Update environment to work (currently pointing to QKeras v3 hls4ml branch) and add a TODO

Author: bo3z

TODO.md

@@ -0,0 +1,21 @@
+1. Understand why hls4ml throws a warning when parsing Add nodes in QONNX (i.e. using bias in Brevitas model)
+
+2. Understand why QONNX needs softmax = legacy (if it's a bug in hls4ml, we should fix it; if not, we should add a note to the notebook)
+
+3. Once QKeras v3 support is merged to hls4ml main from Marius' branch, update environment.yml
+
+4. Add HGQ2 notebook (to replace current HGQ tutorial) in part 2
+
+5. Once PQuant support is merged to hls4ml main, update environment.yml and add notebook to part 2
+
+6. Add more models to `Advanced Models`: From Imperial HGQ CNN, GNN and MLP
+
+7. Add accelerator backend notebooks once the accelerator backends are merged
+
+8. Add detailed README.md with some progression figure indicating what tutorials to follow & update ToC
+
+9. Think about how to source Vivado/Vitis HLS - adding XILINX_HLS to the PATH may not work for all systems (e.g., some clusters use module)
+
+10. Add a notebook (e.g., 3c) on the different implementations of GEMV/Dense layers: Latency, Resource, da4ml
+
+11. Add a part 0 on basics of neural networks, FPGAs, etc., 

environment.yml

@@ -4,22 +4,20 @@ channels:
 dependencies:
   - python=3.10.16
   - notebook==6.4.12
-  - jupyter_contrib_nbextensions
-  - jupyterhub
-  - jupyter-book
-  - jsonschema-with-format-nongpl
+  - jupyter-book==2.1.5
   - pydot==1.4.2
   - graphviz==7.1.0
-  - scikit-learn==1.2.2
-  - tensorflow==2.14.0
-  - tensorflow-datasets==4.8.3
-  - webcolors
-  - widgetsnbextension==3.6.0
-  - pip==23.0.1
+  - onnx==1.17.0
+  - onnxscript==0.7.0
+  - onnxoptimizer==0.4.2
+  - pytorch==2.10.0
+  - tensorflow==2.19.1
+  - scikit-learn==1.4.1
   - pip:
-      - hls4ml[profiling,optimization,sr,HGQ,qkeras]==1.2.0
-      - conifer==1.5
-      - pysr==0.16.3
+      - hls4ml[keras-v3,qkeras-v3,onnx,sr,profiling] @ git+https://github.com/makoeppel/hls4ml.git@feature/add_qkeras_v3
+      - git+https://github.com/fastmachinelearning/qkerasV3.git
+      - brevitas==0.12.1
+      - qonnx==1.0.0
+      - conifer==1.8
+      - pysr==1.5.4
       - xgboost==1.7.5
-      - zstd
-      - tensorflow-model-optimization