active-vision 0.0.1__py3-none-any.whl → 0.0.2__py3-none-any.whl

active_vision/__init__.py

@@ -1 +1,3 @@
-__version__ = "0.0.1"
+__version__ = "0.0.2"
+
+from .core import *

active_vision/core.py

@@ -0,0 +1,149 @@
+import pandas as pd
+from loguru import logger
+from fastai.vision.models import resnet18, resnet34
+from fastai.callback.all import ShowGraphCallback
+from fastai.vision.all import (
+    ImageDataLoaders,
+    aug_transforms,
+    Resize,
+    vision_learner,
+    accuracy,
+    valley,
+    slide,
+    minimum,
+    steep,
+)
+import torch
+import torch.nn.functional as F
+
+import warnings
+
+warnings.filterwarnings("ignore", category=FutureWarning)
+
+
+class ActiveLearner:
+    def __init__(self, model_name: str):
+        self.model = self.load_model(model_name)
+
+    def load_model(self, model_name: str):
+        models = {"resnet18": resnet18, "resnet34": resnet34}
+        logger.info(f"Loading model {model_name}")
+        if model_name not in models:
+            logger.error(f"Model {model_name} not found")
+            raise ValueError(f"Model {model_name} not found")
+        return models[model_name]
+
+    def load_dataset(
+        self,
+        df: pd.DataFrame,
+        filepath_col: str,
+        label_col: str,
+        valid_pct: float = 0.2,
+        batch_size: int = 16,
+        image_size: int = 224,
+    ):
+        logger.info(f"Loading dataset from {filepath_col} and {label_col}")
+        self.train_set = df.copy()
+
+        logger.info("Creating dataloaders")
+        self.dls = ImageDataLoaders.from_df(
+            df,
+            path=".",
+            valid_pct=valid_pct,
+            fn_col=filepath_col,
+            label_col=label_col,
+            bs=batch_size,
+            item_tfms=Resize(image_size),
+            batch_tfms=aug_transforms(size=image_size, min_scale=0.75),
+        )
+        logger.info("Creating learner")
+        self.learn = vision_learner(self.dls, self.model, metrics=accuracy).to_fp16()
+        self.class_names = self.dls.vocab
+        logger.info("Done. Ready to train.")
+
+    def lr_find(self):
+        logger.info("Finding optimal learning rate")
+        self.lrs = self.learn.lr_find(suggest_funcs=(minimum, steep, valley, slide))
+        logger.info(f"Optimal learning rate: {self.lrs.valley}")
+
+    def train(self, epochs: int, lr: float):
+        logger.info(f"Training for {epochs} epochs with learning rate: {lr}")
+        self.learn.fine_tune(epochs, lr, cbs=[ShowGraphCallback()])
+
+    def predict(self, filepaths: list[str], batch_size: int = 16):
+        """
+        Run inference on an unlabeled dataset. Returns a df with filepaths and predicted labels, and confidence scores.
+        """
+        logger.info(f"Running inference on {len(filepaths)} samples")
+        test_dl = self.dls.test_dl(filepaths, bs=batch_size)
+        preds, _, cls_preds = self.learn.get_preds(dl=test_dl, with_decoded=True)
+
+        self.pred_df = pd.DataFrame(
+            {
+                "filepath": filepaths,
+                "pred_label": [self.learn.dls.vocab[i] for i in cls_preds.numpy()],
+                "pred_conf": torch.max(F.softmax(preds, dim=1), dim=1)[0].numpy(),
+            }
+        )
+        return self.pred_df
+
+    def evaluate(self, df: pd.DataFrame, filepath_col: str, label_col: str, batch_size: int = 16):
+        """
+        Evaluate on a labeled dataset. Returns a score.
+        """
+        self.eval_set = df.copy()
+
+        filepaths = self.eval_set[filepath_col].tolist()
+        labels = self.eval_set[label_col].tolist()
+        test_dl = self.dls.test_dl(filepaths, bs=batch_size)
+        preds, _, cls_preds = self.learn.get_preds(dl=test_dl, with_decoded=True)
+
+        self.eval_df = pd.DataFrame(
+            {
+                "filepath": filepaths,
+                "label": labels,
+                "pred_label": [self.learn.dls.vocab[i] for i in cls_preds.numpy()],
+            }
+        )
+
+        accuracy = float((self.eval_df["label"] == self.eval_df["pred_label"]).mean())
+        logger.info(f"Accuracy: {accuracy:.2%}")
+        return accuracy
+
+    def sample_uncertain(self, df: pd.DataFrame, num_samples: int):
+        """
+        Sample top `num_samples` low confidence samples. Returns a df with filepaths and predicted labels, and confidence scores.
+        """
+        uncertain_df = df.sort_values(
+            by="pred_conf", ascending=True
+        ).head(num_samples)
+        return uncertain_df
+
+    def add_to_train_set(self, df: pd.DataFrame):
+        """
+        Add samples to the training set.
+        """
+        new_train_set = df.copy()
+        new_train_set.drop(columns=["pred_conf"], inplace=True)
+        new_train_set.rename(columns={"pred_label": "label"}, inplace=True)
+
+        len_old = len(self.train_set)
+
+        logger.info(f"Adding {len(new_train_set)} samples to training set")
+        self.train_set = pd.concat([self.train_set, new_train_set])
+
+        self.train_set = self.train_set.drop_duplicates(
+            subset=["filepath"], keep="last"
+        )
+        self.train_set.reset_index(drop=True, inplace=True)
+
+
+        if len(self.train_set) == len_old:
+            logger.warning("No new samples added to training set")
+
+        elif len_old + len(new_train_set) < len(self.train_set):
+            logger.warning("Some samples were duplicates and removed from training set")
+
+        else:
+            logger.info("All new samples added to training set")
+            logger.info(f"Training set now has {len(self.train_set)} samples")

{active_vision-0.0.1.dist-info → active_vision-0.0.2.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.2
 Name: active-vision
-Version: 0.0.1
+Version: 0.0.2
 Summary: Active learning for edge vision.
 Requires-Python: >=3.10
 Description-Content-Type: text/markdown
@@ -14,9 +14,11 @@ Requires-Dist: seaborn>=0.13.2
 
 ![Python Version](https://img.shields.io/badge/python-3.10%2B-blue?style=for-the-badge)
 ![License](https://img.shields.io/badge/License-Apache%202.0-green.svg?style=for-the-badge)
+![PyPI](https://img.shields.io/pypi/v/active-vision?style=for-the-badge)
+![Downloads](https://img.shields.io/pepy/dt/active-vision?style=for-the-badge&logo=pypi&logoColor=white&label=Downloads&color=purple)   
 
 <p align="center">
-  <img src="./assets/logo.png" alt="active-vision">
+  <img src="https://github.com/dnth/active-vision/blob/main/assets/logo.png" alt="active-vision">
 </p>
 
 Active learning at the edge for computer vision.
@@ -44,43 +46,37 @@ cd active-vision
 pip install -e .
 ```
 
-## Usage [WIP]
+## Usage
+See the [notebook](./nbs/end-to-end.ipynb) for a complete example.
 
 ```python
-import active_vision as av
+from active_vision import ActiveLearner
+import pandas as pd
 
-# Load a model
-model = av.load_model("resnet18")
+# Create an active learner instance with a model
+al = ActiveLearner("resnet18")
 
-# Load a dataset
-dataset = av.load_dataset(df)
-
-# Inital sampling
-dataset = av.initial_sampling(dataset, n_samples=10)
+# Load the dataset into the active learner
+train_df = pd.read_parquet("training_samples.parquet")
+al.load_dataset(train_df, "filepath", "label")
 
 # Train the model
-model.train()
-
-# Save the model
-model.save()
-
-# Evaluate the model
-model.evaluate(df)
+al.train(epochs=3, lr=1e-3)
 
-# Uncertainty sampling to get the lowest confidence images
-model.uncertainty_sampling()
+# Load evaluation data
+eval_df = pd.read_parquet("evaluation_samples.parquet")
 
-# Diversity sampling to get the most diverse images (outliers)
-model.diversity_sampling()
+# Evaluate the model on a labeled evaluation set
+accuracy = al.evaluate(eval_df, "filepath", "label")
 
-# Random sampling
-model.random_sampling()
+# Get predictions from an unlabeled set
+pred_df = al.predict(filepaths)
 
-# Merge the datasets
-dataset = av.merge_datasets(dataset, dataset_2)
+# Sample low confidence predictions
+uncertain_df = al.sample_uncertain(pred_df, num_samples=10)
 
-# Launch a streamlit app to label the images
-av.label_images(dataset)
+# Add newly labeled data to training set
+al.add_to_train_set(uncertain_df)
 ```
 
 ## Workflow

active_vision-0.0.2.dist-info/RECORD

@@ -0,0 +1,7 @@
+active_vision/__init__.py,sha256=5VE_DRQ_Rgbo7NlPh3-rP2pUClK48jGxPqAcptBscZ8,43
+active_vision/core.py,sha256=RBVabC350wucYl7KJgIp3fc1pS9pxtG14iDb-ZyBJxI,5262
+active_vision-0.0.2.dist-info/LICENSE,sha256=xx0jnfkXJvxRnG63LTGOxlggYnIysveWIZ6H3PNdCrQ,11357
+active_vision-0.0.2.dist-info/METADATA,sha256=7_eqZJnGeIPjb4LLZ-Bqu1AMJ_h77_0bNRyS_COEv5w,8350
+active_vision-0.0.2.dist-info/WHEEL,sha256=In9FTNxeP60KnTkGw7wk6mJPYd_dQSjEZmXdBdMCI-8,91
+active_vision-0.0.2.dist-info/top_level.txt,sha256=7qUQvccN2UU63z5S9vrgJmqK-8sFGrtpf1e9Z86nihE,14
+active_vision-0.0.2.dist-info/RECORD,,

active_vision-0.0.1.dist-info/RECORD

@@ -1,6 +0,0 @@
-active_vision/__init__.py,sha256=sXLh7g3KC4QCFxcZGBTpG2scR7hmmBsMjq6LqRptkRg,22
-active_vision-0.0.1.dist-info/LICENSE,sha256=xx0jnfkXJvxRnG63LTGOxlggYnIysveWIZ6H3PNdCrQ,11357
-active_vision-0.0.1.dist-info/METADATA,sha256=lPOTTVSPAaX3Rn9Q1ci_jgoQOC-HFpQIyTNqrouOYEs,7936
-active_vision-0.0.1.dist-info/WHEEL,sha256=In9FTNxeP60KnTkGw7wk6mJPYd_dQSjEZmXdBdMCI-8,91
-active_vision-0.0.1.dist-info/top_level.txt,sha256=7qUQvccN2UU63z5S9vrgJmqK-8sFGrtpf1e9Z86nihE,14
-active_vision-0.0.1.dist-info/RECORD,,