curiousily/getting-things-done-with-pytorch

Jupyter Notebook tutorials on solving real-world problems with Machine Learning & Deep Learning using PyTorch. Topics: Face detection with Detectron 2, Time Series anomaly detection with LSTM Autoencoders, Object Detection with YOLO v5, Build your first Neural Network, Time Series forecasting for Coronavirus daily cases, Sentiment Analysis with BERT.

Get SH*T Done with PyTorch

Learn how to solve real-world problems with Deep Learning models (NLP, Computer Vision, and Time Series). Go from prototyping to deployment with PyTorch and Python!

Read the book here

📖 Read for FREE

The whole book can be read using the links below. Each part contains a notebook that you can find in this repository.

• Getting Started with PyTorch
• Build Your First Neural Network
• Transfer Learning for Image Classification using Torchvision
• Face Detection on Custom Dataset with Detectron2
• Time Series Forecasting with LSTMs for Daily Coronavirus Cases
• Time Series Anomaly Detection using LSTM Autoencoders
• Create Dataset for Sentiment Analysis by Scraping Google Play App Reviews
• Sentiment Analysis with BERT and Transformers by Hugging Face
• Deploy BERT for Sentiment Analysis as REST API using FastAPI
• Object Detection on Custom Dataset with YOLO (v5)

Consider buying the book if you want to support my work. Thanks for stopping by! 🤗

getting-things-done-with-pytorch's Issues

How to Export 'ONNX' Model?

Hi. Thanks to this code, I was able to write a multi-label classification model well.
By the way, can you tell me how to export the model made like this using torch.onnx? An error occurred when I used the normal torch.onnx.export method.

My code :

test_comment="hello"  
encoding = tokenizer.encode_plus(
    test_comment,
    add_special_tokens=True,
    max_length=63,
    return_token_type_ids=False,
    padding="max_length",
    return_attention_mask=True,
    return_tensors='pt',
  )



torch.onnx.export(trained_model,
                    (encoding["input_ids"], encoding["attention_mask"]),
                    'model.onnx',
                    export_params=True,
                    do_constant_folding=True,
                    opset_version=11,
                    input_names=['input_ids', 'attention_mask'],
                    output_names=['output'],
)

error :

RuntimeError                              Traceback (most recent call last)
<ipython-input-49-9c2e1e064898> in <module>
----> 1 torch.onnx.export(trained_model,
      2                     (encoding["input_ids"], encoding["attention_mask"]),
      3                     'model.onnx',
      4                     export_params=True,
      5                     do_constant_folding=True,

~/anaconda3/envs/myenv1/lib/python3.8/site-packages/torch/onnx/__init__.py in export(model, args, f, export_params, verbose, training, input_names, output_names, aten, export_raw_ir, operator_export_type, opset_version, _retain_param_name, do_constant_folding, example_outputs, strip_doc_string, dynamic_axes, keep_initializers_as_inputs, custom_opsets, enable_onnx_checker, use_external_data_format)
    273 
    274     from torch.onnx import utils
--> 275     return utils.export(model, args, f, export_params, verbose, training,
    276                         input_names, output_names, aten, export_raw_ir,
    277                         operator_export_type, opset_version, _retain_param_name,

~/anaconda3/envs/myenv1/lib/python3.8/site-packages/torch/onnx/utils.py in export(model, args, f, export_params, verbose, training, input_names, output_names, aten, export_raw_ir, operator_export_type, opset_version, _retain_param_name, do_constant_folding, example_outputs, strip_doc_string, dynamic_axes, keep_initializers_as_inputs, custom_opsets, enable_onnx_checker, use_external_data_format)
     86         else:
     87             operator_export_type = OperatorExportTypes.ONNX
---> 88     _export(model, args, f, export_params, verbose, training, input_names, output_names,
     89             operator_export_type=operator_export_type, opset_version=opset_version,
     90             _retain_param_name=_retain_param_name, do_constant_folding=do_constant_folding,

~/anaconda3/envs/myenv1/lib/python3.8/site-packages/torch/onnx/utils.py in _export(model, args, f, export_params, verbose, training, input_names, output_names, operator_export_type, export_type, example_outputs, opset_version, _retain_param_name, do_constant_folding, strip_doc_string, dynamic_axes, keep_initializers_as_inputs, fixed_batch_size, custom_opsets, add_node_names, enable_onnx_checker, use_external_data_format, onnx_shape_inference)
    687 
...
    128             wrapper,
    129             in_vars + module_state,

RuntimeError: output 1 (0
[ CPULongType{} ]) of traced region did not have observable data dependence with trace inputs; this probably indicates your program cannot be understood by the tracer.

Thank you in advance for your reply.

How to handle imbalanced classes ?

Hi, I have a use case wherein there are two classes and are hugely imbalanced. How can i fix this issue.

I have used the source code.

Sentiment Analysis with BERT

DataLoader problem

How to display class name of detected object

Regarding the tutorial you made on face detection using Detectron2, i want to output the bounding box on detected objects with the class name, instead of displaying the scores only. I would be glad if you could explain how to achieve that.

04.first-neural-network.ipynb error

I am getting the following error for 04.first-neural-network.ipynb notebook:

--------------------------------------------------
RuntimeError     Traceback (most recent call last)
<ipython-input-28-8756d5fbee9e> in <module>
     10 
     11     if epoch % 100 == 0:
---> 12       train_acc = calculate_accuracy(y_train, y_pred)
     13 
     14       y_test_pred = net(X_test)

<ipython-input-24-51ba3ab94870> in calculate_accuracy(y_true, y_pred)
      1 def calculate_accuracy(y_true, y_pred):
      2   predicted = y_pred.ge(.5).view(-1)
----> 3   return (y_true == predicted).sum().float() / len(y_true)

RuntimeError: Expected object of scalar type Float but got scalar type Byte for argument #2 'other'

Problem printing the shape of last_hidden_state and pooled_output

Change this line -> bert_model = BertModel.from_pretrained(PRE_TRAINED_MODEL_NAME) to
bert_model = BertModel.from_pretrained(PRE_TRAINED_MODEL_NAME, return_dict=False)

as a result the dict will not return str rather will return tensor.

It will help you with other parts of code as well.

Image files not available anymore

Hi, the dataturks images mentioned in
https://colab.research.google.com/github/curiousily/Getting-Things-Done-with-Pytorch/blob/master/02.face-detection-with-detectron2.ipynb#scrollTo=QEX1UwGV33h4
are not available anymore.
Ex:
http://com.dataturks.a96-i23.open.s3.amazonaws.com/2c9fafb064277d86016431e33e4e003d/8186c3d1-e9d4-4550-8ec1-a062a7628787___0-26.jpg.jpeg

arff2pandas is not supported anymore. Could anyone find alternative for this library?

CUDA out of memory. Tried to allocate 384.00 MiB (GPU 0; 15.90 GiB total capacity; 14.69 GiB already allocated; 291.75 MiB free; 14.76 GiB reserved in total by PyTorch)

any fixes for this?

Please upload code for youtube video

Can you please upload code for time series classification video: https://www.youtube.com/watch?v=PCgrgHgy26c

please share notebook

Hi, can you please share notebook from this your video? https://www.youtube.com/watch?v=ODEGJ_kh2aA
Thank you!

Multi class time series classification

Hello,

Thanks for the good examples in the repo.
I need to classify some time series (similar to what you did in script 06-ecg classification) but I have more than just 2 classes.
I can't seem to make it work. Do you have any input on how it should be done?

Thanks!

ann_viz() not defined in text

AttributeError: 'dict' object has no attribute 'dataset'

I am getting this error while writing the following class code. The error message is attached below the code.

class GPReviewDataset(data.dataset):

def init(self, review, target, tokenizer, max_len):
self.review = review
self.target = target
self.tokenizer = tokenizer
self.max_len = max_len

def len(self):
# return the number of reviews we have
return len(self.reviews)

def getitem(self, item): # takes the index and the reviews
review = str(self.reviews[item])

encoding = self.tokenizer.encode_plus(
  review,
  add_special_tokens=True,
  max_length=self.max_len,
  return_token_type_ids=False,
  pad_to_max_length=True,
  return_attention_mask=True,
  return_tensors='pt'
)

return {
  'review_text': review,
  'input_ids': encoding['input_ids'].flatten(),
  'attention_mask': encoding['attention_mask'].flatten(),
  'targets': torch.tensor(self.target, dtype=torch.long)
}

How to train binary classification

I use sentiment analysis with bert, however it is multiclass classification, how to change for binary class text classification.

LSTM autoencoder vs ANN autoencoder

Hi,
This question is related to notebook 6 on ECG anomaly detection.

I would like to know if there is any study on how much is the accuracy improvement on using LSTM autoencoder versus ANN encoder. Let me know if you have any info.

Regards, Debapriya

weather data permission

!gdown --id 1Q1wUptbNDYdfizk5abhmoFxIQiX19Tn7

Permission denied: https://drive.google.com/uc?id=1Q1wUptbNDYdfizk5abhmoFxIQiX19Tn7
Maybe you need to change permission over 'Anyone with the link'?

Slightly different kind of labels for the input ( Multi-label Text Classification with BERT and PyTorch Lightning)

Great work you did in the Multi-label Text Classification! Thanks!!
I have a similar problem as you except that II only have one column for the labels (for example, you have [1,0,0,0,0,0] as label of the toxic classe but with my data I have only one value as the number of the class. For example with the first class I have 1 in the column and the number goes to 10). This causes a problem in the training (trainer.fit), the code keep telling me that the target size (torch.Size([16]) which is the epoch number) is different from the input size (which is torch.Size([16, 10]) and the 10 here is the number of the classes) Can you please tell me where can I make the changes so the code will run?

Missing Notebook on T5 Fine-tuning

Hi @curiousily I have found your good video about the Fine-Tuning on the T5 model (https://www.youtube.com/watch?v=r6XY80Z9eSA).
Can you please upload the notebook used?

Thank you for your works,
Andrea

Error: ModelCheckpoint(monitor='val_loss') not found in the returned metrics

Hi,
When running the notebook 11.multi-label-text-classification-with-bert.ipynb
I encounter the following error...
pytorch_lightning.utilities.exceptions.MisconfigurationException: ModelCheckpoint(monitor='val_loss') not found in the returned metrics: ['train_loss']. HINT: Did you call self.log('val_loss', tensor) in the LightningModule?

Do you know how I could fix this?

8. sentiment-analysis-with-bert

Hi there,
I was following along with your video guide on this project but with my own dataset.
As I started training for my data I ran into an error.

RuntimeError : stack expects each tensor to be equal size

Our codes are basically identical and the structure of data seems to be identical as well. I'm not sure what's causing this issue or how to resolve it.

More in-depth error:

RuntimeError                              Traceback (most recent call last)
<timed exec> in <module>

<ipython-input-26-8ba1e19dd195> in train_epoch(model, data_loader, loss_fn, optimizer, device, scheduler, n_examples)
      4     correct_predictions = 0
      5 
----> 6     for i in data_loader:
      7         input_ids = i['input_ids'].to(device)
      8         attention_mask = i['attention_mask'].to(device)

~\Anaconda3\lib\site-packages\torch\utils\data\dataloader.py in __next__(self)
    361 
    362     def __next__(self):
--> 363         data = self._next_data()
    364         self._num_yielded += 1
    365         if self._dataset_kind == _DatasetKind.Iterable and \

~\Anaconda3\lib\site-packages\torch\utils\data\dataloader.py in _next_data(self)
    401     def _next_data(self):
    402         index = self._next_index()  # may raise StopIteration
--> 403         data = self._dataset_fetcher.fetch(index)  # may raise StopIteration
    404         if self._pin_memory:
    405             data = _utils.pin_memory.pin_memory(data)

~\Anaconda3\lib\site-packages\torch\utils\data\_utils\fetch.py in fetch(self, possibly_batched_index)
     45         else:
     46             data = self.dataset[possibly_batched_index]
---> 47         return self.collate_fn(data)

~\Anaconda3\lib\site-packages\torch\utils\data\_utils\collate.py in default_collate(batch)
     72         return batch
     73     elif isinstance(elem, container_abcs.Mapping):
---> 74         return {key: default_collate([d[key] for d in batch]) for key in elem}
     75     elif isinstance(elem, tuple) and hasattr(elem, '_fields'):  # namedtuple
     76         return elem_type(*(default_collate(samples) for samples in zip(*batch)))

~\Anaconda3\lib\site-packages\torch\utils\data\_utils\collate.py in <dictcomp>(.0)
     72         return batch
     73     elif isinstance(elem, container_abcs.Mapping):
---> 74         return {key: default_collate([d[key] for d in batch]) for key in elem}
     75     elif isinstance(elem, tuple) and hasattr(elem, '_fields'):  # namedtuple
     76         return elem_type(*(default_collate(samples) for samples in zip(*batch)))

~\Anaconda3\lib\site-packages\torch\utils\data\_utils\collate.py in default_collate(batch)
     53             storage = elem.storage()._new_shared(numel)
     54             out = elem.new(storage)
---> 55         return torch.stack(batch, 0, out=out)
     56     elif elem_type.__module__ == 'numpy' and elem_type.__name__ != 'str_' \
     57             and elem_type.__name__ != 'string_':

RuntimeError: stack expects each tensor to be equal size, but got [160] at entry 0 and [161] at entry 5

ann_viz() does not use the parameter view