diff options
author | Vasil Zlatanov <v@skozl.com> | 2019-02-04 16:52:54 +0000 |
---|---|---|
committer | Vasil Zlatanov <v@skozl.com> | 2019-02-04 16:52:54 +0000 |
commit | 3d55b96e9c5a3abffe4662381b50251f96137805 (patch) | |
tree | 3cc4040f3c79a54b094bd1fc9288d7a0b207dca4 | |
parent | 82d295c4693e17b71c52632740f4f98e141f8d49 (diff) | |
download | e4-vision-3d55b96e9c5a3abffe4662381b50251f96137805.tar.gz e4-vision-3d55b96e9c5a3abffe4662381b50251f96137805.tar.bz2 e4-vision-3d55b96e9c5a3abffe4662381b50251f96137805.zip |
Revise code to match exported format
-rw-r--r-- | evaluate.py | 22 |
1 files changed, 14 insertions, 8 deletions
diff --git a/evaluate.py b/evaluate.py index 6504ec1..309326e 100644 --- a/evaluate.py +++ b/evaluate.py @@ -1,6 +1,7 @@ # EE4 Selected Topics From Computer Vision Coursework # Vasil Zlatanov, Nunzio Pucci +DATA_FILE = 'data.npz' CLUSTER_CNT = 1337 KMEAN_PART = 33 @@ -9,16 +10,21 @@ import matplotlib.pyplot as plt from sklearn.cluster import KMeans -train = [] -test = [] +data = np.load(DATA_FILE) -train_part = np.hstack(train[0:KMEAN_PART]) +train = data['train'] + +# Train part will contain 15 000 descriptors to generate KMeans +part_idx = np.random.random_integers(train.shape[1]) +train_part = np.vstack(train[:][part_idx][300:1300]) kmeans = KMeans(n_clusters=CLUSTER_CNT, random_state=0).fit(train_part) -codewords = [] -i = 0 -for image in train: - codewords.append(np.bincount(kmeans.predict(image) - print codewords[i].shape +histogram = np.zeros((train.shape[0], train.shape[1],CLUSTER_CNT)) + +for i in range(train.shape[0]) + for j in range(train.shape[1]) + histogram[i][j] = np.bincount(kmeans.predict(train[i][j]) + +print(histogram.shape) |