/
train.py
61 lines (46 loc) · 1.64 KB
/
train.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
#!/usr/bin/env python3
"""
Running training.
Source: https://www.analyticsvidhya.com/blog/2022/02/sentiment-analysis-using-transformers/
"""
import numpy as np
import pandas as pd
from transformers import DistilBertTokenizerFast
import torch
from transformers import DistilBertForSequenceClassification
from transformers import Trainer, TrainingArguments
train = pd.read_csv("./data/training.csv")
train.drop("id", axis=1, inplace=True)
tokenizer = DistilBertTokenizerFast.from_pretrained("distilbert-base-uncased")
inputs = tokenizer(train['contents'].tolist(), padding="max_length", truncation=True)
class pdfDataset(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['label'] = torch.tensor(self.labels[idx])
return item
def __len__(self):
return len(self.labels)
train_dataset = pdfDataset(inputs, train['label'].tolist())
model = DistilBertForSequenceClassification.from_pretrained("distilbert-base-uncased")
device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
model.to(device)
training_args = TrainingArguments(
output_dir="./results",
num_train_epochs=10,
per_device_train_batch_size=16,
per_device_eval_batch_size=64,
warmup_steps=500,
weight_decay=0.01,
logging_dir="./logs",
logging_steps=10,
)
trainer = Trainer(
model=model,
args=training_args,
train_dataset=train_dataset,
)
trainer.train()
torch.save(model.state_dict(), 'model_weights.pth')