import json
import sys
from model import ModelWrapper
model_wrapper = ModelWrapper()
model_wrapper.load_model()
data = json.loads(sys.argv[1])
print(f"Data: {data}")
prediction = model_wrapper.predict(data=data)
print(f"Prediction: {prediction}")
for i, batch in enumerate(train_batch):
start_time = time.time()
global_step += 1
loss = model.update(batch)
train_loss += loss
if global_step % opt['log_step'] == 0:
duration = time.time() - start_time
print(format_str.format(datetime.now(), global_step, max_steps, epoch, \
opt['num_epoch'], loss, duration, current_lr))
print("Evaluating on dev set ...")
predictions = []
probabilities = []
dev_loss = 0
for i, batch in enumerate(dev_batch):
preds, probs, loss = model.predict(batch, thres=0.01)
predictions += preds
probabilities += probs
dev_loss += loss
train_loss = train_loss / train_batch.num_examples * opt[
'batch_size'] # avg loss per batch
dev_loss = dev_loss / dev_batch.num_examples * opt['batch_size']
model.scheduler.step(dev_loss)
# _, _, _, dev_f1 = metrics(dev_batch.gold(), predictions)
_, _, _, dev_f1, _ = tune_thres_new(dev_batch.gold(), probabilities)
print("epoch {}: train_loss = {:.6f}, dev_loss = {:.6f}, dev_f1 = {:.4f}".format(epoch,\
train_loss, dev_loss, dev_f1))
model_file = os.path.join(opt['model_save_dir'],
# train_batch = DataLoader(os.path.join(opt['data_dir'], 'train.csv'),
# opt['batch_size'],
# opt,
# weibo2embid=weibo2embid,
# evaluation=False)
dev_batch = DataLoader(os.path.join(opt['data_dir'], 'dev.csv'),
opt['batch_size'],
opt,
weibo2embid=weibo2embid,
evaluation=True)
model = ModelWrapper(opt, weibo2embid, eva=True)
model.load(os.path.join(opt['model_save_dir'], 'best_model.pt'))
all_probs = []
all_preds = []
for i, b in enumerate(dev_batch):
preds, probs, _ = model.predict(b, thres=0.5)
all_probs += probs
all_preds += preds
acc, prec, rec, dev_f1 = metrics(dev_batch.gold(), all_preds)
print('acc: {}, prec: {}, rec: {}, f1: {}\n'.format(acc, prec, rec, dev_f1))
auc, prec, rec, f1, best_thres = tune_thres_new(dev_batch.gold(), all_probs)
print('auc: {}, prec: {}, rec: {}, f1: {}, best_thres: {}'.format(auc, prec, rec, f1, best_thres))
with open('./log.txt', 'a+') as fout:
fout.write('\n' + time.asctime(time.localtime(time.time())))
fout.write(' '.join(sys.argv))
fout.write('\nauc: {}, prec: {}, rec: {}, f1: {}, best_thres: {}\n'.format(auc, prec, rec, f1, best_thres))