feat: update

.gitignore

@@ -165,4 +165,9 @@ data/raw/*
 data/processed/*
 
 !data/raw/.gitkeep
-!data/processed/.gitkeep
+!data/processed/.gitkeep
+
+uvenv/
+runs/
+wandb/
+ubuntu-venv/

.vscode/launch.json

@@ -0,0 +1,16 @@
+{
+  // Use IntelliSense to learn about possible attributes.
+  // Hover to view descriptions of existing attributes.
+  // For more information, visit: https://go.microsoft.com/fwlink/?linkid=830387
+  "version": "0.2.0",
+  "configurations": [
+    {
+      "name": "Python: Current File",
+      "type": "python",
+      "request": "launch",
+      "program": "${file}",
+      "console": "integratedTerminal",
+      "justMyCode": false
+    }
+  ]
+}

.vscode/settings.json

@@ -0,0 +1,10 @@
+{
+  "[python]": {
+    "editor.defaultFormatter": "ms-python.autopep8",
+
+  },
+  "jupyter.debugJustMyCode": false,
+  "debug.allowBreakpointsEverywhere": true,
+  
+  "python.formatting.provider": "none"
+}

Makefile

@@ -1,2 +1,4 @@
 package:
-	pip freeze > requirements.txt
+	pip freeze > requirements.txt
+venv:
+	source /mnt/d/ubuntu/env/mlenv/bin/activate

main.py

@@ -1,5 +1,10 @@
-from src.train import train
-
+from src.train import train_runner
+from src.auto import auto_hyper_parameter
+import os
+# set WANDB_API_KEY=$YOUR_API_KEY
+# os.environ["WANDB_API_KEY"] = '7c0f2b9470a0a5c82bfae5bab4705344cb53288b'
+# os.environ['WANDB_MODE'] = "offline"
 if __name__ == "__main__":
-    train()
-
+    print("Training the model...")
+    # train_runner()
+    auto_hyper_parameter()

model.pth

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:e72cf086b03927eeb2527f1e94fe3fbcdda64d8749746a308108f73bb47d9455
-size 33560199

requirements.txt

@@ -1,68 +1,252 @@
+absl-py==2.0.0
 aiofiles==23.2.1
-altair==5.1.1
-annotated-types==0.5.0
-anyio==3.7.1
+anyio==4.0.0
+anylabeling==0.3.3
+appdirs==1.4.4
+argcomplete==3.1.2
+asgiref==3.7.2
+asttokens==2.4.1
+attr==0.3.1
 attrs==23.1.0
+azure-core==1.29.4
+azure-storage-blob==12.18.3
+beautifulsoup4==4.12.2
+bleach==5.0.1
+boto==2.49.0
+boto3==1.16.28
+botocore==1.19.28
+boxing==0.1.4
+Brotli==1.1.0
+cachetools==5.3.1
 certifi==2022.12.7
+cffi==1.16.0
 charset-normalizer==2.1.1
 click==8.1.7
 colorama==0.4.6
+coloredlogs==15.0.1
+comm==0.2.0
 contourpy==1.1.1
-cycler==0.12.0
+cryptography==41.0.4
+cycler==0.12.1
+dacite==1.7.0
+darkdetect==0.8.0
+debugpy==1.8.0
+decorator==5.1.1
+defusedxml==0.7.1
+Deprecated==1.2.14
+dill==0.3.7
+Django==3.2.20
+django-annoying==0.10.6
+django-cors-headers==3.6.0
+django-debug-toolbar==3.2.1
+django-environ==0.10.0
+django-extensions==3.1.0
+django-filter==2.4.0
+django-model-utils==4.1.1
+django-ranged-fileresponse==0.1.2
+django-rq==2.5.1
+django-storages==1.12.3
+django-user-agents==0.4.0
+djangorestframework==3.13.1
+dnspython==2.4.2
+docker-pycreds==0.4.0
+drf-dynamic-fields==0.3.0
+drf-flex-fields==0.9.5
+drf-generators==0.3.0
 exceptiongroup==1.1.3
-fastapi==0.103.2
-ffmpy==0.3.1
+executing==2.0.1
+expiringdict==1.2.2
+fiftyone==0.22.1
+fiftyone-brain==0.13.2
+fiftyone-db==0.4.0
+fiftyone-desktop==0.29.0
 filelock==3.9.0
-fonttools==4.43.0
-fsspec==2023.9.2
-gradio==3.45.2
-gradio_client==0.5.3
+flatbuffers==23.5.26
+fonttools==4.43.1
+fsspec==2023.4.0
+ftfy==6.1.1
+future==0.18.3
+gitdb==4.0.11
+GitPython==3.1.40
+glob2==0.7
+google-api-core==2.11.0
+google-auth==2.23.4
+google-auth-oauthlib==1.1.0
+google-cloud-appengine-logging==1.1.0
+google-cloud-audit-log==0.2.0
+google-cloud-core==2.3.2
+google-cloud-logging==2.7.2
+google-cloud-storage==2.5.0
+google-crc32c==1.5.0
+google-resumable-media==2.3.3
+googleapis-common-protos==1.56.4
+graphql-core==3.2.3
+grpc-google-iam-v1==0.12.4
+grpcio==1.59.0
+grpcio-status==1.48.2
 h11==0.14.0
+h2==4.1.0
+hpack==4.0.0
+htmlmin==0.1.12
 httpcore==0.18.0
 httpx==0.25.0
-huggingface-hub==0.17.3
+humanfriendly==10.0
+humansignal-drf-yasg==1.21.9
+hypercorn==0.14.4
+hyperframe==6.0.1
 idna==3.4
-importlib-resources==6.1.0
+ijson==3.2.3
+imageio==2.31.5
+imgviz==1.7.4
+inflate64==0.3.1
+inflection==0.5.1
+ipykernel==6.26.0
+ipython==8.17.2
+isodate==0.6.1
+jedi==0.19.1
 Jinja2==3.1.2
-jsonschema==4.19.1
-jsonschema-specifications==2023.7.1
+jmespath==0.10.0
+joblib==1.3.2
+jsonlines==4.0.0
+jsonpatch==1.33
+jsonpointer==2.4
+jsonschema==3.2.0
+jupyter_client==8.6.0
+jupyter_core==5.5.0
+kaleido==0.2.1
 kiwisolver==1.4.5
+label-studio==1.9.1.post0
+label-studio-converter==0.0.57
+label-studio-tools==0.0.3
+launchdarkly-server-sdk==7.5.0
+lazy_loader==0.3
+lockfile==0.12.2
+lxml==4.9.3
+Markdown==3.5.1
 MarkupSafe==2.1.2
 matplotlib==3.8.0
+matplotlib-inline==0.1.6
+mongoengine==0.24.2
+motor==3.3.1
 mpmath==1.3.0
+multivolumefile==0.2.3
+natsort==8.4.0
+nest-asyncio==1.5.8
 networkx==3.0
-numpy==1.24.1
-orjson==3.9.7
-packaging==23.1
+nltk==3.6.7
+numpy==1.24.3
+oauthlib==3.2.2
+onnx==1.13.1
+onnxruntime==1.14.1
+opencv-python==4.8.1.78
+opencv-python-headless==4.8.1.78
+ordered-set==4.0.2
+packaging==23.2
 pandas==2.1.1
-Pillow==9.3.0
-pydantic==2.4.2
-pydantic_core==2.10.1
-pydub==0.25.1
+parso==0.8.3
+pathtools==0.1.2
+Pillow==10.0.1
+platformdirs==4.0.0
+plotly==5.17.0
+pprintpp==0.4.0
+priority==2.0.0
+prompt-toolkit==3.0.40
+proto-plus==1.22.3
+protobuf==3.20.3
+psutil==5.9.5
+psycopg2-binary==2.9.6
+pure-eval==0.2.2
+py-cpuinfo==9.0.0
+py7zr==0.20.6
+pyasn1==0.5.0
+pyasn1-modules==0.3.0
+pybcj==1.0.1
+pycparser==2.21
+pycryptodomex==3.19.0
+pydantic==1.10.13
+Pygments==2.16.1
+pymongo==4.5.0
 pyparsing==3.1.1
+pyppmd==1.0.0
+PyQt5==5.15.10
+PyQt5-Qt5==5.15.2
+PyQt5-sip==12.13.0
+pyreadline3==3.4.1
+pyRFC3339==1.1
+pyrsistent==0.19.3
 python-dateutil==2.8.2
-python-multipart==0.0.6
-pytz==2023.3.post1
+python-json-logger==2.0.4
+pytz==2022.7.1
+pywin32==306
 PyYAML==6.0.1
-referencing==0.30.2
-requests==2.28.1
-rpds-py==0.10.3
-semantic-version==2.10.0
+pyzmq==25.1.1
+pyzstd==0.15.9
+qimage2ndarray==1.10.0
+rarfile==4.1
+redis==3.5.3
+regex==2023.10.3
+requests==2.31.0
+requests-oauthlib==1.3.1
+retrying==1.3.4
+rq==1.10.1
+rsa==4.9
+rules==2.2
+s3transfer==0.3.7
+scikit-image==0.22.0
+scikit-learn==1.3.1
+scipy==1.11.3
+seaborn==0.13.0
+semver==2.13.0
+sentry-sdk==1.32.0
+setproctitle==1.3.3
+simplejson==3.19.2
 six==1.16.0
+smmap==5.0.1
 sniffio==1.3.0
-starlette==0.27.0
+sortedcontainers==2.4.0
+soupsieve==2.5
+sqlparse==0.4.4
+sse-starlette==0.10.3
+sseclient-py==1.8.0
+stack-data==0.6.3
+starlette==0.31.1
+strawberry-graphql==0.138.1
 sympy==1.12
-toolz==0.12.0
-# torch==2.0.1+cu117
-# torchaudio==2.0.2+cu117
-# torchvision==0.15.2+cu117
+tabulate==0.9.0
+tenacity==8.2.3
+tensorboard==2.15.1
+tensorboard-data-server==0.7.2
+termcolor==2.3.0
+texttable==1.7.0
+thop==0.1.1.post2209072238
+threadpoolctl==3.2.0
+tifffile==2023.9.26
+tomli==2.0.1
+torch==2.1.0+cu118
+torchaudio==2.1.0+cu118
+torchsummary==1.5.1
+torchvision==0.16.0+cu118
+tornado==6.3.3
 tqdm==4.66.1
+traitlets==5.13.0
 typing_extensions==4.8.0
 tzdata==2023.3
-urllib3==1.26.13
-uvicorn==0.23.2
-websockets==11.0.3
-
-torch==2.0.1
-torchaudio==2.0.2
-torchvision==0.15.2
+tzlocal==5.1
+ua-parser==0.18.0
+ujson==5.8.0
+ultralytics==8.0.197
+universal-analytics-python3==1.1.1
+uritemplate==4.1.1
+urllib3==1.26.16
+user-agents==2.2.0
+visdom==0.2.4
+voxel51-eta==0.12.0
+wandb==0.16.0
+wcwidth==0.2.8
+webencodings==0.5.1
+websocket-client==1.6.4
+Werkzeug==3.0.1
+wrapt==1.15.0
+wsproto==1.2.0
+xmljson==0.2.0
+xmltodict==0.13.0

src/configs/model_config.py

@@ -5,6 +5,7 @@ class ModelConfig:
         self.learning_rate = 0.001
         self.batch_size = 32
         self.device = "cuda" if torch.cuda.is_available() else "cpu"
-        self.epochs = 20
+        self.epochs = 5
+        self.log_interval = 2 # Log every 2 batches => number of items is 32*2 = 64
     def get_config(self):
         return self

src/data/data_loader.py

@@ -7,7 +7,8 @@ import os
 num_classes = 3
 config = ModelConfig().get_config()
 
-train_dataset = CustomDataset(data_folder=os.path.join("data", 'raw'), transform=data_transform)
+train_dataset = CustomDataset(data_folder=os.path.join(
+    "data", 'raw'), transform=data_transform)
 
 # # Calculate the split point
 # split_index = int(0.8 * len(dataset))
@@ -17,5 +18,11 @@ train_dataset = CustomDataset(data_folder=os.path.join("data", 'raw'), transform
 # test_dataset = dataset[split_index:]
 
 
-train_loader = DataLoader(train_dataset, batch_size=config.batch_size, shuffle=True)
+train_loader = DataLoader(
+    train_dataset, batch_size=config.batch_size, shuffle=True)
+
+
+def get_train_dataset(batch_size):
+    return DataLoader(
+        train_dataset, batch_size=batch_size, shuffle=True)
 

src/models/model.py

@@ -1,17 +1,34 @@
 from torch import nn
 import torch.nn.functional as F
-
+# Ảnh gốc có kích thước 128x128x3
 class ShapeClassifier(nn.Module):
-    def __init__(self, num_classes):
+    def __init__(self, num_classes, hidden_size=128):
         super(ShapeClassifier, self).__init__()
-        self.conv1 = nn.Conv2d(in_channels=3, out_channels=16, kernel_size=3, padding=1)
-        self.pool = nn.MaxPool2d(kernel_size=2, stride=2)
-        self.fc1 = nn.Linear(16 * 64 * 64, 128)
-        self.fc2 = nn.Linear(128, num_classes)
+        # Layer 1: Convolutional layer with 3 input channels (RGB) and 16 output channels, using a 3x3 kernel and padding of 1
+        self.conv1 = nn.Conv2d(in_channels=3, out_channels=16, kernel_size=3, padding=1) # ra 128x128x16
+        
+        # Layer 2: Max pooling layer with a 2x2 kernel and stride of 2 to reduce spatial dimensions
+        self.pool = nn.MaxPool2d(kernel_size=2, stride=2) # ra 64x64x16
+        
+        # Layer 3: Fully connected layer with input size 16 * 64 * 64 (depends on the input image size) and output size 128
+        self.fc1 = nn.Linear(16 * 64 * 64, hidden_size)
+        
+        # Layer 4: Fully connected layer with input size 128 and output size num_classes
+        self.fc2 = nn.Linear(hidden_size, num_classes)
 
     def forward(self, x):
+        # Forward pass through the network
+        
+        # Apply convolution, activation function (ReLU), and max pooling
         x = self.pool(F.relu(self.conv1(x)))
-        x = x.view(-1, 16 * 64 * 64)  # Adjust the dimensions based on your input image size
+        
+        # Adjust the dimensions for the fully connected layer
+        x = x.view(-1, 16 * 64 * 64)
+        
+        # Apply activation function (ReLU) to the first fully connected layer
         x = F.relu(self.fc1(x))
+        
+        # Output layer without activation function (applied later during loss computation)
         x = self.fc2(x)
-        return x
+        
+        return x

src/train.py

@@ -1,51 +1,40 @@
 import torch
 import torch.optim as optim
 import torch.nn.functional as F
-from .models.model import ShapeClassifier
+from src.models.model import ShapeClassifier
 
 from src.configs.model_config import ModelConfig
 from src.data.data_loader import train_loader, num_classes
+from src.utils.logs import writer
+from src.utils.train import train
+from src.utils.test import test
+import wandb
+import json
+wandb.init(project="template-pytorch-model", entity="nguyen")
 
 
-def train():
+def train_runner():
+
     config = ModelConfig().get_config()
     device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
     model = ShapeClassifier(num_classes=num_classes).to(device)
     optimizer = optim.Adam(model.parameters(), lr=config.learning_rate)
     log_interval = 20
+    # log models config to wandb
+    wandb.config.update(config)
+
     for epoch in range(config.epochs):
-        model.train()
-        running_loss = 0.0
-
-        for batch_idx, (inputs, labels) in enumerate(train_loader):
-            inputs, labels = inputs.to(device), labels.to(device)
-            optimizer.zero_grad()
-
-            outputs = model(inputs)
-            loss = F.cross_entropy(outputs, labels)
-            loss.backward()
-            optimizer.step()
-
-            running_loss += loss.item()
-
-            if batch_idx % log_interval == 0:
-                current_loss = running_loss / log_interval
-                print(
-                    f"Epoch [{epoch + 1}/{config.epochs}], Batch [{batch_idx + 1}/{len(train_loader)}], Loss: {current_loss:.4f}")
-                running_loss = 0.0
-
-                # calculate the accuracy on the test set
-
-                with torch.no_grad():
-                    model.eval()
-                    correct = 0
-                    total = 0
-                    for inputs, labels in train_loader:
-                        inputs, labels = inputs.to(device), labels.to(device)
-                        outputs = model(inputs)
-                        predicted = torch.argmax(outputs.data, 1)
-                        total += labels.size(0)
-                        correct += (predicted == labels).sum().item()
-                    print(f"Accuracy of the model on the test images: {100 * correct / total} %")
-                # save the model
-                torch.save(model.state_dict(), "model.pth")
+        print(f"Epoch {epoch+1}\n-------------------------------")
+
+        loss = train(train_loader, model=model, loss_fn=F.cross_entropy,
+                     optimizer=optimizer)
+        test(train_loader, model=model, loss_fn=F.cross_entropy)
+        # 3. Log metrics over time to visualize performance
+        wandb.log({"loss": loss})
+
+        # save model
+        torch.save(model.state_dict(), "model.pth")
+
+        # 4. Log an artifact to W&B
+        wandb.log_artifact("model.pth")
+        # model.train()

src/utils/__init__.py

@@ -0,0 +1,17 @@
+import importlib
+import os
+from inspect import isclass
+
+# import all files under utils/
+utils_dir = os.path.dirname(__file__)
+for file in os.listdir(utils_dir):
+    path = os.path.join(utils_dir, file)
+    if not file.startswith("_") and not file.startswith(".") and (file.endswith(".py") or os.path.isdir(path)):
+        config_name = file[: file.find(".py")] if file.endswith(".py") else file
+        module = importlib.import_module("src.utils." + config_name)
+        for attribute_name in dir(module):
+            attribute = getattr(module, attribute_name)
+
+            if isclass(attribute):
+                # Add the class to this package's variables
+                globals()[attribute_name] = attribute

src/utils/logs.py

@@ -0,0 +1,2 @@
+from torch.utils.tensorboard import SummaryWriter
+writer = SummaryWriter()

src/utils/test.py

@@ -0,0 +1,18 @@
+from src.configs.model_config import ModelConfig
+import torch
+
+def test(dataloader, model, loss_fn):
+    config = ModelConfig().get_config()
+    size = len(dataloader.dataset)
+    num_batches = len(dataloader)
+    model.eval()
+    test_loss, correct = 0, 0
+    with torch.no_grad():
+        for X, y in dataloader:
+            X, y = X.to(config.device), y.to(config.device)
+            pred = model(X)
+            test_loss += loss_fn(pred, y).item()
+            correct += (pred.argmax(1) == y).type(torch.float).sum().item()
+    test_loss /= num_batches
+    correct /= size
+    print(f"Test Error: \n Accuracy: {(100*correct):>0.1f}%, Avg loss: {test_loss:>8f} \n")

src/utils/train.py

@@ -0,0 +1,23 @@
+from src.configs.model_config import ModelConfig
+import torch
+def train(dataloader, model, loss_fn, optimizer):
+    config = ModelConfig().get_config()
+    size = len(dataloader.dataset)
+    model.train()
+    for batch, (X, y) in enumerate(dataloader):
+        X, y = X.to(config.device), y.to(config.device)
+
+        # Compute prediction error
+        pred = model(X)
+        loss = loss_fn(pred, y)
+
+        # Backpropagation
+        loss.backward()
+        optimizer.step()
+        optimizer.zero_grad()
+
+        if batch % config.log_interval == 0:
+            loss, current = loss.item(), (batch + 1) * len(X)
+            print(f"loss: {loss:>7f}  [{current:>5d}/{size:>5d}]")
+    # return loss
+    return loss

src/utils/utils.py

@@ -0,0 +1 @@
+