add performance graph to readme

README.md

@@ -13,6 +13,10 @@ To generate a 4x4 grid of DALL·E Mega images it takes:
 - 48 sec with a P100 in Colab
 - 13 sec with an A100 on Replicate
 
+Here's a more detailed breakdown on the A100 of total inference time vs number of generated images:
+<img src="https://github.com/kuprel/min-dalle/raw/main/performance_a100.png" alt="min-dalle" width="400"/>
+Credit to @technobird22 and his [NeoGen](https://github.com/technobird22/NeoGen) discord bot for the graph
+
 The flax model and code for converting it to torch can be found [here](https://github.com/kuprel/min-dalle-flax).
 
 ## Install

min_dalle/min_dalle.py

@@ -186,7 +186,7 @@ class MinDalle:
         tokens = self.tokenizer.tokenize(text, is_verbose=is_verbose)
         if len(tokens) > self.text_token_count: 
             tokens = tokens[:self.text_token_count]
-        if is_verbose: print("text tokens", tokens)
+        if is_verbose: print("{} text tokens".format(len(tokens)), tokens)
         text_tokens = numpy.ones((2, 64), dtype=numpy.int32)
         text_tokens[0, :2] = [tokens[0], tokens[-1]]
         text_tokens[1, :len(tokens)] = tokens