|
|
import json |
|
|
from copy import deepcopy |
|
|
import os |
|
|
from pathlib import Path |
|
|
import re |
|
|
from multiprocessing import Pool |
|
|
from PIL import Image |
|
|
from tqdm import tqdm |
|
|
import multiprocessing |
|
|
|
|
|
def resize_image(image, scale=0.75): |
|
|
""" |
|
|
Resize image to have its shorter edge equal to 720 pixels while maintaining aspect ratio. |
|
|
|
|
|
Args: |
|
|
image: PIL Image object |
|
|
|
|
|
Returns: |
|
|
Resized PIL Image |
|
|
""" |
|
|
|
|
|
width, height = image.size |
|
|
|
|
|
|
|
|
new_width = int(width * scale) |
|
|
new_height = int(height * scale) |
|
|
|
|
|
|
|
|
resized_image = image.resize((new_width, new_height), Image.LANCZOS) |
|
|
return resized_image |
|
|
|
|
|
def merge_convs(conversations): |
|
|
""" |
|
|
Merge all successive 'human' conversations comprehensively. |
|
|
|
|
|
Args: |
|
|
conversations (list): List of conversation dictionaries |
|
|
|
|
|
Returns: |
|
|
list: Processed conversations with all successive human messages merged |
|
|
|
|
|
Raises: |
|
|
ValueError: If input is not a list or contains invalid conversation dictionaries |
|
|
""" |
|
|
|
|
|
if not isinstance(conversations, list): |
|
|
raise ValueError("Input must be a list of conversation dictionaries") |
|
|
|
|
|
|
|
|
for conv in conversations: |
|
|
if not isinstance(conv, dict): |
|
|
raise ValueError("Each conversation must be a dictionary") |
|
|
if 'from' not in conv or 'value' not in conv: |
|
|
raise ValueError("Each conversation must have 'from' and 'value' keys") |
|
|
|
|
|
processed_conversations = [] |
|
|
i = 0 |
|
|
while i < len(conversations): |
|
|
current_conv = conversations[i] |
|
|
|
|
|
|
|
|
if current_conv['from'] == 'human': |
|
|
|
|
|
merged_value = current_conv['value'] |
|
|
j = i + 1 |
|
|
while j < len(conversations) and conversations[j]['from'] == 'human': |
|
|
merged_value += '\n\n' + conversations[j]['value'] |
|
|
j += 1 |
|
|
|
|
|
|
|
|
current_conv['value'] = merged_value |
|
|
|
|
|
|
|
|
i = j |
|
|
else: |
|
|
|
|
|
i += 1 |
|
|
|
|
|
processed_conversations.append(current_conv) |
|
|
|
|
|
return processed_conversations |
|
|
|
|
|
def parse_reasoning(input_string): |
|
|
input_string = input_string.strip() |
|
|
if not input_string.endswith("```"): |
|
|
input_string += "```" |
|
|
|
|
|
pattern = r'```([ABC])\n(.*?)```' |
|
|
|
|
|
|
|
|
matches = re.findall(pattern, input_string, re.DOTALL) |
|
|
|
|
|
|
|
|
parsed_texts = [] |
|
|
|
|
|
|
|
|
for _, text in matches: |
|
|
parsed_texts.append(text.strip()) |
|
|
|
|
|
if len(parsed_texts) != 3: |
|
|
|
|
|
return None, None, None |
|
|
|
|
|
caption, instruction, reasoning = parsed_texts |
|
|
|
|
|
return caption, instruction.replace("Task: ", ""), reasoning |
|
|
|
|
|
def encode_action(action_json): |
|
|
""" |
|
|
Encode different types of actions into human-readable descriptions. |
|
|
|
|
|
Args: |
|
|
action_json (dict): A dictionary containing action details |
|
|
|
|
|
Returns: |
|
|
str: A human-readable description of the action |
|
|
""" |
|
|
action_type = action_json.get("action", "") |
|
|
|
|
|
if action_type == "SWIPE": |
|
|
|
|
|
if len(action_json.get("info", [])) >= 2: |
|
|
start_y = action_json["touch_coord"][1] |
|
|
end_y = action_json["lift_coord"][1] |
|
|
|
|
|
|
|
|
if start_y > end_y: |
|
|
return "SCROLL UP" |
|
|
elif start_y < end_y: |
|
|
return "SCROLL DOWN" |
|
|
|
|
|
elif action_type == "TYPE": |
|
|
text_to_type = action_json.get("type_text", "") |
|
|
return f'TYPE "{text_to_type}"' |
|
|
|
|
|
elif action_type == "CLICK": |
|
|
|
|
|
if action_json.get("info") == "KEY_HOME": |
|
|
return "go to the home screen" |
|
|
elif action_json.get("info") == "KEY_BACK": |
|
|
return "go to the previous screen" |
|
|
elif action_json.get("info") == "KEY_RECENT": |
|
|
return "go to the previous App" |
|
|
|
|
|
|
|
|
return f"Perform {action_type} action" |
|
|
|
|
|
grounding_step_prompt = "<|img|>Step {step_idx}. Given a GUI image, what are the relative (0-1000) pixel point coordinates for the element corresponding to the following instruction or description: {instruction}" |
|
|
grounding_step_ans = "```\n{point_str}\n```" |
|
|
act_step_prompt = "<|img|>Step {step_idx}. Instruction: {prev_instruction}" |
|
|
act_step_ans = "The agent's action: {prev_action}" |
|
|
user_start_prompt = "The agent is performing the ultimate task: {ultimate_task}." |
|
|
user_history_instr_prompt = "History of the agent's steps:\n{history_list}." |
|
|
|
|
|
resize_ratios_per_window_size = { |
|
|
1: 0.25, |
|
|
2: 0.25, |
|
|
3: 0.25, |
|
|
} |
|
|
|
|
|
def process_android_episodes(data, window_size=2, img_dir="./AMEX/screenshot"): |
|
|
""" |
|
|
Process Android episodes and extract steps with click or long_press actions. |
|
|
|
|
|
Args: |
|
|
data (list): List of episode dictionaries |
|
|
window_size (int, optional): Number of recent image-included conversations to include. |
|
|
Defaults to 3 (current image + 2 previous image-included steps). |
|
|
|
|
|
Returns: |
|
|
dict: Dictionary with episode_id as key and list of filtered steps as value |
|
|
""" |
|
|
instructions = [] |
|
|
for episode in data: |
|
|
episode_id = episode["episode_id"] |
|
|
|
|
|
for i, step in enumerate(episode["steps"]): |
|
|
is_grounding = step["is_grounding"] |
|
|
|
|
|
if not is_grounding: |
|
|
continue |
|
|
|
|
|
if window_size > 0 and i == 0: |
|
|
continue |
|
|
|
|
|
convs = [ |
|
|
{ |
|
|
"from": "human", |
|
|
"value": user_start_prompt.format( |
|
|
ultimate_task=episode["instruction"] |
|
|
), |
|
|
}, |
|
|
] |
|
|
|
|
|
cur_img_list = [Path(img_dir) / Path(step["image_path"]).name] |
|
|
|
|
|
if window_size > 0: |
|
|
window_steps = episode["steps"][i-window_size:i] if i >= window_size else episode["steps"][:i] |
|
|
|
|
|
if i > window_size: |
|
|
convs.append( |
|
|
{ |
|
|
"from": "human", |
|
|
"value": user_history_instr_prompt.format( |
|
|
history_list="\n".join( |
|
|
[ |
|
|
f"\t{j+1}. " + prev_step["step_instruction"] |
|
|
for j, prev_step in enumerate(episode["steps"][:i-window_size]) |
|
|
] |
|
|
) |
|
|
), |
|
|
}, |
|
|
) |
|
|
|
|
|
convs.append( |
|
|
{ |
|
|
"from": "human", |
|
|
"value": "The recent steps with the GUI images are as follows:\n", |
|
|
} |
|
|
) |
|
|
|
|
|
for j, win_step_i in enumerate(window_steps): |
|
|
if win_step_i["is_grounding"]: |
|
|
convs.append( |
|
|
{ |
|
|
"from": "human", |
|
|
"value": grounding_step_prompt.format( |
|
|
instruction=win_step_i["step_instruction"], step_idx=i+1-(len(window_steps)-j) |
|
|
), |
|
|
} |
|
|
) |
|
|
convs.append( |
|
|
{ |
|
|
"from": "gpt", |
|
|
"value": grounding_step_ans.format(point_str=f"({win_step_i['coord_norm'][0]}, {win_step_i['coord_norm'][1]})"), |
|
|
} |
|
|
) |
|
|
else: |
|
|
convs.append( |
|
|
{ |
|
|
"from": "human", |
|
|
"value": act_step_prompt.format( |
|
|
prev_instruction=encode_action(win_step_i), step_idx=i+1-(len(window_steps)-j) |
|
|
), |
|
|
} |
|
|
) |
|
|
convs.append( |
|
|
{ |
|
|
"from": "human", |
|
|
"value": act_step_ans.format( |
|
|
prev_action=encode_action(win_step_i) |
|
|
), |
|
|
} |
|
|
) |
|
|
|
|
|
win_img_list = [ |
|
|
(Path(img_dir) / Path(win_step["image_path"]).name) for win_step in window_steps |
|
|
] |
|
|
|
|
|
if not all([img_path.exists() for img_path in win_img_list+cur_img_list]): |
|
|
print(f"Image not found for episode {episode_id}, step {i+1}. Skipping...") |
|
|
continue |
|
|
|
|
|
has_img_broken = False |
|
|
for img_path in win_img_list+cur_img_list: |
|
|
try: |
|
|
Image.open(str(img_path)) |
|
|
except Exception as e: |
|
|
print(f"Error opening image {img_path}: {e}") |
|
|
has_img_broken = True |
|
|
break |
|
|
if has_img_broken: |
|
|
print(f"Image broken for episode {episode_id}, step {i+1}. Skipping...") |
|
|
continue |
|
|
|
|
|
resize_scale = resize_ratios_per_window_size[window_size] |
|
|
win_img_list_resized = [] |
|
|
for img_path in win_img_list: |
|
|
new_save_name = img_path.stem + f"_{resize_scale}x" + img_path.suffix |
|
|
new_save_dir = img_path.parent.parent / f"images_resized" |
|
|
new_save_dir.mkdir(parents=True, exist_ok=True) |
|
|
new_save_path = new_save_dir / new_save_name |
|
|
if new_save_path.exists(): |
|
|
try: |
|
|
Image.open(str(new_save_path)) |
|
|
except Exception as e: |
|
|
print(f"Error opening image {new_save_path}: {e}") |
|
|
os.remove(new_save_path) |
|
|
else: |
|
|
win_img_list_resized.append(new_save_path) |
|
|
continue |
|
|
win_img = Image.open(str(img_path)) |
|
|
win_img = resize_image(win_img, scale=resize_scale) |
|
|
win_img.save(str(new_save_path)) |
|
|
win_img_list_resized.append(new_save_path) |
|
|
else: |
|
|
convs.append( |
|
|
{ |
|
|
"from": "human", |
|
|
"value": user_history_instr_prompt.format( |
|
|
history_list="\n".join( |
|
|
[ |
|
|
f"\t{j+1}. " + prev_step["step_instruction"] |
|
|
for j, prev_step in enumerate(episode["steps"][:i-window_size]) |
|
|
] |
|
|
) |
|
|
), |
|
|
}, |
|
|
) |
|
|
|
|
|
cur_img_list_resized = [] |
|
|
for img_path in cur_img_list: |
|
|
new_save_name = img_path.stem + f"_{0.5}x" + img_path.suffix |
|
|
new_save_dir = img_path.parent.parent / f"images_resized" |
|
|
new_save_dir.mkdir(parents=True, exist_ok=True) |
|
|
new_save_path = new_save_dir / new_save_name |
|
|
if new_save_path.exists(): |
|
|
try: |
|
|
Image.open(str(new_save_path)) |
|
|
except Exception as e: |
|
|
print(f"Error opening image {new_save_path}: {e}") |
|
|
os.remove(new_save_path) |
|
|
else: |
|
|
cur_img_list_resized.append(new_save_path) |
|
|
continue |
|
|
cur_img = Image.open(str(img_path)) |
|
|
cur_img = resize_image(cur_img, scale=0.5) |
|
|
cur_img.save(str(new_save_path)) |
|
|
cur_img_list_resized.append(new_save_path) |
|
|
|
|
|
if window_size > 0: |
|
|
img_list = win_img_list_resized + cur_img_list_resized |
|
|
else: |
|
|
img_list = cur_img_list_resized |
|
|
|
|
|
convs.append( |
|
|
{ |
|
|
"from": "human", |
|
|
"value": grounding_step_prompt.format(instruction=step["step_instruction"], step_idx=i+1), |
|
|
} |
|
|
) |
|
|
convs.append( |
|
|
{ |
|
|
"from": "gpt", |
|
|
"value": grounding_step_ans.format(point_str=f"({step['coord_norm'][0]}, {step['coord_norm'][1]})"), |
|
|
} |
|
|
) |
|
|
|
|
|
convs = merge_convs(convs) |
|
|
|
|
|
instructions.append( |
|
|
{ |
|
|
"image": [str(img_path) for img_path in img_list], |
|
|
"conversations": convs, |
|
|
} |
|
|
) |
|
|
|
|
|
return instructions |
|
|
|
|
|
|
|
|
if __name__ == "__main__": |
|
|
|
|
|
data = [] |
|
|
|
|
|
episode_files = list(Path("./episodes_grounding_inner_reasoning_v3").glob("*.json")) |
|
|
for episode_file in episode_files: |
|
|
with open(episode_file, "r", encoding="utf-8") as file: |
|
|
episode_data = json.load(file) |
|
|
for i, step_i in enumerate(episode_data["steps"]): |
|
|
if not step_i.get("grounding_reasoning"): |
|
|
step_i["step_instruction"] = encode_action(step_i) |
|
|
step_i["is_grounding"] = False |
|
|
continue |
|
|
_, step_instruction, _ = parse_reasoning(step_i["grounding_reasoning"]) |
|
|
step_i["step_instruction"] = step_instruction |
|
|
step_i["is_grounding"] = bool(step_instruction) and step_i["action"] == "TAP" |
|
|
data.append(episode_data) |
|
|
|
|
|
def preprocess_coord_norm(episode): |
|
|
for step in episode["steps"]: |
|
|
image_path = Path("./AMEX/screenshot") / step["image_path"] |
|
|
if not image_path.exists(): |
|
|
print(f"Image not found: {image_path}") |
|
|
if step["action"] == "TAP": |
|
|
image = Image.open(image_path) |
|
|
image_width, image_height = image.size |
|
|
step["coord_norm"] = [int(step["touch_coord"][0]/image_width*1000), int(step["touch_coord"][1]/image_height*1000)] |
|
|
return episode |
|
|
|
|
|
with Pool() as pool: |
|
|
data = pool.map(preprocess_coord_norm, data) |
|
|
|
|
|
|
|
|
|
|
|
window_size_list = [0,1,2,3] |
|
|
|
|
|
def process_episode(args): |
|
|
episode, window_size = args |
|
|
return process_android_episodes([episode], window_size) |
|
|
|
|
|
instructions = [] |
|
|
for window_size in window_size_list: |
|
|
tasks = [(episode, window_size) for episode in data] |
|
|
with multiprocessing.Pool(processes=multiprocessing.cpu_count()) as pool: |
|
|
results = list(tqdm(pool.imap(process_episode, tasks), total=len(tasks), desc=f"Window Size {window_size}")) |
|
|
for result in results: |
|
|
instructions.extend(result) |
|
|
|
|
|
print(f"Number of context aware train instructions: {len(instructions)}") |
|
|
|
|
|
with open(f"amex_train_window_{'-'.join([str(e) for e in window_size_list])}_{len(instructions)//1000}k.json", "w", encoding="utf-8") as file: |
|
|
json.dump(instructions, file, ensure_ascii=False, indent=4) |