docs: Add steps to reproduce

Document steps to reproduce, and correct the wrong filename for the dataset.

Author: jo-elimu

README.md

@@ -23,3 +23,55 @@ We present REBUS: a collection of 333 hand-crafted rebuses spanning 13 diverse c
 | CogVLM            | 0.9           | 1.6           | 0.0           | 0.0          |
 | QWEN              | 0.9           | 1.6           | 0.0           | 0.0          |
 | InstructBLIP      | 0.6           | 0.5           | 0.9           | 0.0          |
+
+## Steps to Reproduce
+
+1. Install [Python](https://www.python.org).
+
+1. Install the [package installer for Python](https://pypi.org/project/pip/) (pip).
+
+1. Install the imported packages:
+
+   - torch
+  
+        `pip install torch`
+   
+   - requests
+
+        `pip install requests`
+
+   - PIL
+
+        `pip install Pillow`
+
+   - transformers
+
+        `pip install transformers`
+
+   - openai
+  
+        `pip install openai`
+
+   - matplotlib
+
+        `pip install matplotlib`
+
+   - tiktoken
+
+        `pip install tiktoken`
+
+   - transformers_stream_generator
+
+        `pip install transformers_stream_generator`
+
+   - einops
+
+        `pip install einops`
+
+   - torchvision
+
+        `pip install tochvision`
+
+1. Run the scripts in `eval_scripts/`, for example `blip.py`:
+
+   - `python ./eval_scripts/blip.py`

eval_scripts/blip.py

@@ -7,11 +7,11 @@
 from transformers import Blip2Processor, Blip2ForConditionalGeneration
 
 
-# 1. Load datasaet
+# 1. Load dataset
 
 rows = []
 
-with open('dataset.csv', mode='r', encoding='utf-8') as file:
+with open('data.csv', mode='r', encoding='utf-8') as file:
     csv_reader = csv.reader(file)
 
     headers = next(csv_reader)

eval_scripts/coglvm.py

@@ -5,11 +5,11 @@
 from PIL import Image
 from transformers import LlamaTokenizer, AutoModelForCausalLM
 
-# 1. Load datasaet
+# 1. Load dataset
 
 rows = []
 
-with open('dataset.csv', mode='r', encoding='utf-8') as file:
+with open('data.csv', mode='r', encoding='utf-8') as file:
     csv_reader = csv.reader(file)
 
     headers = next(csv_reader)

eval_scripts/gemini.py

@@ -8,7 +8,7 @@
 
 rows = []
 
-with open('dataset.csv', mode='r', encoding='utf-8') as file:
+with open('data.csv', mode='r', encoding='utf-8') as file:
     csv_reader = csv.reader(file)
 
     headers = next(csv_reader)

eval_scripts/gpt-4v.py

@@ -6,7 +6,7 @@
 
 rows = []
 
-with open('dataset.csv', mode='r', encoding='utf-8') as file:
+with open('data.csv', mode='r', encoding='utf-8') as file:
     csv_reader = csv.reader(file)
 
     headers = next(csv_reader)

eval_scripts/instructblip.py

@@ -7,11 +7,11 @@
 import requests
 
 
-# 1. Load datasaet
+# 1. Load dataset
 
 rows = []
 
-with open('dataset.csv', mode='r', encoding='utf-8') as file:
+with open('data.csv', mode='r', encoding='utf-8') as file:
     csv_reader = csv.reader(file)
 
     headers = next(csv_reader)

eval_scripts/llava_13b.py

@@ -6,11 +6,11 @@
 import torch
 from transformers import pipeline
 
-# 1. Load datasaet
+# 1. Load dataset
 
 rows = []
 
-with open('dataset.csv', mode='r', encoding='utf-8') as file:
+with open('data.csv', mode='r', encoding='utf-8') as file:
     csv_reader = csv.reader(file)
 
     headers = next(csv_reader)

eval_scripts/llava_7b.py

@@ -6,11 +6,11 @@
 import torch
 from transformers import pipeline
 
-# 1. Load datasaet
+# 1. Load dataset
 
 rows = []
 
-with open('dataset.csv', mode='r', encoding='utf-8') as file:
+with open('data.csv', mode='r', encoding='utf-8') as file:
     csv_reader = csv.reader(file)
 
     headers = next(csv_reader)

eval_scripts/qwen.py

@@ -3,11 +3,11 @@
 
 from transformers import AutoModelForCausalLM, AutoTokenizer
 
-# 1. Load datasaet
+# 1. Load dataset
 
 rows = []
 
-with open('dataset.csv', mode='r', encoding='utf-8') as file:
+with open('data.csv', mode='r', encoding='utf-8') as file:
     csv_reader = csv.reader(file)
 
     headers = next(csv_reader)