Upload folder using huggingface_hub

.gitattributes

@@ -44,6 +44,7 @@ assets/image_106_1.png filter=lfs diff=lfs merge=lfs -text
 assets/image_107_2.png filter=lfs diff=lfs merge=lfs -text
 assets/image_108_0.png filter=lfs diff=lfs merge=lfs -text
 assets/image_109_1.png filter=lfs diff=lfs merge=lfs -text
+assets/image_10_1.png filter=lfs diff=lfs merge=lfs -text
 assets/image_110_2.png filter=lfs diff=lfs merge=lfs -text
 assets/image_111_0.png filter=lfs diff=lfs merge=lfs -text
 assets/image_112_1.png filter=lfs diff=lfs merge=lfs -text
@@ -55,6 +56,9 @@ assets/image_117_0.png filter=lfs diff=lfs merge=lfs -text
 assets/image_118_1.png filter=lfs diff=lfs merge=lfs -text
 assets/image_119_2.png filter=lfs diff=lfs merge=lfs -text
 assets/image_11_2.png filter=lfs diff=lfs merge=lfs -text
+assets/image_120_0.png filter=lfs diff=lfs merge=lfs -text
+assets/image_121_1.png filter=lfs diff=lfs merge=lfs -text
+assets/image_122_2.png filter=lfs diff=lfs merge=lfs -text
 assets/image_123_0.png filter=lfs diff=lfs merge=lfs -text
 assets/image_124_1.png filter=lfs diff=lfs merge=lfs -text
 assets/image_125_2.png filter=lfs diff=lfs merge=lfs -text
@@ -62,6 +66,7 @@ assets/image_126_0.png filter=lfs diff=lfs merge=lfs -text
 assets/image_127_1.png filter=lfs diff=lfs merge=lfs -text
 assets/image_128_2.png filter=lfs diff=lfs merge=lfs -text
 assets/image_129_0.png filter=lfs diff=lfs merge=lfs -text
+assets/image_12_0.png filter=lfs diff=lfs merge=lfs -text
 assets/image_130_1.png filter=lfs diff=lfs merge=lfs -text
 assets/image_131_2.png filter=lfs diff=lfs merge=lfs -text
 assets/image_132_0.png filter=lfs diff=lfs merge=lfs -text
@@ -72,6 +77,7 @@ assets/image_136_1.png filter=lfs diff=lfs merge=lfs -text
 assets/image_137_2.png filter=lfs diff=lfs merge=lfs -text
 assets/image_138_0.png filter=lfs diff=lfs merge=lfs -text
 assets/image_139_1.png filter=lfs diff=lfs merge=lfs -text
+assets/image_13_1.png filter=lfs diff=lfs merge=lfs -text
 assets/image_140_2.png filter=lfs diff=lfs merge=lfs -text
 assets/image_141_0.png filter=lfs diff=lfs merge=lfs -text
 assets/image_142_1.png filter=lfs diff=lfs merge=lfs -text
@@ -82,6 +88,7 @@ assets/image_146_2.png filter=lfs diff=lfs merge=lfs -text
 assets/image_147_0.png filter=lfs diff=lfs merge=lfs -text
 assets/image_148_1.png filter=lfs diff=lfs merge=lfs -text
 assets/image_149_2.png filter=lfs diff=lfs merge=lfs -text
+assets/image_14_2.png filter=lfs diff=lfs merge=lfs -text
 assets/image_150_0.png filter=lfs diff=lfs merge=lfs -text
 assets/image_151_1.png filter=lfs diff=lfs merge=lfs -text
 assets/image_152_2.png filter=lfs diff=lfs merge=lfs -text
@@ -107,7 +114,6 @@ assets/image_16_1.png filter=lfs diff=lfs merge=lfs -text
 assets/image_170_2.png filter=lfs diff=lfs merge=lfs -text
 assets/image_171_0.png filter=lfs diff=lfs merge=lfs -text
 assets/image_172_1.png filter=lfs diff=lfs merge=lfs -text
-assets/image_173_2.png filter=lfs diff=lfs merge=lfs -text
 assets/image_174_0.png filter=lfs diff=lfs merge=lfs -text
 assets/image_175_1.png filter=lfs diff=lfs merge=lfs -text
 assets/image_176_2.png filter=lfs diff=lfs merge=lfs -text
@@ -126,8 +132,103 @@ assets/image_187_1.png filter=lfs diff=lfs merge=lfs -text
 assets/image_188_2.png filter=lfs diff=lfs merge=lfs -text
 assets/image_189_0.png filter=lfs diff=lfs merge=lfs -text
 assets/image_18_0.png filter=lfs diff=lfs merge=lfs -text
-assets/image_190_1.png filter=lfs diff=lfs merge=lfs -text
 assets/image_191_2.png filter=lfs diff=lfs merge=lfs -text
+assets/image_19_1.png filter=lfs diff=lfs merge=lfs -text
+assets/image_1_1.png filter=lfs diff=lfs merge=lfs -text
+assets/image_20_2.png filter=lfs diff=lfs merge=lfs -text
+assets/image_21_0.png filter=lfs diff=lfs merge=lfs -text
+assets/image_22_1.png filter=lfs diff=lfs merge=lfs -text
+assets/image_23_2.png filter=lfs diff=lfs merge=lfs -text
+assets/image_24_0.png filter=lfs diff=lfs merge=lfs -text
+assets/image_25_1.png filter=lfs diff=lfs merge=lfs -text
+assets/image_26_2.png filter=lfs diff=lfs merge=lfs -text
+assets/image_27_0.png filter=lfs diff=lfs merge=lfs -text
+assets/image_28_1.png filter=lfs diff=lfs merge=lfs -text
+assets/image_29_2.png filter=lfs diff=lfs merge=lfs -text
+assets/image_2_2.png filter=lfs diff=lfs merge=lfs -text
+assets/image_30_0.png filter=lfs diff=lfs merge=lfs -text
+assets/image_31_1.png filter=lfs diff=lfs merge=lfs -text
+assets/image_32_2.png filter=lfs diff=lfs merge=lfs -text
+assets/image_33_0.png filter=lfs diff=lfs merge=lfs -text
+assets/image_34_1.png filter=lfs diff=lfs merge=lfs -text
+assets/image_35_2.png filter=lfs diff=lfs merge=lfs -text
+assets/image_36_0.png filter=lfs diff=lfs merge=lfs -text
+assets/image_37_1.png filter=lfs diff=lfs merge=lfs -text
+assets/image_38_2.png filter=lfs diff=lfs merge=lfs -text
+assets/image_39_0.png filter=lfs diff=lfs merge=lfs -text
+assets/image_3_0.png filter=lfs diff=lfs merge=lfs -text
+assets/image_40_1.png filter=lfs diff=lfs merge=lfs -text
+assets/image_41_2.png filter=lfs diff=lfs merge=lfs -text
+assets/image_42_0.png filter=lfs diff=lfs merge=lfs -text
+assets/image_43_1.png filter=lfs diff=lfs merge=lfs -text
+assets/image_44_2.png filter=lfs diff=lfs merge=lfs -text
+assets/image_45_0.png filter=lfs diff=lfs merge=lfs -text
+assets/image_46_1.png filter=lfs diff=lfs merge=lfs -text
+assets/image_47_2.png filter=lfs diff=lfs merge=lfs -text
+assets/image_48_0.png filter=lfs diff=lfs merge=lfs -text
+assets/image_49_1.png filter=lfs diff=lfs merge=lfs -text
+assets/image_4_1.png filter=lfs diff=lfs merge=lfs -text
+assets/image_50_2.png filter=lfs diff=lfs merge=lfs -text
+assets/image_51_0.png filter=lfs diff=lfs merge=lfs -text
+assets/image_52_1.png filter=lfs diff=lfs merge=lfs -text
+assets/image_53_2.png filter=lfs diff=lfs merge=lfs -text
+assets/image_54_0.png filter=lfs diff=lfs merge=lfs -text
+assets/image_55_1.png filter=lfs diff=lfs merge=lfs -text
+assets/image_56_2.png filter=lfs diff=lfs merge=lfs -text
+assets/image_57_0.png filter=lfs diff=lfs merge=lfs -text
+assets/image_58_1.png filter=lfs diff=lfs merge=lfs -text
+assets/image_59_2.png filter=lfs diff=lfs merge=lfs -text
+assets/image_5_2.png filter=lfs diff=lfs merge=lfs -text
+assets/image_60_0.png filter=lfs diff=lfs merge=lfs -text
+assets/image_61_1.png filter=lfs diff=lfs merge=lfs -text
+assets/image_62_2.png filter=lfs diff=lfs merge=lfs -text
+assets/image_63_0.png filter=lfs diff=lfs merge=lfs -text
+assets/image_64_1.png filter=lfs diff=lfs merge=lfs -text
+assets/image_65_2.png filter=lfs diff=lfs merge=lfs -text
+assets/image_66_0.png filter=lfs diff=lfs merge=lfs -text
+assets/image_67_1.png filter=lfs diff=lfs merge=lfs -text
+assets/image_68_2.png filter=lfs diff=lfs merge=lfs -text
+assets/image_69_0.png filter=lfs diff=lfs merge=lfs -text
+assets/image_6_0.png filter=lfs diff=lfs merge=lfs -text
+assets/image_70_1.png filter=lfs diff=lfs merge=lfs -text
+assets/image_71_2.png filter=lfs diff=lfs merge=lfs -text
+assets/image_72_0.png filter=lfs diff=lfs merge=lfs -text
+assets/image_73_1.png filter=lfs diff=lfs merge=lfs -text
+assets/image_74_2.png filter=lfs diff=lfs merge=lfs -text
+assets/image_75_0.png filter=lfs diff=lfs merge=lfs -text
+assets/image_76_1.png filter=lfs diff=lfs merge=lfs -text
+assets/image_77_2.png filter=lfs diff=lfs merge=lfs -text
+assets/image_78_0.png filter=lfs diff=lfs merge=lfs -text
+assets/image_79_1.png filter=lfs diff=lfs merge=lfs -text
+assets/image_7_1.png filter=lfs diff=lfs merge=lfs -text
+assets/image_80_2.png filter=lfs diff=lfs merge=lfs -text
+assets/image_81_0.png filter=lfs diff=lfs merge=lfs -text
+assets/image_82_1.png filter=lfs diff=lfs merge=lfs -text
+assets/image_83_2.png filter=lfs diff=lfs merge=lfs -text
+assets/image_84_0.png filter=lfs diff=lfs merge=lfs -text
+assets/image_85_1.png filter=lfs diff=lfs merge=lfs -text
+assets/image_87_0.png filter=lfs diff=lfs merge=lfs -text
+assets/image_88_1.png filter=lfs diff=lfs merge=lfs -text
+assets/image_89_2.png filter=lfs diff=lfs merge=lfs -text
+assets/image_8_2.png filter=lfs diff=lfs merge=lfs -text
+assets/image_90_0.png filter=lfs diff=lfs merge=lfs -text
+assets/image_91_1.png filter=lfs diff=lfs merge=lfs -text
+assets/image_92_2.png filter=lfs diff=lfs merge=lfs -text
+assets/image_93_0.png filter=lfs diff=lfs merge=lfs -text
+assets/image_94_1.png filter=lfs diff=lfs merge=lfs -text
+assets/image_95_2.png filter=lfs diff=lfs merge=lfs -text
+assets/image_96_0.png filter=lfs diff=lfs merge=lfs -text
+assets/image_97_1.png filter=lfs diff=lfs merge=lfs -text
+assets/image_98_2.png filter=lfs diff=lfs merge=lfs -text
+assets/image_99_0.png filter=lfs diff=lfs merge=lfs -text
+assets/image_9_0.png filter=lfs diff=lfs merge=lfs -text
+training_state-photo-concept-bucket.json filter=lfs diff=lfs merge=lfs -text
+assets/image_173_2.png filter=lfs diff=lfs merge=lfs -text
+assets/image_190_1.png filter=lfs diff=lfs merge=lfs -text
+assets/image_86_2.png filter=lfs diff=lfs merge=lfs -text
+training_state-dalle3.json filter=lfs diff=lfs merge=lfs -text
+training_state-nijijourney-v6-520k-raw.json filter=lfs diff=lfs merge=lfs -text
+training_state-midjourney-v6-520k-raw.json filter=lfs diff=lfs merge=lfs -text
 assets/image_192_0.png filter=lfs diff=lfs merge=lfs -text
 assets/image_193_1.png filter=lfs diff=lfs merge=lfs -text
 assets/image_194_2.png filter=lfs diff=lfs merge=lfs -text
@@ -136,8 +237,6 @@ assets/image_196_1.png filter=lfs diff=lfs merge=lfs -text
 assets/image_197_2.png filter=lfs diff=lfs merge=lfs -text
 assets/image_198_0.png filter=lfs diff=lfs merge=lfs -text
 assets/image_199_1.png filter=lfs diff=lfs merge=lfs -text
-assets/image_19_1.png filter=lfs diff=lfs merge=lfs -text
-assets/image_1_1.png filter=lfs diff=lfs merge=lfs -text
 assets/image_200_2.png filter=lfs diff=lfs merge=lfs -text
 assets/image_201_0.png filter=lfs diff=lfs merge=lfs -text
 assets/image_202_1.png filter=lfs diff=lfs merge=lfs -text
@@ -146,8 +245,8 @@ assets/image_204_0.png filter=lfs diff=lfs merge=lfs -text
 assets/image_205_1.png filter=lfs diff=lfs merge=lfs -text
 assets/image_206_2.png filter=lfs diff=lfs merge=lfs -text
 assets/image_207_0.png filter=lfs diff=lfs merge=lfs -text
+assets/image_208_1.png filter=lfs diff=lfs merge=lfs -text
 assets/image_209_2.png filter=lfs diff=lfs merge=lfs -text
-assets/image_20_2.png filter=lfs diff=lfs merge=lfs -text
 assets/image_210_0.png filter=lfs diff=lfs merge=lfs -text
 assets/image_211_1.png filter=lfs diff=lfs merge=lfs -text
 assets/image_212_2.png filter=lfs diff=lfs merge=lfs -text
@@ -158,7 +257,6 @@ assets/image_216_0.png filter=lfs diff=lfs merge=lfs -text
 assets/image_217_1.png filter=lfs diff=lfs merge=lfs -text
 assets/image_218_2.png filter=lfs diff=lfs merge=lfs -text
 assets/image_219_0.png filter=lfs diff=lfs merge=lfs -text
-assets/image_21_0.png filter=lfs diff=lfs merge=lfs -text
 assets/image_220_1.png filter=lfs diff=lfs merge=lfs -text
 assets/image_221_2.png filter=lfs diff=lfs merge=lfs -text
 assets/image_222_0.png filter=lfs diff=lfs merge=lfs -text
@@ -169,7 +267,6 @@ assets/image_226_1.png filter=lfs diff=lfs merge=lfs -text
 assets/image_227_2.png filter=lfs diff=lfs merge=lfs -text
 assets/image_228_0.png filter=lfs diff=lfs merge=lfs -text
 assets/image_229_1.png filter=lfs diff=lfs merge=lfs -text
-assets/image_22_1.png filter=lfs diff=lfs merge=lfs -text
 assets/image_230_2.png filter=lfs diff=lfs merge=lfs -text
 assets/image_231_0.png filter=lfs diff=lfs merge=lfs -text
 assets/image_232_1.png filter=lfs diff=lfs merge=lfs -text
@@ -180,7 +277,6 @@ assets/image_236_2.png filter=lfs diff=lfs merge=lfs -text
 assets/image_237_0.png filter=lfs diff=lfs merge=lfs -text
 assets/image_238_1.png filter=lfs diff=lfs merge=lfs -text
 assets/image_239_2.png filter=lfs diff=lfs merge=lfs -text
-assets/image_23_2.png filter=lfs diff=lfs merge=lfs -text
 assets/image_240_0.png filter=lfs diff=lfs merge=lfs -text
 assets/image_241_1.png filter=lfs diff=lfs merge=lfs -text
 assets/image_242_2.png filter=lfs diff=lfs merge=lfs -text
@@ -191,7 +287,6 @@ assets/image_246_0.png filter=lfs diff=lfs merge=lfs -text
 assets/image_247_1.png filter=lfs diff=lfs merge=lfs -text
 assets/image_248_2.png filter=lfs diff=lfs merge=lfs -text
 assets/image_249_0.png filter=lfs diff=lfs merge=lfs -text
-assets/image_24_0.png filter=lfs diff=lfs merge=lfs -text
 assets/image_250_1.png filter=lfs diff=lfs merge=lfs -text
 assets/image_251_2.png filter=lfs diff=lfs merge=lfs -text
 assets/image_252_0.png filter=lfs diff=lfs merge=lfs -text
@@ -202,7 +297,6 @@ assets/image_256_1.png filter=lfs diff=lfs merge=lfs -text
 assets/image_257_2.png filter=lfs diff=lfs merge=lfs -text
 assets/image_258_0.png filter=lfs diff=lfs merge=lfs -text
 assets/image_259_1.png filter=lfs diff=lfs merge=lfs -text
-assets/image_25_1.png filter=lfs diff=lfs merge=lfs -text
 assets/image_260_2.png filter=lfs diff=lfs merge=lfs -text
 assets/image_261_0.png filter=lfs diff=lfs merge=lfs -text
 assets/image_262_1.png filter=lfs diff=lfs merge=lfs -text
@@ -213,7 +307,6 @@ assets/image_266_2.png filter=lfs diff=lfs merge=lfs -text
 assets/image_267_0.png filter=lfs diff=lfs merge=lfs -text
 assets/image_268_1.png filter=lfs diff=lfs merge=lfs -text
 assets/image_269_2.png filter=lfs diff=lfs merge=lfs -text
-assets/image_26_2.png filter=lfs diff=lfs merge=lfs -text
 assets/image_270_0.png filter=lfs diff=lfs merge=lfs -text
 assets/image_271_1.png filter=lfs diff=lfs merge=lfs -text
 assets/image_272_2.png filter=lfs diff=lfs merge=lfs -text
@@ -224,104 +317,5 @@ assets/image_276_0.png filter=lfs diff=lfs merge=lfs -text
 assets/image_277_1.png filter=lfs diff=lfs merge=lfs -text
 assets/image_278_2.png filter=lfs diff=lfs merge=lfs -text
 assets/image_279_0.png filter=lfs diff=lfs merge=lfs -text
-assets/image_27_0.png filter=lfs diff=lfs merge=lfs -text
 assets/image_280_1.png filter=lfs diff=lfs merge=lfs -text
 assets/image_281_2.png filter=lfs diff=lfs merge=lfs -text
-assets/image_282_0.png filter=lfs diff=lfs merge=lfs -text
-assets/image_283_1.png filter=lfs diff=lfs merge=lfs -text
-assets/image_284_2.png filter=lfs diff=lfs merge=lfs -text
-assets/image_285_0.png filter=lfs diff=lfs merge=lfs -text
-assets/image_286_1.png filter=lfs diff=lfs merge=lfs -text
-assets/image_287_2.png filter=lfs diff=lfs merge=lfs -text
-assets/image_288_0.png filter=lfs diff=lfs merge=lfs -text
-assets/image_289_1.png filter=lfs diff=lfs merge=lfs -text
-assets/image_28_1.png filter=lfs diff=lfs merge=lfs -text
-assets/image_290_2.png filter=lfs diff=lfs merge=lfs -text
-assets/image_291_0.png filter=lfs diff=lfs merge=lfs -text
-assets/image_292_1.png filter=lfs diff=lfs merge=lfs -text
-assets/image_293_2.png filter=lfs diff=lfs merge=lfs -text
-assets/image_294_0.png filter=lfs diff=lfs merge=lfs -text
-assets/image_295_1.png filter=lfs diff=lfs merge=lfs -text
-assets/image_296_2.png filter=lfs diff=lfs merge=lfs -text
-assets/image_297_0.png filter=lfs diff=lfs merge=lfs -text
-assets/image_298_1.png filter=lfs diff=lfs merge=lfs -text
-assets/image_299_2.png filter=lfs diff=lfs merge=lfs -text
-assets/image_29_2.png filter=lfs diff=lfs merge=lfs -text
-assets/image_2_2.png filter=lfs diff=lfs merge=lfs -text
-assets/image_300_0.png filter=lfs diff=lfs merge=lfs -text
-assets/image_301_1.png filter=lfs diff=lfs merge=lfs -text
-assets/image_302_2.png filter=lfs diff=lfs merge=lfs -text
-assets/image_30_0.png filter=lfs diff=lfs merge=lfs -text
-assets/image_31_1.png filter=lfs diff=lfs merge=lfs -text
-assets/image_32_2.png filter=lfs diff=lfs merge=lfs -text
-assets/image_33_0.png filter=lfs diff=lfs merge=lfs -text
-assets/image_34_1.png filter=lfs diff=lfs merge=lfs -text
-assets/image_35_2.png filter=lfs diff=lfs merge=lfs -text
-assets/image_36_0.png filter=lfs diff=lfs merge=lfs -text
-assets/image_37_1.png filter=lfs diff=lfs merge=lfs -text
-assets/image_38_2.png filter=lfs diff=lfs merge=lfs -text
-assets/image_39_0.png filter=lfs diff=lfs merge=lfs -text
-assets/image_3_0.png filter=lfs diff=lfs merge=lfs -text
-assets/image_40_1.png filter=lfs diff=lfs merge=lfs -text
-assets/image_41_2.png filter=lfs diff=lfs merge=lfs -text
-assets/image_42_0.png filter=lfs diff=lfs merge=lfs -text
-assets/image_43_1.png filter=lfs diff=lfs merge=lfs -text
-assets/image_44_2.png filter=lfs diff=lfs merge=lfs -text
-assets/image_45_0.png filter=lfs diff=lfs merge=lfs -text
-assets/image_46_1.png filter=lfs diff=lfs merge=lfs -text
-assets/image_47_2.png filter=lfs diff=lfs merge=lfs -text
-assets/image_48_0.png filter=lfs diff=lfs merge=lfs -text
-assets/image_49_1.png filter=lfs diff=lfs merge=lfs -text
-assets/image_4_1.png filter=lfs diff=lfs merge=lfs -text
-assets/image_50_2.png filter=lfs diff=lfs merge=lfs -text
-assets/image_51_0.png filter=lfs diff=lfs merge=lfs -text
-assets/image_52_1.png filter=lfs diff=lfs merge=lfs -text
-assets/image_53_2.png filter=lfs diff=lfs merge=lfs -text
-assets/image_54_0.png filter=lfs diff=lfs merge=lfs -text
-assets/image_55_1.png filter=lfs diff=lfs merge=lfs -text
-assets/image_56_2.png filter=lfs diff=lfs merge=lfs -text
-assets/image_57_0.png filter=lfs diff=lfs merge=lfs -text
-assets/image_58_1.png filter=lfs diff=lfs merge=lfs -text
-assets/image_59_2.png filter=lfs diff=lfs merge=lfs -text
-assets/image_5_2.png filter=lfs diff=lfs merge=lfs -text
-assets/image_60_0.png filter=lfs diff=lfs merge=lfs -text
-assets/image_61_1.png filter=lfs diff=lfs merge=lfs -text
-assets/image_62_2.png filter=lfs diff=lfs merge=lfs -text
-assets/image_63_0.png filter=lfs diff=lfs merge=lfs -text
-assets/image_64_1.png filter=lfs diff=lfs merge=lfs -text
-assets/image_65_2.png filter=lfs diff=lfs merge=lfs -text
-assets/image_67_1.png filter=lfs diff=lfs merge=lfs -text
-assets/image_68_2.png filter=lfs diff=lfs merge=lfs -text
-assets/image_69_0.png filter=lfs diff=lfs merge=lfs -text
-assets/image_6_0.png filter=lfs diff=lfs merge=lfs -text
-assets/image_70_1.png filter=lfs diff=lfs merge=lfs -text
-assets/image_71_2.png filter=lfs diff=lfs merge=lfs -text
-assets/image_72_0.png filter=lfs diff=lfs merge=lfs -text
-assets/image_73_1.png filter=lfs diff=lfs merge=lfs -text
-assets/image_74_2.png filter=lfs diff=lfs merge=lfs -text
-assets/image_75_0.png filter=lfs diff=lfs merge=lfs -text
-assets/image_76_1.png filter=lfs diff=lfs merge=lfs -text
-assets/image_77_2.png filter=lfs diff=lfs merge=lfs -text
-assets/image_78_0.png filter=lfs diff=lfs merge=lfs -text
-assets/image_79_1.png filter=lfs diff=lfs merge=lfs -text
-assets/image_7_1.png filter=lfs diff=lfs merge=lfs -text
-assets/image_80_2.png filter=lfs diff=lfs merge=lfs -text
-assets/image_81_0.png filter=lfs diff=lfs merge=lfs -text
-assets/image_82_1.png filter=lfs diff=lfs merge=lfs -text
-assets/image_83_2.png filter=lfs diff=lfs merge=lfs -text
-assets/image_84_0.png filter=lfs diff=lfs merge=lfs -text
-assets/image_85_1.png filter=lfs diff=lfs merge=lfs -text
-assets/image_86_2.png filter=lfs diff=lfs merge=lfs -text
-assets/image_8_2.png filter=lfs diff=lfs merge=lfs -text
-assets/image_90_0.png filter=lfs diff=lfs merge=lfs -text
-assets/image_91_1.png filter=lfs diff=lfs merge=lfs -text
-assets/image_92_2.png filter=lfs diff=lfs merge=lfs -text
-assets/image_93_0.png filter=lfs diff=lfs merge=lfs -text
-assets/image_94_1.png filter=lfs diff=lfs merge=lfs -text
-assets/image_95_2.png filter=lfs diff=lfs merge=lfs -text
-assets/image_96_0.png filter=lfs diff=lfs merge=lfs -text
-assets/image_97_1.png filter=lfs diff=lfs merge=lfs -text
-assets/image_98_2.png filter=lfs diff=lfs merge=lfs -text
-assets/image_99_0.png filter=lfs diff=lfs merge=lfs -text
-assets/image_9_0.png filter=lfs diff=lfs merge=lfs -text
-training_state-dalle3.json filter=lfs diff=lfs merge=lfs -text

model_index.json

@@ -0,0 +1,24 @@
+{
+  "_class_name": "PixArtSigmaPipeline",
+  "_diffusers_version": "0.28.0.dev0",
+  "scheduler": [
+    "diffusers",
+    "UniPCMultistepScheduler"
+  ],
+  "text_encoder": [
+    "transformers",
+    "T5EncoderModel"
+  ],
+  "tokenizer": [
+    "transformers",
+    "T5Tokenizer"
+  ],
+  "transformer": [
+    "diffusers",
+    "Transformer2DModel"
+  ],
+  "vae": [
+    "diffusers",
+    "AutoencoderKL"
+  ]
+}

scheduler/scheduler_config.json

@@ -0,0 +1,25 @@
+{
+  "_class_name": "DPMSolverMultistepScheduler",
+  "_diffusers_version": "0.22.0.dev0",
+  "algorithm_type": "dpmsolver++",
+  "beta_end": 0.02,
+  "beta_schedule": "linear",
+  "beta_start": 0.0001,
+  "dynamic_thresholding_ratio": 0.995,
+  "euler_at_final": false,
+  "lambda_min_clipped": -Infinity,
+  "lower_order_final": true,
+  "num_train_timesteps": 1000,
+  "prediction_type": "epsilon",
+  "sample_max_value": 1.0,
+  "solver_order": 2,
+  "solver_type": "midpoint",
+  "steps_offset": 0,
+  "thresholding": false,
+  "timestep_spacing": "linspace",
+  "trained_betas": null,
+  "use_karras_sigmas": false,
+  "use_lu_lambdas": false,
+  "variance_type": null
+}
+

text_encoder/config.json

@@ -0,0 +1,31 @@
+{
+  "_name_or_path": "data/t5_ckpts",
+  "architectures": [
+    "T5EncoderModel"
+  ],
+  "d_ff": 10240,
+  "d_kv": 64,
+  "d_model": 4096,
+  "decoder_start_token_id": 0,
+  "dense_act_fn": "gelu_new",
+  "dropout_rate": 0.1,
+  "eos_token_id": 1,
+  "feed_forward_proj": "gated-gelu",
+  "initializer_factor": 1.0,
+  "is_encoder_decoder": true,
+  "is_gated_act": true,
+  "layer_norm_epsilon": 1e-06,
+  "model_type": "t5",
+  "num_decoder_layers": 24,
+  "num_heads": 64,
+  "num_layers": 24,
+  "output_past": true,
+  "pad_token_id": 0,
+  "relative_attention_max_distance": 128,
+  "relative_attention_num_buckets": 32,
+  "tie_word_embeddings": false,
+  "torch_dtype": "float32",
+  "transformers_version": "4.30.0",
+  "use_cache": true,
+  "vocab_size": 32128
+}

text_encoder/model-00001-of-00002.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:eb529f693f4b17773a24e787fcba29486d5e1700dadcc20bb91e4c8b00212d08
+size 9989150328

text_encoder/model-00002-of-00002.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:58f33f4acba7a5502b6b3e4cb7863f8fd503d5012016428099bb460c98e2fe77
+size 9060119392

text_encoder/model.safetensors.index.json

@@ -0,0 +1,226 @@
+{
+  "metadata": {
+    "total_size": 19049242624
+  },
+  "weight_map": {
+    "encoder.block.0.layer.0.SelfAttention.k.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.0.layer.0.SelfAttention.o.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.0.layer.0.SelfAttention.q.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.0.layer.0.SelfAttention.relative_attention_bias.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.0.layer.0.SelfAttention.v.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.0.layer.0.layer_norm.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.0.layer.1.DenseReluDense.wi_0.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.0.layer.1.DenseReluDense.wi_1.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.0.layer.1.DenseReluDense.wo.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.0.layer.1.layer_norm.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.1.layer.0.SelfAttention.k.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.1.layer.0.SelfAttention.o.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.1.layer.0.SelfAttention.q.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.1.layer.0.SelfAttention.v.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.1.layer.0.layer_norm.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.1.layer.1.DenseReluDense.wi_0.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.1.layer.1.DenseReluDense.wi_1.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.1.layer.1.DenseReluDense.wo.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.1.layer.1.layer_norm.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.10.layer.0.SelfAttention.k.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.10.layer.0.SelfAttention.o.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.10.layer.0.SelfAttention.q.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.10.layer.0.SelfAttention.v.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.10.layer.0.layer_norm.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.10.layer.1.DenseReluDense.wi_0.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.10.layer.1.DenseReluDense.wi_1.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.10.layer.1.DenseReluDense.wo.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.10.layer.1.layer_norm.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.11.layer.0.SelfAttention.k.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.11.layer.0.SelfAttention.o.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.11.layer.0.SelfAttention.q.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.11.layer.0.SelfAttention.v.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.11.layer.0.layer_norm.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.11.layer.1.DenseReluDense.wi_0.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.11.layer.1.DenseReluDense.wi_1.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.11.layer.1.DenseReluDense.wo.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.11.layer.1.layer_norm.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.12.layer.0.SelfAttention.k.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.12.layer.0.SelfAttention.o.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.12.layer.0.SelfAttention.q.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.12.layer.0.SelfAttention.v.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.12.layer.0.layer_norm.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.12.layer.1.DenseReluDense.wi_0.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.12.layer.1.DenseReluDense.wi_1.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.12.layer.1.DenseReluDense.wo.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.12.layer.1.layer_norm.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.13.layer.0.SelfAttention.k.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.13.layer.0.SelfAttention.o.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.13.layer.0.SelfAttention.q.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.13.layer.0.SelfAttention.v.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.13.layer.0.layer_norm.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.13.layer.1.DenseReluDense.wi_0.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.13.layer.1.DenseReluDense.wi_1.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.13.layer.1.DenseReluDense.wo.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.13.layer.1.layer_norm.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.14.layer.0.SelfAttention.k.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.14.layer.0.SelfAttention.o.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.14.layer.0.SelfAttention.q.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.14.layer.0.SelfAttention.v.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.14.layer.0.layer_norm.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.14.layer.1.DenseReluDense.wi_0.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.14.layer.1.DenseReluDense.wi_1.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.14.layer.1.DenseReluDense.wo.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.14.layer.1.layer_norm.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.15.layer.0.SelfAttention.k.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.15.layer.0.SelfAttention.o.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.15.layer.0.SelfAttention.q.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.15.layer.0.SelfAttention.v.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.15.layer.0.layer_norm.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.15.layer.1.DenseReluDense.wi_0.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.15.layer.1.DenseReluDense.wi_1.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.15.layer.1.DenseReluDense.wo.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.15.layer.1.layer_norm.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.16.layer.0.SelfAttention.k.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.16.layer.0.SelfAttention.o.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.16.layer.0.SelfAttention.q.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.16.layer.0.SelfAttention.v.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.16.layer.0.layer_norm.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.16.layer.1.DenseReluDense.wi_0.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.16.layer.1.DenseReluDense.wi_1.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.16.layer.1.DenseReluDense.wo.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.16.layer.1.layer_norm.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.17.layer.0.SelfAttention.k.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.17.layer.0.SelfAttention.o.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.17.layer.0.SelfAttention.q.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.17.layer.0.SelfAttention.v.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.17.layer.0.layer_norm.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.17.layer.1.DenseReluDense.wi_0.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.17.layer.1.DenseReluDense.wi_1.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.17.layer.1.DenseReluDense.wo.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.17.layer.1.layer_norm.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.18.layer.0.SelfAttention.k.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.18.layer.0.SelfAttention.o.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.18.layer.0.SelfAttention.q.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.18.layer.0.SelfAttention.v.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.18.layer.0.layer_norm.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.18.layer.1.DenseReluDense.wi_0.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.18.layer.1.DenseReluDense.wi_1.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.18.layer.1.DenseReluDense.wo.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.18.layer.1.layer_norm.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.19.layer.0.SelfAttention.k.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.19.layer.0.SelfAttention.o.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.19.layer.0.SelfAttention.q.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.19.layer.0.SelfAttention.v.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.19.layer.0.layer_norm.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.19.layer.1.DenseReluDense.wi_0.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.19.layer.1.DenseReluDense.wi_1.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.19.layer.1.DenseReluDense.wo.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.19.layer.1.layer_norm.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.2.layer.0.SelfAttention.k.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.2.layer.0.SelfAttention.o.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.2.layer.0.SelfAttention.q.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.2.layer.0.SelfAttention.v.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.2.layer.0.layer_norm.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.2.layer.1.DenseReluDense.wi_0.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.2.layer.1.DenseReluDense.wi_1.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.2.layer.1.DenseReluDense.wo.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.2.layer.1.layer_norm.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.20.layer.0.SelfAttention.k.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.20.layer.0.SelfAttention.o.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.20.layer.0.SelfAttention.q.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.20.layer.0.SelfAttention.v.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.20.layer.0.layer_norm.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.20.layer.1.DenseReluDense.wi_0.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.20.layer.1.DenseReluDense.wi_1.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.20.layer.1.DenseReluDense.wo.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.20.layer.1.layer_norm.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.21.layer.0.SelfAttention.k.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.21.layer.0.SelfAttention.o.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.21.layer.0.SelfAttention.q.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.21.layer.0.SelfAttention.v.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.21.layer.0.layer_norm.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.21.layer.1.DenseReluDense.wi_0.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.21.layer.1.DenseReluDense.wi_1.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.21.layer.1.DenseReluDense.wo.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.21.layer.1.layer_norm.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.22.layer.0.SelfAttention.k.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.22.layer.0.SelfAttention.o.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.22.layer.0.SelfAttention.q.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.22.layer.0.SelfAttention.v.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.22.layer.0.layer_norm.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.22.layer.1.DenseReluDense.wi_0.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.22.layer.1.DenseReluDense.wi_1.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.22.layer.1.DenseReluDense.wo.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.22.layer.1.layer_norm.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.23.layer.0.SelfAttention.k.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.23.layer.0.SelfAttention.o.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.23.layer.0.SelfAttention.q.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.23.layer.0.SelfAttention.v.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.23.layer.0.layer_norm.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.23.layer.1.DenseReluDense.wi_0.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.23.layer.1.DenseReluDense.wi_1.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.23.layer.1.DenseReluDense.wo.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.23.layer.1.layer_norm.weight": "model-00002-of-00002.safetensors",
+    "encoder.block.3.layer.0.SelfAttention.k.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.3.layer.0.SelfAttention.o.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.3.layer.0.SelfAttention.q.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.3.layer.0.SelfAttention.v.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.3.layer.0.layer_norm.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.3.layer.1.DenseReluDense.wi_0.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.3.layer.1.DenseReluDense.wi_1.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.3.layer.1.DenseReluDense.wo.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.3.layer.1.layer_norm.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.4.layer.0.SelfAttention.k.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.4.layer.0.SelfAttention.o.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.4.layer.0.SelfAttention.q.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.4.layer.0.SelfAttention.v.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.4.layer.0.layer_norm.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.4.layer.1.DenseReluDense.wi_0.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.4.layer.1.DenseReluDense.wi_1.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.4.layer.1.DenseReluDense.wo.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.4.layer.1.layer_norm.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.5.layer.0.SelfAttention.k.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.5.layer.0.SelfAttention.o.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.5.layer.0.SelfAttention.q.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.5.layer.0.SelfAttention.v.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.5.layer.0.layer_norm.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.5.layer.1.DenseReluDense.wi_0.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.5.layer.1.DenseReluDense.wi_1.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.5.layer.1.DenseReluDense.wo.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.5.layer.1.layer_norm.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.6.layer.0.SelfAttention.k.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.6.layer.0.SelfAttention.o.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.6.layer.0.SelfAttention.q.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.6.layer.0.SelfAttention.v.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.6.layer.0.layer_norm.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.6.layer.1.DenseReluDense.wi_0.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.6.layer.1.DenseReluDense.wi_1.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.6.layer.1.DenseReluDense.wo.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.6.layer.1.layer_norm.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.7.layer.0.SelfAttention.k.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.7.layer.0.SelfAttention.o.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.7.layer.0.SelfAttention.q.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.7.layer.0.SelfAttention.v.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.7.layer.0.layer_norm.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.7.layer.1.DenseReluDense.wi_0.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.7.layer.1.DenseReluDense.wi_1.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.7.layer.1.DenseReluDense.wo.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.7.layer.1.layer_norm.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.8.layer.0.SelfAttention.k.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.8.layer.0.SelfAttention.o.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.8.layer.0.SelfAttention.q.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.8.layer.0.SelfAttention.v.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.8.layer.0.layer_norm.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.8.layer.1.DenseReluDense.wi_0.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.8.layer.1.DenseReluDense.wi_1.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.8.layer.1.DenseReluDense.wo.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.8.layer.1.layer_norm.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.9.layer.0.SelfAttention.k.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.9.layer.0.SelfAttention.o.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.9.layer.0.SelfAttention.q.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.9.layer.0.SelfAttention.v.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.9.layer.0.layer_norm.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.9.layer.1.DenseReluDense.wi_0.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.9.layer.1.DenseReluDense.wi_1.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.9.layer.1.DenseReluDense.wo.weight": "model-00001-of-00002.safetensors",
+    "encoder.block.9.layer.1.layer_norm.weight": "model-00001-of-00002.safetensors",
+    "encoder.final_layer_norm.weight": "model-00002-of-00002.safetensors",
+    "shared.weight": "model-00001-of-00002.safetensors"
+  }
+}

tokenizer/special_tokens_map.json

@@ -0,0 +1,107 @@
+{
+  "additional_special_tokens": [
+    "<extra_id_0>",
+    "<extra_id_1>",
+    "<extra_id_2>",
+    "<extra_id_3>",
+    "<extra_id_4>",
+    "<extra_id_5>",
+    "<extra_id_6>",
+    "<extra_id_7>",
+    "<extra_id_8>",
+    "<extra_id_9>",
+    "<extra_id_10>",
+    "<extra_id_11>",
+    "<extra_id_12>",
+    "<extra_id_13>",
+    "<extra_id_14>",
+    "<extra_id_15>",
+    "<extra_id_16>",
+    "<extra_id_17>",
+    "<extra_id_18>",
+    "<extra_id_19>",
+    "<extra_id_20>",
+    "<extra_id_21>",
+    "<extra_id_22>",
+    "<extra_id_23>",
+    "<extra_id_24>",
+    "<extra_id_25>",
+    "<extra_id_26>",
+    "<extra_id_27>",
+    "<extra_id_28>",
+    "<extra_id_29>",
+    "<extra_id_30>",
+    "<extra_id_31>",
+    "<extra_id_32>",
+    "<extra_id_33>",
+    "<extra_id_34>",
+    "<extra_id_35>",
+    "<extra_id_36>",
+    "<extra_id_37>",
+    "<extra_id_38>",
+    "<extra_id_39>",
+    "<extra_id_40>",
+    "<extra_id_41>",
+    "<extra_id_42>",
+    "<extra_id_43>",
+    "<extra_id_44>",
+    "<extra_id_45>",
+    "<extra_id_46>",
+    "<extra_id_47>",
+    "<extra_id_48>",
+    "<extra_id_49>",
+    "<extra_id_50>",
+    "<extra_id_51>",
+    "<extra_id_52>",
+    "<extra_id_53>",
+    "<extra_id_54>",
+    "<extra_id_55>",
+    "<extra_id_56>",
+    "<extra_id_57>",
+    "<extra_id_58>",
+    "<extra_id_59>",
+    "<extra_id_60>",
+    "<extra_id_61>",
+    "<extra_id_62>",
+    "<extra_id_63>",
+    "<extra_id_64>",
+    "<extra_id_65>",
+    "<extra_id_66>",
+    "<extra_id_67>",
+    "<extra_id_68>",
+    "<extra_id_69>",
+    "<extra_id_70>",
+    "<extra_id_71>",
+    "<extra_id_72>",
+    "<extra_id_73>",
+    "<extra_id_74>",
+    "<extra_id_75>",
+    "<extra_id_76>",
+    "<extra_id_77>",
+    "<extra_id_78>",
+    "<extra_id_79>",
+    "<extra_id_80>",
+    "<extra_id_81>",
+    "<extra_id_82>",
+    "<extra_id_83>",
+    "<extra_id_84>",
+    "<extra_id_85>",
+    "<extra_id_86>",
+    "<extra_id_87>",
+    "<extra_id_88>",
+    "<extra_id_89>",
+    "<extra_id_90>",
+    "<extra_id_91>",
+    "<extra_id_92>",
+    "<extra_id_93>",
+    "<extra_id_94>",
+    "<extra_id_95>",
+    "<extra_id_96>",
+    "<extra_id_97>",
+    "<extra_id_98>",
+    "<extra_id_99>"
+  ],
+  "eos_token": "</s>",
+  "pad_token": "<pad>",
+  "unk_token": "<unk>"
+}

tokenizer/spiece.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d60acb128cf7b7f2536e8f38a5b18a05535c9e14c7a355904270e15b0945ea86
+size 791656

tokenizer/tokenizer_config.json

@@ -0,0 +1,112 @@
+{
+  "additional_special_tokens": [
+    "<extra_id_0>",
+    "<extra_id_1>",
+    "<extra_id_2>",
+    "<extra_id_3>",
+    "<extra_id_4>",
+    "<extra_id_5>",
+    "<extra_id_6>",
+    "<extra_id_7>",
+    "<extra_id_8>",
+    "<extra_id_9>",
+    "<extra_id_10>",
+    "<extra_id_11>",
+    "<extra_id_12>",
+    "<extra_id_13>",
+    "<extra_id_14>",
+    "<extra_id_15>",
+    "<extra_id_16>",
+    "<extra_id_17>",
+    "<extra_id_18>",
+    "<extra_id_19>",
+    "<extra_id_20>",
+    "<extra_id_21>",
+    "<extra_id_22>",
+    "<extra_id_23>",
+    "<extra_id_24>",
+    "<extra_id_25>",
+    "<extra_id_26>",
+    "<extra_id_27>",
+    "<extra_id_28>",
+    "<extra_id_29>",
+    "<extra_id_30>",
+    "<extra_id_31>",
+    "<extra_id_32>",
+    "<extra_id_33>",
+    "<extra_id_34>",
+    "<extra_id_35>",
+    "<extra_id_36>",
+    "<extra_id_37>",
+    "<extra_id_38>",
+    "<extra_id_39>",
+    "<extra_id_40>",
+    "<extra_id_41>",
+    "<extra_id_42>",
+    "<extra_id_43>",
+    "<extra_id_44>",
+    "<extra_id_45>",
+    "<extra_id_46>",
+    "<extra_id_47>",
+    "<extra_id_48>",
+    "<extra_id_49>",
+    "<extra_id_50>",
+    "<extra_id_51>",
+    "<extra_id_52>",
+    "<extra_id_53>",
+    "<extra_id_54>",
+    "<extra_id_55>",
+    "<extra_id_56>",
+    "<extra_id_57>",
+    "<extra_id_58>",
+    "<extra_id_59>",
+    "<extra_id_60>",
+    "<extra_id_61>",
+    "<extra_id_62>",
+    "<extra_id_63>",
+    "<extra_id_64>",
+    "<extra_id_65>",
+    "<extra_id_66>",
+    "<extra_id_67>",
+    "<extra_id_68>",
+    "<extra_id_69>",
+    "<extra_id_70>",
+    "<extra_id_71>",
+    "<extra_id_72>",
+    "<extra_id_73>",
+    "<extra_id_74>",
+    "<extra_id_75>",
+    "<extra_id_76>",
+    "<extra_id_77>",
+    "<extra_id_78>",
+    "<extra_id_79>",
+    "<extra_id_80>",
+    "<extra_id_81>",
+    "<extra_id_82>",
+    "<extra_id_83>",
+    "<extra_id_84>",
+    "<extra_id_85>",
+    "<extra_id_86>",
+    "<extra_id_87>",
+    "<extra_id_88>",
+    "<extra_id_89>",
+    "<extra_id_90>",
+    "<extra_id_91>",
+    "<extra_id_92>",
+    "<extra_id_93>",
+    "<extra_id_94>",
+    "<extra_id_95>",
+    "<extra_id_96>",
+    "<extra_id_97>",
+    "<extra_id_98>",
+    "<extra_id_99>"
+  ],
+  "clean_up_tokenization_spaces": true,
+  "eos_token": "</s>",
+  "extra_ids": 100,
+  "model_max_length": 512,
+  "pad_token": "<pad>",
+  "sp_model_kwargs": {},
+  "tokenizer_class": "T5Tokenizer",
+  "unk_token": "<unk>"
+}

vae/config.json

@@ -0,0 +1,32 @@
+{
+  "_class_name": "AutoencoderKL",
+  "_diffusers_version": "0.18.0.dev0",
+  "_name_or_path": ".",
+  "act_fn": "silu",
+  "block_out_channels": [
+    128,
+    256,
+    512,
+    512
+  ],
+  "down_block_types": [
+    "DownEncoderBlock2D",
+    "DownEncoderBlock2D",
+    "DownEncoderBlock2D",
+    "DownEncoderBlock2D"
+  ],
+  "in_channels": 3,
+  "latent_channels": 4,
+  "layers_per_block": 2,
+  "norm_num_groups": 32,
+  "out_channels": 3,
+  "sample_size": 512,
+  "scaling_factor": 0.13025,
+  "up_block_types": [
+    "UpDecoderBlock2D",
+    "UpDecoderBlock2D",
+    "UpDecoderBlock2D",
+    "UpDecoderBlock2D"
+  ],
+  "force_upcast": false
+}

vae/diffusion_pytorch_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1b909373b28f2137098b0fd9dbc6f97f8410854f31f84ddc9fa04b077b0ace2c
+size 334643238