Merge pull request #519 from microsoft/PreRelease

Classification module debugging

.gitignore

@@ -10,4 +10,10 @@ PytorchWildlife.egg-info/
 *setup/*
 *dist/*
 *VSCodeCounter*
-*build/*
+*build/*
+*bee*
+*Bee*
+*debug*
+*annotations.csv
+*cropped_resized*
+*log*

PW_FT_classification/README.md

@@ -88,8 +88,9 @@ Before training your model, you need to configure the training and data paramete
   - `num_workers`: Number of subprocesses to use for data loading.
 
 - **Model Parameters:**
+  - `num_classes`: The number of classes. 
   - `model_name`: The name of the model architecture to use. The current version only supports PlainResNetClassifier.
-  - `num_layers`: Number of layers in the model. Currently only supports 18 and 50.
+  - `num_layers`: Number of layers in the resnet model. Currently only supports 18 and 50.
   - `weights_init`: Initial weights setting for the model. Currently only supports "ImageNet".
 
 - **Optimization Parameters:**

PW_FT_classification/configs/Raw/Crop_res50_plain_082723.yaml → PW_FT_classification/configs/Raw/Crop_res18_plain_071824.yaml

@@ -1,8 +1,8 @@
 # training
-conf_id: Crop_Res50_plain_082723
+conf_id: Crop_Res18_plain_071824
 algorithm: Plain
 log_dir: Crop
-num_epochs: 60
+num_epochs: 30
 log_interval: 10
 parallel: 0
 
@@ -18,11 +18,12 @@ val_size: 0.2
 split_data: True
 split_type: location # options are: random, location, sequence
 # data loading
-batch_size: 256
-num_workers: 0 #40
+batch_size: 32
+num_workers: 4 #40
 # model
+num_classes: 2 
 model_name: PlainResNetClassifier
-num_layers: 50
+num_layers: 18
 weights_init: ImageNet
 
 # optim
@@ -35,6 +36,6 @@ lr_classifier: 0.01
 momentum_classifier: 0.9
 weight_decay_classifier: 0.0005
 ## lr_scheduler
-step_size: 20
+step_size: 10
 gamma: 0.1
 

PW_FT_classification/main.py

@@ -16,9 +16,10 @@
 # %%
 from src.utils import batch_detection_cropping
 from src.utils import data_splitting
-#app = typer.Typer()
+
+app = typer.Typer(pretty_exceptions_short=True, pretty_exceptions_show_locals=False)
 # %%
-#@app.command()
+@app.command()
 def main(
         config:str='./configs/Raw/Crop_res50_plain_082723.yaml',
         project:str='Custom-classification',
@@ -34,7 +35,7 @@ def main(
         predict_root:str=""
     ):
     """
-    Main function for training or evaluating a ResNet-50 model using PyTorch Lightning.
+    Main function for training or evaluating a ResNet model (50 or 18) using PyTorch Lightning.
     It loads configurations, initializes the model, logger, and other components based on provided arguments.
 
     Args:
@@ -56,7 +57,7 @@ def main(
     gpus = gpus if torch.cuda.is_available() else None
     gpus = [int(i) for i in gpus.split(',')]
 
-    # Environment variable setup for numpy multi-threading
+    # Environment variable setup for numpy multi-threading. It is important to avoid cpu and ram issues.
     os.environ["OMP_NUM_THREADS"] = str(np_threads)
     os.environ["OPENBLAS_NUM_THREADS"] = str(np_threads)
     os.environ["MKL_NUM_THREADS"] = str(np_threads)
@@ -73,7 +74,6 @@ def main(
     # Set a global seed for reproducibility
     pl.seed_everything(seed)
 
-
     # If the annotation directory does not have a data split, split the data first
     if conf.split_data:
         # Replace annotation dir from config with the directory containing the split files
@@ -92,19 +92,18 @@ def main(
     train_annotations = os.path.join(conf.dataset_root, 'train_annotations.csv')
     test_annotations = os.path.join(conf.dataset_root, 'test_annotations.csv')
     val_annotations = os.path.join(conf.dataset_root, 'val_annotations.csv')
-    # Split training data
-
+    # Crop training data
     batch_detection_cropping.batch_detection_cropping(conf.dataset_root, os.path.join(conf.dataset_root, "cropped_resized"), train_annotations)
-    # Split validation and test data
+    # Crop validation data
     batch_detection_cropping.batch_detection_cropping(conf.dataset_root, os.path.join(conf.dataset_root, "cropped_resized"), val_annotations)
+    # Crop test data (most likely we don't need this)
     batch_detection_cropping.batch_detection_cropping(conf.dataset_root, os.path.join(conf.dataset_root, "cropped_resized"), test_annotations)
 
-
     # Dataset and algorithm loading based on the configuration
     dataset = datasets.__dict__[conf.dataset_name](conf=conf)
     learner = algorithms.__dict__[conf.algorithm](conf=conf,
-                            train_class_counts=dataset.train_class_counts, 
-                            id_to_labels=dataset.id_to_labels)
+                                                  train_class_counts=dataset.train_class_counts, 
+                                                  id_to_labels=dataset.id_to_labels)
 
     # Logger setup based on the specified logger type
     log_folder = 'log_dev' if dev else 'log'
@@ -155,7 +154,7 @@ def main(
         devices=gpus,
         logger=None if evaluate is not None else logger,
         callbacks=[lr_monitor, checkpoint_callback],
-        strategy='ddp',
+        strategy='auto',
         num_sanity_val_steps=0,
         profiler=None
     )
@@ -171,8 +170,6 @@ def main(
         trainer.fit(learner, datamodule=dataset)
 # %%
 if __name__ == '__main__':
-    main()
-
-
+    app()
 
 # %%

PW_FT_classification/requirements.txt

@@ -1,125 +1,5 @@
-absl-py==2.1.0
-aiofiles==23.2.1
-aiohttp==3.9.3
-aiosignal==1.3.1
-altair==5.2.0
-annotated-types==0.6.0
-anyio==4.2.0
-asttokens==2.4.1
-async-timeout==4.0.3
-attrs==23.2.0
-backcall==0.2.0
-cachetools==5.3.2
-certifi==2023.11.17
-charset-normalizer==3.3.2
-click==8.1.7
-colorama==0.4.6
-contourpy==1.1.1
-cycler==0.12.1
-decorator==5.1.1
-exceptiongroup==1.2.0
-executing==2.0.1
-fastapi==0.109.0
-ffmpy==0.3.1
-filelock==3.13.1
-fire==0.5.0
-fonttools==4.47.2
-frozenlist==1.4.1
-fsspec==2023.12.2
-google-auth==2.27.0
-google-auth-oauthlib==1.0.0
-gradio
-grpcio==1.60.0
-h11==0.14.0
-httpcore==1.0.2
-httpx==0.26.0
-huggingface-hub==0.20.3
-idna==3.6
-importlib-metadata==7.0.1
-importlib-resources==6.1.1
-ipython==8.12.3
-jedi==0.19.1
-jinja2==3.1.3
-joblib==1.3.2
-jsonschema==4.21.1
-jsonschema-specifications==2023.12.1
-kiwisolver==1.4.5
-lightning-utilities==0.10.1
-markdown==3.5.2
-markdown-it-py==3.0.0
-markupsafe==2.1.4
-matplotlib==3.7.4
-matplotlib-inline==0.1.6
-mdurl==0.1.2
-multidict==6.0.4
-munch==2.5.0
-numpy==1.24.4
-oauthlib==3.2.2
-opencv-python==4.9.0.80
-opencv-python-headless==4.9.0.80
-orjson==3.9.12
-packaging==23.2
-pandas==2.0.3
-parso==0.8.3
-pexpect==4.9.0
-pickleshare==0.7.5
-pillow==10.1.0
-pkgutil-resolve-name==1.3.10
-prompt-toolkit==3.0.43
-protobuf==3.20.1
-psutil==5.9.8
-ptyprocess==0.7.0
-pure-eval==0.2.2
-pyasn1==0.5.1
-pyasn1-modules==0.3.0
-pydantic==2.6.0
-pydantic-core==2.16.1
-pydub==0.25.1
-pygments==2.17.2
-pyparsing==3.1.1
-python-dateutil==2.8.2
-python-multipart==0.0.6
-pytorch-lightning==1.9.0
-pytorchwildlife
-pytz==2023.4
-pyyaml==6.0.1
-referencing==0.33.0
-requests==2.31.0
-requests-oauthlib==1.3.1
-rich==13.7.0
-rpds-py==0.17.1
-rsa==4.9
-scikit-learn==1.2.0
-scipy==1.10.1
-seaborn==0.13.2
-semantic-version==2.10.0
-shellingham==1.5.4
-six==1.16.0
-sniffio==1.3.0
-stack-data==0.6.3
-starlette==0.35.1
-supervision==0.16.0
-tensorboard==2.14.0
-tensorboard-data-server==0.7.2
-termcolor==2.4.0
-thop==0.1.1-2209072238
-threadpoolctl==3.2.0
-tomlkit==0.12.0
-toolz==0.12.1
-torch==1.10.1
-torchaudio==0.10.1
-torchmetrics==1.3.0.post0
-torchvision==0.11.2
-tqdm==4.66.1
-traitlets==5.14.1
-typer==0.9.0
-typing-extensions==4.9.0
-tzdata==2023.4
-ultralytics-yolov5==0.1.1
-urllib3==2.2.0
-uvicorn==0.27.0.post1
-wcwidth==0.2.13
-websockets==11.0.3
-werkzeug==3.0.1
-yarl==1.9.4
-zipp==3.17.0
+PytorchWildlife
+scikit_learn
+lightning
+munch
+typer

PW_FT_classification/src/algorithms/plain.py

@@ -41,7 +41,7 @@ def __init__(self, conf, train_class_counts, id_to_labels, **kwargs):
         self.save_hyperparameters(ignore=['conf', 'train_class_counts'])
         self.train_class_counts = train_class_counts
         self.id_to_labels = id_to_labels
-        self.net = models.__dict__[self.hparams.model_name](num_cls=1, 
+        self.net = models.__dict__[self.hparams.model_name](num_cls=self.hparams.num_classes, 
                                                             num_layers=self.hparams.num_layers)
 
     def configure_optimizers(self):
@@ -93,9 +93,8 @@ def training_step(self, batch, batch_idx):
         # Forward pass
         feats = self.net.feature(data)
         logits = self.net.classifier(feats)
-        logits = logits.squeeze(1)
         # Calculate loss
-        loss = self.net.criterion_cls(logits, label_ids.float())
+        loss = self.net.criterion_cls(logits, label_ids)
         self.log("train_loss", loss)
 
         return loss
@@ -117,8 +116,8 @@ def validation_step(self, batch, batch_idx):
         data, label_ids = batch[0], batch[1]
         # Forward pass
         feats = self.net.feature(data)
-        logits = self.net.classifier(feats).squeeze(1)
-        preds = logits>0.5
+        logits = self.net.classifier(feats)
+        preds = logits.argmax(dim=1)
 
         self.val_st_outs.append((preds.detach().cpu().numpy(),
                                  label_ids.detach().cpu().numpy()))
@@ -148,8 +147,8 @@ def test_step(self, batch, batch_idx):
         data, label_ids, labels, file_ids = batch
         # Forward pass
         feats = self.net.feature(data)
-        logits = torch.sigmoid(self.net.classifier(feats))
-        preds = logits>0.5
+        logits = self.net.classifier(feats)
+        preds = logits.argmax(dim=1)
 
         self.te_st_outs.append((preds.detach().cpu().numpy(),
                                label_ids.detach().cpu().numpy(),
@@ -198,8 +197,8 @@ def predict_step(self, batch, batch_idx):
         data, file_ids = batch
         # Forward pass
         feats = self.net.feature(data)
-        logits = torch.sigmoid(self.net.classifier(feats))
-        preds = logits>0.5
+        logits = self.net.classifier(feats)
+        preds = logits.argmax(dim=1)
 
         self.pr_st_outs.append((preds.detach().cpu().numpy(),
                                feats.detach().cpu().numpy(),

PW_FT_classification/src/datasets/custom.py

@@ -122,6 +122,7 @@ def __getitem__(self, index):
 
         return sample, label_id, label, file_dir
 
+
 class Custom_Crop_DS(Custom_Base_DS):
     """
     Dataset class for handling custom cropped datasets.
@@ -146,6 +147,7 @@ def __init__(self, rootdir, dset='train', transform=None):
                                                 .format('test' if dset == 'test' else dset)))
         self.load_data()
 
+
 class Custom_Base(pl.LightningDataModule):
     """
     Base data module for handling custom datasets in PyTorch Lightning.
@@ -213,6 +215,7 @@ def predict_dataloader(self):
         """
         return DataLoader(self.dset_pr, batch_size=64, shuffle=False, pin_memory=True, num_workers=self.conf.num_workers, drop_last=False)
 
+
 class Custom_Crop(Custom_Base):
     """
     Custom data module specifically for cropped datasets in PyTorch Lightning.

PW_FT_classification/src/models/plain_resnet.py

@@ -144,7 +144,7 @@ def setup_criteria(self):
         Set up the criterion for the classifier.
         """
         # Criterion for binary classification
-        self.criterion_cls = nn.BCEWithLogitsLoss()
+        self.criterion_cls = nn.CrossEntropyLoss()
 
     def feat_init(self):
         """
@@ -167,4 +167,3 @@ def feat_init(self):
         unused_keys = load_keys - self_keys
         print('missing keys: {}'.format(sorted(list(missing_keys))))
         print('unused_keys: {}'.format(sorted(list(unused_keys))))
-        pass

PW_FT_classification/src/utils/data_splitting.py

@@ -50,9 +50,6 @@ def create_splits(csv_path, output_folder, test_size=0.2, val_size=0.1):
     # Return the dataframes
     return train_set, val_set, test_set
 
-import pandas as pd
-from sklearn.model_selection import train_test_split
-
 def split_by_location(csv_path, output_folder, val_size=0.15, test_size=0.15, random_state=None):
     """
     Splits the dataset into train, validation, and test sets based on location, ensuring that:
@@ -95,10 +92,6 @@ def split_by_location(csv_path, output_folder, val_size=0.15, test_size=0.15, ra
     return train_data, val_data, test_data
 
 
-
-import pandas as pd
-from sklearn.model_selection import train_test_split
-
 def split_by_seq(csv_path, output_folder, val_size=0.15, test_size=0.15, random_state=None):
     """
     Splits the dataset into train, validation, and test sets based on sequence ID, ensuring that: