import torch


from transformers import AutoProcessor, AutoModelForVision2Seq, AwqConfig

import numpy as np
import pyttsx3


START_TO_COUCH = np.array([[0.5, 0], [0.5, 0.5]]).ravel()
COUCH_TO_KITCHEN = np.array([[0.5, -0.5], [1.0, -1.0]]).ravel()
KITCHEN_TO_START = np.array([[0.5, -0.5], [0, 0]]).ravel()

engine = pyttsx3.init("espeak")
voices = engine.getProperty("voices")
engine.setProperty("voice", voices[3].id)


def speak(text):
    print(f"said {text}", flush=True)
    engine.say(text)
    engine.runAndWait()


MODE = "quantized"
DEVICE = "cuda"
PROCESSOR = AutoProcessor.from_pretrained("HuggingFaceM4/idefics2-tfrm-compatible")
BAD_WORDS_IDS = PROCESSOR.tokenizer(
    ["<image>", "<fake_token_around_image>"], add_special_tokens=False
).input_ids
EOS_WORDS_IDS = PROCESSOR.tokenizer(
    "<end_of_utterance>", add_special_tokens=False
).input_ids + [PROCESSOR.tokenizer.eos_token_id]

# Load model
if MODE == "regular":
    model = AutoModelForVision2Seq.from_pretrained(
        "HuggingFaceM4/idefics2-tfrm-compatible",
        torch_dtype=torch.float16,
        trust_remote_code=True,
        _attn_implementation="flash_attention_2",
        revision="3dc93be345d64fb6b1c550a233fe87ddb36f183d",
    ).to(DEVICE)
elif MODE == "quantized":
    quant_path = "HuggingFaceM4/idefics2-tfrm-compatible-AWQ"
    model = AutoModelForVision2Seq.from_pretrained(
        quant_path, trust_remote_code=True
    ).to(DEVICE)
elif MODE == "fused_quantized":
    quant_path = "HuggingFaceM4/idefics2-tfrm-compatible-AWQ"
    quantization_config = AwqConfig(
        bits=4,
        fuse_max_seq_len=4096,
        modules_to_fuse={
            "attention": ["q_proj", "k_proj", "v_proj", "o_proj"],
            "mlp": ["gate_proj", "up_proj", "down_proj"],
            "layernorm": ["input_layernorm", "post_attention_layernorm", "norm"],
            "use_alibi": False,
            "num_attention_heads": 32,
            "num_key_value_heads": 8,
            "hidden_size": 4096,
        },
    )
    model = AutoModelForVision2Seq.from_pretrained(
        quant_path, quantization_config=quantization_config, trust_remote_code=True
    ).to(DEVICE)
else:
    raise ValueError("Unknown mode")


def ask_vlm(image, instruction):
    prompts = [
        "User:",
        image,
        f"{instruction}.<end_of_utterance>\n",
        "Assistant:",
    ]
    inputs = PROCESSOR(prompts)
    inputs = {k: torch.tensor(v).to(DEVICE) for k, v in inputs.items()}

    generated_ids = model.generate(
        **inputs, bad_words_ids=BAD_WORDS_IDS, max_new_tokens=10
    )
    generated_texts = PROCESSOR.batch_decode(generated_ids, skip_special_tokens=True)
    return generated_texts[0].split("\nAssistant: ")[1]