Create files.py
Browse files
files.py
ADDED
|
@@ -0,0 +1,28 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
from transformers import AutoTokenizer, AutoModelForSequenceClassification
|
| 2 |
+
|
| 3 |
+
model_name = "distilbert-base-uncased"
|
| 4 |
+
|
| 5 |
+
tokenizer = AutoTokenizer.from_pretrained(model_name)
|
| 6 |
+
model = AutoModelForSequenceClassification.from_pretrained(
|
| 7 |
+
model_name,
|
| 8 |
+
num_labels=6,
|
| 9 |
+
id2label={
|
| 10 |
+
0: "Australopithecus",
|
| 11 |
+
1: "Homo habilis",
|
| 12 |
+
2: "Homo erectus",
|
| 13 |
+
3: "Homo heidelbergensis",
|
| 14 |
+
4: "Homo neanderthalensis",
|
| 15 |
+
5: "Homo sapiens"
|
| 16 |
+
},
|
| 17 |
+
label2id={
|
| 18 |
+
"Australopithecus": 0,
|
| 19 |
+
"Homo habilis": 1,
|
| 20 |
+
"Homo erectus": 2,
|
| 21 |
+
"Homo heidelbergensis": 3,
|
| 22 |
+
"Homo neanderthalensis": 4,
|
| 23 |
+
"Homo sapiens": 5
|
| 24 |
+
}
|
| 25 |
+
)
|
| 26 |
+
|
| 27 |
+
model.save_pretrained("human-evolution-classifier")
|
| 28 |
+
tokenizer.save_pretrained("human-evolution-classifier")
|