cache-dit 0.2.4__py3-none-any.whl → 0.2.6__py3-none-any.whl

{cache_dit-0.2.4.dist-info → cache_dit-0.2.6.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: cache_dit
-Version: 0.2.4
+Version: 0.2.6
 Summary: 🤗 CacheDiT: A Training-free and Easy-to-use Cache Acceleration Toolbox for Diffusion Transformers
 Author: DefTruth, vipshop.com, etc.
 Maintainer: DefTruth, vipshop.com, etc
@@ -13,6 +13,8 @@ Requires-Dist: packaging
 Requires-Dist: torch>=2.5.1
 Requires-Dist: transformers>=4.51.3
 Requires-Dist: diffusers>=0.33.1
+Requires-Dist: scikit-image
+Requires-Dist: scipy
 Provides-Extra: all
 Provides-Extra: dev
 Requires-Dist: pre-commit; extra == "dev"
@@ -28,6 +30,8 @@ Requires-Dist: protobuf; extra == "dev"
 Requires-Dist: sentencepiece; extra == "dev"
 Requires-Dist: opencv-python-headless; extra == "dev"
 Requires-Dist: ftfy; extra == "dev"
+Requires-Dist: scikit-image; extra == "dev"
+Requires-Dist: pytorch-fid; extra == "dev"
 Dynamic: license-file
 Dynamic: provides-extra
 Dynamic: requires-dist
@@ -44,7 +48,7 @@ Dynamic: requires-python
       <img src=https://img.shields.io/badge/PyPI-pass-brightgreen.svg >
       <img src=https://static.pepy.tech/badge/cache-dit >
       <img src=https://img.shields.io/badge/Python-3.10|3.11|3.12-9cf.svg >
-      <img src=https://img.shields.io/badge/Release-v0.2.2-brightgreen.svg >
+      <img src=https://img.shields.io/badge/Release-v0.2-brightgreen.svg >
  </div>
   <p align="center">
     DeepCache is for UNet not DiT. Most DiT cache speedups are complex and not training-free. CacheDiT offers <br>a set of training-free cache accelerators for DiT: <b>🔥<a href="#dbcache">DBCache</a>, <a href="#dbprune">DBPrune</a>, <a href="#taylorseer">TaylorSeer</a>, <a href="#fbcache">FBCache</a></b>, etc🔥
@@ -159,6 +163,7 @@ The **CacheDiT** codebase is adapted from [FBCache](https://github.com/chengzeyi
 - [⚡️Dynamic Block Prune](#dbprune)
 - [🎉Context Parallelism](#context-parallelism)  
 - [🔥Torch Compile](#compile)
+- [⚙️Metrics CLI](#metrics)
 - [👋Contribute](#contribute)
 - [©️License](#license)
 
@@ -169,7 +174,7 @@ The **CacheDiT** codebase is adapted from [FBCache](https://github.com/chengzeyi
 You can install the stable release of `cache-dit` from PyPI:
 
 ```bash
-pip3 install cache-dit
+pip3 install -U cache-dit
 ```
 Or you can install the latest develop version from GitHub:
 
@@ -181,11 +186,13 @@ pip3 install git+https://github.com/vipshop/cache-dit.git
 
 <div id="supported"></div>
 
-- [🚀FLUX.1](https://github.com/vipshop/cache-dit/raw/main/examples)  
-- [🚀Mochi](https://github.com/vipshop/cache-dit/raw/main/examples)
+- [🚀FLUX.1-dev](https://github.com/vipshop/cache-dit/raw/main/examples)  
+- [🚀FLUX.1-Fill-dev](https://github.com/vipshop/cache-dit/raw/main/examples)  
+- [🚀mochi-1-preview](https://github.com/vipshop/cache-dit/raw/main/examples)
 - [🚀CogVideoX](https://github.com/vipshop/cache-dit/raw/main/examples)
 - [🚀CogVideoX1.5](https://github.com/vipshop/cache-dit/raw/main/examples)
-- [🚀Wan2.1](https://github.com/vipshop/cache-dit/raw/main/examples)
+- [🚀Wan2.1-T2V](https://github.com/vipshop/cache-dit/raw/main/examples)
+- [🚀Wan2.1-FLF2V](https://github.com/vipshop/cache-dit/raw/main/examples)
 - [🚀HunyuanVideo](https://github.com/vipshop/cache-dit/raw/main/examples)
 
 
@@ -281,7 +288,7 @@ cache_options = {
     "taylorseer_kwargs": {
         "n_derivatives": 2, # default is 2.
     },
-    "warmup_steps": 3, # n_derivatives + 1
+    "warmup_steps": 3, # prefer: >= n_derivatives + 1
     "residual_diff_threshold": 0.12,
 }
 ``` 
@@ -304,12 +311,23 @@ cache_options = {
 
 <div id="cfg"></div>
 
-CacheDiT supports caching for CFG (classifier-free guidance). For models that fuse CFG and non-CFG into a single forward step, or models that do not include CFG (classifier-free guidance) in the forward step, please set `do_separate_classifier_free_guidance` param to False. Otherwise, set it to True. Wan 2.1: True. FLUX.1, HunyunVideo, CogVideoX, Mochi: False. 
+CacheDiT supports caching for **CFG (classifier-free guidance)**. For models that fuse CFG and non-CFG into a single forward step, or models that do not include CFG (classifier-free guidance) in the forward step, please set `do_separate_classifier_free_guidance` param to **False (default)**. Otherwise, set it to True. For examples:
 
 ```python
 cache_options = {
+    # CFG: classifier free guidance or not
+    # For model that fused CFG and non-CFG into single forward step,
+    # should set do_separate_classifier_free_guidance as False.
+    # For example, set it as True for Wan 2.1 and set it as False 
+    # for FLUX.1, HunyuanVideo, CogVideoX, Mochi.
     "do_separate_classifier_free_guidance": True,  # Wan 2.1
+    # Compute cfg forward first or not, default False, namely, 
+    # 0, 2, 4, ..., -> non-CFG step; 1, 3, 5, ... -> CFG step.
     "cfg_compute_first": False,
+    # Compute spearate diff values for CFG and non-CFG step, 
+    # default True. If False, we will use the computed diff from 
+    # current non-CFG transformer step for current CFG step.
+    "cfg_diff_compute_separate": True,
 }
 ```
 
@@ -487,6 +505,39 @@ torch._dynamo.config.accumulated_recompile_limit = 2048  # default is 256
 
 Please check [bench.py](./bench/bench.py) for more details.
 
+
+## ⚙️Metrics CLI
+
+<div id="metrics"></div>    
+
+You can utilize the APIs provided by CacheDiT to quickly evaluate the accuracy losses caused by different cache configurations. For example:
+
+```python
+from cache_dit.metrics import compute_psnr
+from cache_dit.metrics import compute_video_psnr
+from cache_dit.metrics import FrechetInceptionDistance  # FID
+
+FID = FrechetInceptionDistance()
+image_psnr, n = compute_psnr("true.png", "test.png") # Num: n
+image_fid,  n = FID.compute_fid("true.png", "test.png")
+video_psnr, n = compute_video_psnr("true.mp4", "test.mp4")
+```
+
+Please check [test_metrics.py](./tests/test_metrics.py) for more details. Or, you can use `cache-dit-metrics-cli` tool. For examples: 
+
+```bash
+cache-dit-metrics-cli -h  # show usage
+cache-dit-metrics-cli all -v1 true.mp4 -v2 test.mp4  # compare video
+cache-dit-metrics-cli all -i1 true.png -i2 test.png  # compare image
+cache-dit-metrics-cli all -i1 true_dir -i2 test_dir  # compare image dir
+cache-dit-metrics-cli all -i1 BASELINE -i2 OPTIMIZED # compare image dir
+
+INFO 07-09 20:59:40 [metrics.py:295] BASELINE vs OPTIMIZED, Num: 1000, PSNR: 38.742413478199005
+INFO 07-09 21:00:32 [metrics.py:300] BASELINE vs OPTIMIZED, Num: 1000, SSIM: 0.9863484896791567
+INFO 07-09 21:00:45 [metrics.py:305] BASELINE vs OPTIMIZED, Num: 1000,  MSE: 12.287594770695606
+INFO 07-09 21:01:04 [metrics.py:311] BASELINE vs OPTIMIZED, Num: 1000,  FID: 5.983550108647762
+```
+
 ## 👋Contribute 
 <div id="contribute"></div>
 

{cache_dit-0.2.4.dist-info → cache_dit-0.2.6.dist-info}/RECORD

@@ -1,12 +1,12 @@
 cache_dit/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-cache_dit/_version.py,sha256=1LUN_sRKOiFInoB6AlW6TYoQMCh1Z4KutwcHNvHcfB0,511
+cache_dit/_version.py,sha256=nObnONsicQ3YX6SG5MVBxmIp5dmRacXDauSqZijWQbY,511
 cache_dit/logger.py,sha256=0zsu42hN-3-rgGC_C29ms1IvVpV4_b4_SwJCKSenxBE,4304
 cache_dit/primitives.py,sha256=A2iG9YLot3gOsZSPp-_gyjqjLgJvWQRx8aitD4JQ23Y,3877
 cache_dit/cache_factory/__init__.py,sha256=5RNuhWakvvqrOV4vkqrEBA7d-V1LwcNSsjtW14mkqK8,5255
 cache_dit/cache_factory/taylorseer.py,sha256=LKSNo2ode69EVo9xrxjxAMEjz0yDGiGADeDYnEqddA8,3987
 cache_dit/cache_factory/utils.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 cache_dit/cache_factory/dual_block_cache/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-cache_dit/cache_factory/dual_block_cache/cache_context.py,sha256=H7u5zIAdEjiYU0QvWYIMj3lKYI4D8cmDLy7eZ9tyoyU,66848
+cache_dit/cache_factory/dual_block_cache/cache_context.py,sha256=wE_xYp7DRbgB-fD8dpr75o4Cvvl2s-jnT2fRyqWm_RM,71286
 cache_dit/cache_factory/dual_block_cache/diffusers_adapters/__init__.py,sha256=krNAICf-aS3JLmSG8vOB9tpLa04uYRcABsC8PMbVUKY,1870
 cache_dit/cache_factory/dual_block_cache/diffusers_adapters/cogvideox.py,sha256=fibkeU-FHa30BNT-uPV2Eqcd5IRli07EKb25tMDp23c,2270
 cache_dit/cache_factory/dual_block_cache/diffusers_adapters/flux.py,sha256=fddSpTHXU24COMGAY-Z21EmHHAEArZBv_-XLRFD6ADU,2625
@@ -22,7 +22,7 @@ cache_dit/cache_factory/dynamic_block_prune/diffusers_adapters/hunyuan_video.py,
 cache_dit/cache_factory/dynamic_block_prune/diffusers_adapters/mochi.py,sha256=zXgoRDDjus3a2WSjtNh4ERtQp20ceb6nzohHMDlo2zY,2265
 cache_dit/cache_factory/dynamic_block_prune/diffusers_adapters/wan.py,sha256=PA7nuLgfAelnaI8usQx0Kxi8XATzMapyR1WndEdFoZA,2604
 cache_dit/cache_factory/first_block_cache/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-cache_dit/cache_factory/first_block_cache/cache_context.py,sha256=NeAfDJlJVVUAL4btax5_iOLTuue1x4qeXwk0pM-QH28,23219
+cache_dit/cache_factory/first_block_cache/cache_context.py,sha256=tTPwhPLEA7LqGupps1Zy2MycCtLzs22wsW0yUhiiF-U,23217
 cache_dit/cache_factory/first_block_cache/diffusers_adapters/__init__.py,sha256=-FFgA2MoudEo7uDacg4aWgm1KwfLZFsEDTVxatgbq9M,2146
 cache_dit/cache_factory/first_block_cache/diffusers_adapters/cogvideox.py,sha256=qO5CWyurtwW30mvOe6cxeQPTSXLDlPJcezm72zEjDq8,2375
 cache_dit/cache_factory/first_block_cache/diffusers_adapters/flux.py,sha256=Dcd4OzABCtyQCZNX2KNnUTdVoO1E1ApM7P8gcVYzcK0,2733
@@ -33,8 +33,13 @@ cache_dit/compile/__init__.py,sha256=DfMdPleFFGADXLsr7zXui8BTz_y9futY6rNmNdh9y7k
 cache_dit/compile/utils.py,sha256=KU60xc474Anbj7Y_FLRFmNxEjVYLLXkhbtCLXO7o_Tc,3699
 cache_dit/custom_ops/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 cache_dit/custom_ops/triton_taylorseer.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-cache_dit-0.2.4.dist-info/licenses/LICENSE,sha256=Dqb07Ik2dV41s9nIdMUbiRWEfDqo7-dQeRiY7kPO8PE,3769
-cache_dit-0.2.4.dist-info/METADATA,sha256=1oDgkkUwGVfwX_jCyU0jHbQTVQDfL59OEbrUb_9SVF4,25442
-cache_dit-0.2.4.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
-cache_dit-0.2.4.dist-info/top_level.txt,sha256=ZJDydonLEhujzz0FOkVbO-BqfzO9d_VqRHmZU-3MOZo,10
-cache_dit-0.2.4.dist-info/RECORD,,
+cache_dit/metrics/__init__.py,sha256=5yavk_6b50EhaHij0C-7nQegOA0uD79-qX96dI1i_8s,461
+cache_dit/metrics/fid.py,sha256=RIC-wW0RFv6W9IW6nc6Ih4dAxunUTnzcugMbYZjdqRM,12891
+cache_dit/metrics/inception.py,sha256=pBVe2X6ylLPIXTG4-GWDM9DWnCviMJbJ45R3ulhktR0,12759
+cache_dit/metrics/metrics.py,sha256=FUrpc58ofg0LKyM3Y_7kfUVbevMfCFFcaeaxeOrj9iY,10498
+cache_dit-0.2.6.dist-info/licenses/LICENSE,sha256=Dqb07Ik2dV41s9nIdMUbiRWEfDqo7-dQeRiY7kPO8PE,3769
+cache_dit-0.2.6.dist-info/METADATA,sha256=aTTyx_2gM7Z-mcIYd3_LlGljiZOwIq9iadDXOFPDzwA,27848
+cache_dit-0.2.6.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
+cache_dit-0.2.6.dist-info/entry_points.txt,sha256=FX2gysXaZx6NeK1iCLMcIdP8Q4_qikkIHtEmi3oWn8o,65
+cache_dit-0.2.6.dist-info/top_level.txt,sha256=ZJDydonLEhujzz0FOkVbO-BqfzO9d_VqRHmZU-3MOZo,10
+cache_dit-0.2.6.dist-info/RECORD,,

cache_dit-0.2.6.dist-info/entry_points.txt

@@ -0,0 +1,2 @@
+[console_scripts]
+cache-dit-metrics-cli = cache_dit.metrics:main