Figure 5
From: Learning deep features for dead and living breast cancer cell classification without staining
Visualization of the Inception-v3 learned feature space for our No treatment testing dataset. The 2048-dimension features were projected to a 2D space using t-SNE, and colored according to ground truth labels (a), and predicted labels (b). Cells with the same state tend to cluster together. Visual inspection of the images in each cluster further reveals the shared characteristics within each group. All images were rendered using Seaborn 0.11.1 https://seaborn.pydata.org/ and Matplotlib 3.3.3 https://matplotlib.org/.
