muhammadravi251001
commited on
Commit
•
b84e1be
1
Parent(s):
69bd761
Update README.md
Browse files
README.md
CHANGED
@@ -32,4 +32,72 @@ test_dataset = Dataset.from_dict(df_test)
|
|
32 |
|
33 |
dataset = DatasetDict({"train": train_dataset, "validation": validation_dataset, "test": test_dataset})
|
34 |
dataset
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
35 |
```
|
|
|
32 |
|
33 |
dataset = DatasetDict({"train": train_dataset, "validation": validation_dataset, "test": test_dataset})
|
34 |
dataset
|
35 |
+
```
|
36 |
+
|
37 |
+
If you want to download keep-invalid-data-dataset:
|
38 |
+
```
|
39 |
+
from datasets import load_dataset, Dataset, DatasetDict
|
40 |
+
import pandas as pd
|
41 |
+
|
42 |
+
data_files = {"train": "data_nli_train_df_keep.csv",
|
43 |
+
"validation": "data_nli_val_df_keep.csv",
|
44 |
+
"test": "data_nli_test_df_keep.csv"}
|
45 |
+
|
46 |
+
dataset = load_dataset("muhammadravi251001/debug-entailment", data_files=data_files)
|
47 |
+
|
48 |
+
# selected_columns = ["premise", "hypothesis", "label"]
|
49 |
+
selected_columns = dataset.column_names['train'] # Uncomment this line to retrieve all of the columns
|
50 |
+
|
51 |
+
df_train = pd.DataFrame(dataset["train"])
|
52 |
+
df_train = df_train[selected_columns]
|
53 |
+
|
54 |
+
df_val = pd.DataFrame(dataset["validation"])
|
55 |
+
df_val = df_val[selected_columns]
|
56 |
+
|
57 |
+
df_test = pd.DataFrame(dataset["test"])
|
58 |
+
df_test = df_test[selected_columns]
|
59 |
+
|
60 |
+
train_dataset = Dataset.from_dict(df_train)
|
61 |
+
validation_dataset = Dataset.from_dict(df_val)
|
62 |
+
test_dataset = Dataset.from_dict(df_test)
|
63 |
+
|
64 |
+
dataset = DatasetDict({"train": train_dataset, "validation": validation_dataset, "test": test_dataset})
|
65 |
+
|
66 |
+
data_qas_train_df = pd.DataFrame(dataset["train"])
|
67 |
+
data_qas_val_df = pd.DataFrame(dataset["validation"])
|
68 |
+
data_qas_test_df = pd.DataFrame(dataset["test"])
|
69 |
+
```
|
70 |
+
|
71 |
+
If you want to download drop-invalid-data-dataset:
|
72 |
+
```
|
73 |
+
from datasets import load_dataset, Dataset, DatasetDict
|
74 |
+
import pandas as pd
|
75 |
+
|
76 |
+
data_files = {"train": "data_nli_train_df_drop.csv",
|
77 |
+
"validation": "data_nli_val_df_drop.csv",
|
78 |
+
"test": "data_nli_test_df_drop.csv"}
|
79 |
+
|
80 |
+
dataset = load_dataset("muhammadravi251001/debug-entailment", data_files=data_files)
|
81 |
+
|
82 |
+
# selected_columns = ["premise", "hypothesis", "label"]
|
83 |
+
selected_columns = dataset.column_names['train'] # Uncomment this line to retrieve all of the columns
|
84 |
+
|
85 |
+
df_train = pd.DataFrame(dataset["train"])
|
86 |
+
df_train = df_train[selected_columns]
|
87 |
+
|
88 |
+
df_val = pd.DataFrame(dataset["validation"])
|
89 |
+
df_val = df_val[selected_columns]
|
90 |
+
|
91 |
+
df_test = pd.DataFrame(dataset["test"])
|
92 |
+
df_test = df_test[selected_columns]
|
93 |
+
|
94 |
+
train_dataset = Dataset.from_dict(df_train)
|
95 |
+
validation_dataset = Dataset.from_dict(df_val)
|
96 |
+
test_dataset = Dataset.from_dict(df_test)
|
97 |
+
|
98 |
+
dataset = DatasetDict({"train": train_dataset, "validation": validation_dataset, "test": test_dataset})
|
99 |
+
|
100 |
+
data_qas_train_df = pd.DataFrame(dataset["train"])
|
101 |
+
data_qas_val_df = pd.DataFrame(dataset["validation"])
|
102 |
+
data_qas_test_df = pd.DataFrame(dataset["test"])
|
103 |
```
|