I'm trying to create a question and answering AI, I would like it to be as accurate as possible without having to train the model myself.
I can create a simple AI using the existing base models like so via their documentation:
from transformers import AlbertTokenizer, AlbertForQuestionAnswering
import torch
tokenizer = AlbertTokenizer.from_pretrained('albert-base-v2')
model = AlbertForQuestionAnswering.from_pretrained('albert-base-v2')
question, text = "What does He like?", "He likes bears"
inputs = tokenizer(question, text, return_tensors='pt')
start_positions = torch.tensor([1])
end_positions = torch.tensor([3])
outputs = model(**inputs, start_positions=start_positions, end_positions=end_positions)
loss = outputs.loss
start_scores = outputs.start_logits
end_scores = outputs.end_logits
answer_start = torch.argmax(start_scores) # get the most likely beginning of answer with the argmax of the score
answer_end = torch.argmax(end_scores) + 1
tokenizer.convert_tokens_to_string(tokenizer.convert_ids_to_tokens(inputs["input_ids"][0][answer_start:answer_end]))
However this model doesn't answer questions as accurate as others. On the HuggingFace site I've found an example that I'd like to use of a fine-tuned model
However the instructions show how to train a model like so. The example works on the page so clearly a pretrained model of the exists.
Does anyone know how I can reuse the existing models so I don't have to train one from scratch?
Turns out I just needed to grab an additional identifier when trying to request the model:
from transformers import AlbertTokenizer, AlbertForQuestionAnswering
import torch
MODEL_PATH = 'ktrapeznikov/albert-xlarge-v2-squad-v2';
tokenizer = AlbertTokenizer.from_pretrained(MODEL_PATH)
model = AlbertForQuestionAnswering.from_pretrained(MODEL_PATH)
For future reference this information can be grabbed from the transformers use button. Seem in the image below.
Related
I'm using Tensorflow Lite to train an image classifier. I now have a bunch of *.tflite models stored, and I'm trying to write some code that allows me to pick a tflite model file, pick a dataset, and test that model on that dataset (inference).
When I train a model using:
model = image_classifier.create(trainData, validation_data=valData, shuffle=True, use_augmentation=False)
I am able to easily test this model on a test dataset right away because the model is actually stored in the variable 'model', by using:
model.evaluate_tflite('model.tflite', test_data)
or
loss, accuracy = model.evaluate(test_data)
However, if I simply want to load an already existing *.tflite model, without having trained it in the same run, I can't figure out a simple way to do that.
Following these instructions, it seems to be a lot of steps for what I'm trying to do. In other Machine Learning libraries (like PyTorch), you are able to define the model and then quickly load the saved weights and then get to testing, like:
model = models.densenet201(progress=True, pretrained=pretrained)
model.load_state_dict(torch.load("models/model.pt"))
Is there a simple way for me to initialise the model into the 'model' variable, load the saved weights from a *.tflite file, and then run inference?
Thank you for your help
A simple example of image classification:
import tensorflow as tf
import numpy as np
import cv2
class TFLiteModel:
def __init__(self, model_path: str):
self.interpreter = tf.lite.Interpreter(model_path)
self.interpreter.allocate_tensors()
self.input_details = self.interpreter.get_input_details()
self.output_details = self.interpreter.get_output_details()
def predict(self, *data_args):
assert len(data_args) == len(self.input_details)
for data, details in zip(data_args, self.input_details):
self.interpreter.set_tensor(details["index"], data)
self.interpreter.invoke()
return self.interpreter.get_tensor(self.output_details[0]["index"])
model = TFLiteModel("mobilenet_v2_1.0_224_1_default_1.tflite")
image = cv2.imread("hand_blower.png")
image = cv2.resize(image, (224, 224))
image = image.astype(np.float32)[np.newaxis]
image = (image - 127.5) / 127.5
label = model.predict(image)[0].argmax()
print(label)
Please refer to the official documentation for detailed information:
https://www.tensorflow.org/api_docs/python/tf/lite/Interpreter
The model was loaded from:
https://tfhub.dev/tensorflow/lite-model/mobilenet_v2_1.0_224/1/default/1
I have trained a fastai model using Kaggle notebook, it has saved the model but how to load the model is the problem, i have tried different methods like the method given below. Even it does load the model it doesn't have any predict function only thing I can see is model.eval().
The second problem is when the model was trained on google collab it didn't even get the single image, I did try to convert the image to NumPy way and another way but both didn't work out.
I am attaching the kaggle link of model training, the saved model and the test images in last after this code
#Code for Loading model
from fastai import *
from fastai.vision import *
import torch
loc = torch.load('/content/gdrive/MyDrive/Data Exports/35k data/stage-1.pth')
body = create_body(models.resnet18, True, None)
data_classes = 4
nf = callbacks.hooks.num_features_model(body) * 2
head = create_head(nf, data_classes, None, ps=0.5, bn_final=False)
model = nn.Sequential(body, head)
Kaggle Model
Test Images From Kaggle Dataset
Saved Model
How to load pytorch models:
loc = torch.load('/content/gdrive/MyDrive/Data Exports/35k data/stage-1.pth')
model = ... # build your model
model.load_state_dict(loc)
model.eval()
Now you should be able to simply use the forward pass to generate your predictions:
input = ... # your input image
pred = model(input) # your class predictions
Don't forget to convert your inputs to torch tensors first, you might want to use a DataLoader for ease of use.
I am trying to use this huggingface model and have been following the example provided, but I am getting an error when loading the tokenizer:
from transformers import AutoTokenizer
task = 'sentiment'
MODEL = f"cardiffnlp/twitter-roberta-base-{task}"
tokenizer = AutoTokenizer.from_pretrained(MODEL)
OSError: Can't load tokenizer for 'cardiffnlp/twitter-roberta-base-sentiment'. Make sure that:
'cardiffnlp/twitter-roberta-base-sentiment' is a correct model identifier listed on 'https://huggingface.co/models'
or 'cardiffnlp/twitter-roberta-base-sentiment' is the correct path to a directory containing relevant tokenizer files
What I find very weird is that I was able to run my script several times but ran into an error after some time, while I don't recall changing anything in the meantime. Does anyone know what's the solution here?
EDIT: Here is my entire script:
from transformers import AutoTokenizer
from transformers import AutoModelForSequenceClassification
from transformers import TFAutoModelForSequenceClassification
import numpy as np
from scipy.special import softmax
import csv
import urllib.request
task = 'sentiment'
MODEL = f"nlptown/bert-base-multilingual-uncased-{task}"
tokenizer = AutoTokenizer.from_pretrained(MODEL)
labels = ['very_negative', 'negative', 'neutral', 'positive', 'very_positive']
model = AutoModelForSequenceClassification.from_pretrained(MODEL)
model.save_pretrained(MODEL)
text = "I love you"
encoded_input = tokenizer(text, return_tensors='pt')
output = model(**encoded_input)
scores = output[0][0].detach().numpy()
scores = softmax(scores)
print(scores)
The error seems to start happening when I run model.save_pretrained(MODEL), but this might be a coincidence.
I just came across this same issue. It seems like a bug with model.save_pretrained(), as you noted.
I was able to resolve by deleting the directory where the model had been saved (cardiffnlp/) and running again without model.save_pretrained().
Not sure what your application is. For me, re-downloading the model each time takes ~5s and that is acceptable.
I have problem with fastai library. My code below:
import fastai
from fastai.text import *
import os
import pandas as pd
import fastai
from fastai import *
lab = df.columns[0]
data_lm = TextLMDataBunch.from_csv(r'/AWD', 'data.csv', label_cols = lab, text_cols = ['text'])
data_clas = TextClasDataBunch.from_csv(r'/AWD', 'data.csv', vocab = data_lm.train_ds.vocab, bs = 256,label_cols = lab, text_cols=['text'])
data_lm.save('data_lm_export.pkl')
data_clas.save('data_clas.pkl')
learn = language_model_learner(data_lm,AWD_LSTM,drop_mult = 0.3)
learn.lr_find()
learn.recorder.plot(skip_end=10)
learn.fit_one_cycle(10,1e-2,moms=(0.8,0.7))
learn.save('fit_head')
learn.load('fit_head')
My data is quite big, so each epoch in fit_one_cycle lasts about 6h. My resources enables me only to train model in SLURM JOB 70h, so my whole script will be cancelled. I wanted to divide my script into pieces and the first longest part has to learn and save fit_head. Everything was ok, and after that I wanted to load my model to train it again, but i got this error:
**RuntimeError: Error(s) in loading state_dict for SequentialRNN:
size mismatch for 0.encoder.weight: copying a param with shape torch.Size([54376, 400]) from checkpoint, the shape in current model is torch.Size([54720, 400]).
**
I have checked similar problems on github/stack posts and I tried those solutions like this below, but i cannot find anything usefull.
data_clas.vocab.stoi = data_lm.vocab.stoi
data_clas.vocab.itos = data_lm.vocab.itos
Is there any possibility to load trained model without having this issue ?
When you do learner.save() only the model weights are saved on your disk and not the model state dict which contains the model architecture information.
To train the model in a different session you must first define the model itself. Remember to use the same code to define your new model. Since your data is quite heavy as you mentioned you can use a very small subset (~16 records) of your data to create this new model and then do learn.load(model_path) and you should be able to resume training.
you can modify the training data with learn.data.train_dl = new_dl
I have trouble getting the DeepBeliefTrainer to work on my data in PyBrain/Python. Since I can't find any examples other than unsupervised on how to use the deep learning in PyBrain, I hope that someone can give examples that would show a basic concept of usage.
I have tried to initialize using:
epochs = 100
layerDims = [768,100,100,1]
net = buildNetwork(*layerDims)
dataset = self.dataset
trainer = DeepBeliefTrainer(net, dataset=dataSet)
trainer.trainEpochs(epochs)
I try to use a SupervisedDataset for regression, but the training just fails. Have anyone succeded with using deeplearning trainer for supervised machine learning? And how did you do it?
Error I get:
File "/Library/Python/2.7/site-packages/PyBrain-0.3.1-py2.7.egg/pybrain/structure/networks/rbm.py", line 39, in __init__
self.con = self.net.connections[self.visible][0]
KeyError: None
It's because your initial network:
net = buildNetwork(*layerDims) doesn't have a layer with the name of the visible layer in your deep belief network, which is 'visible'. So, in order to find it mapped in the initial network, you can do something like:
net.addInputModule(LinearLayer(input_dim, 'visible'))
[...]
trainer = DeepBeliefTrainer(net, dataset=dataSet)