File size: 2,248 Bytes
44b7441
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
import gradio as gr
import numpy as np, pandas as pd
from huggingface_hub import hf_hub_download
from sentence_transformers import SentenceTransformer

DATASET_REPO = "samvlad/dream-decoder-dataset"

def _dl(name):
    return hf_hub_download(repo_id=DATASET_REPO, filename=name, repo_type="dataset")

# Load embeddings + a table (meta if present, else full dataset)
emb_path = _dl("data/embeddings.npy")
try:
    table_path = _dl("data/meta.parquet")
    meta = pd.read_parquet(table_path)[["dream_text","interpretation"]]
except Exception:
    table_path = _dl("data/dreams.parquet")
    meta = pd.read_parquet(table_path)[["dream_text","interpretation"]]

emb = np.load(emb_path)  # already normalized vectors
encoder = SentenceTransformer("sentence-transformers/all-MiniLM-L6-v2")

def top3_similar(text):
    t = (text or "").strip()
    if len(t) < 10:
        return "Please paste a longer dream (>= 10 characters)."
    q = encoder.encode([t], normalize_embeddings=True)
    scores = (emb @ q.T).squeeze()
    idx = np.argsort(-scores)[:3]
    out = []
    for i, j in enumerate(idx, 1):
        row = meta.iloc[j]
        out.append(
            f"### #{i} • Similarity: {scores[j]:.3f}\n"
            f"**Dream:** {row['dream_text']}\n\n"
            f"**Interpretation:** {row['interpretation']}\n"
        )
    return "\n---\n".join(out)

with gr.Blocks(fill_height=True) as demo:
    gr.Markdown("# 🌙 Dream Decoder\nPaste a dream. We’ll find the 3 most similar dreams and show their interpretations.")
    dream = gr.Textbox(label="Your dream", lines=7, placeholder="I was on a rooftop, running from a shadow...")
    btn = gr.Button("Find 3 similar dreams")
    out = gr.Markdown()
    gr.Examples(
        examples=[
            "I was in a school hallway, hiding from a teacher. A mirror cracked while I was waiting, and I felt anxious yet hopeful.",
            "I was at a beach at night, searching for my phone. A storm rolled in and I felt excited yet confused.",
            "I was in an old house, escaping a fire. A baby cried and I felt lonely yet relieved.",
        ],
        inputs=[dream],
        label="Try examples",
    )
    btn.click(top3_similar, dream, out)

if __name__ == "__main__":
    demo.launch()