Duplicate from yuvalkirstain/PickScore

Co-authored-by: Yuval Kirstain <[email protected]>

.gitattributes

@@ -0,0 +1,35 @@
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ckpt filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.mlmodel filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.npy filter=lfs diff=lfs merge=lfs -text
+*.npz filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pickle filter=lfs diff=lfs merge=lfs -text
+*.pkl filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+*.safetensors filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.xz filter=lfs diff=lfs merge=lfs -text
+*.zip filter=lfs diff=lfs merge=lfs -text
+*.zst filter=lfs diff=lfs merge=lfs -text
+*tfevents* filter=lfs diff=lfs merge=lfs -text
+*.png filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,14 @@
+---
+title: PickScore
+emoji: 😻
+colorFrom: yellow
+colorTo: blue
+sdk: gradio
+sdk_version: 3.28.3
+app_file: app.py
+pinned: false
+license: mit
+duplicated_from: yuvalkirstain/PickScore
+---
+
+Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference

app.py

@@ -0,0 +1,97 @@
+import time
+from PIL import Image
+import gradio as gr
+from glob import glob
+import torch
+from transformers import AutoModel, AutoProcessor
+
+DEFAULT_EXAMPLE_PATH = f'examples/example_0'
+
+device = "cuda" if torch.cuda.is_available() else "cpu"
+weight_dtype = torch.bfloat16 if device == "cuda" else torch.float32
+print(f"Using device: {device} ({weight_dtype})")
+print("Loading model...")
+model_pretrained_name_or_path = "yuvalkirstain/PickScore_v1"
+processor = AutoProcessor.from_pretrained(model_pretrained_name_or_path)
+model = AutoModel.from_pretrained(model_pretrained_name_or_path, torch_dtype=weight_dtype).eval().to(device)
+print("Model loaded.")
+
+
+def calc_probs(prompt, images):
+    print("Processing inputs...")
+    image_inputs = processor(
+        images=images,
+        padding=True,
+        truncation=True,
+        max_length=77,
+        return_tensors="pt",
+    ).to(device)
+
+    image_inputs = {k: v.to(weight_dtype) for k, v in image_inputs.items()}
+
+    text_inputs = processor(
+        text=prompt,
+        padding=True,
+        truncation=True,
+        max_length=77,
+        return_tensors="pt",
+    ).to(device)
+
+    with torch.no_grad():
+        print("Embedding images and text...")
+        image_embs = model.get_image_features(**image_inputs)
+        image_embs = image_embs / torch.norm(image_embs, dim=-1, keepdim=True)
+
+        text_embs = model.get_text_features(**text_inputs)
+        text_embs = text_embs / torch.norm(text_embs, dim=-1, keepdim=True)
+
+        print("Calculating scores...")
+        scores = model.logit_scale.exp() * (text_embs.float() @ image_embs.float().T)[0]
+
+        print("Calculating probabilities...")
+        probs = torch.softmax(scores, dim=-1)
+
+    return probs.cpu().tolist()
+
+
+def predict(prompt, image_1, image_2):
+    print(f"Starting prediction for prompt: {prompt}")
+    start_time = time.time()
+    probs = calc_probs(prompt, [image_1, image_2])
+    print(f"Prediction: {probs} ({time.time() - start_time:.2f} seconds, ) ")
+    if device == "cuda":
+        print(f"GPU mem used: {round(torch.cuda.max_memory_allocated(device) / 1024 / 1024 / 1024, 2)}/{round(torch.cuda.get_device_properties(device).total_memory / 1024 / 1024 / 1024, 2)} GB")
+    return str(round(probs[0], 3)), str(round(probs[1], 3))
+
+
+with gr.Blocks(title="PickScore v1") as demo:
+    gr.Markdown("# PickScore v1")
+    gr.Markdown(
+        "This is a demo for the PickScore model - see [paper](https://arxiv.org/abs/2305.01569), [code](https://github.com/yuvalkirstain/PickScore), [dataset](https://huggingface.co/datasets/pickapic-anonymous/pickapic_v1), and [model](https://huggingface.co/yuvalkirstain/PickScore_v1).")
+    gr.Markdown("## Instructions")
+    gr.Markdown("Write a prompt, place two images, and press run to get their PickScore!")
+    with gr.Row():
+        prompt = gr.inputs.Textbox(lines=1, label="Prompt",
+                                   default=open(f'{DEFAULT_EXAMPLE_PATH}/prompt.txt').readline())
+    with gr.Row():
+        image_1 = gr.components.Image(type="pil", label="image 1",
+                                      value=Image.open(f'{DEFAULT_EXAMPLE_PATH}/image_1.png'))
+        image_2 = gr.components.Image(type="pil", label="image 2",
+                                      value=Image.open(f'{DEFAULT_EXAMPLE_PATH}/image_2.png'))
+    with gr.Row():
+        pred_1 = gr.outputs.Textbox(label="Probability 1")
+        pred_2 = gr.outputs.Textbox(label="Probability 2")
+
+    btn = gr.Button("Run")
+    btn.click(fn=predict, inputs=[prompt, image_1, image_2], outputs=[pred_1, pred_2])
+    prompt.change(lambda: ("", ""), inputs=[], outputs=[pred_1, pred_2])
+
+    gr.Examples(
+        [[open(f'{path}/prompt.txt').readline(), f'{path}/image_1.png', f'{path}/image_2.png'] for path in
+         glob(f'examples/*')],
+        [prompt, image_1, image_2],
+        [pred_1, pred_2],
+        predict
+    )
+
+demo.queue(concurrency_count=5).launch()

examples/example_0/prompt.txt

@@ -0,0 +1 @@
+A sign that says PICK A PIC

examples/example_1/prompt.txt

@@ -0,0 +1 @@
+A bee devouring the world

examples/example_2/prompt.txt

@@ -0,0 +1 @@
+Crazy frog on one wheel

examples/example_3/prompt.txt

@@ -0,0 +1 @@
+A cute fluffy easter bunny singing

examples/example_4/prompt.txt

@@ -0,0 +1 @@
+Silver furred Lion man hybrid with few leather clothes

examples/example_5/prompt.txt

@@ -0,0 +1 @@
+Superman's corpse found

examples/example_6/prompt.txt

@@ -0,0 +1 @@
+a girl

examples/example_7/prompt.txt

@@ -0,0 +1 @@
+A gloomy rabbit drinks wine

examples/example_8/prompt.txt

@@ -0,0 +1 @@
+a beautiful landscape

examples/example_9/prompt.txt

@@ -0,0 +1 @@
+A gloomy rabbit drinks wine

requirements.txt

@@ -0,0 +1,2 @@
+torch
+transformers