formatting

README.md

@@ -251,65 +251,46 @@ model-index:
 
 - create the summarizer object:
 
-```
+```python
 from transformers import AutoModelForSeq2SeqLM, AutoTokenizer
 from transformers import pipeline
 
-_model = AutoModelForSeq2SeqLM.from_pretrained(
-                "pszemraj/bigbird-pegasus-large-K-booksum",
-                low_cpu_mem_usage=True,
-            )
+model = AutoModelForSeq2SeqLM.from_pretrained(
+    "pszemraj/bigbird-pegasus-large-K-booksum",
+    low_cpu_mem_usage=True,
+)
+
+tokenizer = AutoTokenizer.from_pretrained(
+    "pszemraj/bigbird-pegasus-large-K-booksum",
+)
 
-_tokenizer = AutoTokenizer.from_pretrained(
-                "pszemraj/bigbird-pegasus-large-K-booksum",
-            )
-                                           
 
 summarizer = pipeline(
-                    "summarization", 
-                    model=_model, 
-                    tokenizer=_tokenizer
-                )
-             
+    "summarization",
+    model=model,
+    tokenizer=tokenizer,
+)          
  ```
 
 - define text to be summarized, and pass it through the pipeline. Boom done. 
 
-```
+```python
 wall_of_text = "your text to be summarized goes here."
 
 result = summarizer(
-           wall_of_text,
-           min_length=16, 
-           max_length=256,
-           no_repeat_ngram_size=3, 
-           clean_up_tokenization_spaces=True,
-    )
+    wall_of_text,
+    min_length=16,
+    max_length=256,
+    no_repeat_ngram_size=3,
+    clean_up_tokenization_spaces=True,
+)
 
-print(result[0]['summary_text'])
+print(result[0]["summary_text"])
 ```
 
 ## Alternate Checkpoint
 
 - if experiencing runtime/memory issues, try [this earlier checkpoint](https://huggingface.co/pszemraj/bigbird-pegasus-large-booksum-40k-K) at 40,000 steps which is almost as good at the explanatory summarization task but runs faster.
+- see similar summarization models fine-tuned on booksum but using different architectures: [long-t5 base](https://huggingface.co/pszemraj/long-t5-tglobal-base-16384-book-summary) and [LED-Large](https://huggingface.co/pszemraj/led-large-book-summary)
 
 ---
-
-# Results
-
-- note that while the dataset has three subsets (chapter, book, paragraph) - see the [paper](https://arxiv.org/abs/2105.08209). the scores below are run in aggregate. The paper has some benchmark scores listed, which this model competes with.
-- note that eval generations are run & computed at a length of 128 tokens.
-
-
-```
-'eval_gen_len': 126.9791,
- 'eval_loss': 4.00944709777832,
- 'eval_rouge1': 27.6028,
- 'eval_rouge2': 4.6556,
- 'eval_rougeL': 14.5259,
- 'eval_rougeLsum': 25.6632,
- 'eval_runtime': 29847.4812,
- 'eval_samples_per_second': 0.05,
- 'eval_steps_per_second': 0.05}
-
-```