@@ -4,7 +4,6 @@ scikit-learn
|
||||
seqeval
|
||||
psutil
|
||||
sacrebleu >= 1.4.12
|
||||
git+https://github.com/huggingface/accelerate@main#egg=accelerate
|
||||
rouge-score
|
||||
tensorflow_datasets
|
||||
matplotlib
|
||||
|
||||
@@ -14,8 +14,7 @@
|
||||
# See the License for the specific language governing permissions and
|
||||
# limitations under the License.
|
||||
"""
|
||||
Fine-tuning a 🤗 Transformers model on token classification tasks (NER, POS, CHUNKS) relying on the accelerate library
|
||||
without using a Trainer.
|
||||
Fine-tuning a 🤗 Transformers model on token classification tasks (NER, POS, CHUNKS)
|
||||
"""
|
||||
|
||||
import json
|
||||
|
||||
Reference in New Issue
Block a user