import collections import heapq import json import os import logging import gradio as gr import numpy as np import torch import torch.nn.functional as F from open_clip import create_model, get_tokenizer from torchvision import transforms from templates import openai_imagenet_template from huggingface_hub import webhook_endpoint, WebhookPayload log_format = "[%(asctime)s] [%(levelname)s] [%(name)s] %(message)s" logging.basicConfig(level=logging.INFO, format=log_format) logger = logging.getLogger() hf_token = os.getenv("HF_TOKEN") model_str = "hf-hub:imageomics/bioclip" tokenizer_str = "ViT-B-16" txt_emb_npy = "txt_emb_species.npy" txt_names_json = "txt_emb_species.json" min_prob = 1e-9 k = 5 device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu") preprocess_img = transforms.Compose( [ transforms.ToTensor(), transforms.Resize((224, 224), antialias=True), transforms.Normalize( mean=(0.48145466, 0.4578275, 0.40821073), std=(0.26862954, 0.26130258, 0.27577711), ), ] ) ranks = ("Kingdom", "Phylum", "Class", "Order", "Family", "Genus", "Species") open_domain_examples = [ ["examples/Ursus-arctos.jpeg", "Species"], ["examples/Phoca-vitulina.png", "Species"], ["examples/Felis-catus.jpeg", "Genus"], ["examples/Sarcoscypha-coccinea.jpeg", "Order"], ] zero_shot_examples = [ [ "examples/Ursus-arctos.jpeg", "brown bear\nblack bear\npolar bear\nkoala bear\ngrizzly bear", ], ["examples/milk-snake.png", "coral snake\nmilk snake"], ["examples/coral-snake.jpeg", "coral snake\nmilk snake"], [ "examples/Carnegiea-gigantea.png", "Carnegiea gigantea\nSchlumbergera opuntioides\nMammillaria albicoma", ], [ "examples/Amanita-muscaria.jpeg", "Amanita fulva\nAmanita vaginata (grisette)\nAmanita calyptrata (coccoli)\nAmanita crocea\nAmanita rubescens (blusher)\nAmanita caesarea (Caesar's mushroom)\nAmanita jacksonii (American Caesar's mushroom)\nAmanita muscaria (fly agaric)\nAmanita pantherina (panther cap)", ], [ "examples/Actinostola-abyssorum.png", "Animalia Cnidaria Hexacorallia Actiniaria Actinostolidae Actinostola abyssorum\nAnimalia Cnidaria Hexacorallia Actiniaria Actinostolidae Actinostola bulbosa\nAnimalia Cnidaria Hexacorallia Actiniaria Actinostolidae Actinostola callosa\nAnimalia Cnidaria Hexacorallia Actiniaria Actinostolidae Actinostola capensis\nAnimalia Cnidaria Hexacorallia Actiniaria Actinostolidae Actinostola carlgreni", ], [ "examples/Sarcoscypha-coccinea.jpeg", "scarlet elf cup (coccinea)\nscharlachroter kelchbecherling (austriaca)\ncrimson cup (dudleyi)\nstalked scarlet cup (occidentalis)", ], [ "examples/Onoclea-hintonii.jpg", "Onoclea attenuata\nOnoclea boryana\nOnoclea hintonii\nOnoclea intermedia\nOnoclea sensibilis", ], [ "examples/Onoclea-sensibilis.jpg", "Onoclea attenuata\nOnoclea boryana\nOnoclea hintonii\nOnoclea intermedia\nOnoclea sensibilis", ], ] def indexed(lst, indices): return [lst[i] for i in indices] @torch.no_grad() def get_txt_features(classnames, templates): all_features = [] for classname in classnames: txts = [template(classname) for template in templates] txts = tokenizer(txts).to(device) txt_features = model.encode_text(txts) txt_features = F.normalize(txt_features, dim=-1).mean(dim=0) txt_features /= txt_features.norm() all_features.append(txt_features) all_features = torch.stack(all_features, dim=1) return all_features @torch.no_grad() def zero_shot_classification(img, cls_str: str) -> dict[str, float]: classes = [cls.strip() for cls in cls_str.split("\n") if cls.strip()] txt_features = get_txt_features(classes, openai_imagenet_template) img = preprocess_img(img).to(device) img_features = model.encode_image(img.unsqueeze(0)) img_features = F.normalize(img_features, dim=-1) logits = (model.logit_scale.exp() * img_features @ txt_features).squeeze() probs = F.softmax(logits, dim=0).to("cpu").tolist() return {cls: prob for cls, prob in zip(classes, probs)} def format_name(taxon, common): taxon = " ".join(taxon) if not common: return taxon return f"{taxon} ({common})" @torch.no_grad() def open_domain_classification(img, rank: int) -> dict[str, float]: """ Predicts from the entire tree of life. If targeting a higher rank than species, then this function predicts among all species, then sums up species-level probabilities for the given rank. """ img = preprocess_img(img).to(device) img_features = model.encode_image(img.unsqueeze(0)) img_features = F.normalize(img_features, dim=-1) logits = (model.logit_scale.exp() * img_features @ txt_emb).squeeze() probs = F.softmax(logits, dim=0) # If predicting species, no need to sum probabilities. if rank + 1 == len(ranks): topk = probs.topk(k) return { format_name(*txt_names[i]): prob for i, prob in zip(topk.indices, topk.values) } # Sum up by the rank output = collections.defaultdict(float) for i in torch.nonzero(probs > min_prob).squeeze(): output[" ".join(txt_names[i][0][: rank + 1])] += probs[i] topk_names = heapq.nlargest(k, output, key=output.get) return {name: output[name] for name in topk_names} @torch.no_grad() def api_classification(img, rank: int): # -> dict[str, float]: """ Predicts from the entire tree of life. If targeting a higher rank than species, then this function predicts among all species, then sums up species-level probabilities for the given rank. """ img = preprocess_img(img).to(device) img_features = model.encode_image(img.unsqueeze(0)) img_features = F.normalize(img_features, dim=-1) logits = (model.logit_scale.exp() * img_features @ txt_emb).squeeze() probs = F.softmax(logits, dim=0) # If predicting species, no need to sum probabilities. if rank + 1 == len(ranks): topk = probs.topk(k) return { format_name(*txt_names[i]): prob for i, prob in zip(topk.indices, topk.values) } # Sum up by the rank output = collections.defaultdict(float) for i in torch.nonzero(probs > min_prob).squeeze(): output[" ".join(txt_names[i][0][: rank + 1])] += probs[i] topk_names = heapq.nlargest(k, output, key=output.get) exampleB = topk_names[0] logger.info("exampleB") logger.info(exampleB) logger.info("output(exampleB)") logger.info(output(exampleB)) return {name: output[name] for name in topk_names} def change_output(choice): return gr.Label(num_top_classes=k, label=ranks[choice], show_label=True, value=None) @webhook_endpoint # https://huggingface.co/docs/huggingface_hub/guides/webhooks_server async def trigger_test(payload: WebhookPayload): logger.info(payload) if payload.repo.type == "dataset" and payload.event.action == "update": logger.info("oo") return {"message": "hello"} if __name__ == "__main__": logger.info("Starting.") model = create_model(model_str, output_dict=True, require_pretrained=True) model = model.to(device) logger.info("Created model.") model = torch.compile(model) logger.info("Compiled model.") tokenizer = get_tokenizer(tokenizer_str) txt_emb = torch.from_numpy(np.load(txt_emb_npy, mmap_mode="r")).to(device) with open(txt_names_json) as fd: txt_names = json.load(fd) done = txt_emb.any(axis=0).sum().item() total = txt_emb.shape[1] status_msg = "" if done != total: status_msg = f"{done}/{total} ({done / total * 100:.1f}%) indexed" with gr.Blocks() as app: img_input = gr.Image() with gr.Tab("Open-Ended"): with gr.Row(): with gr.Column(): rank_dropdown = gr.Dropdown( label="Taxonomic Rank", info="Which taxonomic rank to predict. Fine-grained ranks (genus, species) are more challenging.", choices=ranks, value="Species", type="index", ) open_domain_btn = gr.Button("Submit", variant="primary") api_classification_btn = gr.Button("API", variant="primary") with gr.Column(): open_domain_output = gr.Label( num_top_classes=k, label="Prediction", show_label=True, value=None, ) # open_domain_flag_btn = gr.Button("Flag Mistake", variant="primary") api_classification_output = gr.Label( num_top_classes=k, label="API Output", show_label=True, value=None, ) with gr.Row(): gr.Examples( examples=open_domain_examples, inputs=[img_input, rank_dropdown], cache_examples=True, fn=open_domain_classification, outputs=[open_domain_output], ) # open_domain_callback = gr.HuggingFaceDatasetSaver( # hf_token, "imageomics/bioclip-demo-open-domain-mistakes", private=True # ) # open_domain_callback.setup( # [img_input, rank_dropdown, open_domain_output], # flagging_dir="logs/flagged", # ) # open_domain_flag_btn.click( # lambda *args: open_domain_callback.flag(args), # [img_input, rank_dropdown, open_domain_output], # None, # preprocess=False, # ) with gr.Tab("Zero-Shot"): with gr.Row(): with gr.Column(): classes_txt = gr.Textbox( placeholder="Canis familiaris (dog)\nFelis catus (cat)\n...", lines=3, label="Classes", show_label=True, info="Use taxonomic names where possible; include common names if possible.", ) zero_shot_btn = gr.Button("Submit", variant="primary") with gr.Column(): zero_shot_output = gr.Label( num_top_classes=k, label="Prediction", show_label=True ) # zero_shot_flag_btn = gr.Button("Flag Mistake", variant="primary") with gr.Row(): gr.Examples( examples=zero_shot_examples, inputs=[img_input, classes_txt], cache_examples=True, fn=zero_shot_classification, outputs=[zero_shot_output], ) # zero_shot_callback = gr.HuggingFaceDatasetSaver( # hf_token, "imageomics/bioclip-demo-zero-shot-mistakes", private=True # ) # zero_shot_callback.setup( # [img_input, zero_shot_output], flagging_dir="logs/flagged" # ) # zero_shot_flag_btn.click( # lambda *args: zero_shot_callback.flag(args), # [img_input, zero_shot_output], # None, # preprocess=False, # ) rank_dropdown.change( fn=change_output, inputs=rank_dropdown, outputs=[open_domain_output] ) open_domain_btn.click( fn=open_domain_classification, inputs=[img_input, rank_dropdown], outputs=[open_domain_output], ) api_classification_btn.click( fn=api_classification, inputs=[img_input, rank_dropdown], outputs=[api_classification_output], ) zero_shot_btn.click( fn=zero_shot_classification, inputs=[img_input, classes_txt], outputs=zero_shot_output, ) app.queue(max_size=20) app.launch() # app.py