Upload README.md with huggingface_hub

README.md

@@ -154,8 +154,11 @@ from huggingface_hub import hf_hub_download
 from transformers import PreTrainedTokenizerFast, Gemma3ForConditionalGeneration
 
 repo = "shibatch/tinygemma3ocr2m"
-tok = PreTrainedTokenizerFast.from_pretrained(repo)
-model = Gemma3ForConditionalGeneration.from_pretrained(repo, torch_dtype=torch.bfloat16).cuda().eval()
+
+tok = PreTrainedTokenizerFast.from_pretrained(repo, subfolder="hf")
+model = Gemma3ForConditionalGeneration.from_pretrained(
+    repo, subfolder="hf", torch_dtype=torch.bfloat16
+).cuda().eval()
 
 path = hf_hub_download(repo, "sample_images/sample_00_6235317.png")
 img = Image.open(path).convert("RGB")
@@ -164,7 +167,19 @@ pix = torch.from_numpy(np.asarray(img, dtype=np.float32) / 127.5 - 1).permute(2,
 ids = [tok.bos_token_id] + [model.config.image_token_index] * model.config.mm_tokens_per_image
 ids += tok.encode("\nRead the digits.\n", add_special_tokens=False)
 
-out = model.generate(input_ids=torch.tensor([ids]).cuda(), pixel_values=pix, max_new_tokens=12, do_sample=False)
+input_ids = torch.tensor([ids], device="cuda")
+attention_mask = torch.ones_like(input_ids)
+
+out = model.generate(
+    input_ids=input_ids,
+    attention_mask=attention_mask,
+    pixel_values=pix,
+    max_new_tokens=12,
+    do_sample=False,
+    pad_token_id=tok.bos_token_id,
+    eos_token_id=tok.eos_token_id,
+)
+
 print(tok.decode(out[0][len(ids):], skip_special_tokens=True))
 ```
 
@@ -192,11 +207,13 @@ import torch
 from transformers import PreTrainedTokenizerFast, Gemma3ForConditionalGeneration
 
 repo = "shibatch/tinygemma3ocr2m"
-tok = PreTrainedTokenizerFast.from_pretrained(repo)
-model = Gemma3ForConditionalGeneration.from_pretrained(repo, torch_dtype=torch.bfloat16).cuda().eval()
+tok = PreTrainedTokenizerFast.from_pretrained(repo, subfolder="hf")
+model = Gemma3ForConditionalGeneration.from_pretrained(
+    repo, subfolder="hf", torch_dtype=torch.bfloat16
+).cuda().eval()
 
 ids = [tok.bos_token_id] + tok.encode("Once upon", add_special_tokens=False)
-x = torch.tensor([ids]).cuda()
+x = torch.tensor([ids], device="cuda")
 
 for _ in range(50):
     h = model.model.language_model(input_ids=x, use_cache=False, return_dict=True).last_hidden_state