(input_ids, attention_mask=attention_mask, labels=label)
        loss = outputs[0]
        logits=outputs[1]
        #被迫节省内存
        total_loss += float(loss.item())
        #这里是计算train的acc
        pred = torch.argmax(logits, dim=-1)
        preds.extend(pred.cpu().tolist())
        labels.extend(batch['labels'].cpu().tolist())
        metric = accuracy_score(preds, labels)
#         print(metric)
        total_metric += float(metric.item())
        loss.requires_grad_(True)
        tmp_op.step()        
#     logger.info("Epoch {}, loss: {:.4f}".format(epoch, total_loss))
#     print("Epoch {}, loss: {:.4f}".format(epoch, total_loss))
    return total_loss/step, total_metric/step

def evaluate(tmp_model,dataloader, split, post_evaluate_hook=None):
        
        preds = []
        labels = []
        tmp_model.eval()
        
        val_total_loss = 0.0
        val_total_metric=0.0
        
        with torch.no_grad():
            for step,batch in enumerate(dataloader,1):
                input_ids = batch['input_ids'].to(device)
                attention_mask = batch['attention_mask'].to(device)
                label = batch['labels'].to(device)
                outputs = tmp_model(input_ids, attention_mask=attention_mask, labels=label)
                loss = outputs[0]
                logits=outputs[1]
                
                pred = torch.argmax(logits, dim=-1)
                preds.extend(pred.cpu().tolist())
                labels.extend(batch['labels'].cpu().tolist())

                val_total_loss += loss.item()

#                 print(val_total_loss)
                metric = accuracy_score(preds, labels)
                val_total_metric += float(metric.item())
                
                
        tmp_model.train()
        return val_total_loss/step,val_total_metric/step

model.device

device(type='cpu')

model=model.to(device)

model.device

device(type='cuda', index=0)

epochs = 10
dfhistory = pd.DataFrame(columns = ["epoch","loss",metric_name,"val_loss","val_"+metric_name]) 
print("Start Training...")
nowtime = datetime.datetime.now().strftime('%Y-%m-%d %H:%M:%S')
print("=========="*8 + "%s"%nowtime)

# with torch.no_grad():
for epoch in range(1,epochs+1):  
    train_loss,train_metric= train_epoch(model,train_data_loader,optimizer,epoch)
    val_loss,val_metric=  evaluate(model,val_data_loader, "Valid")
    # 3，记录日志-------------------------------------------------
    info = (epoch, train_loss, train_metric, val_loss,val_metric)
    dfhistory.loc[epoch-1] = info
    # 打印epoch级别日志
    print(("\nEPOCH = %d, loss = %.4f,"+ metric_name + \
          "  = %.4f, val_loss = %.4f, "+"val_"+ metric_name+" = %.4f") 
          %info)
    nowtime = datetime.datetime.now().strftime('%Y-%m-%d %H:%M:%S')
    print("\n"+"=========="*8 + "%s"%nowtime)
        
print('Finished Training...')
    
# evaluate(model,test_data_loader, "Test")

Start Training...
================================================================================2022-04-26 10:51:26


C:\Users\22274\AppData\Local\Temp\ipykernel_18928\4205912457.py:7: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).
  item = {key: torch.tensor(val[idx]) for key, val in self.encodings.items()}
C:\Users\22274\AppData\Local\Temp\ipykernel_18928\4205912457.py:7: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).
  item = {key: torch.tensor(val[idx]) for key, val in self.encodings.items()}
C:\Users\22274\AppData\Local\Temp\ipykernel_18928\4205912457.py:7: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).
  item = {key: torch.tensor(val[idx]) for key, val in self.encodings.items()}
C:\Users\22274\AppData\Local\Temp\ipykernel_18928\4205912457.py:7: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).
  item = {key: torch.tensor(val[idx]) for key, val in self.encodings.items()}
C:\Users\22274\AppData\Local\Temp\ipykernel_18928\4205912457.py:7: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).
  item = {key: torch.tensor(val[idx]) for key, val in self.encodings.items()}
C:\Users\22274\AppData\Local\Temp\ipykernel_18928\4205912457.py:7: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).
  item = {key: torch.tensor(val[idx]) for key, val in self.encodings.items()}
C:\Users\22274\AppData\Local\Temp\ipykernel_18928\4205912457.py:7: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).
  item = {key: torch.tensor(val[idx]) for key, val in self.encodings.items()}
C:\Users\22274\AppData\Local\Temp\ipykernel_18928\4205912457.py:7: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).
  item = {key: torch.tensor(val[idx]) for key, val in self.encodings.items()}
C:\Users\22274\AppData\Local\Temp\ipykernel_18928\4205912457.py:7: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).
  item = {key: torch.tensor(val[idx]) for key, val in self.encodings.items()}
C:\Users\22274\AppData\Local\Temp\ipykernel_18928\4205912457.py:7: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).
  item = {key: torch.tensor(val[idx]) for key, val in self.encodings.items()}
C:\Users\22274\AppData\Local\Temp\ipykernel_18928\4205912457.py:7: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).
  item = {key: torch.tensor(val[idx]) for key, val in self.encodings.items()}
C:\Users\22274\AppData\Local\Temp\ipykernel_18928\4205912457.py:7: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).
  item = {key: torch.tensor(val[idx]) for key, val in self.encodings.items()}
C:\Users\22274\AppData\Local\Temp\ipykernel_18928\4205912457.py:7: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).
  item = {key: torch.tensor(val[idx]) for key, val in self.encodings.items()}
C:\Users\22274\AppData\Local\Temp\ipykernel_18928\4205912457.py:7: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).
  item = {key: torch.tensor(val[idx]) for key, val in self.encodings.items()}
C:\Users\22274\AppData\Local\Temp\ipykernel_18928\4205912457.py:7: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).
  item = {key: torch.tensor(val[idx]) for key, val in self.encodings.items()}
C:\Users\22274\AppData\Local\Temp\ipykernel_18928\4205912457.py:7: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).
  item = {key: torch.tensor(val[idx]) for key, val in self.encodings.items()}
C:\Users\22274\AppData\Local\Temp\ipykernel_18928\4205912457.py:7: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).
  item = {key: torch.tensor(val[idx]) for key, val in self.encodings.items()}



EPOCH = 1, loss = 0.6709,accuracy  = 0.6237, val_loss = 0.6608, val_accuracy = 0.6641

================================================================================2022-04-26 10:51:30

EPOCH = 2, loss = 0.6715,accuracy  = 0.4870, val_loss = 0.6608, val_accuracy = 0.6484

================================================================================2022-04-26 10:51:30

EPOCH = 3, loss = 0.6704,accuracy  = 0.5827, val_loss = 0.6608, val_accuracy = 0.6641

================================================================================2022-04-26 10:51:30


C:\Users\22274\AppData\Local\Temp\ipykernel_18928\4205912457.py:7: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).
  item = {key: torch.tensor(val[idx]) for key, val in self.encodings.items()}
C:\Users\22274\AppData\Local\Temp\ipykernel_18928\4205912457.py:7: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).
  item = {key: torch.tensor(val[idx]) for key, val in self.encodings.items()}
C:\Users\22274\AppData\Local\Temp\ipykernel_18928\4205912457.py:7: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).
  item = {key: torch.tensor(val[idx]) for key, val in self.encodings.items()}
C:\Users\22274\AppData\Local\Temp\ipykernel_18928\4205912457.py:7: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).
  item = {key: torch.tensor(val[idx]) for key, val in self.encodings.items()}
C:\Users\22274\AppData\Local\Temp\ipykernel_18928\4205912457.py:7: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).
  item = {key: torch.tensor(val[idx]) for key, val in self.encodings.items()}
C:\Users\22274\AppData\Local\Temp\ipykernel_18928\4205912457.py:7: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).
  item = {key: torch.tensor(val[idx]) for key, val in self.encodings.items()}
C:\Users\22274\AppData\Local\Temp\ipykernel_18928\4205912457.py:7: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).
  item = {key: torch.tensor(val[idx]) for key, val in self.encodings.items()}
C:\Users\22274\AppData\Local\Temp\ipykernel_18928\4205912457.py:7: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).
  item = {key: torch.tensor(val[idx]) for key, val in self.encodings.items()}
C:\Users\22274\AppData\Local\Temp\ipykernel_18928\4205912457.py:7: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).
  item = {key: torch.tensor(val[idx]) for key, val in self.encodings.items()}
C:\Users\22274\AppData\Local\Temp\ipykernel_18928\4205912457.py:7: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).
  item = {key: torch.tensor(val[idx]) for key, val in self.encodings.items()}
C:\Users\22274\AppData\Local\Temp\ipykernel_18928\4205912457.py:7: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).
  item = {key: torch.tensor(val[idx]) for key, val in self.encodings.items()}
C:\Users\22274\AppData\Local\Temp\ipykernel_18928\4205912457.py:7: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).
  item = {key: torch.tensor(val[idx]) for key, val in self.encodings.items()}
C:\Users\22274\AppData\Local\Temp\ipykernel_18928\4205912457.py:7: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).
  item = {key: torch.tensor(val[idx]) for key, val in self.encodings.items()}
C:\Users\22274\AppData\Local\Temp\ipykernel_18928\4205912457.py:7: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).
  item = {key: torch.tensor(val[idx]) for key, val in self.encodings.items()}
C:\Users\22274\AppData\Local\Temp\ipykernel_18928\4205912457.py:7: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).
  item = {key: torch.tensor(val[idx]) for key, val in self.encodings.items()}
C:\Users\22274\AppData\Local\Temp\ipykernel_18928\4205912457.py:7: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).
  item = {key: torch.tensor(val[idx]) for key, val in self.encodings.items()}
C:\Users\22274\AppData\Local\Temp\ipykernel_18928\4205912457.py:7: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).
  item = {key: torch.tensor(val[idx]) for key, val in self.encodings.items()}



EPOCH = 4, loss = 0.6777,accuracy  = 0.5762, val_loss = 0.6608, val_accuracy = 0.6641

================================================================================2022-04-26 10:51:30

EPOCH = 5, loss = 0.6793,accuracy  = 0.6530, val_loss = 0.6608, val_accuracy = 0.6953

================================================================================2022-04-26 10:51:30

EPOCH = 6, loss = 0.6817,accuracy  = 0.5553, val_loss = 0.6608, val_accuracy = 0.7109

================================================================================2022-04-26 10:51:30


C:\Users\22274\AppData\Local\Temp\ipykernel_18928\4205912457.py:7: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).
  item = {key: torch.tensor(val[idx]) for key, val in self.encodings.items()}
C:\Users\22274\AppData\Local\Temp\ipykernel_18928\4205912457.py:7: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).
  item = {key: torch.tensor(val[idx]) for key, val in self.encodings.items()}
C:\Users\22274\AppData\Local\Temp\ipykernel_18928\4205912457.py:7: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).
  item = {key: torch.tensor(val[idx]) for key, val in self.encodings.items()}
C:\Users\22274\AppData\Local\Temp\ipykernel_18928\4205912457.py:7: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).
  item = {key: torch.tensor(val[idx]) for key, val in self.encodings.items()}
C:\Users\22274\AppData\Local\Temp\ipykernel_18928\4205912457.py:7: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).
  item = {key: torch.tensor(val[idx]) for key, val in self.encodings.items()}
C:\Users\22274\AppData\Local\Temp\ipykernel_18928\4205912457.py:7: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).
  item = {key: torch.tensor(val[idx]) for key, val in self.encodings.items()}
C:\Users\22274\AppData\Local\Temp\ipykernel_18928\4205912457.py:7: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).
  item = {key: torch.tensor(val[idx]) for key, val in self.encodings.items()}
C:\Users\22274\AppData\Local\Temp\ipykernel_18928\4205912457.py:7: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).
  item = {key: torch.tensor(val[idx]) for key, val in self.encodings.items()}
C:\Users\22274\AppData\Local\Temp\ipykernel_18928\4205912457.py:7: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).
  item = {key: torch.tensor(val[idx]) for key, val in self.encodings.items()}
C:\Users\22274\AppData\Local\Temp\ipykernel_18928\4205912457.py:7: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).
  item = {key: torch.tensor(val[idx]) for key, val in self.encodings.items()}
C:\Users\22274\AppData\Local\Temp\ipykernel_18928\4205912457.py:7: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).
  item = {key: torch.tensor(val[idx]) for key, val in self.encodings.items()}
C:\Users\22274\AppData\Local\Temp\ipykernel_18928\4205912457.py:7: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).
  item = {key: torch.tensor(val[idx]) for key, val in self.encodings.items()}
C:\Users\22274\AppData\Local\Temp\ipykernel_18928\4205912457.py:7: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).
  item = {key: torch.tensor(val[idx]) for key, val in self.encodings.items()}
C:\Users\22274\AppData\Local\Temp\ipykernel_18928\4205912457.py:7: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).
  item = {key: torch.tensor(val[idx]) for key, val in self.encodings.items()}
C:\Users\22274\AppData\Local\Temp\ipykernel_18928\4205912457.py:7: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).
  item = {key: torch.tensor(val[idx]) for key, val in self.encodings.items()}
C:\Users\22274\AppData\Local\Temp\ipykernel_18928\4205912457.py:7: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).
  item = {key: torch.tensor(val[idx]) for key, val in self.encodings.items()}
C:\Users\22274\AppData\Local\Temp\ipykernel_18928\4205912457.py:7: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).
  item = {key: torch.tensor(val[idx]) for key, val in self.encodings.items()}



EPOCH = 7, loss = 0.6757,accuracy  = 0.6016, val_loss = 0.6608, val_accuracy = 0.6484

================================================================================2022-04-26 10:51:30

EPOCH = 8, loss = 0.6775,accuracy  = 0.5996, val_loss = 0.6608, val_accuracy = 0.6953

================================================================================2022-04-26 10:51:30

EPOCH = 9, loss = 0.6936,accuracy  = 0.6035, val_loss = 0.6608, val_accuracy = 0.7266

================================================================================2022-04-26 10:51:30

EPOCH = 10, loss = 0.6795,accuracy  = 0.6204, val_loss = 0.6608, val_accuracy = 0.6484

================================================================================2022-04-26 10:51:30
Finished Training...


C:\Users\22274\AppData\Local\Temp\ipykernel_18928\4205912457.py:7: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).
  item = {key: torch.tensor(val[idx]) for key, val in self.encodings.items()}
C:\Users\22274\AppData\Local\Temp\ipykernel_18928\4205912457.py:7: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).
  item = {key: torch.tensor(val[idx]) for key, val in self.encodings.items()}
C:\Users\22274\AppData\Local\Temp\ipykernel_18928\4205912457.py:7: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).
  item = {key: torch.tensor(val[idx]) for key, val in self.encodings.items()}
C:\Users\22274\AppData\Local\Temp\ipykernel_18928\4205912457.py:7: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).
  item = {key: torch.tensor(val[idx]) for key, val in self.encodings.items()}
C:\Users\22274\AppData\Local\Temp\ipykernel_18928\4205912457.py:7: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).
  item = {key: torch.tensor(val[idx]) for key, val in self.encodings.items()}
C:\Users\22274\AppData\Local\Temp\ipykernel_18928\4205912457.py:7: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).
  item = {key: torch.tensor(val[idx]) for key, val in self.encodings.items()}
C:\Users\22274\AppData\Local\Temp\ipykernel_18928\4205912457.py:7: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).
  item = {key: torch.tensor(val[idx]) for key, val in self.encodings.items()}
C:\Users\22274\AppData\Local\Temp\ipykernel_18928\4205912457.py:7: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).
  item = {key: torch.tensor(val[idx]) for key, val in self.encodings.items()}
C:\Users\22274\AppData\Local\Temp\ipykernel_18928\4205912457.py:7: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).
  item = {key: torch.tensor(val[idx]) for key, val in self.encodings.items()}