julien-c HF staff commited on
Commit
a5d869e
1 Parent(s): 7b65715

Migrate model card from transformers-repo

Browse files

Read announcement at https://discuss.huggingface.co/t/announcement-all-model-cards-will-be-migrated-to-hf-co-model-repos/2755
Original file history: https://github.com/huggingface/transformers/commits/master/model_cards/lvwerra/gpt2-imdb/README.md

Files changed (1) hide show
  1. README.md +27 -0
README.md ADDED
@@ -0,0 +1,27 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # GPT2-IMDB
2
+
3
+ ## What is it?
4
+ A GPT2 (`gpt2`) language model fine-tuned on the [IMDB dataset](https://www.kaggle.com/lakshmi25npathi/imdb-dataset-of-50k-movie-reviews).
5
+
6
+ ## Training setting
7
+
8
+ The GPT2 language model was fine-tuned for 1 epoch on the IMDB dataset. All comments were joined into a single text file separated by the EOS token:
9
+
10
+ ```
11
+ import pandas as pd
12
+ df = pd.read_csv("imdb-dataset.csv")
13
+ imdb_str = " <|endoftext|> ".join(df['review'].tolist())
14
+
15
+ with open ('imdb.txt', 'w') as f:
16
+ f.write(imdb_str)
17
+ ```
18
+
19
+ To train the model the `run_language_modeling.py` script in the `transformer` library was used:
20
+
21
+ ```
22
+ python run_language_modeling.py
23
+ --train_data_file imdb.txt
24
+ --output_dir gpt2-imdb
25
+ --model_type gpt2
26
+ --model_name_or_path gpt2
27
+ ```