add mscoco generative benchmark (#63)

* add initial generative benchmark

* add pycocoeval dep

* muse generate_beamsearch

* update to generate

* clean generate

---------

Co-authored-by: Romain Beaumont <romain.rom1@gmail.com>

clip_benchmark/cli.py

@@ -7,7 +7,7 @@
 from copy import copy
 import os
 from clip_benchmark.datasets.builder import build_dataset, get_dataset_collate_fn, get_dataset_default_task, dataset_collection, get_dataset_collection_from_file
-from clip_benchmark.metrics import zeroshot_classification, zeroshot_retrieval, linear_probe
+from clip_benchmark.metrics import zeroshot_classification, zeroshot_retrieval, linear_probe, mscoco_generative
 from clip_benchmark.model_collection import get_model_collection_from_file, model_collection
 from clip_benchmark.models import load_clip, MODEL_TYPES
 
@@ -22,7 +22,7 @@ def get_parser_args():
     parser_eval.add_argument('--model', type=str, default="ViT-B-32-quickgelu", help="Model architecture to use from OpenCLIP")
     parser_eval.add_argument('--pretrained', type=str, default="laion400m_e32", help="Model checkpoint name to use from OpenCLIP")
     parser_eval.add_argument('--pretrained_model', type=str, default="", nargs="+", help="Pre-trained model(s) to use. Can be the full model name where `model` and `pretrained` are comma separated (e.g., --pretrained_model='ViT-B-32-quickgelu,laion400m_e32'), a model collection name ('openai' or 'openclip_base' or 'openclip_multilingual' or 'openclip_all'), or path of a text file where each line is a model fullname where model and pretrained are comma separated (e.g., ViT-B-32-quickgelu,laion400m_e32). --model and --pretrained are ignored if --pretrained_model is used.")
-    parser_eval.add_argument('--task', type=str, default="auto", choices=["zeroshot_classification", "zeroshot_retrieval", "linear_probe", "auto"], help="Task to evaluate on. With --task=auto, the task is automatically inferred from the dataset.")
+    parser_eval.add_argument('--task', type=str, default="auto", choices=["zeroshot_classification", "zeroshot_retrieval", "linear_probe", "mscoco_generative", "auto"], help="Task to evaluate on. With --task=auto, the task is automatically inferred from the dataset.")
     parser_eval.add_argument('--amp', default=True, action="store_true", help="whether to use mixed precision")
     parser_eval.add_argument('--num_workers', default=4, type=int)
     parser_eval.add_argument('--recall_k', default=[5], type=int, help="for retrieval, select the k for Recall@K metric. ", nargs="+",)
@@ -278,6 +278,17 @@ def run(args):
             amp=args.amp,
             verbose=args.verbose,
         )
+    elif task == "mscoco_generative":
+        metrics = mscoco_generative.evaluate(
+            model=model, 
+            dataloader=dataloader, 
+            batch_size=args.batch_size,
+            num_workers=args.num_workers,
+            device=args.device, 
+            amp=args.amp,
+            verbose=args.verbose,
+            transform=transform
+        )
     else:
         raise ValueError("Unsupported task: {}. task should `zeroshot_classification` or `zeroshot_retrieval`".format(task))
     dump = {

clip_benchmark/metrics/mscoco_generative.py

@@ -0,0 +1,31 @@
+import json
+from pycocoevalcap.eval import COCOEvalCap
+from open_clip import tokenize
+from tqdm.auto import tqdm
+from open_clip.tokenizer import _tokenizer
+
+def evaluate(model, dataloader, batch_size, device, transform, train_dataloader=None, num_workers=None, amp=True, verbose=False):
+    coco = dataloader.dataset.coco
+    indexer = dataloader.dataset.ids
+    results = []
+    for idx, (img, _) in enumerate(tqdm(dataloader)):
+        n_samples = img.shape[0] # for last batch
+        idxs = [indexer[idx * batch_size + id] for id in range(n_samples)]
+        out = model.generate(img.to(device))
+        decoded = [_tokenizer.decode(i).split("<end_of_text>")[0].replace("<start_of_text>", "").strip() for i in out.cpu().numpy()]
+        for image_id, caption in zip(idxs, decoded):
+            results.append({"image_id":image_id, "caption":caption})
+    temp_res_file = "temp_results.json"
+    with open(temp_res_file, "w") as jf:
+        json.dump(results, jf)
+
+    coco_result = coco.loadRes(temp_res_file)
+    coco_eval = COCOEvalCap(coco, coco_result)
+    coco_eval.evaluate()
+    metrics = coco_eval.eval
+
+    # print output evaluation scores
+    for metric, score in metrics.items():
+        print(f'{metric}: {score:.3f}')
+
+    return metrics

requirements.txt

@@ -3,5 +3,6 @@ torchvision>=0.8.9,<2
 tqdm>=2
 scikit-learn>=1.0,<2
 open_clip_torch>=0.2.1
+pycocoevalcap
 webdataset>=0.2.31
-transformers
+transformers