Spaces:
Runtime error
Runtime error
polinaeterna
commited on
Commit
·
a3b4e99
1
Parent(s):
b858233
fix
Browse files
app.py
CHANGED
|
@@ -75,9 +75,9 @@ def plot_and_df(texts, preds):
|
|
| 75 |
# counts.reset_index(inplace=True)
|
| 76 |
return (
|
| 77 |
gr.BarPlot(counts_df, x="quality", y="count", sort=None),
|
| 78 |
-
texts_df[texts_df["quality"] == "Low"][["text"]][:20],
|
| 79 |
-
texts_df[texts_df["quality"] == "Medium"][["text"]][:20],
|
| 80 |
-
texts_df[texts_df["quality"] == "High"][["text"]][:20],
|
| 81 |
)
|
| 82 |
|
| 83 |
|
|
@@ -99,6 +99,7 @@ def run_quality_check(dataset, config, split, column, batch_size, num_examples):
|
|
| 99 |
return
|
| 100 |
|
| 101 |
try:
|
|
|
|
| 102 |
data = pl.read_parquet(f"hf://datasets/{dataset}@~parquet/{filename}", columns=[column])
|
| 103 |
except Exception as error:
|
| 104 |
yield f"❌ {error}", gr.BarPlot(), pd.DataFrame(), pd.DataFrame(), pd.DataFrame(), pd.DataFrame(),
|
|
|
|
| 75 |
# counts.reset_index(inplace=True)
|
| 76 |
return (
|
| 77 |
gr.BarPlot(counts_df, x="quality", y="count", sort=None),
|
| 78 |
+
texts_df[texts_df["quality"] == "Low"][["text"]][:min(texts_df.shape[0], 20)],
|
| 79 |
+
texts_df[texts_df["quality"] == "Medium"][["text"]][:min(texts_df.shape[0], 20)],
|
| 80 |
+
texts_df[texts_df["quality"] == "High"][["text"]][:min(texts_df.shape[0], 20)],
|
| 81 |
)
|
| 82 |
|
| 83 |
|
|
|
|
| 99 |
return
|
| 100 |
|
| 101 |
try:
|
| 102 |
+
logging.info(f"Loading hf://datasets/{dataset}@~parquet/{filename}")
|
| 103 |
data = pl.read_parquet(f"hf://datasets/{dataset}@~parquet/{filename}", columns=[column])
|
| 104 |
except Exception as error:
|
| 105 |
yield f"❌ {error}", gr.BarPlot(), pd.DataFrame(), pd.DataFrame(), pd.DataFrame(), pd.DataFrame(),
|