comments

app.py

@@ -1,5 +1,5 @@
 
-
+import gradio as gr
 
 import os
 os.system("git clone https://huggingface.co/Cene655/ImagenT5-3B")
@@ -48,7 +48,7 @@ from gfpgan import GFPGANer
 has_cuda = th.cuda.is_available()
 device = th.device('cpu' if not has_cuda else 'cuda')
 
-Setting Up
+#Setting Up
 
 def model_fn(x_t, ts, **kwargs):
     guidance_scale = 5
@@ -97,17 +97,17 @@ model.to(device)
 model.load_state_dict(_fix_path('/content/ImagenT5-3B/model.pt'))
 print('total base parameters', sum(x.numel() for x in model.parameters()))
 
-total base parameters 1550556742
+#total base parameters 1550556742
 
 num_params = sum(param.numel() for param in model.parameters())
 num_params
 
-1550556742
+#1550556742
 
 realesrgan_model = RRDBNet(num_in_ch=3, num_out_ch=3, num_feat=64,
                            num_block=23, num_grow_ch=32, scale=4)
 
-netscale = 4
+#netscale = 4
 
 upsampler = RealESRGANer(
     scale=netscale,
@@ -129,12 +129,7 @@ face_enhancer = GFPGANer(
 
 tokenizer = AutoTokenizer.from_pretrained(options['t5_name'])
 
-/usr/local/lib/python3.7/dist-packages/transformers/models/t5/tokenization_t5_fast.py:161: FutureWarning: This tokenizer was incorrectly instantiated with a model max length of 512 which will be corrected in Transformers v5.
-For now, this behavior is kept to avoid breaking backwards compatibility when padding/encoding with `truncation is True`.
-- Be aware that you SHOULD NOT rely on t5-3b automatically truncating your input to 512 when padding/encoding.
-- If you want to encode/pad to sequences longer than 512 you can either instantiate this tokenizer with `model_max_length` or pass `max_length` when encoding/padding.
-- To avoid this warning, please instantiate this tokenizer with `model_max_length` set to your preferred value.
-  FutureWarning,
+
 
 #@title What do you want to generate?
 
@@ -174,7 +169,7 @@ def gen_img(prompt):
     model_kwargs["mask"] = th.cat((cond_attention_mask,
                                 uncond_attention_mask)).to(device)
 
-    Generation
+    #Generation
 
     model.del_cache()
     sample = diffusion.p_sample_loop(
@@ -189,6 +184,9 @@ def gen_img(prompt):
 
     return sample
 
+
+
+
 demo = gr.Blocks()
 
 with demo: