Upload folder using huggingface_hub
Browse files- yolksac_human.py +3 -3
yolksac_human.py
CHANGED
|
@@ -32,7 +32,7 @@ class RNAExp(datasets.ArrowBasedBuilder):
|
|
| 32 |
self.batch = 1000
|
| 33 |
|
| 34 |
# create a dictionary of features where raw_counts are ints and the rest are strings
|
| 35 |
-
features = {"raw_counts": datasets.features.Sequence(datasets.features.Value("
|
| 36 |
for feature in FEATURES_TO_INCLUDE:
|
| 37 |
if not features.get(feature):
|
| 38 |
features[feature] = datasets.Value("string")
|
|
@@ -71,9 +71,9 @@ class RNAExp(datasets.ArrowBasedBuilder):
|
|
| 71 |
|
| 72 |
# raw counts
|
| 73 |
if RAW_COUNTS == "X":
|
| 74 |
-
chunk = adata.X[batch:batch+batch_size].tolil().astype('
|
| 75 |
elif RAW_COUNTS == "raw.X":
|
| 76 |
-
chunk = adata.raw.X[batch:batch+batch_size].tolil().astype('
|
| 77 |
else:
|
| 78 |
raise("Not valid raw_counts")
|
| 79 |
df = pd.DataFrame([chunk.data,chunk.rows]).T
|
|
|
|
| 32 |
self.batch = 1000
|
| 33 |
|
| 34 |
# create a dictionary of features where raw_counts are ints and the rest are strings
|
| 35 |
+
features = {"raw_counts": datasets.features.Sequence(datasets.features.Value("uint32")),"rows": datasets.features.Sequence(datasets.features.Value("uint32")),"size":datasets.Value("uint32")}
|
| 36 |
for feature in FEATURES_TO_INCLUDE:
|
| 37 |
if not features.get(feature):
|
| 38 |
features[feature] = datasets.Value("string")
|
|
|
|
| 71 |
|
| 72 |
# raw counts
|
| 73 |
if RAW_COUNTS == "X":
|
| 74 |
+
chunk = adata.X[batch:batch+batch_size].tolil().astype('uint32')
|
| 75 |
elif RAW_COUNTS == "raw.X":
|
| 76 |
+
chunk = adata.raw.X[batch:batch+batch_size].tolil().astype('uint32')
|
| 77 |
else:
|
| 78 |
raise("Not valid raw_counts")
|
| 79 |
df = pd.DataFrame([chunk.data,chunk.rows]).T
|